# FlowRunner — full page index > FlowRunner is a visual AI agent orchestration platform (Orchestration as a Service): workflows run fully autonomous, or pause to ask a human for judgment at designated decision points, via email, Slack, WhatsApp, or phone. AI agents are native, first-class nodes in a flow, and human-in-the-loop is a callable tool an agent invokes mid-execution. Built by Midnight Coders, Inc. (DBA FlowRunner), Texas, USA. This file covers every page on https://flowrunner.ai. Product, concept, comparison, solution, and blog pages are inlined below in full; section hubs, workflow guides and integration pages are listed as links to their Markdown twins. Every page on the site has a Markdown twin at the same path plus ".md" (homepage: https://flowrunner.ai/index.md). A shorter orientation file is at https://flowrunner.ai/llms.txt. --- ## FlowRunner | Orchestration as a Service Source: https://flowrunner.ai/ FlowRunner is the orchestration layer for production AI agents. Every model provider and tool is a verified connector, you build it no-code or with code, and when an agent hits a decision that carries weight it stops and asks a human. Audit-ready. Human-in-the-loop. [Start Building Free](https://app.flowrunner.ai) [Book a demo](https://calendly.com/flowrunner/intro) Free plan, no card required. Every new account starts with 14 days of Professional. ![A FlowRunner run paused on a fraud alert that needs a human decision](https://flowrunner.ai/assets/hero-image.webp) ### One platform, every team. #### [Developers](https://flowrunner.ai/teams/developers) Every model provider, vector store, and tool your agent needs, with a human gate on publish, production writes, and spend. See the stack #### [Security & IT](https://flowrunner.ai/teams/security) Identity, triage, and response, with a human on every action that changes access. See the stack #### [Finance & Operations](https://flowrunner.ai/teams/finance-ops) Invoicing, reconciliation, and close, with a human on every payment and posting. See the stack #### [RevOps](https://flowrunner.ai/teams/revops) Lead enrichment, CRM hygiene, and handoffs, with a human on every mass send and merge. See the stack #### [Support & CX](https://flowrunner.ai/teams/support) Triage, drafting, and follow-up, with a human on every customer-facing reply. See the stack #### [Data & Platform](https://flowrunner.ai/teams/data) Syncs, loads, and migrations, with a human on every production write. See the stack ### What your agents handle while you sleep. Live workflow [Explore Developers](https://flowrunner.ai/teams/developers) ![Every document embedded and indexed by FlowRunner, with a pause for you before the index rebuilds.](https://flowrunner.ai/images/flows/developers.svg) ![Every sales call logged to your CRM by FlowRunner, with a nudge for you on the one hot prospect.](https://flowrunner.ai/images/flows/leads.svg) ![Every phishing report triaged by FlowRunner, with a pause for you before any mailbox purge.](https://flowrunner.ai/images/flows/security.svg) ![Every support ticket answered by FlowRunner, with a hold for you before the reply reaches the customer.](https://flowrunner.ai/images/flows/support.svg) ![Every bill three-way matched by FlowRunner, with a stop for you before any money moves.](https://flowrunner.ai/images/flows/finance.svg) ![Every data batch validated by FlowRunner, with a stop for you on the rows that fail.](https://flowrunner.ai/images/flows/data.svg) ### Your team only touches what matters. 1. ### Build the flow. Open the visual builder or drop to code. Compose your systems as verified connectors, define the logic, and mark where a human should weigh in. 2. ### Deploy your agent. Connect your model providers and tools. Your agent goes live in minutes, not months, with no glue code to maintain. 3. ### Handle only exceptions. The agent runs the process end to end. When it hits the andon cord, a decision that carries weight, it asks via Slack, email, or WhatsApp. You decide. It continues. 1,800+ Verified integrations agent-ready 30+ AI providers BYOK, no markup 1 day Time to first agent not one quarter ### Automation that knows when to stop and ask. #### Human-in-the-loop. Not an afterthought. Most platforms bolt on approval gates. FlowRunner agents invoke human oversight as callable tools. They pause execution, contact your team through their preferred channel, provide full context, and resume seamlessly after you decide. "Sounds like a digital andon cord." CEO · Automotive services #### No-code where you want it, code where you need it Your team builds and modifies agents in a visual editor, and drops to code for the edge cases. No engineering backlog. New connectors in 30 minutes. Production-ready agents in a day, not a quarter. #### Compliance built in, not bolted on Audit trails, SLA tracking, and role-based access starting at $299/mo. Competitors charge $800+ or require Enterprise pricing for the same capabilities. #### Every provider, your keys Bring your own keys across 30+ AI providers with no markup on inference. 1,800+ verified connectors your agents call as tools. Cloud or self-hosted, full feature parity either way. ### Why the andon cord lands. > "Where AI actually stops itself and says, okay, I don't know, I'm going to pull up here and get guidance." CEO, Automotive services, on first seeing human-in-the-loop --- ## About FlowRunner Source: https://flowrunner.ai/about FlowRunner is a visual orchestration platform for AI agents. We help teams automate the routine and bring people in for the judgment calls. ### The company FlowRunner is the product of **Midnight Coders, Inc.**, a Texas-based software company. The platform is bootstrapped, built by a small team that has spent the last two decades building backend infrastructure used by millions of developers. **[Midnight Flow](https://midnightflow.ai) is FlowRunner's partner consultancy**, founded by the same team and operated by Midnight Coders, Inc. Midnight Flow handles complex workflow implementation services for organizations that need expert help getting started. When you see a Midnight Flow case study cited in FlowRunner content, it's a first-party engagement we're transparently disclosing, not a third-party validation. ### The founder **Mark Piller** is FlowRunner's founder and CEO. He previously founded Backendless, a mobile/web backend platform used by tens of thousands of developers. His work on FlowRunner is informed by years of building infrastructure that has to work reliably at scale, with humans in the loop for the decisions that actually matter. ### The product belief AI agents that act without judgment are a liability. AI agents that escalate everything are useless. The interesting middle ground is agents that know the difference. FlowRunner is built around that principle: workflows run autonomously when the path is clear, and route to a human when there's ambiguity, risk, or a decision worth a person's attention. We call it **Orchestration as a Service** — a layer above agent builders that coordinates, governs, and manages multi-agent environments with explicit human oversight points. ### Editorial process The default byline on FlowRunner content is Mark Piller, Founder of FlowRunner. As the team grows, additional named authors will publish here under their own bylines. We do not invent author identities or attribute content to people who did not produce or substantively review it. A growing share of our content is produced with AI assistance, then reviewed by a human editor before publishing. AI assistance does not change the byline — the named author remains accountable for what gets published. We disclose AI assistance openly because it's how the work actually gets done. The full policy is published at [/editorial-policy](https://flowrunner.ai/editorial-policy). ### Where to reach us For sales, demos, or general inquiries: [contact us](https://flowrunner.ai/contact) through the form, or email [sales@flowrunner.ai](mailto:sales@flowrunner.ai). The founder is reachable directly at [mark@flowrunner.ai](mailto:mark@flowrunner.ai). --- ## Contact Us Source: https://flowrunner.ai/contact Have questions about FlowRunner? Whether you're exploring its features, need assistance, or want to share feedback, our team is here to help. #### [Sales and inquiries](mailto:sales@flowrunner.ai) sales@flowrunner.ai #### Book a demo A 30-minute walkthrough with the team. [Book a demo](https://calendly.com/flowrunner/intro) #### [Help and support](https://docs.flowrunner.ai) Documentation and guides at docs.flowrunner.ai ### Send us a message ### Frequently asked questions #### What is the best way to get support? For technical support, visit our documentation at [docs.flowrunner.ai](https://docs.flowrunner.ai). You can browse guides or reach out to our team directly through the contact form above. #### How quickly can I expect a response? We respond to all inquiries within 1-2 business days. For urgent matters, indicate the priority in your message. #### Who should I contact for a demo? Book a demo directly through our Calendly link above, or email sales@flowrunner.ai to schedule a walkthrough. #### Can I request new features? We value user feedback. Share feature requests through the contact form or email us directly. We review every suggestion. #### Where can I learn more about using FlowRunner? Visit [docs.flowrunner.ai](https://docs.flowrunner.ai) for detailed guides, sample workflows, and video tutorials to get started. #### What should I do if I encounter a bug? Report bugs through the contact form or email us. Include as much detail as possible, including steps to reproduce the issue. --- ## Data Processing Agreement Source: https://flowrunner.ai/data-processing Last Updated: February 28, 2026 ### Applicability This Data Processing Agreement ("DPA") applies between Midnight Coders, Inc. ("Company," "Processor," "we," or "us"), a Texas corporation with its principal office at 539 W. Commerce St, Suite 2023, Dallas, TX 75208, and any Customer who has agreed to the FlowRunner [Terms of Service](https://flowrunner.ai/terms) (the "Agreement"), to the extent that the Company Processes Customer Personal Data subject to applicable Data Protection Laws. This DPA is incorporated into and forms part of the Agreement. By agreeing to the Agreement, the Customer agrees to this DPA. In the event of a conflict between this DPA and the Agreement, this DPA shall prevail with respect to the Processing and protection of Personal Data. This DPA applies when the Processing of Customer Personal Data is subject to the European General Data Protection Regulation (EU) 2016/679 ("GDPR"), the United Kingdom General Data Protection Regulation ("UK GDPR"), the Swiss Federal Act on Data Protection ("FADP"), the California Consumer Privacy Act as amended by the California Privacy Rights Act ("CCPA/CPRA"), the Texas Data Privacy and Security Act ("TDPSA"), or any other applicable data protection legislation (collectively, "Data Protection Laws"). Where the Customer has entered into a separate Master Service Agreement or Enterprise Agreement with the Company, this DPA supplements that agreement. Enterprise Customers who require a countersigned copy of this DPA for their compliance records may contact [legal@flowrunner.ai](mailto:legal@flowrunner.ai) to request an executable version. Capitalized terms not defined in this DPA have the meanings given to them in the Agreement. ### 1\. Definitions **"Customer Personal Data"** means any Personal Data that is uploaded to, transmitted through, stored in, or processed using the Service by or on behalf of the Customer, including data processed through Workflows, AI Agents, Human-in-Loop interactions, and MCP server connections. Customer Personal Data does not include Account Data (as defined in our Privacy Policy) that the Company collects as a controller. **"Data Protection Laws"** means all applicable laws and regulations relating to the processing of Personal Data, including GDPR, UK GDPR, FADP, CCPA/CPRA, TDPSA, and any other applicable data protection or privacy legislation, as amended from time to time. **"Data Subject"** means an identified or identifiable natural person to whom Customer Personal Data relates. **"Personal Data"** means any information relating to an identified or identifiable natural person, as defined in the applicable Data Protection Laws. **"Personal Data Breach"** means a breach of security leading to the accidental or unlawful destruction, loss, alteration, unauthorized disclosure of, or access to Customer Personal Data. **"Processing"** (and "Process") means any operation or set of operations performed on Personal Data, including collection, recording, organization, structuring, storage, adaptation, alteration, retrieval, consultation, use, disclosure by transmission, dissemination, alignment, combination, restriction, erasure, or destruction. **"SCCs"** means the Standard Contractual Clauses for the transfer of personal data to processors established in third countries, as approved by the European Commission in Implementing Decision (EU) 2021/914 of 4 June 2021, and any successor or replacement clauses. **"Sub-Processor"** means any third party engaged by the Company to Process Customer Personal Data on behalf of the Customer in connection with the Service. **"Supervisory Authority"** means an independent public authority established by an EU/EEA Member State, the UK Information Commissioner's Office, or any other competent data protection authority under applicable Data Protection Laws. **"Technical and Organizational Measures"** or **"TOMs"** means the security measures described in Annex II of this DPA. ### 2\. Roles and Scope of Processing #### 2.1 Roles For the purposes of this DPA: (a) The **Customer is the Controller** (or, where the Customer itself acts as a processor on behalf of a third-party controller, the Customer is a Processor) of Customer Personal Data. (b) The **Company is the Processor** (or, where the Customer acts as a Processor, the Company is a Sub-Processor) of Customer Personal Data. Where the Customer acts as a Processor, the Customer represents and warrants that: (i) it has obtained all necessary authorizations from the relevant Controller to engage the Company as a Sub-Processor; (ii) its instructions to the Company comply with the instructions of the relevant Controller; and (iii) it has entered into a data processing agreement with the relevant Controller that is compliant with applicable Data Protection Laws. #### 2.2 Scope of Processing The subject matter, duration, nature and purpose of processing, types of Customer Personal Data, and categories of Data Subjects are described in **Annex I** of this DPA. The Company shall Process Customer Personal Data only: (a) to provide, operate, and maintain the Service as described in the Agreement; (b) in accordance with the Customer's documented instructions as set forth in the Agreement, this DPA, and any subsequent written instructions agreed to by the parties; and (c) as required by applicable law, in which case the Company shall (to the extent permitted by law) inform the Customer of the legal requirement before Processing. #### 2.3 Customer Obligations The Customer is responsible for: (a) Ensuring that it has a lawful basis for Processing Customer Personal Data and for instructing the Company to Process Customer Personal Data on its behalf; (b) Ensuring that all necessary consents, notices, and authorizations have been obtained from Data Subjects or other relevant parties as required by applicable Data Protection Laws; (c) Determining the lawfulness and appropriateness of any Customer Personal Data transmitted through the Service, including through Workflows, AI Agents, Human-in-Loop interactions, and MCP server connections; (d) Configuring the Service appropriately for the sensitivity and regulatory classification of the Customer Personal Data being processed, including selecting a Subscription Plan with adequate compliance features. #### 2.4 Company Obligations The Company shall: (a) Process Customer Personal Data only on documented instructions from the Customer, unless required to do so by applicable law; (b) Immediately inform the Customer if, in the Company's reasonable opinion, an instruction from the Customer infringes applicable Data Protection Laws; (c) Ensure that persons authorized to Process Customer Personal Data have committed themselves to confidentiality or are under an appropriate statutory obligation of confidentiality; (d) Implement and maintain the Technical and Organizational Measures described in Annex II; (e) Comply with the Sub-Processor requirements set forth in Section 4; (f) Assist the Customer in responding to requests from Data Subjects exercising their rights under applicable Data Protection Laws, as described in Section 6; (g) Assist the Customer in ensuring compliance with its obligations under Articles 32 through 36 of the GDPR (and equivalent provisions under other applicable Data Protection Laws), including obligations relating to security of processing, breach notification, data protection impact assessments, and prior consultation with Supervisory Authorities, taking into account the nature of the Processing and the information available to the Company; (h) At the Customer's choice, delete or return all Customer Personal Data to the Customer upon termination of the Agreement, as described in Section 8; (i) Make available to the Customer all information necessary to demonstrate compliance with the obligations set forth in this DPA and applicable Data Protection Laws, and allow for and contribute to audits as described in Section 7. ### 3\. Security Measures The Company shall implement and maintain appropriate Technical and Organizational Measures to ensure a level of security appropriate to the risk, as described in **Annex II** of this DPA. These measures are designed to protect Customer Personal Data against accidental or unlawful destruction, loss, alteration, unauthorized disclosure, or access. The Company shall regularly test, assess, and evaluate the effectiveness of its Technical and Organizational Measures. The Customer acknowledges that security measures are subject to technical progress and development, and the Company may update its measures from time to time, provided that any update does not materially decrease the overall level of protection. The Customer is responsible for reviewing the Technical and Organizational Measures described in Annex II to determine whether they meet the Customer's security requirements. The Customer is also responsible for implementing its own security measures for systems and credentials under its control, including BYOK API keys, MCP server configurations, and Self-Hosted Deployment infrastructure. ### 4\. Sub-Processors #### 4.1 General Authorization The Customer provides a general written authorization to the Company to engage Sub-Processors to Process Customer Personal Data. The current list of Sub-Processors is set forth in **Annex III** of this DPA and is also maintained at flowrunner.ai/sub-processors. #### 4.2 Notification of Changes The Company shall notify the Customer at least thirty (30) days before engaging any new Sub-Processor or replacing an existing Sub-Processor that Processes Customer Personal Data. Notification shall be provided by email to the address associated with the Customer's Account or by posting an update at flowrunner.ai/sub-processors. #### 4.3 Right to Object If the Customer objects to a new or replacement Sub-Processor on reasonable data protection grounds, the Customer shall notify the Company in writing within thirty (30) days of receiving the Company's notification. Upon receipt of such objection: (a) The Company shall use commercially reasonable efforts to make available to the Customer a change in the Service or recommend a commercially reasonable alternative to avoid Processing of Customer Personal Data by the objected-to Sub-Processor; (b) If the Company is unable to provide such an alternative within thirty (30) days of receiving the Customer's objection, the Customer may terminate the affected portion of the Agreement (or the Agreement in its entirety) by providing written notice to the Company. The Company will refund any prepaid fees covering the remainder of the term following the effective date of termination. #### 4.4 Sub-Processor Obligations The Company shall: (a) Enter into a written agreement with each Sub-Processor that imposes data protection obligations no less protective than those set forth in this DPA; (b) Ensure that each Sub-Processor provides sufficient guarantees to implement appropriate Technical and Organizational Measures; (c) Remain fully liable to the Customer for the performance of each Sub-Processor's obligations. #### 4.5 BYOK and Customer-Directed Third Parties (a) Where the Customer elects to use third-party AI service providers, communication channels, MCP servers, or other third-party services through bring-your-own-key ("BYOK") functionality or other Customer-configured integrations, such third parties are engaged directly by the Customer and not by the Company. (b) In such cases, the Company acts solely as a conduit for Customer-directed transmissions and does not independently determine the purposes or means of Processing performed by such third parties. The Customer acknowledges that it enters into a direct relationship with such third parties and is solely responsible for executing any required data processing agreements with them. (c) To the extent Customer Personal Data is transmitted through Company infrastructure solely for routing purposes to a Customer-designated third party, the Company's Processing is limited to facilitating such transmission in accordance with the Customer's instructions and does not render the third party a Sub-Processor of the Company. (d) The Customer is responsible for assessing whether such third parties provide adequate data protection safeguards under applicable Data Protection Laws. ### 5\. Personal Data Breach Notification #### 5.1 Notification The Company shall notify the Customer without undue delay and, where feasible, within seventy-two (72) hours after becoming aware of a confirmed Personal Data Breach affecting Customer Personal Data. Notification shall be provided to the Customer's Account administrator email address and, where available, through any dedicated security contact designated by the Customer. #### 5.2 Notification Content The Company's notification shall include, to the extent reasonably available: (a) A description of the nature of the Personal Data Breach, including the categories and approximate number of Data Subjects and records concerned; (b) The name and contact details of the Company's point of contact for further information; (c) A description of the likely consequences of the Personal Data Breach; (d) A description of the measures taken or proposed to be taken to address the breach, including measures to mitigate its possible adverse effects. If it is not possible to provide all information simultaneously, the Company may provide it in phases without undue further delay. #### 5.3 Company Cooperation The Company shall cooperate with the Customer and take commercially reasonable steps to assist in the investigation, mitigation, and remediation of any Personal Data Breach. The Company shall also assist the Customer in fulfilling its own breach notification obligations under applicable Data Protection Laws. #### 5.4 Customer Responsibilities The Customer is solely responsible for determining whether a Personal Data Breach triggers notification obligations under applicable Data Protection Laws and for fulfilling those obligations, including notifications to Supervisory Authorities and affected Data Subjects. #### 5.5 No Admission; Privileged Communications Any notifications, communications, or information provided pursuant to this Section 5 are made without prejudice and shall not constitute an admission or acknowledgment of fault or liability. Nothing in this Section shall require the disclosure of information protected by attorney-client privilege, work product doctrine, or other applicable legal protections. ### 6\. Data Subject Rights The Company shall, taking into account the nature of the Processing, provide reasonable assistance to the Customer in fulfilling its obligations to respond to requests from Data Subjects exercising their rights under applicable Data Protection Laws, including rights of access, rectification, erasure, restriction, portability, and objection. If the Company receives a request directly from a Data Subject regarding Customer Personal Data, the Company shall promptly redirect the Data Subject to the Customer and notify the Customer of the request, unless prohibited by law. The Company shall not respond to the Data Subject directly unless instructed to do so by the Customer or required by applicable law. The Service provides built-in tools that enable the Customer to access, export, correct, and delete Customer Personal Data directly. The Customer should use these self-service tools as the primary means of responding to Data Subject requests. Where the self-service tools are insufficient, the Company shall provide additional assistance upon the Customer's written request, subject to a reasonable fee for extraordinary requests that require significant manual effort. ### 7\. Audit Rights The Company shall make available to the Customer all information reasonably necessary to demonstrate compliance with the obligations laid down in this DPA and applicable Data Protection Laws. #### 7.1 Documentation and Reports The Company shall, upon the Customer's written request and subject to appropriate confidentiality obligations: (a) Provide summaries or copies of relevant audit reports, certifications, or independent third-party security assessments (such as SOC 2 Type II reports, if and when available) that relate to the Company's Processing of Customer Personal Data; (b) Provide written responses to reasonable information requests regarding the Company's data protection practices and Technical and Organizational Measures. #### 7.2 On-Site Audits If the Customer reasonably determines that the documentation provided under Section 7.1 is insufficient to verify the Company's compliance, the Customer may conduct or commission an independent third-party audit of the Company's Processing activities, subject to the following conditions: (a) The Customer shall provide at least thirty (30) days' prior written notice of any audit; (b) Audits shall be conducted during normal business hours, no more than once per twelve-month period, unless a Personal Data Breach or regulatory requirement necessitates an additional audit; (c) The auditor must execute a confidentiality agreement reasonably acceptable to the Company before commencing the audit; (d) The audit scope shall be limited to the Company's Processing of Customer Personal Data and shall not extend to data or systems of other customers; (e) The Customer shall bear the costs of the audit and any third-party auditor engaged by the Customer. The Company shall bear its own internal costs of compliance unless the audit reveals a material breach of this DPA, in which case the Company shall reimburse the Customer for reasonable audit costs directly attributable to such material breach; (f) The Customer shall provide the Company with a copy of audit findings before disclosing them to any third party (other than the Customer's advisors or a Supervisory Authority as required by law). #### 7.3 Supervisory Authority Audits The Company shall cooperate with any audit or inspection by a Supervisory Authority to the extent relating to the Company's Processing of Customer Personal Data under this DPA, provided the Supervisory Authority has the legal authority to conduct such audit or inspection. ### 8\. Data Return and Deletion Upon termination or expiration of the Agreement, the Customer may: (a) **Export** Customer Personal Data using the Service's built-in export tools during the thirty (30) day post-termination retention period described in the Agreement; (b) **Request return** of Customer Personal Data in a structured, commonly used, machine-readable format by contacting the Company at [legal@flowrunner.ai](mailto:legal@flowrunner.ai) during the thirty-day retention period; (c) **Request deletion** of Customer Personal Data by contacting the Company at [legal@flowrunner.ai](mailto:legal@flowrunner.ai). If the Customer does not export, request return, or request deletion during the thirty (30) day retention period, the Company shall delete Customer Personal Data from its active systems. Residual copies may persist in encrypted backups for up to ninety (90) days following deletion from active systems, after which they will be permanently purged. Residual backup copies are not actively processed and are retained solely for disaster recovery and business continuity purposes. Such copies are subject to strict access controls and will be automatically overwritten or securely deleted in accordance with the Company's backup retention schedule. The Company may retain Customer Personal Data to the extent required by applicable law, and only for the period and purposes required by such law. The Company shall inform the Customer of any such retention requirement (to the extent permitted by law) and shall ensure that the retained data is processed only for the purposes required by law and protected by appropriate Technical and Organizational Measures. Execution logs and audit trail data are retained for the periods specified for the Customer's Subscription Plan (7 days for Free and Starter, 30 days for Growth, 90 days for Professional and Business, and as configured for Enterprise). The period for which such data is visible in the Service may be shorter than the retention period, as described on the pricing page; data outside the visible window remains stored until the retention period ends. These retention periods apply regardless of the thirty-day post-termination window for Customer Content. To the extent audit logs contain Customer Personal Data, such logs are retained for security, fraud prevention, and regulatory compliance purposes. Where feasible, the Company shall restrict, redact, or anonymize personal data within logs in response to valid erasure requests, unless retention is required by applicable law or necessary to establish, exercise, or defend legal claims. ### 9\. International Data Transfers The Service is hosted in the United States. Where the Customer is located in the EEA, United Kingdom, or Switzerland, the transfer of Customer Personal Data to the Company in the United States shall be governed by the following mechanisms: #### 9.1 Standard Contractual Clauses The parties agree that the Standard Contractual Clauses (SCCs) approved by the European Commission in Implementing Decision (EU) 2021/914 are incorporated into this DPA by reference and shall apply to transfers of Customer Personal Data from the EEA to the United States. Specifically: (a) **Module Two (Controller to Processor)** applies where the Customer is a Controller and the Company is a Processor; (b) **Module Three (Processor to Sub-Processor)** applies where the Customer is a Processor and the Company is a Sub-Processor; (c) The details required under Annex I and Annex II of the SCCs are set forth in Annex I and Annex II of this DPA, respectively; (d) The optional Clause 7 (Docking Clause) is included to allow additional parties to accede to the SCCs; (e) For Clause 9 (Use of Sub-Processors), Option 2 (General Written Authorization) is selected, with a notification period of thirty (30) days as described in Section 4.2 of this DPA; (f) For Clause 11 (Redress), the optional language regarding independent dispute resolution is not included; (g) For Clause 17 (Governing Law), the SCCs shall be governed by the laws of Ireland or, where required by applicable law, the Member State in which the Customer is established; (h) For Clause 18 (Choice of Forum and Jurisdiction), disputes shall be resolved before the courts of Ireland. #### 9.2 UK International Data Transfer Addendum For transfers of Customer Personal Data from the United Kingdom, the International Data Transfer Addendum to the EU Standard Contractual Clauses (as issued by the UK Information Commissioner's Office under section 119A of the Data Protection Act 2018) is incorporated into this DPA by reference and shall apply to such transfers. #### 9.3 Swiss Data Transfers For transfers of Customer Personal Data from Switzerland, the SCCs shall apply with the modifications required by the Swiss Federal Data Protection and Information Commissioner, including that references to GDPR shall be interpreted as references to the Swiss Federal Act on Data Protection (FADP). #### 9.4 Alternative Transfer Mechanisms To the extent an alternative lawful transfer mechanism becomes available (such as an adequacy decision covering transfers to the United States, or the Company's self-certification under the EU-US Data Privacy Framework), the parties may rely on such mechanism as an alternative to or in addition to the SCCs. The Company may elect to self-certify under the EU-US Data Privacy Framework or any successor framework. If and when self-certification becomes effective, the Company shall notify the Customer and may rely on such mechanism as a primary lawful transfer mechanism. ### 10\. Data Protection Impact Assessments The Company shall provide reasonable assistance to the Customer, upon written request, in conducting data protection impact assessments ("DPIAs") and prior consultations with Supervisory Authorities to the extent required under applicable Data Protection Laws, taking into account the nature of the Processing and the information available to the Company. The Company's assistance under this section shall be limited to providing information about the Company's Processing activities, Technical and Organizational Measures, and Sub-Processors. The Company may charge reasonable fees for assistance beyond what is necessary to fulfill its obligations under Article 28(3)(f) of the GDPR. ### 11\. CCPA/CPRA Service Provider Addendum To the extent that the Company Processes Customer Personal Data that constitutes "Personal Information" (as defined in the CCPA/CPRA) on behalf of a Customer who is a "Business" (as defined in the CCPA/CPRA), the Company acts as a "Service Provider" (as defined in the CCPA/CPRA) and the following additional provisions apply: (a) The Company shall not sell or share (as those terms are defined in the CCPA/CPRA) Customer Personal Data; (b) The Company shall not retain, use, or disclose Customer Personal Data for any purpose other than providing the Service as specified in the Agreement, or as otherwise permitted under the CCPA/CPRA; (c) The Company shall not retain, use, or disclose Customer Personal Data outside of the direct business relationship between the Company and the Customer; (d) The Company shall not combine Customer Personal Data with Personal Information received from or on behalf of other persons, or collected from its own interactions with Data Subjects, except as permitted under the CCPA/CPRA; (e) The Company shall assist the Customer in responding to verifiable consumer requests as described in Section 6 of this DPA; (f) The Company shall notify the Customer if it determines that it can no longer meet its obligations under the CCPA/CPRA; (g) The Company grants the Customer the right to take reasonable and appropriate steps to ensure that the Company uses Customer Personal Data in a manner consistent with the Customer's obligations under the CCPA/CPRA; (h) The Company certifies that it understands and will comply with the restrictions set forth in this Section 11; (i) The Company shall permit the Customer to take reasonable and appropriate steps to monitor the Company's compliance with this Section 11, consistent with Section 7 (Audit Rights) of this DPA. ### 12\. Relationship to Business Associate Agreement (HIPAA) Where the Customer has executed a Business Associate Agreement ("BAA") with the Company under the Health Insurance Portability and Accountability Act ("HIPAA"): (a) The BAA shall govern the Processing of Protected Health Information ("PHI") as defined under HIPAA; (b) This DPA shall govern the Processing of Customer Personal Data that does not constitute PHI and is subject to Data Protection Laws; (c) Where Customer Personal Data constitutes both PHI and Personal Data subject to Data Protection Laws (e.g., GDPR), both the BAA and this DPA shall apply, with the more protective substantive data protection provision prevailing in the event of a conflict; provided, however, that breach notification timelines for Protected Health Information shall be governed by the applicable Business Associate Agreement to the extent required by HIPAA, and nothing herein shall limit obligations under applicable non-HIPAA Data Protection Laws; (d) The execution of this DPA does not satisfy the Customer's obligation to execute a BAA prior to Processing PHI through the Service. A separate BAA is required as described in the Agreement and Privacy Policy. ### 13\. General Provisions #### 13.1 Term This DPA shall remain in effect for the duration of the Agreement and shall automatically terminate upon termination or expiration of the Agreement, subject to the Company's obligations regarding data return and deletion under Section 8, which shall survive termination. #### 13.2 Order of Precedence In the event of a conflict between this DPA and the Agreement, this DPA shall prevail with respect to the Processing and protection of Customer Personal Data. In the event of a conflict between this DPA and the SCCs, the SCCs shall prevail. #### 13.3 Liability Each party's liability under this DPA is subject to the limitations and exclusions of liability set forth in the Agreement, except that the limitations of liability shall not apply to the extent prohibited by applicable Data Protection Laws. Nothing in this DPA shall exclude or limit liability where such exclusion or limitation is prohibited by applicable Data Protection Laws, including Article 82 of the GDPR. #### 13.4 Governing Law Except as otherwise specified in the SCCs (which are governed by the law of Ireland or, where applicable, the Member State in which the Customer is established), this DPA shall be governed by and construed in accordance with the governing law provisions of the Agreement (State of Texas). #### 13.5 Severability If any provision of this DPA is found to be invalid or unenforceable, the remaining provisions shall remain in full force and effect. The parties shall negotiate in good faith to replace the invalid provision with a valid provision that achieves the same or substantially similar purpose. #### 13.6 Amendments This DPA may be amended only in accordance with the modification procedures set forth in the Agreement. The Company may update Annex II (Technical and Organizational Measures) and Annex III (Sub-Processors) in accordance with Sections 3 and 4 of this DPA, respectively. #### 13.7 Contact Questions regarding this DPA should be directed to: Midnight Coders, Inc. Attn: Legal / Data Protection 539 W. Commerce St, Suite 2023 Dallas, TX 75208 Email: [legal@flowrunner.ai](mailto:legal@flowrunner.ai) ### Annex I: Description of Processing #### A. List of Parties **Data Exporter (Controller):** The Customer as identified in the Agreement. **Data Importer (Processor):** Midnight Coders, Inc., 539 W. Commerce St, Suite 2023, Dallas, TX 75208, USA. Contact: [legal@flowrunner.ai](mailto:legal@flowrunner.ai). #### B. Description of Processing **Subject matter:** Processing of Customer Personal Data through the FlowRunner workflow automation and AI agent orchestration platform. **Duration:** For the term of the Agreement, plus post-termination retention periods as described in Section 8 of this DPA. **Nature and purpose:** Automated workflow execution, AI agent orchestration, Human-in-Loop communication facilitation, data transformation, integration with Customer-designated third-party services, execution logging, and audit trail generation — all as instructed by the Customer through Workflow configuration. **Types of Personal Data:** Determined by the Customer based on Workflow configuration. May include: names, email addresses, phone numbers, postal addresses, IP addresses, employee identifiers, customer/patient identifiers, financial account references, health-related data (if BAA is in place), and any other personal data the Customer transmits through Workflows. **Categories of Data Subjects:** Determined by the Customer. May include: Customer's employees, customers, patients (if BAA in place), vendors, contractors, business contacts, and any other individuals whose data is processed through Customer's Workflows. **Sensitive data (if applicable):** The Customer may process special categories of data (health data, financial data) if appropriate agreements (BAA, etc.) and Subscription Plan features (audit trails, RBAC) are in place. The Company does not determine whether sensitive data is processed — the Customer does. #### C. Competent Supervisory Authority The competent Supervisory Authority shall be determined in accordance with Clause 13 of the SCCs. Where the Customer is established in the EEA, the Supervisory Authority of the Customer's Member State of establishment shall be the competent authority. ### Annex II: Technical and Organizational Measures The Company implements and maintains the following Technical and Organizational Measures to protect Customer Personal Data: #### 1\. Encryption (a) **Encryption in transit:** All data transmitted between the Customer and the Service is encrypted using TLS 1.2 or higher. (b) **Encryption at rest:** Customer Personal Data is encrypted at rest using AES-256 or equivalent encryption. (c) **API key encryption:** BYOK API keys provided by the Customer are encrypted at rest and are not accessible to Company personnel in plaintext. #### 2\. Access Controls (a) **Authentication:** Multi-factor authentication (MFA) is required for all Company personnel accessing systems that store or process Customer Personal Data. (b) **Authorization:** Access to Customer Personal Data is restricted to authorized personnel on a need-to-know basis, using role-based access controls. (c) **Customer RBAC:** The Service provides role-based access control features (available on Professional, Business, and Enterprise Subscription Plans) enabling Customers to control which Authorized Users can create, edit, view, or execute Workflows. (d) **SSO/SAML:** The Service supports single sign-on authentication via SAML 2.0 (available on Business and Enterprise Subscription Plans) for centralized Customer access management. #### 3\. Infrastructure Security (a) **Hosting:** Cloud Deployment infrastructure is hosted by DigitalOcean in SOC 2-audited data centers. (b) **Network security:** Firewalls, intrusion detection systems, and network segmentation are used to protect against unauthorized access. (c) **Patching:** Security patches are applied to production systems on a regular schedule, with critical patches applied promptly upon release. (d) **SOC 2 certification:** The Company is pursuing SOC 2 Type II certification and will update Customers upon completion. #### 4\. Data Segregation (a) Customer data is logically segregated to prevent unauthorized cross-customer access. (b) Each Customer's Workflows, configurations, and execution data are accessible only through that Customer's authenticated Account. #### 5\. Monitoring and Logging (a) **Security monitoring:** Continuous monitoring of infrastructure and application logs for anomalous activity and potential security incidents. (b) **Audit logging:** The Service maintains audit trails of administrative actions and Workflow executions (retention periods vary by Subscription Plan). (c) **Incident response:** Documented incident response procedures are maintained and tested. #### 6\. Personnel Security (a) **Confidentiality:** All Company employees and contractors with access to Customer Personal Data are bound by confidentiality obligations. (b) **Training:** Personnel with access to Customer Personal Data receive data protection and security awareness training. (c) **Background checks:** Background checks are conducted for personnel in roles with access to Customer Personal Data, to the extent permitted by applicable law. #### 7\. Business Continuity and Disaster Recovery (a) **Backups:** Customer Personal Data is backed up regularly. Backups are encrypted and stored in geographically separate locations. (b) **Recovery:** Disaster recovery procedures are documented and tested to ensure the availability of Customer Personal Data. #### 8\. Vendor Management (a) Sub-Processors are evaluated for security practices before engagement. (b) Sub-Processor agreements include data protection obligations no less protective than those in this DPA. ### Annex III: Sub-Processors As of the effective date of this DPA, the Company engages the following Sub-Processors to Process Customer Personal Data: **DigitalOcean, Inc.** — Cloud infrastructure hosting and compute. Location: United States. Data processed: All Customer Personal Data stored in the Service (encrypted at rest). **Stripe, Inc.** — Payment processing. Location: United States. Data processed: Transaction confirmations, Stripe customer identifiers (no payment card data stored by Company). **MailerSend** — Transactional email delivery. Data processed: Email addresses, notification content. **Alphabet** — Website analytics (marketing site only). Data processed: Device/access info, anonymized usage patterns (no Customer Personal Data). The current list is also maintained at flowrunner.ai/sub-processors. The Customer acknowledges that third-party AI service providers accessed through the Customer's BYOK API keys and Customer-registered MCP servers are not Sub-Processors of the Company, as described in Section 4.5 of this DPA. --- ## Editorial Policy Source: https://flowrunner.ai/editorial-policy Last updated: 2026-05-13 ### 1\. Author attribution The default byline on FlowRunner content is **Mark Piller, Founder of FlowRunner**. The byline links to our [About page](https://flowrunner.ai/about), which describes the company, the team, and the leadership credentials behind the content we publish. Mark is named because someone real signs off on what gets published. Readers can hold a named human accountable for accuracy, framing, and follow-up. If a piece you read here gets something wrong, the right person to email is [mark@flowrunner.ai](mailto:mark@flowrunner.ai). #### When a different byline appears As FlowRunner grows, additional people will author content under their own names. When that happens, the byline reflects the actual author and links to a profile or About-page section that identifies them. We do not invent author personas, use stock photos for fictional contributors, or attribute content to people who did not write or substantively review it. Some pages on the site — integration reference pages, product catalog pages — carry no byline. Those are reference documentation, not authored articles. They are owned by the FlowRunner organization and the company is the author of record (visible in JSON-LD as the publisher). ### 2\. AI-assisted content A growing share of our content is produced with AI assistance, then reviewed by a human editor before publishing. The default editor is Mark Piller. We disclose AI assistance because it is how the work actually gets done; pretending otherwise would misrepresent the production process. AI assistance does not change the author byline. The named author is the person responsible for the published work, regardless of which tools were used to produce it — the same way an author who used a research assistant or a copy editor still takes the byline. ### 3\. What editorial review checks Before any AI-assisted content is published: - Factual claims are checked against source material (product documentation, customer evidence, cited research) - Quantitative claims trace to a verifiable source per our E-E-A-T contract; framing matches the strength of the underlying evidence - Brand voice and positioning are aligned with our published guidelines - The piece must be one the named author would defend if questioned about it ### 4\. What we do not do - We do not invent author identities or attribute content to people who did not produce or substantively review it. - We do not fabricate quantitative claims, customer logos, or case studies. - We do not present anonymous AI-generated content under a fake "team" or "staff" byline to obscure the production model. - We do not republish AI output without human review. ### 5\. Disclosures **Cross-site relationship.** FlowRunner and Midnight Flow are operated by the same team ([Midnight Coders, Inc.](https://flowrunner.ai/about)). When FlowRunner content cites a Midnight Flow case study, we disclose the relationship inline so readers know it is a first-party engagement, not third-party validation. **BYOK and third-party AI.** FlowRunner uses a Bring Your Own Keys model for AI providers in the product. The content on this site may also be produced with assistance from third-party AI tools. Those tools do not change the editorial responsibility — the named author and editor are accountable for what gets published. ### 6\. Corrections and contact If you find a factual error, a misleading framing, or a citation that does not support the claim it is attached to, email [editorial@flowrunner.ai](mailto:editorial@flowrunner.ai) (or directly to [mark@flowrunner.ai](mailto:mark@flowrunner.ai)). We correct substantive errors promptly and note the correction in the page where appropriate. ### 7\. When this policy changes We update this policy as the team grows, as new content production patterns are introduced, and as regulatory guidance on AI-assisted content evolves (FTC AI guidance, Google's Spam Policy on Spammy Automatically-Generated Content, FTC endorsement guidelines). The "last updated" date at the top of this page reflects the most recent material change. --- ## Industries Source: https://flowrunner.ai/industries Healthcare compliance violations. Financial audit failures. Missed shipments. FlowRunner agents handle the routine work and escalate the exceptions that need human judgment. #### [Healthcare & Life Sciences](https://flowrunner.ai/industries/healthcare) Lab results processing, prior authorization, clinical document intake, and revenue cycle management with HIPAA-ready compliance. 94% first-pass approval rate View details #### [Financial Services](https://flowrunner.ai/industries/financial-services) Loan document processing, KYC/AML intake, invoice validation, and data reconciliation with complete audit trails. 70% faster loan processing View details #### [Logistics & 3PL](https://flowrunner.ai/industries/logistics) 3PL receipt tracking, stock-out risk detection, order validation, and freight claim management across your supply chain. 73% fewer stock-outs View details --- ## Financial Services Source: https://flowrunner.ai/industries/financial-services Process loans, validate documents, and manage compliance workflows with full audit trails and human oversight on exceptions. ### The Compliance-Operations Tension Your operations team processes hundreds of documents daily: loan applications, KYC forms, transaction monitoring alerts. Each one needs validation, documentation, and audit trails. Manual processes create risk. Automation without oversight creates liability. The result? Documents sit in queues waiting for review. Compliance asks for reports that take days to compile. And you're always one audit away from finding gaps in your process. ### Financial Workflows We Automate #### Loan Document Processing Extract data from applications, validate supporting docs, check credit criteria, route for approval. - 70% faster loan processing - Missing documents caught upfront - Credit memo auto-generated #### KYC/AML Document Intake Collect identity docs, validate against watchlists, flag high-risk entities for review. - 94% of cases auto-cleared - Suspicious activity flagged immediately - Complete audit trail for regulators #### Invoice Validation Match invoices to POs, check pricing, validate tax calculations, route exceptions. - 94% auto-processed - Duplicate detection - Early payment discount capture [View details](https://flowrunner.ai/use-cases/invoice-validation) #### Data Reconciliation Match transactions across systems, identify discrepancies, flag for investigation. - 99.9% accuracy rate - 3 days faster month-end close - Audit-ready documentation [View details](https://flowrunner.ai/use-cases/data-reconciliation) ### Built for Financial Compliance #### Complete Audit Trail Every decision logged with user, timestamp, and reasoning. Ready for examiner review. #### Human-in-the-Loop AI handles routine cases. Exceptions route to compliance staff with full context. #### Model Risk Management Document AI decision criteria, monitor accuracy, maintain model inventory. #### Data Residency Deploy in your region. Full control over where data is processed and stored. > "We cut loan processing time from 5 days to 36 hours. The agent handles document validation and data entry, so our underwriters focus on credit decisions. Compliance loves the complete audit trail." \- Chief Operating Officer, commercial lender --- ## Healthcare & Life Sciences Source: https://flowrunner.ai/industries/healthcare Process lab results, clinical documents, and prior authorizations without hiring more staff. HIPAA-ready compliance built in. ### The Healthcare Operations Challenge Your clinical and administrative staff spend 40% of their time on manual document processing: lab results sitting in queues, prior authorization forms rejected for missing attachments, patient intake packets incomplete after three follow-ups. Meanwhile, compliance requirements get stricter and labor costs rise. The result? Patients wait longer for results. Prior authorizations delay critical medications. And your team is stuck doing data entry instead of patient care. ### Healthcare Workflows We Automate #### Lab Results Processing Extract data from lab reports, route abnormal results to physicians, file in patient records automatically. - 400-700 results processed daily per agent - Abnormal values flagged for immediate review - Integration with Epic, Cerner, Allscripts #### Prior Authorization Pre-populate payer forms, validate clinical documentation, track submissions through approval. - 94% first-pass approval rate - 5 days faster patient access - Missing attachments caught before submission [View details](https://flowrunner.ai/use-cases/prior-authorization) #### Clinical Document Intake Collect patient histories, consent forms, and insurance documentation with validation and completeness checking. - 73% faster patient onboarding - Automated reminder sequences - Digital signature integration [View details](https://flowrunner.ai/use-cases/client-intake) #### Revenue Cycle Management Validate claims before submission, track denials, manage appeals with full documentation. - 23% reduction in claim denials - Automated eligibility verification - Denial root cause analysis ### Built for Healthcare Compliance #### HIPAA Ready End-to-end encryption, access controls, audit logging. BAAs available. #### Complete Audit Trail Every action logged with user, timestamp, and data accessed. Ready for OCR review. #### Human-in-the-Loop AI handles routine cases. Exceptions route to clinical staff with full context. #### On-Premise Option Deploy entirely within your infrastructure for maximum data control. > "We process 600+ lab results daily with two agents. Abnormal values are flagged and routed to physicians within minutes. Our medical assistants finally have time to focus on patients instead of data entry." \- Chief Medical Officer, regional health system --- ## Logistics & 3PL Source: https://flowrunner.ai/industries/logistics Validate warehouse orders, track inventory across locations, and automate fulfillment without the back-and-forth. ### The Logistics Visibility Gap You have inventory in 6 warehouses managed by 4 different 3PLs. Every day, transfer orders move between them. And every day, your team sends emails asking: "Did you receive PO-2847?" "Where's the shipment from Dallas?" The result? Inventory records don't match reality. Customers get told products are in stock that aren't. And you spend 3 hours daily on status calls instead of optimizing your supply chain. ### Logistics Workflows We Automate #### 3PL Receipt Tracking Monitor transfer orders across all warehouses. Auto-reconcile received quantities against ASNs. - 100% visibility across all locations - Discrepancies flagged within hours - Zero manual status emails [View details](https://flowrunner.ai/use-cases/3pl-receipt-tracking) #### Stock-Out Risk Detection AI monitors inventory levels and velocity. Alerts when you have time to act, not when it's a crisis. - 73% fewer stock-outs - 4 days earlier warning - $2M+ sales protected annually [View details](https://flowrunner.ai/use-cases/stock-out-risk-detection) #### Order Validation Check orders against inventory, credit limits, and shipping rules before release. - 94% of orders auto-approved - Problem orders flagged instantly - Integration with WMS and OMS #### Freight Claim Management Auto-detect damaged shipments, file claims with carriers, track through resolution. - 47% faster claim filing - Documentation auto-assembled - $50K+ average annual recovery > "We eliminated the daily 3PL status call. The agent catches discrepancies within hours instead of weeks. Last month it flagged a $23K inventory mismatch we would have never found." \- Director of Operations, omnichannel retailer --- ## FlowRunner Pricing | Free Plan, Paid Plans from $5/mo Source: https://flowrunner.ai/pricing Start on the Free plan. No credit card required. Every new account gets 14 days of Professional. #### Free $0.00 / month executions: 100 One execution = one complete workflow run from start to finish - Unlimited workflows - Unlimited users - 1 concurrent execution - Full AI Agents (BYOK) - All integrations - Human-in-the-loop orchestration - Pause and wait up to 30 days per run - 24-hour log history - Single-flow export [Start Free](https://app.flowrunner.ai) No credit card, no time limit #### Starter $5.00 / month executions: 300 3,000 One execution = one complete workflow run from start to finish - Everything in Free - 300 or 3,000 executions a month - 1 concurrent execution - 24-hour log history - Single-flow export [Get Started](https://app.flowrunner.ai) Per-run price falls as volume grows #### Growth $45.00 / month executions: 12,000 30,000 60,000 One execution = one complete workflow run from start to finish - Everything in Starter - 20 concurrent executions - Pause and wait up to 30 days per run - 1 hour max active runtime per run - 7-day log history - Bulk workspace and API export [Get Started](https://app.flowrunner.ai) Upgrade from Free anytime Popular #### Professional $299.00 / month executions: 75,000 One execution = one complete workflow run from start to finish - Everything in Growth - 50 concurrent executions - Pause and wait up to 1 year per run - 4 hours max active runtime per run - 30-day audit trails - Role-based access control (RBAC) - Priority support [Start 14-Day Trial](https://app.flowrunner.ai) Free for 14 days on signup. No credit card required #### Business $999.00 / month executions: 250,000 One execution = one complete workflow run from start to finish - Everything in Professional - 200 concurrent executions - Pause and wait up to 1 year per run - 12 hours max active runtime per run - SLA tracking and monitoring - SSO/SAML authentications - Advanced compliance reporting - 90-day audit trails [Get Started](https://app.flowrunner.ai) Upgrade from any plan #### Enterprise Custom pricing - Unlimited workflows - Unlimited executions - Concurrency with a queue - Pause and wait up to 1 year, extendable - Unlimited active runtime - Unlimited audit trail retention - Cloud or on-premise deployment - Custom SLA with guaranteed uptime - SSO/SAML/LDAP authentication - White-glove onboarding - Architecture review and optimization - Priority feature requests [Contact Us](https://flowrunner.ai/contact) Coming Soon #### Self-Hosted Community Free - Unlimited executions - Unlimited workflows - Full AI Agents - Runtime and wait bounded only by your infrastructure - Single node deployment - No audit trails - No RBAC - No SSO ### Feature comparison | Feature | Free $0 | Starter $5-15 | Growth $45-149 | Professional $299 | Business $999 | Enterprise | | --- | --- | --- | --- | --- | --- | --- | | Unlimited Workflows Build as many flows as you need. No cap on flow count, no per-flow charge, on any plan. | Included | Included | Included | Included | Included | Included | | Unlimited Users Invite your whole team from day one. No per-seat pricing on any plan, including Free. | Included | Included | Included | Included | Included | Included | | All integrations The full connector catalog on every plan. Nothing is gated behind a higher tier. | Included | Included | Included | Included | Included | Included | | AI Agents (BYOK) AI agents run as native nodes inside a flow, not as a bolt-on. You connect your own provider keys (OpenAI, Anthropic, Google, and others) and pay those providers directly for model usage. | Included | Included | Included | Included | Included | Included | | Human-in-the-loop Agents pause a run and pull in a person over email, Slack, WhatsApp, or phone, then resume with full context. Included on every plan. | Included | Included | Included | Included | Included | Included | | Executions/month One execution is one complete run of a flow, start to finish. A flow with 40 steps that runs once counts as 1. A run that pauses to wait, for a person or for another system, is still 1. | 100 | 300-3,000 | 12k-60k | 75,000 | 250,000 | Unlimited | | Concurrent executions How many runs can be in flight at the same time. Runs beyond the limit wait in the queue until a slot opens. A flow parked on a human decision or an external system keeps its slot while it waits. | 1 | 1 | 20-50 | 50 | 200 | Configurable | | Max active runtime per run The time a run spends actually computing. This is a safety ceiling to catch a runaway flow, not a meter you spend down. Time spent parked waiting on a person or another system does not count toward it. | 1 hour | 1 hour | 1 hour | 4 hours | 12 hours | Unlimited | | Max wait per run How long a single run can sit parked before it has to resume. A flow enters wait mode when it reaches a step that listens for a trigger: an approval reply from a person, an inbound webhook, a callback from another system. The clock starts the moment the flow parks, not when the run began, and stops when the trigger fires. While parked, the run holds its full state and one concurrent slot, but uses no runtime and no executions. When the trigger fires, the flow resumes at the next step with the context it had before it paused. | Up to 30 days | Up to 30 days | Up to 30 days | Up to 1 year | Up to 1 year | Up to 1 year, extendable | | Execution backlog queue When you are at your concurrency limit, new runs wait in a backlog and start automatically as slots free up, instead of being dropped. | Not included | Not included | Not included | Not included | Included | Included | | Log history How far back run history and analytics are visible in the app, measured from when the flow completes. Older runs are hidden, not deleted, and upgrading reveals them. | 24 hours | 24 hours | 7 days | 30 days | 90 days | Unlimited | | Export Any single flow can be exported on every plan. Bulk workspace export and API export are available from Growth. | Single flow | Single flow | Full workspace + API | Full workspace + API | Full workspace + API | Full workspace + API | | Audit Trails A durable record of who approved what, when, and with what context in front of them, exportable for auditors. The audit window follows the log history row above. | Not included | Not included | Not included | Included | Included | Included | | Role-Based Access Control Control who can build, run, view, and approve, across Admin, Builder, Viewer, and Operator roles. | Not included | Not included | Not included | Included | Included | Included | | Compliance Reporting Prebuilt exports and reports covering run history, approvals, and access events, for internal audit and regulatory review. | Not included | Not included | Not included | Not included | Included | Included | | SLA Tracking Track how long human decision steps take against your target response times, and see which ones ran late. | Not included | Not included | Not included | Not included | Included | Included | | Support How you reach us. Enterprise adds a dedicated customer success manager, a shared Slack channel, and phone. | Forum | Forum | Forum | Forum (priority) | Forum + Email | Dedicated CSM, Slack, Phone | A flow can pause for a person or for another system. Paused flows use no runtime and no executions. Log history windows are measured from the moment a flow completes. Older runs are hidden, not deleted. ### Frequently asked questions Running a security review? Encryption, certifications, sub-processors, and data handling are covered on the [Trust & Security](https://flowrunner.ai/trust) page. #### Is the Free plan really free? Yes. $0, no time limit, no credit card. The Free plan runs 100 executions a month, one execution at a time, with 24 hours of log history, and includes everything else the platform has: every integration, AI agents with your own keys, human-in-the-loop, unlimited users, and unlimited workflows. One Free workspace per account. #### How does the 14-day trial work? Every new account starts its first workspace on the Professional plan for 14 days: 75,000 executions, 30-day audit trails, and RBAC. No credit card required. If the trial ends without a payment method on file, the workspace moves to the Free plan and your flows keep running within Free limits. Nothing is paused or deleted. #### What email address can I sign up with? Any real address, including personal domains such as Gmail or Outlook. Disposable and throwaway email domains are not accepted. #### What counts as an execution? One execution equals one complete run of a workflow from start to finish. A workflow with 10 steps that runs once equals 1 execution. A run that pauses to wait, whether for a person or for another system, is still one execution, no matter how long the wait lasts. #### What is the difference between Free, Starter, and Growth? Volume, concurrency, log history, and export. Free runs 100 executions a month, Starter 300 or 3,000, Growth 12,000 to 60,000. Free and Starter run one execution at a time and show 24 hours of run history. Growth runs 20 to 50 at once, shows 7 days, and adds bulk workspace and API export. The platform itself, including every integration, AI agents, and human-in-the-loop, is the same on all three. #### Why does the per-run price fall as I move up? Starter at $5 works out to $0.0167 per execution. Starter at $15 is $0.005. Growth at $45 is $0.00375, and Growth at $149 is $0.00248. Every step up buys more volume per dollar than the plan you are leaving. #### What's the difference between active runtime and wait time? Active runtime is the time a run spends actually computing. It has a per-plan ceiling: 1 hour on Free, Starter, and Growth, 4 hours on Professional, 12 hours on Business, unlimited on Enterprise. Wait time is separate. It is the time a flow sits parked waiting on something outside the flow, and it costs no runtime and no executions. A flow can wait up to 30 days on Free, Starter, and Growth, and up to 1 year on Professional, Business, and Enterprise. #### What can a flow wait for? A person, through email, Slack, WhatsApp, or phone. A webhook or callback from another system. A batch job finishing. The wait ceiling is the same regardless of what the flow is waiting on. #### Does waiting use up my executions or runtime? No. A parked flow is not computing, so it consumes no runtime, and it stays a single execution however long the wait runs. It does hold one of your concurrent execution slots while it is parked, which is why concurrency scales with the plan: 1 on Free and Starter, 20 to 50 on Growth, 50 on Professional, 200 on Business. #### What happens when the response finally arrives? The flow resumes exactly where it stopped, with the full context it had before it paused, whether the response came from a person or from another system. Waiting never cancels or drops a run. #### Will my flows hit the active runtime ceiling? Almost certainly not. Finance and operations flows finish in seconds or minutes. The ceiling exists to stop a flow caught in a loop or hung on an unresponsive endpoint, not to meter your work. Time spent parked waiting on a person or another system never counts toward it, because a parked flow is not computing. #### How much run history can I see? 24 hours on Free and Starter, 7 days on Growth, 30 days on Professional, 90 days on Business, unlimited on Enterprise. The window counts from the moment a flow completes, so a flow that waited three weeks for an approval still gets the full window once it finishes. #### What happens to runs older than my log history window? They are hidden, not deleted. The app shows how many runs sit outside your window, and upgrading reveals them. The periods after which run data is actually deleted are stated in the Privacy Policy. #### Are users really unlimited on all plans? All plans, including Free, include unlimited user accounts with no per-seat pricing. Team members can be invited from day one. #### Are workflows really unlimited? All plans include unlimited workflow creation without artificial restrictions. #### What does BYOK (Bring Your Own Keys) mean? BYOK means you provide your own API keys for AI services like OpenAI, Anthropic, Google, and others. You pay the AI providers directly based on your usage. #### What happens if I hit my execution limit? When you reach 100% of your monthly execution limit, workflows pause immediately. They resume automatically when your next billing cycle begins, or you can upgrade for immediate reactivation. No surprise overage charges. #### What can I export on Free and Starter? Any single flow, on every plan. Bulk workspace export and API export are available from Growth. Export is not available while a workspace is on the 14-day trial. #### What's the difference between Growth and Professional? Growth ($45) targets teams without compliance requirements. Professional ($299) adds audit trails and RBAC for regulated industries. SLA tracking starts at Business ($999). #### What compliance features are included? Professional: 30-day audit trails and RBAC. Business: 90-day audit trails, SLA tracking, SSO/SAML, advanced compliance reporting. Enterprise: Unlimited audit retention, self-hosted option, LDAP support. #### What does Role-Based Access Control (RBAC) do? RBAC controls workflow access permissions across Admin, Builder, Viewer, and Operator roles. Essential for compliance documentation. #### What is SSO/SAML and why do I need it? SSO (Single Sign-On) lets employees log into FlowRunner using company credentials from Okta, Azure AD, Google Workspace, or OneLogin. #### What self-hosted options are available? Community Edition (free, unlimited executions, single node) or Enterprise Self-Hosted (custom pricing, multi-instance clustering, full compliance features). #### What are human-in-loop features? Human-in-loop allows AI agents to pause workflow execution, contact a human via their preferred channel (email, Slack, WhatsApp, or phone), wait for a response with full context preserved, and then continue processing. #### Are AI agents included in all plans? Yes. All plans, including Free, include full AI agent orchestration: multi-agent coordination, human-in-loop subflows, and native MCP (Model Context Protocol) support. #### How do I upgrade to a higher plan? Upgrade anytime through the in-app billing page. Billing is prorated for remaining cycle days. Moving to any paid plan requires a payment method on file; the Free plan never does. #### What payment methods do you accept? Credit cards (Visa, MasterCard, Amex, Discover). Enterprise can use ACH or invoice billing. #### What is Midnight Flow consulting? Our consulting arm providing implementation services. Included with Enterprise tier or available separately starting at $5,000. #### How does FlowRunner compare to n8n? Both bill per execution, so the comparison is direct. FlowRunner's Free plan runs 100 executions a month with every integration, AI agents, and human-in-the-loop included, and paid plans start at $5. n8n Cloud starts at €20/mo, and its free option is self-hosting. FlowRunner includes the full cloud compliance suite (audit trails and RBAC) at $299/mo, where n8n's cloud equivalent requires custom-priced Enterprise. #### What integrations are included? All plans, including Free, include all integrations: Email, Slack, WhatsApp, Phone, and connections to business tools via native integrations, REST APIs, webhooks, and MCP. #### What are concurrent executions? The maximum number of workflows running simultaneously. When the limit is exceeded, additional workflows queue until a slot opens. #### What is your refund policy? Subscription fees are non-refundable. You can cancel anytime through your account settings, and your plan stays active through the end of the current billing period. The Free plan is always available to move down to. --- ## Privacy Policy Source: https://flowrunner.ai/privacy Last Updated: February 28, 2025 ### 1\. Introduction This Privacy Policy describes how Midnight Coders, Inc. ("Company," "we," "us," or "our"), a Texas corporation, collects, uses, discloses, and protects information in connection with the FlowRunner platform and related services (the "Service") available at flowrunner.ai. This Policy applies to all visitors to our website, registered account holders, authorized users, and any individuals whose personal information is processed through the Service. It applies to both our cloud-hosted and self-hosted deployment options, though data handling responsibilities differ based on deployment model as described in Section 10. Capitalized terms not defined in this Policy have the meanings given to them in our [Terms of Service](https://flowrunner.ai/terms). #### 1.1 Our Dual Role With respect to personal data, we operate in two distinct capacities: **As a data controller:** We determine the purposes and means of processing your Account Data (registration information, billing details, usage analytics, and communications with us). Sections 2 through 9 of this Policy describe our controller activities. **As a data processor:** When you use the Service to process your own data through Workflows, AI Agents, and Human-in-Loop interactions, we act as a processor on your behalf. You, as the Customer, are the controller of that data. Section 10 describes our processor obligations. ### 2\. Information We Collect #### 2.1 Account Data When you register for an account, we collect: - **Identity information:** Full name, email address. - **Authentication credentials:** Password (stored in hashed form) and, where applicable, single sign-on tokens from identity providers (Okta, Azure AD, Google Workspace). - **Billing information:** We do not collect or store payment card numbers, expiration dates, or billing addresses. All payment processing is handled entirely by Stripe, Inc., our third-party payment processor. We receive only a transaction confirmation and Stripe customer identifier — no payment card details ever touch our systems. - **Account preferences:** Notification settings, Subscription Plan selection, and feature configurations. #### 2.2 Usage Data We automatically collect information about how you interact with the Service: - **Platform activity:** Workflows created, Executions run, AI Agents configured, features used, errors encountered, and performance metrics. - **Device and access information:** IP address, browser type and version, operating system, device identifiers, referring URL, pages visited, and access timestamps. - **Execution metadata:** Workflow execution start and end times, Running Time consumed, Waiting Time periods, execution status (success, failure, paused), and Execution counts against tier limits. #### 2.3 Communication Data When you contact us or we contact you, we collect: - **Support communications:** Emails, chat messages, and support tickets, including any attachments or screenshots you provide. - **Sales communications:** Records of discovery calls, demo sessions, and business correspondence. #### 2.4 Cookie and Tracking Data We use cookies and similar technologies as described in Section 7. #### 2.5 Information We Do Not Collect We do not intentionally collect: - Personal information from individuals under 18 years of age. - Social Security numbers, government-issued identification numbers, or biometric data. - Information from disposable or temporary email addresses — registrations from such providers are not accepted. #### 2.6 Customer Content **Customer Content is data that you, as a Customer, transmit through or store within the Service through your Workflows, AI Agents, and Human-in-Loop interactions.** We process Customer Content on your behalf as a data processor. We do not determine the categories or types of personal information contained in Customer Content — you do. See Section 10 for our processor obligations regarding Customer Content. ### 3\. How We Use Your Information #### 3.1 To Provide and Operate the Service We use your Account Data and Usage Data to: - Create and manage your account. - Authenticate your access and enforce role-based access controls. - Process payments and manage billing. - Deliver the Service, including Workflow execution, AI Agent orchestration, and Human-in-Loop communications. - Monitor and enforce Subscription Plan limits (Execution counts, Running Time, concurrent Executions). - Provide customer support and respond to inquiries. #### 3.2 To Improve and Develop the Service We use Usage Data in aggregated and anonymized form to: - Analyze platform performance and identify areas for improvement. - Understand feature adoption patterns and usage trends. - Develop new features and capabilities. - Conduct internal research and analytics. We do not use Customer Content for product development, model training, or any purpose other than providing the Service to you. #### 3.3 To Communicate with You We use your contact information to: - Send transactional notifications (account confirmation, password resets, Execution limit warnings, trial expiry notices, billing receipts). - Notify you of material changes to the Service, this Policy, or our Terms of Service. - Send product updates, feature announcements, and educational content (you may opt out of non-transactional communications at any time). - Respond to your support requests and inquiries. #### 3.4 To Protect and Secure the Service We use Account Data, Usage Data, and Device and Access Information to: - Detect and prevent fraud, abuse, and unauthorized access. - Monitor for security threats and vulnerabilities. - Enforce our Terms of Service and Acceptable Use policies. - Comply with legal obligations and respond to lawful requests. #### 3.5 How We Do Not Use Your Information We do not: - Sell your personal information to third parties. - Share your personal information for third-party advertising purposes. - Use Customer Content for AI model training, product development, or any purpose beyond providing the Service. - Use personal information for automated decision-making that produces legal or similarly significant effects without human involvement. ### 4\. Legal Bases for Processing (EEA, UK, and Swiss Individuals) If you are located in the European Economic Area (EEA), United Kingdom, or Switzerland, we process your personal data under the following legal bases: **Performance of a contract (Article 6(1)(b)):** Processing your Account Data to provide the Service, manage your subscription, process payments, and deliver support. This is the primary basis for most processing activities. **Legitimate interests (Article 6(1)(f)):** Processing Usage Data for security, fraud prevention, service improvement, and analytics, where our interests do not override your fundamental rights. Our legitimate interests include operating and improving the Service, protecting against misuse, and understanding how the Service is used. **Consent (Article 6(1)(a)):** Processing for non-essential cookies, marketing communications, and any other activities for which we specifically request your consent. You may withdraw consent at any time. **Legal obligation (Article 6(1)(c)):** Processing required to comply with applicable laws, regulations, court orders, or governmental requests. ### 5\. Information Sharing and Disclosure We do not sell your personal information. We share personal information only in the following circumstances: #### 5.1 Service Providers (Sub-Processors) We engage the following third-party service providers to help us deliver the Service. These providers process data only on our instructions and are bound by contractual data protection obligations: - **DigitalOcean** — Cloud infrastructure hosting. Data processed: All data stored in the Service (encrypted at rest). - **Stripe, Inc.** — Payment processing. Data processed: Transaction confirmations, Stripe customer identifiers (no payment card data). - **MailerSend** — Transactional email delivery. Data processed: Email addresses, notification content. - **Alphabet, Inc.** — Website analytics. Data processed: Device/access info, anonymized usage patterns. We maintain a current list of sub-processors at flowrunner.ai/sub-processors. We will notify you at least thirty (30) days before adding a new sub-processor that processes Customer Content, as described in our Terms of Service. #### 5.2 Third-Party AI Services (BYOK) When you provide your own API keys for third-party AI services (the BYOK model), data from your Workflows is transmitted directly to those third-party services using your credentials. **We do not control, monitor, or have visibility into data transmitted to third-party AI providers through your API keys.** Your use of those services is governed by your own agreements with those providers. We recommend reviewing the privacy policies of any AI service you integrate. #### 5.3 Human-in-Loop Communication Channels When your Workflows use Human-in-Loop features, communications are transmitted through third-party channels (email infrastructure, Slack, WhatsApp, telephone carriers). We facilitate the delivery of these communications on your behalf but do not independently collect or retain the personal information of Human-in-Loop recipients beyond what is necessary to deliver the communication and maintain audit logs where enabled. #### 5.4 Legal and Compliance Disclosures We may disclose personal information if required to do so by law, or in the good-faith belief that such disclosure is necessary to: - Comply with a legal obligation, court order, subpoena, or governmental request. - Protect and defend our rights or property. - Prevent or investigate possible wrongdoing in connection with the Service. - Protect the personal safety of users of the Service or the public. #### 5.5 Business Transfers If Midnight Coders, Inc. is involved in a merger, acquisition, reorganization, or sale of assets, your personal information may be transferred as part of that transaction. We will notify you via email or prominent notice on the Service before your personal information becomes subject to a different privacy policy. #### 5.6 With Your Consent We may share your personal information for other purposes with your explicit consent. #### 5.7 Midnight Flow Consulting If you engage Midnight Flow consulting services, relevant Account Data and project information may be shared between the FlowRunner platform team and Midnight Flow consultants to deliver the consulting engagement. This sharing is governed by your consulting agreement with Midnight Flow. ### 6\. Data Retention We retain personal information only for as long as reasonably necessary to fulfill the purposes described in this Policy, unless a longer retention period is required by law. Specific retention periods are: - **Account Data** — Duration of active subscription + 30 days. Basis: Contract performance; 30-day reactivation window per ToS. - **Billing records** — 7 years after transaction. Basis: Tax and financial reporting obligations. - **Execution logs (Free, Starter)** — 7 days. Visible in the Service for 24 hours; retained for 7 days and visible again on upgrade. Basis: Per Subscription Plan. - **Execution logs (Growth)** — 30 days. Visible in the Service for 7 days. Basis: Per Subscription Plan. - **Execution logs and audit trails (Professional)** — 90 days. Visible in the Service for 30 days. Basis: Per Subscription Plan. - **Execution logs and audit trails (Business)** — 90 days. Basis: Per Subscription Plan. - **Audit trails (Enterprise)** — Unlimited, configurable. Basis: Per Enterprise agreement. - **Customer Content (post-termination)** — 30 days active + 90 days in encrypted backups. Basis: Per ToS Section 8.4. - **Usage Data (aggregated)** — Indefinite, anonymized. Basis: Legitimate interest in service improvement. - **Support correspondence** — 3 years after resolution. Basis: Legitimate interest; legal defense. - **Cookie data** — See Section 7. Basis: Per cookie category. After the applicable retention period, we delete or anonymize personal information. You may request earlier deletion subject to our legal retention obligations (see Section 8). ### 7\. Cookies and Tracking Technologies #### 7.1 Cookies We Use **Strictly Necessary Cookies:** Required for the Service to function (session management, authentication, security tokens). These cannot be disabled. Duration: session or up to 30 days for "keep me signed in." **Functional Cookies:** Remember your preferences (language, notification settings, interface customizations). Duration: up to 12 months. **Analytics Cookies:** Help us understand how visitors interact with our website and Service (page views, navigation patterns, feature usage). Duration: up to 24 months. **Marketing Cookies:** Used to track visitors across websites for the purpose of displaying relevant advertisements. Duration: up to 12 months. #### 7.2 Cookie Consent When you first visit our website, you will be presented with a cookie consent banner that allows you to accept or decline non-essential cookies (functional, analytics, and marketing). Strictly necessary cookies do not require consent. You can change your cookie preferences at any time through our cookie settings page at flowrunner.ai/cookie-preferences. #### 7.3 Do Not Track We currently do not respond to "Do Not Track" browser signals, as there is no industry-standard interpretation of these signals. We will update this Policy if a standard emerges that we decide to follow. #### 7.4 Application vs. Marketing Website Marketing cookies and remarketing pixels operate only on our marketing website (flowrunner.ai public pages). **We do not deploy marketing cookies, remarketing pixels, or third-party advertising trackers within the authenticated FlowRunner application environment.** This separation is maintained to protect the privacy of Customer Content and to support our customers' compliance requirements. ### 8\. Your Privacy Rights #### 8.1 Rights for All Users Regardless of your location, you may: - **Access** your personal information by logging into your Account settings. - **Update or correct** your Account Data through your Account settings. - **Delete your account** by contacting [support@flowrunner.ai](mailto:support@flowrunner.ai) or through Account settings. - **Opt out** of marketing communications by clicking the unsubscribe link in any marketing email or updating your notification preferences. - **Export your data** (Workflows, configurations) through the Service's built-in export features. #### 8.2 Additional Rights for EEA, UK, and Swiss Residents (GDPR) If you are located in the European Economic Area, United Kingdom, or Switzerland, you have the following additional rights under GDPR: - **Right of access (Article 15):** Request a copy of all personal data we hold about you in a structured, commonly used, machine-readable format. - **Right to rectification (Article 16):** Request correction of inaccurate or incomplete personal data. - **Right to erasure (Article 17):** Request deletion of your personal data, subject to our legal retention obligations. - **Right to restriction (Article 18):** Request that we restrict processing of your personal data in certain circumstances. - **Right to data portability (Article 20):** Receive your personal data in a machine-readable format and transmit it to another controller. - **Right to object (Article 21):** Object to processing based on legitimate interests, including profiling. - **Right to withdraw consent:** Where processing is based on consent, withdraw that consent at any time without affecting the lawfulness of processing performed before withdrawal. - **Right to lodge a complaint:** File a complaint with your local data protection supervisory authority. We will respond to GDPR requests within thirty (30) days. If we require additional time (up to an additional sixty days for complex requests), we will inform you of the extension and the reasons for the delay. #### 8.3 Additional Rights for California Residents (CCPA/CPRA) If you are a California resident, the California Consumer Privacy Act (CCPA) and California Privacy Rights Act (CPRA) provide the following rights: - **Right to know:** You may request disclosure of the categories and specific pieces of personal information we have collected about you, the categories of sources, the business purposes for collection, and the categories of third parties with whom we share it. - **Right to delete:** You may request deletion of personal information we have collected, subject to certain exceptions. - **Right to correct:** You may request correction of inaccurate personal information. - **Right to opt out of sale or sharing:** We do not sell personal information or share it for cross-context behavioral advertising. No opt-out is necessary, but we honor Global Privacy Control (GPC) signals as a valid opt-out request. - **Right to non-discrimination:** We will not discriminate against you for exercising your CCPA/CPRA rights. We will respond to verified CCPA requests within forty-five (45) days. You may submit requests up to twice per twelve-month period. #### 8.4 Additional Rights for Texas Residents The Texas Data Privacy and Security Act (TDPSA) provides Texas residents with rights including access, correction, deletion, data portability, and the right to opt out of targeted advertising, sale of personal data, and profiling. We honor these rights consistent with the procedures described above. #### 8.5 How to Exercise Your Rights To exercise any of the rights described above, contact us at: **Email:** [privacy@flowrunner.ai](mailto:privacy@flowrunner.ai) **Mail:** Midnight Coders, Inc., Attn: Privacy, 539 W. Commerce St, Suite 2023, Dallas, TX 75208 We may need to verify your identity before processing your request. For account holders, we will verify through your authenticated Account. For non-account holders, we may request additional information to confirm your identity. If you are a Customer exercising rights regarding personal data contained within your **Customer Content** (i.e., data processed through your Workflows), you should use the Service's built-in tools to access, export, correct, or delete that data directly, as you are the data controller for Customer Content. ### 9\. Data Security We implement reasonable administrative, technical, and organizational security measures designed to protect personal information from unauthorized access, disclosure, alteration, and destruction. These measures include: - **Encryption in transit:** All data transmitted between your browser and the Service is encrypted using TLS 1.2 or higher. - **Encryption at rest:** Customer Content and sensitive Account Data are encrypted at rest using AES-256 or equivalent encryption. - **Access controls:** Internal access to personal data is restricted to authorized personnel on a need-to-know basis, with multi-factor authentication required. - **Infrastructure security:** Cloud infrastructure is hosted in SOC 2-audited data centers operated by DigitalOcean. - **Monitoring and logging:** We maintain security monitoring and logging to detect and respond to potential incidents. - **Credential handling:** BYOK API keys are encrypted at rest and are not accessible to Company personnel in plaintext. While we strive to protect your personal information, no method of transmission over the Internet or electronic storage is completely secure. We cannot guarantee absolute security. #### 9.1 Data Breach Notification In the event of a data breach affecting your personal information, we will: - Notify affected individuals without undue delay and, where required by GDPR, within seventy-two (72) hours of becoming aware of the breach. - Notify the relevant supervisory authority as required by applicable law. - Provide details of the breach, the likely consequences, and the measures taken or proposed to address it. - For Customers with active subscriptions, notify the Account administrator via email. ### 10\. Customer Content and Data Processing #### 10.1 Customer as Controller When you use the Service to process data through your Workflows, AI Agents, and Human-in-Loop interactions, you are the data controller for that Customer Content. You determine what data is processed, the purposes of processing, and the lawful basis for processing. We process Customer Content solely on your instructions as documented in our Terms of Service and any applicable Data Processing Agreement. #### 10.2 Data Processing Agreement For Customers who require a [Data Processing Agreement (DPA)](https://flowrunner.ai/data-processing) under GDPR or other applicable data protection laws, we offer a standard DPA that covers our processor obligations, including data security measures, sub-processor management, data breach notification, audit rights, and cross-border transfer mechanisms. Contact [legal@flowrunner.ai](mailto:legal@flowrunner.ai) to execute a DPA. #### 10.3 HIPAA and Protected Health Information If you are a Covered Entity or Business Associate under HIPAA and intend to use the Service to process Protected Health Information (PHI), you must execute a Business Associate Agreement (BAA) with us **before** transmitting any PHI through the Service. Contact [legal@flowrunner.ai](mailto:legal@flowrunner.ai) to initiate a BAA. Without a BAA in place, you must not use the Service to process PHI. The availability of compliance features (audit trails, RBAC, SLA tracking) varies by Subscription Plan — you are responsible for selecting a plan that meets your HIPAA obligations. #### 10.4 Cloud vs. Self-Hosted Deployment **Cloud Deployment:** For cloud-hosted Customers, we manage the infrastructure and act as a data processor. The security, availability, and backup practices described in this Policy apply to cloud deployments. **Self-Hosted Deployment:** For self-hosted Customers (Community Edition or Enterprise self-hosted), you are responsible for the security, availability, and data protection of your own infrastructure. We do not have access to or visibility into Customer Content on self-hosted installations. This Policy's data security and breach notification commitments do not apply to self-hosted deployments except to the extent we provide managed support services under an Enterprise agreement. #### 10.5 International Data Transfers The Service is hosted in the United States. If you are located outside the United States, your personal data will be transferred to and processed in the United States. For transfers of personal data from the EEA, UK, or Switzerland to the United States, we rely on: - **Standard Contractual Clauses (SCCs)** approved by the European Commission, incorporated into our Data Processing Agreement. - Any applicable adequacy decisions. - Other lawful transfer mechanisms as may be available under applicable data protection laws. You may obtain a copy of the SCCs by contacting [legal@flowrunner.ai](mailto:legal@flowrunner.ai). ### 11\. Children's Privacy The Service is designed for business use and is not directed at individuals under eighteen (18) years of age. We do not knowingly collect personal information from children under 18. Our Terms of Service require users to be at least 18 years old, and our corporate email requirement is designed to prevent registration by minors. If we learn that we have collected personal information from a child under 18, we will take steps to delete that information promptly. If you believe a child under 18 has provided us with personal information, please contact us at [privacy@flowrunner.ai](mailto:privacy@flowrunner.ai). ### 12\. Third-Party Links and Services The Service may contain links to third-party websites and integrate with Third-Party Services. This Policy does not apply to the practices of third parties. We encourage you to review the privacy policies of any third-party services you connect to the Service, including AI model providers (OpenAI, Anthropic, Google, Cohere), communication platforms (Slack, WhatsApp), and enterprise applications you integrate through your Workflows. ### 13\. Changes to This Policy We may update this Privacy Policy from time to time. If we make material changes, we will notify you by email (to the address associated with your Account) or by posting a prominent notice on the Service at least thirty (30) days before the changes take effect. We will also update the "Last Updated" date at the top of this Policy. Your continued use of the Service after the effective date of any changes constitutes your acceptance of the updated Policy. We encourage you to review this Policy periodically. ### 14\. Contact Us If you have questions, concerns, or requests regarding this Privacy Policy or our data practices, please contact us: Midnight Coders, Inc. Attn: Privacy 539 W. Commerce St, Suite 2023 Dallas, TX 75208 **Privacy inquiries:** [privacy@flowrunner.ai](mailto:privacy@flowrunner.ai) **Legal inquiries:** [legal@flowrunner.ai](mailto:legal@flowrunner.ai) **Security inquiries:** [security@flowrunner.ai](mailto:security@flowrunner.ai) **Data subject access requests:** [privacy@flowrunner.ai](mailto:privacy@flowrunner.ai) **BAA/DPA requests:** [legal@flowrunner.ai](mailto:legal@flowrunner.ai) --- ## Solutions Source: https://flowrunner.ai/solutions Pick the problem you're solving. We'll show you the integrations, workflows, and human-in-the-loop pattern that fit. 10 solutions published No solutions match your current filters. - [Accounting Automation](https://flowrunner.ai/solutions/accounting-automation) - [Close Management Software](https://flowrunner.ai/solutions/close-management-software) - [Journal Entry Automation](https://flowrunner.ai/solutions/journal-entry-automation) - [Marketing Workflows](https://flowrunner.ai/solutions/marketing-workflows) - [Procurement Automation](https://flowrunner.ai/solutions/procurement-automation) - [Property Management Automation Software](https://flowrunner.ai/solutions/property-management-automation-software) - [Purchase Order Automation Software](https://flowrunner.ai/solutions/purchase-order-automation-software) - [Purchase Request Software](https://flowrunner.ai/solutions/purchase-request-software) - [Supplier Portal Software](https://flowrunner.ai/solutions/supplier-portal-software) - [Supplier Relationship Management Software](https://flowrunner.ai/solutions/supplier-relationship-management-software) --- ## FlowRunner by team Source: https://flowrunner.ai/teams Whatever your team runs on, an agent can drive it, and a human stays on the decisions that carry weight. Pick your stack, or build your own on the largest verified tool library. 2142 integrations 56,186 verified actions Every action verified against the vendor's official API ### [Developers](https://flowrunner.ai/teams/developers) 164 integrations Ship agents and LLM apps on a platform that gives you every model provider, every vector store, and the tool library your agent needs, with a human gate on the moves that publish, write to production, or cost real money. Explore the stack ### [Security & IT](https://flowrunner.ai/teams/security) 45 integrations Automate the identity, triage, and response work that eats your team's day, with an agent that runs the investigation and a human who owns every action that changes access or cannot be undone. Explore the stack ### [Finance & Operations](https://flowrunner.ai/teams/finance-ops) 326 integrations Automate the invoicing, reconciliation, and close work that swallows your month, with an agent that runs the matching and the routine and a human who owns every payment, posting, and refund that touches the ledger or leaves the building. Explore the stack ### [RevOps](https://flowrunner.ai/teams/revops) 281 integrations Automate the lead enrichment, CRM hygiene, and handoff work that decides whether pipeline moves or stalls, with an agent that runs the routine and a human who owns every move that touches a named account, a big deal, or the whole database. Explore the stack ### [Support & CX](https://flowrunner.ai/teams/support) 87 integrations Automate the triage, drafting, and follow-up work that buries your queue, with an agent that runs the routine and a human who owns every reply, escalation, and bulk action that reaches a customer. Explore the stack ### [Data & Platform](https://flowrunner.ai/teams/data) 66 integrations Automate the syncs, loads, migrations, and index rebuilds that fill your on-call rotation, with an agent that runs the pipeline and a human who owns every write that hits production or cannot be rolled back. Explore the stack ### Built for people who build The orchestration layer between your models, your tools, and your team, with the plumbing and the guardrails already solved. #### Ship in a fraction of the time, no glue code Every integration is a block, not an API you wrap. Build visually where you want and drop to code where you need it. Compose flows and subflows, call a flow as a tool or an action, and reuse patterns across projects instead of rewriting the plumbing. #### Human-in-the-loop, native The digital andon cord is a first-class primitive, not a bolt-on approval node. Ship agents that run autonomously and stop for a person at the decision that carries weight, over Slack, email, WhatsApp, or phone, then resume with the answer in the audit trail. #### On your terms Bring your own model keys and pay the provider directly. Self-host or run on-prem so sensitive data never leaves your environment. No external platform dependencies, and unlimited users and workflows on every tier. #### Do more with a lean team Non-technical teammates run and monitor what you build, and unlimited seats mean the whole team collaborates from day one. Scale what your automations cover without scaling the headcount that maintains them. --- ## FlowRunner for data teams Source: https://flowrunner.ai/teams/data Automate the syncs, loads, migrations, and index rebuilds that fill your on-call rotation, with an agent that runs the pipeline and a human who owns every write that hits production or cannot be rolled back. [Start free](https://app.flowrunner.ai) See the stack The verified stack [Airtable](https://flowrunner.ai/integrations/airtable "Airtable") [AITable](https://flowrunner.ai/integrations/apitable "AITable") [Amazon DynamoDB](https://flowrunner.ai/integrations/dynamodb-service "Amazon DynamoDB") [Amazon Redshift](https://flowrunner.ai/integrations/aws-redshift "Amazon Redshift") [ApptiveGrid](https://flowrunner.ai/integrations/apptivegrid "ApptiveGrid") [Azure AI Search](https://flowrunner.ai/integrations/azure-ai-search "Azure AI Search") [Azure Cosmos DB](https://flowrunner.ai/integrations/azure-cosmos-db "Azure Cosmos DB") [Azure Table Storage](https://flowrunner.ai/integrations/azure-table-storage "Azure Table Storage") [Baserow](https://flowrunner.ai/integrations/baserow "Baserow") [Caspio](https://flowrunner.ai/integrations/caspio "Caspio") [Chroma](https://flowrunner.ai/integrations/chroma "Chroma") [Coveo](https://flowrunner.ai/integrations/coveo "Coveo") [CrateDB](https://flowrunner.ai/integrations/cratedb "CrateDB") [Databricks](https://flowrunner.ai/integrations/databricks "Databricks") [Elasticsearch](https://flowrunner.ai/integrations/elasticsearch "Elasticsearch") [Feishu Base](https://flowrunner.ai/integrations/feishubase "Feishu Base") [FileMaker](https://flowrunner.ai/integrations/filemaker "FileMaker") [Fusioo](https://flowrunner.ai/integrations/fusioo "Fusioo") [Google Analytics](https://flowrunner.ai/integrations/google-analytics "Google Analytics") [Google BigQuery](https://flowrunner.ai/integrations/bigquery "Google BigQuery") [Google Firestore](https://flowrunner.ai/integrations/google-firestore "Google Firestore") [Grist](https://flowrunner.ai/integrations/grist "Grist") [Jestor](https://flowrunner.ai/integrations/jestor "Jestor") [Jodoo](https://flowrunner.ai/integrations/jodoo "Jodoo") [Keboola](https://flowrunner.ai/integrations/keboola "Keboola") [Kintone](https://flowrunner.ai/integrations/kintone "Kintone") [Knack](https://flowrunner.ai/integrations/knack "Knack") [Ksaar](https://flowrunner.ai/integrations/ksaar "Ksaar") [Lark Base](https://flowrunner.ai/integrations/larksuitebase "Lark Base") [LEAV Engine](https://flowrunner.ai/integrations/leav-engine "LEAV Engine") [Mem0](https://flowrunner.ai/integrations/mem0 "Mem0") [Microsoft SQL Server](https://flowrunner.ai/integrations/sql-server "Microsoft SQL Server") [Milvus](https://flowrunner.ai/integrations/milvus "Milvus") [MongoDB](https://flowrunner.ai/integrations/mongodb "MongoDB") [MySQL](https://flowrunner.ai/integrations/mysql "MySQL") [Ninox](https://flowrunner.ai/integrations/ninox "Ninox") [NocoDB](https://flowrunner.ai/integrations/nocodb "NocoDB") [NoCodeTable](https://flowrunner.ai/integrations/nocodetable "NoCodeTable") [Oracle Database](https://flowrunner.ai/integrations/oracle-database "Oracle Database") [Origami](https://flowrunner.ai/integrations/origami "Origami") [PGVector](https://flowrunner.ai/integrations/pgvector "PGVector") [Pickaform](https://flowrunner.ai/integrations/pickaform "Pickaform") [Pinecone](https://flowrunner.ai/integrations/pinecone "Pinecone") [PostgreSQL](https://flowrunner.ai/integrations/postgresql "PostgreSQL") [Qdrant](https://flowrunner.ai/integrations/qdrant "Qdrant") [QuestDB](https://flowrunner.ai/integrations/questdb "QuestDB") [Quick Base](https://flowrunner.ai/integrations/quickbase "Quick Base") [QuintaDB](https://flowrunner.ai/integrations/quintadb "QuintaDB") [Ragic](https://flowrunner.ai/integrations/ragic "Ragic") [Redis](https://flowrunner.ai/integrations/redis "Redis") [restdb.io](https://flowrunner.ai/integrations/restdb-io "restdb.io") [Retable](https://flowrunner.ai/integrations/retable "Retable") [SeaTable](https://flowrunner.ai/integrations/seatable "SeaTable") [Segment](https://flowrunner.ai/integrations/segment "Segment") [Snowflake](https://flowrunner.ai/integrations/snowflake "Snowflake") [Stackby](https://flowrunner.ai/integrations/stackby "Stackby") [Supabase](https://flowrunner.ai/integrations/supabase "Supabase") [Tabidoo](https://flowrunner.ai/integrations/tabidoo "Tabidoo") [Tadabase](https://flowrunner.ai/integrations/tadabase "Tadabase") [TimescaleDB](https://flowrunner.ai/integrations/timescaledb "TimescaleDB") [TimeTonic](https://flowrunner.ai/integrations/timetonic "TimeTonic") [Weaviate](https://flowrunner.ai/integrations/weaviate "Weaviate") [Workiom](https://flowrunner.ai/integrations/workiom "Workiom") [Xata](https://flowrunner.ai/integrations/xata "Xata") [Zep](https://flowrunner.ai/integrations/zep "Zep") [Zite DB](https://flowrunner.ai/integrations/zite "Zite DB") 66 integrations 1,532 verified actions Verified against each vendor's official API ### The pipelines nobody wants to babysit Your day runs on jobs that mostly work until they do not. A reverse-ETL sync drifts and half the accounts in the warehouse go stale. A nightly load ingests a malformed file and the numbers in every downstream report are quietly wrong by morning. A backfill you kicked off by hand runs long, and now you are watching a terminal at 2am hoping the row counts land where they should. The routine is boring right up to the moment it is not, and the not is always expensive. The connectors are not the problem. You already run Postgres and MySQL for the operational side, Snowflake or BigQuery for the warehouse, a vector store for the AI features, and analytics on top. The problem is the manual glue and the judgment calls in the middle. A DELETE without a WHERE, a bulk write past the point of easy recovery, a schema change, a drop-and-rebuild of a live index. That is exactly the work you cannot hand to a cron job that has no idea when it is about to do something irreversible. - A reverse-ETL sync drifts and the warehouse goes stale. - A malformed file loads and every downstream report is wrong by morning. - Backfills run long while you watch row counts at 2am. - A schema change ships without anyone signing off on it. - The one destructive write is exactly what a blind cron job should not run. ### What you can automate Automations Each of these runs as a flow. The agent does the extraction, validation, and routine loads on its own. It stops and asks a human at the one point where a wrong write is expensive or impossible to undo. Every tool named here is a FlowRunner connector, built and verified against the vendor's official API. 1. ### Reverse-ETL sync to the warehouse The operational data lands in the warehouse. The production write-back waits for a human. 1. Trigger A schedule fires, or a change-data-capture event lands from the operational database. 2. Agent The agent reads the changed rows from PostgreSQL, reconciles them against the current state in Snowflake, maps the fields, and stages the delta so the warehouse mirrors the source without a full reload. 3. Result The warehouse stays in step with the operational store on every run, and a write back into production always has a named human who saw the diff first. Run 4400 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/postgresql "PostgreSQL")[](https://flowrunner.ai/integrations/snowflake "Snowflake")[](https://flowrunner.ai/integrations/segment "Segment")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Reads and staging happen on their own. When the sync would write back into the production PostgreSQL tables, or the reconciled delta is larger than the row threshold you set, it stops and hands the engineer the exact diff before anything is committed. 2. ### Validation before the load Bad data never reaches the warehouse. The agent quarantines it and asks. 1. Trigger A new file or batch arrives on the ingestion path, or an upstream job signals it is ready. 2. Agent The agent profiles the batch against the expected schema, checks row counts, null rates, types, and referential keys against MySQL, and separates the clean rows from the ones that fail the contract. 3. Result Clean data flows through untouched and a bad batch is caught before the load, not discovered three reports later when the numbers already went out. Run 4437 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/mysql "MySQL")[](https://flowrunner.ai/integrations/bigquery "Google BigQuery")[](https://flowrunner.ai/integrations/google-analytics "Google Analytics")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Clean batches load into BigQuery on their own. When a batch fails validation, or the failure rate crosses the bar you set, it holds the load, quarantines the bad rows, and routes the profile to the owner rather than poisoning the warehouse. 3. ### Bulk migration and backfill The migration plans and dry-runs itself. The cutover write waits for a go. 1. Trigger A migration ticket or a backfill request kicks it off. 2. Agent The agent reads the source records from Oracle Database, transforms them to the target shape, runs the mapping against a sample, and reports the row counts, the diffs, and any records that will not map cleanly. 3. Result A migration that used to mean a hand-watched terminal at 2am becomes a reviewed plan with the destructive bulk write gated behind a human go. Run 4474 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/oracle-database "Oracle Database")[](https://flowrunner.ai/integrations/snowflake "Snowflake")[](https://flowrunner.ai/integrations/sql-server "Microsoft SQL Server")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint It never bulk-writes past your threshold on autopilot. The agent stages the full backfill into Snowflake and stops for a sign-off, showing the exact volume and the failed-mapping list, before it commits a write of that size. 4. ### RAG index build and refresh The vector index rebuilds itself from source. The drop-and-swap waits for a person. 1. Trigger Source content changes, or a scheduled refresh of the knowledge base fires. 2. Agent The agent pulls the changed documents from MongoDB, chunks and embeds them with your model keys, and stages the new vectors so it can upsert incrementally into Pinecone without touching what is already serving. 3. Result The vector index tracks the source without a manual reindex, and a full drop-and-rebuild that would interrupt retrieval always has a person who chose the moment. Run 4511 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/mongodb "MongoDB")[](https://flowrunner.ai/integrations/pinecone "Pinecone")[](https://flowrunner.ai/integrations/weaviate "Weaviate")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Incremental upserts run on their own. When a change calls for dropping and rebuilding the whole index, which takes the RAG feature offline while it repopulates, the agent stops and waits for an engineer to confirm the rebuild rather than swapping a live index blind. 5. ### Ad-hoc data pull and report The pull assembles itself across systems. The write to production waits. 1. Trigger A stakeholder request lands, or a recurring report is due. 2. Agent The agent runs the read queries across Redis and PostgreSQL, joins them against Databricks, shapes the result into the requested cut, and drafts the report with the figures and the queries that produced them. 3. Result One-off pulls stop landing on whoever knows the schemas, and the only step that touches production is the one a human explicitly approved. Run 4548 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/redis "Redis")[](https://flowrunner.ai/integrations/postgresql "PostgreSQL")[](https://flowrunner.ai/integrations/databricks "Databricks")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Read-only pulls it completes and delivers on its own. If the request asks it to persist the result back into a production table or overwrite an existing dataset, it stops and shows the target and the payload to an owner before writing. ### The pattern that makes it safe to automate Data automation fails when a job writes something it should not. FlowRunner's answer is the digital andon cord: the agent runs the pipeline and pulls it the instant a step carries real consequence. It reads, validates, transforms, and stages the routine work on its own. A DELETE or UPDATE without a WHERE, a bulk write past your threshold, a schema or DDL change, a write into a production database, or a drop-and-rebuild of a live index always stops and routes to a person through the channel they already watch, with the diff, the row counts, and the query attached. Read and staging happen immediately; the write that cannot be rolled back waits for a human. That is the difference between automation you trust with production data and automation you do not. [The digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord) ### What your team gets Outcomes #### No unsupervised destructive writes An agent never runs a DELETE without a WHERE, bulk-writes past your threshold, or drops a live index on its own. The write that cannot be undone always has a named human behind it. #### Bad data caught before the load Every batch is profiled against its contract before it lands, so a malformed file is quarantined at the door instead of being discovered three reports downstream. #### The warehouse stays in step Reverse-ETL syncs reconcile the operational store and the warehouse on every run, so downstream numbers stop drifting between manual reloads. #### Migrations you can review Backfills and migrations arrive as a dry-run plan with row counts and failed mappings, so the cutover write is a reviewed decision, not a hand-watched terminal at 2am. #### Vector indexes that track the source RAG indexes refresh incrementally from the source database, and a full rebuild that would interrupt retrieval only happens when a person picks the moment. #### No coverage gaps The agent connects to the operational databases, warehouses, and vector stores you already run, every connector verified against the vendor API, so there is no store it cannot reach. ### Built for data teams' requirements Controls The controls you would demand of any automation with credentials to your production databases and warehouses are native to the platform, not add-ons. #### Complete audit trail Every query, every write, the user or agent that ran it, the approver, and the timestamp, recorded and exportable, so you can answer exactly who changed what data and when. #### RBAC and SSO/SAML Role-based access to the platform itself, with SSO and SAML, so who can approve a production write is governed by your identity provider, not a shared credential. #### Bring your own keys Your model providers and your database credentials, your keys. Embeddings and inference run on keys you control, and sensitive rows never route through a vendor's account. #### Self-hosted option Deploy inside your own infrastructure so regulated and proprietary data never leaves your environment on its way through a pipeline. > _I'm tired of exporting information from the systems manually._ Director of Operations, beauty brand ### The complete data and platform stack The stack Operational databases, warehouses, vector stores, and analytics, from PostgreSQL, MySQL, and Oracle to Snowflake, BigQuery, and Databricks, alongside Pinecone, Weaviate, and the rest of the vector layer. Every connector is built and verified against the vendor's official API, so an agent queries and writes the way the vendor's API actually allows. The tools the automations above run on. [PostgreSQL](https://flowrunner.ai/integrations/postgresql "PostgreSQL") [Snowflake](https://flowrunner.ai/integrations/snowflake "Snowflake") [Segment](https://flowrunner.ai/integrations/segment "Segment") [MySQL](https://flowrunner.ai/integrations/mysql "MySQL") [Google BigQuery](https://flowrunner.ai/integrations/bigquery "Google BigQuery") [Google Analytics](https://flowrunner.ai/integrations/google-analytics "Google Analytics") [Oracle Database](https://flowrunner.ai/integrations/oracle-database "Oracle Database") [Microsoft SQL Server](https://flowrunner.ai/integrations/sql-server "Microsoft SQL Server") [MongoDB](https://flowrunner.ai/integrations/mongodb "MongoDB") [Pinecone](https://flowrunner.ai/integrations/pinecone "Pinecone") [Weaviate](https://flowrunner.ai/integrations/weaviate "Weaviate") [Redis](https://flowrunner.ai/integrations/redis "Redis") [Databricks](https://flowrunner.ai/integrations/databricks "Databricks") [Databases & Warehouses 54 integrations](https://flowrunner.ai/integrations/category/databases-warehouses) [Vector Stores & AI Infra 10 integrations](https://flowrunner.ai/integrations/category/vector-stores-ai-infra) [Analytics & Data 91 integrations](https://flowrunner.ai/integrations/category/analytics-data) Every connector verified against each vendor's official API. --- ## FlowRunner for developers Source: https://flowrunner.ai/teams/developers Ship agents and LLM apps on a platform that gives you every model provider, every vector store, and the tool library your agent needs, with a human gate on the moves that publish, write to production, or cost real money. [Start free](https://app.flowrunner.ai) See the stack The verified stack [Agent.ai](https://flowrunner.ai/integrations/agent-ai "Agent.ai") [AI Image Generator](https://flowrunner.ai/integrations/ai-image-generator "AI Image Generator") [AI Vision](https://flowrunner.ai/integrations/ai-vision "AI Vision") [AI/ML API](https://flowrunner.ai/integrations/aimlapi "AI/ML API") [AI21 Labs](https://flowrunner.ai/integrations/ai21-labs "AI21 Labs") [Airtop](https://flowrunner.ai/integrations/airtop "Airtop") [AIVOOV](https://flowrunner.ai/integrations/aivoov "AIVOOV") [Algebras AI](https://flowrunner.ai/integrations/algebras "Algebras AI") [All-Images.ai](https://flowrunner.ai/integrations/all-images-ai "All-Images.ai") [AltTextLab](https://flowrunner.ai/integrations/alttextlab "AltTextLab") [Anthropic Claude](https://flowrunner.ai/integrations/anthropic-ai "Anthropic Claude") [APIFRAME](https://flowrunner.ai/integrations/apiframe "APIFRAME") [AssemblyAI](https://flowrunner.ai/integrations/assembly-ai "AssemblyAI") [AutoContent API](https://flowrunner.ai/integrations/auto-content-api "AutoContent API") [AvatarTalk AI](https://flowrunner.ai/integrations/avatartalk-ai "AvatarTalk AI") [AWS Bedrock](https://flowrunner.ai/integrations/aws-bedrock "AWS Bedrock") [AWS Comprehend](https://flowrunner.ai/integrations/aws-comprehend "AWS Comprehend") [AWS Rekognition](https://flowrunner.ai/integrations/aws-rekognition "AWS Rekognition") [AWS Textract](https://flowrunner.ai/integrations/aws-textract "AWS Textract") [AWS Transcribe](https://flowrunner.ai/integrations/aws-transcribe "AWS Transcribe") [Azure AI Foundry](https://flowrunner.ai/integrations/azure-ai-foundry "Azure AI Foundry") [Azure AI Search](https://flowrunner.ai/integrations/azure-ai-search "Azure AI Search") [Azure OpenAI](https://flowrunner.ai/integrations/azure-openai "Azure OpenAI") [Beatoven.ai](https://flowrunner.ai/integrations/beatoven-ai "Beatoven.ai") [BigML](https://flowrunner.ai/integrations/bigml "BigML") [Bookoly](https://flowrunner.ai/integrations/bookoly "Bookoly") [Botpress](https://flowrunner.ai/integrations/botpress "Botpress") [Cerebras](https://flowrunner.ai/integrations/cerebras-ai "Cerebras") [Chat Data](https://flowrunner.ai/integrations/chatdata "Chat Data") [Chatbase](https://flowrunner.ai/integrations/chatbase "Chatbase") [ChatBot](https://flowrunner.ai/integrations/chatbot "ChatBot") [ChatNode](https://flowrunner.ai/integrations/chatnode "ChatNode") [Chatsistant](https://flowrunner.ai/integrations/chatsistant "Chatsistant") [Chatsonic](https://flowrunner.ai/integrations/chatsonic "Chatsonic") [Chatvolt AI](https://flowrunner.ai/integrations/chatvolt-ai "Chatvolt AI") [CherryIN](https://flowrunner.ai/integrations/cherryin "CherryIN") [Chroma](https://flowrunner.ai/integrations/chroma "Chroma") [Clarifai](https://flowrunner.ai/integrations/clarifai "Clarifai") [Clipdrop](https://flowrunner.ai/integrations/clipdrop "Clipdrop") [Cohere](https://flowrunner.ai/integrations/cohere "Cohere") [CometAPI](https://flowrunner.ai/integrations/cometapi "CometAPI") [Copy.ai](https://flowrunner.ai/integrations/copy-ai "Copy.ai") [Coveo](https://flowrunner.ai/integrations/coveo "Coveo") [Coze](https://flowrunner.ai/integrations/coze "Coze") [Creatify](https://flowrunner.ai/integrations/creatify-ai "Creatify") [DataRobot](https://flowrunner.ai/integrations/datarobot "DataRobot") [Deep-Image.ai](https://flowrunner.ai/integrations/deep-image-ai "Deep-Image.ai") [DeepAI](https://flowrunner.ai/integrations/deep-ai "DeepAI") [Deepgram](https://flowrunner.ai/integrations/deepgram "Deepgram") [DeepInfra](https://flowrunner.ai/integrations/deepinfra "DeepInfra") [DeepL](https://flowrunner.ai/integrations/deepl "DeepL") [DeepSeek](https://flowrunner.ai/integrations/deepseek "DeepSeek") [Detecting-AI](https://flowrunner.ai/integrations/detecting-ai "Detecting-AI") [DeutschlandGPT](https://flowrunner.ai/integrations/deutschlandgpt "DeutschlandGPT") [Dialogflow](https://flowrunner.ai/integrations/google-cloud-dialogflow "Dialogflow") [DialoX](https://flowrunner.ai/integrations/dialox "DialoX") [Diffbot](https://flowrunner.ai/integrations/diffbot "Diffbot") [Dify](https://flowrunner.ai/integrations/dify "Dify") [DonnaJames](https://flowrunner.ai/integrations/donnajames "DonnaJames") [Doubao](https://flowrunner.ai/integrations/doubao "Doubao") [Dumpling AI](https://flowrunner.ai/integrations/dumplingai "Dumpling AI") [Dust](https://flowrunner.ai/integrations/dust "Dust") [E2B](https://flowrunner.ai/integrations/e2b "E2B") [Eden AI](https://flowrunner.ai/integrations/edenai "Eden AI") [ElevenLabs](https://flowrunner.ai/integrations/elevenlabs "ElevenLabs") [Exa](https://flowrunner.ai/integrations/exa-ai "Exa") [fal.ai](https://flowrunner.ai/integrations/fal-ai "fal.ai") [Firecrawl](https://flowrunner.ai/integrations/firecrawl "Firecrawl") [Fliki](https://flowrunner.ai/integrations/fliki "Fliki") [Freepik](https://flowrunner.ai/integrations/freepik "Freepik") [Frosty AI](https://flowrunner.ai/integrations/frosty-ai "Frosty AI") [Gemini AI](https://flowrunner.ai/integrations/gemini-ai "Gemini AI") [GetTranscribe](https://flowrunner.ai/integrations/gettranscribe "GetTranscribe") [Google Cloud Natural Language](https://flowrunner.ai/integrations/google-natural-language "Google Cloud Natural Language") [Google Cloud Speech-to-Text](https://flowrunner.ai/integrations/google-cloud-speech "Google Cloud Speech-to-Text") [Google Cloud Text-to-Speech](https://flowrunner.ai/integrations/google-cloud-tts "Google Cloud Text-to-Speech") [Google Cloud Vision](https://flowrunner.ai/integrations/googlecloudvision "Google Cloud Vision") [Google Translate](https://flowrunner.ai/integrations/google-translate "Google Translate") [Google Vertex AI](https://flowrunner.ai/integrations/google-vertex-ai "Google Vertex AI") [GPT Maker](https://flowrunner.ai/integrations/gpt-maker "GPT Maker") [GPT-trainer](https://flowrunner.ai/integrations/gpt-trainer "GPT-trainer") [Groq](https://flowrunner.ai/integrations/groq "Groq") [HeyGen](https://flowrunner.ai/integrations/heygen "HeyGen") [Hugging Face](https://flowrunner.ai/integrations/huggingface "Hugging Face") [IBM watsonx.ai](https://flowrunner.ai/integrations/watsonx-ai "IBM watsonx.ai") [Ideogram](https://flowrunner.ai/integrations/ideogram "Ideogram") [Jasper](https://flowrunner.ai/integrations/jasper-ai "Jasper") [Jina AI](https://flowrunner.ai/integrations/jina "Jina AI") [JoggAI](https://flowrunner.ai/integrations/joggai "JoggAI") [Kimi](https://flowrunner.ai/integrations/kimi "Kimi") [Langdock](https://flowrunner.ai/integrations/langdock "Langdock") [Language Weaver](https://flowrunner.ai/integrations/language-weaver "Language Weaver") [Leafy Plant](https://flowrunner.ai/integrations/leafy-plant "Leafy Plant") [Legnext.ai](https://flowrunner.ai/integrations/legnext "Legnext.ai") [Leonardo.Ai](https://flowrunner.ai/integrations/leonardo-ai "Leonardo.Ai") [LingvaNex](https://flowrunner.ai/integrations/lingvanex "LingvaNex") [Linkup](https://flowrunner.ai/integrations/linkup "Linkup") [LiveChatAI](https://flowrunner.ai/integrations/live-chat-ai "LiveChatAI") [Luma AI](https://flowrunner.ai/integrations/luma-ai "Luma AI") [Maybe Your AI](https://flowrunner.ai/integrations/your-ai "Maybe Your AI") [Mem0](https://flowrunner.ai/integrations/mem0 "Mem0") [Meta Llama](https://flowrunner.ai/integrations/llama "Meta Llama") [Milvus](https://flowrunner.ai/integrations/milvus "Milvus") [Mindee](https://flowrunner.ai/integrations/mindee "Mindee") [MindStudio](https://flowrunner.ai/integrations/mindstudio-ai "MindStudio") [Mistral AI](https://flowrunner.ai/integrations/mistral-ai "Mistral AI") [Murf AI](https://flowrunner.ai/integrations/murf-ai "Murf AI") [MusicGPT](https://flowrunner.ai/integrations/musicgpt "MusicGPT") [nele.ai](https://flowrunner.ai/integrations/nele-ai "nele.ai") [Nero AI](https://flowrunner.ai/integrations/nero-ai "Nero AI") [NVIDIA NIM](https://flowrunner.ai/integrations/nvidia "NVIDIA NIM") [Officely](https://flowrunner.ai/integrations/officely "Officely") [Ollama](https://flowrunner.ai/integrations/ollama "Ollama") [OpenAI](https://flowrunner.ai/integrations/openai-ai "OpenAI") [OpenRouter](https://flowrunner.ai/integrations/openrouter "OpenRouter") [Orsay AI](https://flowrunner.ai/integrations/orsay "Orsay AI") [Perplexity](https://flowrunner.ai/integrations/perplexity "Perplexity") [Perspective](https://flowrunner.ai/integrations/perspective "Perspective") [PGVector](https://flowrunner.ai/integrations/pgvector "PGVector") [PhotoRoom](https://flowrunner.ai/integrations/photoroom "PhotoRoom") [PiAPI](https://flowrunner.ai/integrations/piapi "PiAPI") [Picsart](https://flowrunner.ai/integrations/picsart "Picsart") [Pictory](https://flowrunner.ai/integrations/pictory "Pictory") [Pinecone](https://flowrunner.ai/integrations/pinecone "Pinecone") [PlayHT](https://flowrunner.ai/integrations/playht "PlayHT") [Productify.ai](https://flowrunner.ai/integrations/productify-ai "Productify.ai") [Qdrant](https://flowrunner.ai/integrations/qdrant "Qdrant") [Qwen](https://flowrunner.ai/integrations/qwen-ai "Qwen") [Relevance AI](https://flowrunner.ai/integrations/relevance "Relevance AI") [remove.bg](https://flowrunner.ai/integrations/removebg "remove.bg") [Retell AI](https://flowrunner.ai/integrations/retell-ai "Retell AI") [Romulus](https://flowrunner.ai/integrations/romulus "Romulus") [Runware](https://flowrunner.ai/integrations/runware-ai "Runware") [Runway](https://flowrunner.ai/integrations/runway "Runway") [SambaNova](https://flowrunner.ai/integrations/sambanova "SambaNova") [ScrapeGraphAI](https://flowrunner.ai/integrations/scrapegraphai "ScrapeGraphAI") [SectorFlow](https://flowrunner.ai/integrations/sectorflow "SectorFlow") [Segmind](https://flowrunner.ai/integrations/segmind "Segmind") [ShareAI](https://flowrunner.ai/integrations/share-ai "ShareAI") [Simpleen](https://flowrunner.ai/integrations/simpleen-translation "Simpleen") [Snack Prompt](https://flowrunner.ai/integrations/snack-prompt "Snack Prompt") [Speechmatics](https://flowrunner.ai/integrations/speechmatics "Speechmatics") [Stability AI](https://flowrunner.ai/integrations/stability-ai "Stability AI") [StealthGPT](https://flowrunner.ai/integrations/stealthgpt "StealthGPT") [Synthesia](https://flowrunner.ai/integrations/synthesia "Synthesia") [Synthflow AI](https://flowrunner.ai/integrations/synthflow-ai "Synthflow AI") [Tavily](https://flowrunner.ai/integrations/tavily "Tavily") [Temi](https://flowrunner.ai/integrations/temi "Temi") [Tess AI](https://flowrunner.ai/integrations/tess-ai "Tess AI") [TextCortex](https://flowrunner.ai/integrations/textcortex-ai "TextCortex") [Tisane](https://flowrunner.ai/integrations/tisane "Tisane") [Together AI](https://flowrunner.ai/integrations/together-ai "Together AI") [Trint](https://flowrunner.ai/integrations/trint "Trint") [Typecast](https://flowrunner.ai/integrations/typecast "Typecast") [uClassify](https://flowrunner.ai/integrations/uclassify "uClassify") [URL to Text](https://flowrunner.ai/integrations/urltotext "URL to Text") [Vapi](https://flowrunner.ai/integrations/vapi "Vapi") [Vercel AI Gateway](https://flowrunner.ai/integrations/vercel-ai-gateway "Vercel AI Gateway") [Weaviate](https://flowrunner.ai/integrations/weaviate "Weaviate") [WizyChat](https://flowrunner.ai/integrations/wizychat "WizyChat") [xAI Grok](https://flowrunner.ai/integrations/xai-grok "xAI Grok") [You.com](https://flowrunner.ai/integrations/you "You.com") [YourGPT Chatbot](https://flowrunner.ai/integrations/yourgpt-chatbot "YourGPT Chatbot") [Zep](https://flowrunner.ai/integrations/zep "Zep") 164 integrations 2,402 verified actions Verified against each vendor's official API ### The gap between a demo and production Your prototype works. The agent retrieves, reasons, and answers. Then you try to ship it and the real work appears: routing across providers so one outage does not take you down, keeping an eval and a guardrail between the model and the user, catching the low-confidence extraction before it becomes a wrong fact in a customer record. The model was the easy part. The orchestration around it is where the weeks go. You already know your stack. OpenAI, Anthropic Claude, Gemini, a vector store, an embeddings model, an OCR service. The problem is the plumbing between them, and the fact that the consequential steps, publishing generated content, letting a tool call write to a production system, kicking off a batch job that burns tokens, deleting a namespace, are exactly the steps you cannot hand to an autonomous loop that does not know when to stop. - The demo works; the production plumbing takes the weeks. - One provider outage takes your whole app down. - A low-confidence extraction gets written as if it were true. - Generated content ships with no one reviewing it first. - An agent tool call writes to production with nothing watching. ### What you can build Automations Each of these runs as a flow. The agent does the retrieval, the routing, and the routine steps on its own. It stops and asks a human at the one point where publishing, a production write, a real bill, or a wrong fact is expensive or irreversible. Every tool named here is a FlowRunner connector, built and verified against the provider's official API. 1. ### Grounded RAG over your own sources Answers come cited and grounded. The reindex of your knowledge base waits for a human. 1. Trigger A question hits your app, or new source documents land in the pipeline. 2. Agent The agent chunks and embeds the source with Cohere or OpenAI embeddings, upserts the vectors into Pinecone, retrieves the top matches for the query, and passes the grounded context to Anthropic Claude for an answer that cites what it used. 3. Result Users get answers grounded in your own sources with citations, and the index that everything depends on never gets rebuilt unsupervised. Run 4400 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/cohere "Cohere")[](https://flowrunner.ai/integrations/pinecone "Pinecone")[](https://flowrunner.ai/integrations/anthropic-ai "Anthropic Claude")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Read-only retrieval and answering run on their own. Rebuilding the index or overwriting a production namespace is destructive and hard to undo, so the agent stages the reindex and waits for an engineer to confirm before it replaces the live vectors. 2. ### Document extraction into structured data Clean documents post themselves. The low-confidence read waits for a person. 1. Trigger An invoice, form, or contract arrives by email, upload, or into a watched bucket. 2. Agent The agent runs the document through AWS Textract for layout and Mindee for typed fields, reconciles the two, and maps the result to your schema with a confidence score on every field. 3. Result Clean documents flow straight into structured data, and the ambiguous read gets a human eye before it becomes a wrong number someone trusts downstream. Run 4437 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/aws-textract "AWS Textract")[](https://flowrunner.ai/integrations/mindee "Mindee")[](https://flowrunner.ai/integrations/aws-comprehend "AWS Comprehend")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint When every field clears the confidence bar you set, it writes the record and moves on. When a total, an account number, or a date comes back low-confidence, it stops and routes the page plus the extracted values to a human, rather than writing a guess into the system of record as if it were fact. 3. ### Multi-provider routing and fallback Requests route to the right model. The expensive batch run waits for a go-ahead. 1. Trigger A request enters your app, or a scheduled batch of work is ready to process. 2. Agent The agent routes each request by task: cheap and fast to Groq, long-context reasoning to Anthropic Claude, and it fails over through OpenRouter the moment a provider errors or rate-limits, so a single outage never takes the app down. 3. Result Traffic always reaches a working model at the right price and latency, and the runs that cost real money have a named human behind the start button. Run 4474 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/groq "Groq")[](https://flowrunner.ai/integrations/anthropic-ai "Anthropic Claude")[](https://flowrunner.ai/integrations/openrouter "OpenRouter")[](https://flowrunner.ai/integrations/openai-ai "OpenAI")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Live traffic routes and fails over automatically. A large batch reprocess, the kind that burns real tokens across thousands of items, stops and shows an engineer the estimated scope and cost before it kicks off, so no run surprises you on the bill. 4. ### Content and media generation with a publish gate Drafts and assets generate on demand. Nothing ships to the public without a sign-off. 1. Trigger A content request comes in, or a scheduled generation job fires. 2. Agent The agent drafts the copy with OpenAI, generates the accompanying visual with AI Image Generator, produces a voiceover with ElevenLabs, and assembles the finished asset ready to review. 3. Result Content and media get produced at speed, and a person always owns the moment anything generated reaches the public. Run 4511 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/openai-ai "OpenAI")[](https://flowrunner.ai/integrations/ai-image-generator "AI Image Generator")[](https://flowrunner.ai/integrations/elevenlabs "ElevenLabs")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Generation runs on its own. Publishing to an audience is public and hard to unsay, so the agent stops and hands the full package to a human, who approves, edits, or rejects before a single asset goes out under your brand. 5. ### Agent memory with an eval and guardrail step The agent remembers and gets checked. The write that changes memory or fails a check waits. 1. Trigger An agent finishes a turn, or a response is ready to return to the user. 2. Agent The agent stores and recalls long-term context in Zep, screens each candidate output through Perspective for toxicity, and runs an OpenAI eval pass to score the response against your rubric before anything is returned. 3. Result The agent carries context across sessions and every output passes a guardrail, with a person owning the writes to memory and the outputs that fail the bar. Run 4548 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/zep "Zep")[](https://flowrunner.ai/integrations/perspective "Perspective")[](https://flowrunner.ai/integrations/openai-ai "OpenAI")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Responses that clear the eval and the guardrail go back to the user. When a response fails a check, or when the agent wants to commit a durable fact to long-term memory that will shape every future answer, it stops and routes to a human rather than trusting itself unsupervised. ### The pattern that makes it safe to ship AI apps fail in production when an autonomous loop does something it should not: publishes a hallucination, writes a low-confidence guess as truth, deletes an index, or burns a fortune in tokens on a runaway batch. FlowRunner's answer is the digital andon cord: the agent runs the line and pulls it the instant a step carries real consequence. It retrieves, routes, extracts, and generates on its own. Publishing to the public, writing to a system of record, starting a high-cost run, committing to long-term memory, or dropping a vector namespace always stops and routes to a person through the channel they already watch, with everything the agent produced attached. Read-only work happens immediately; the consequential write waits for a human. That is the difference between a demo and an agent you trust in production. [The digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord) ### What your team gets Outcomes #### Skip the orchestration plumbing Routing, fallback, retrieval, and eval steps are flow components you compose, not glue code you write and maintain between every provider and store. #### No single point of failure Route and fail over across providers, so one vendor outage or rate limit never takes your whole app down with it. #### Wrong facts get caught first A low-confidence extraction or a failed eval stops for a human instead of flowing into your system of record as if it were true. #### Nothing ships unreviewed Generated content and media always pass a human before they reach the public, so the model never publishes under your brand alone. #### No runaway costs High-cost batch runs stop and show the scope before they start, so a loop never burns tokens at scale without a person behind it. #### Every provider and store, one platform The model providers, embeddings services, and vector stores you already use, each verified against its official API, so there is no tool your agent cannot reach. ### Built for developers' requirements Controls The controls you would demand of anything running your models against real data and real users are native to the platform, not add-ons. #### Bring your own keys Your model providers, your keys. Inference runs on credentials you control, you keep your provider rate limits and pricing, and the platform never sits between you and your vendor bill. #### Complete audit trail Every model call, tool call, retrieval, and human decision recorded with the actor and timestamp, so you can trace exactly what your agent did and why. #### RBAC and SSO/SAML Role-based access to the platform itself, with SSO and SAML so your identity provider stays the source of truth for who can touch which flows and keys. #### Self-hosted option Deploy inside your own infrastructure so prompts, embeddings, retrieved context, and customer data never leave your environment. > _It sounds super interesting. I'm thinking about so many scenarios._ Operations Manager, CPG brand ### Every model provider and the tools around them The stack Full-surface coverage of the providers you build on, from OpenAI, Anthropic Claude, Gemini, and AWS Bedrock to the vector stores that ground them: Pinecone, Weaviate, Qdrant, Chroma, and PGVector. Alongside them sit the tools your agent actually needs: OCR, vision, translation, transcription, embeddings, and speech. Every connector is built and verified against the provider's official API, so an agent calls it the way the provider actually allows. The tools the automations above run on. [Cohere](https://flowrunner.ai/integrations/cohere "Cohere") [Pinecone](https://flowrunner.ai/integrations/pinecone "Pinecone") [Anthropic Claude](https://flowrunner.ai/integrations/anthropic-ai "Anthropic Claude") [AWS Textract](https://flowrunner.ai/integrations/aws-textract "AWS Textract") [Mindee](https://flowrunner.ai/integrations/mindee "Mindee") [AWS Comprehend](https://flowrunner.ai/integrations/aws-comprehend "AWS Comprehend") [Groq](https://flowrunner.ai/integrations/groq "Groq") [OpenRouter](https://flowrunner.ai/integrations/openrouter "OpenRouter") [OpenAI](https://flowrunner.ai/integrations/openai-ai "OpenAI") [AI Image Generator](https://flowrunner.ai/integrations/ai-image-generator "AI Image Generator") [ElevenLabs](https://flowrunner.ai/integrations/elevenlabs "ElevenLabs") [Zep](https://flowrunner.ai/integrations/zep "Zep") [Perspective](https://flowrunner.ai/integrations/perspective "Perspective") [AI & LLMs 154 integrations](https://flowrunner.ai/integrations/category/ai-llms) [Vector Stores & AI Infra 10 integrations](https://flowrunner.ai/integrations/category/vector-stores-ai-infra) Every connector verified against each vendor's official API. --- ## FlowRunner for finance operations Source: https://flowrunner.ai/teams/finance-ops Automate the invoicing, reconciliation, and close work that swallows your month, with an agent that runs the matching and the routine and a human who owns every payment, posting, and refund that touches the ledger or leaves the building. [Start free](https://app.flowrunner.ai) See the stack The verified stack [29 Next](https://flowrunner.ai/integrations/twentyninenext "29 Next") [3dcart](https://flowrunner.ai/integrations/a3dcart "3dcart") [7Loc](https://flowrunner.ai/integrations/seven-loc "7Loc") [Abby](https://flowrunner.ai/integrations/abby "Abby") [ABRA Flexi](https://flowrunner.ai/integrations/flexibee "ABRA Flexi") [Acumatica](https://flowrunner.ai/integrations/acumatica "Acumatica") [AfterShip](https://flowrunner.ai/integrations/aftership "AfterShip") [AirMenu](https://flowrunner.ai/integrations/airmenu "AirMenu") [Alegra](https://flowrunner.ai/integrations/alegra "Alegra") [AlphaInsider](https://flowrunner.ai/integrations/alphainsider "AlphaInsider") [Altoviz](https://flowrunner.ai/integrations/altoviz "Altoviz") [Amazon Creators API](https://flowrunner.ai/integrations/amazon-pa-api-v5 "Amazon Creators API") [Amazon Seller Central](https://flowrunner.ai/integrations/amazon-seller-central "Amazon Seller Central") [AmeriCommerce (Cart.com)](https://flowrunner.ai/integrations/americommerce "AmeriCommerce (Cart.com)") [Anaplan](https://flowrunner.ai/integrations/anaplan "Anaplan") [Anchor](https://flowrunner.ai/integrations/anchor "Anchor") [Apaleo](https://flowrunner.ai/integrations/apaleo "Apaleo") [Asaas](https://flowrunner.ai/integrations/asaas "Asaas") [Askara](https://flowrunner.ai/integrations/askara "Askara") [Astara Connect](https://flowrunner.ai/integrations/astara-connect "Astara Connect") [Auchan](https://flowrunner.ai/integrations/auchan "Auchan") [Aurora Solar](https://flowrunner.ai/integrations/aurora-solar "Aurora Solar") [Authvia](https://flowrunner.ai/integrations/authvia "Authvia") [Autovit](https://flowrunner.ai/integrations/autovit "Autovit") [Avalara AvaTax](https://flowrunner.ai/integrations/avalara-avatax "Avalara AvaTax") [Axelor](https://flowrunner.ai/integrations/axelor "Axelor") [Axonaut](https://flowrunner.ai/integrations/axonaut "Axonaut") [BaseLinker](https://flowrunner.ai/integrations/baselinker "BaseLinker") [Bcon](https://flowrunner.ai/integrations/becon "Bcon") [Beds24](https://flowrunner.ai/integrations/beds24 "Beds24") [Best Buy](https://flowrunner.ai/integrations/best-buy "Best Buy") [bexio](https://flowrunner.ai/integrations/bexio "bexio") [Big Cartel](https://flowrunner.ai/integrations/big-cartel "Big Cartel") [BigCommerce](https://flowrunner.ai/integrations/bigcommerce "BigCommerce") [BILL](https://flowrunner.ai/integrations/billcom "BILL") [Billit](https://flowrunner.ai/integrations/billit "Billit") [Billomat](https://flowrunner.ai/integrations/billomat "Billomat") [Billplz](https://flowrunner.ai/integrations/billplz "Billplz") [Billsby](https://flowrunner.ai/integrations/billsby "Billsby") [Binance](https://flowrunner.ai/integrations/binance "Binance") [Bind ERP](https://flowrunner.ai/integrations/bind-erp "Bind ERP") [Biyo POS](https://flowrunner.ai/integrations/biyopos "Biyo POS") [Bleez](https://flowrunner.ai/integrations/bleez "Bleez") [Bluebarry](https://flowrunner.ai/integrations/bluebarry "Bluebarry") [Bluestone PIM](https://flowrunner.ai/integrations/bluestone-api-public "Bluestone PIM") [Bolta](https://flowrunner.ai/integrations/bolta "Bolta") [BoondManager](https://flowrunner.ai/integrations/boondmanager "BoondManager") [Booqable](https://flowrunner.ai/integrations/booqable "Booqable") [Braintree](https://flowrunner.ai/integrations/braintree "Braintree") [Brex](https://flowrunner.ai/integrations/brex "Brex") [Brightflag](https://flowrunner.ai/integrations/brightflag "Brightflag") [Cargoboard](https://flowrunner.ai/integrations/cargoboard "Cargoboard") [Cargus](https://flowrunner.ai/integrations/cargus-romania "Cargus") [Carix](https://flowrunner.ai/integrations/carix "Carix") [Carrefour](https://flowrunner.ai/integrations/carrefour "Carrefour") [Channex](https://flowrunner.ai/integrations/channex "Channex") [Chargebee](https://flowrunner.ai/integrations/chargebee "Chargebee") [ChargeOver](https://flowrunner.ai/integrations/chargeover "ChargeOver") [Checkbook.io](https://flowrunner.ai/integrations/checkbook "Checkbook.io") [Cin7 Core (DEAR)](https://flowrunner.ai/integrations/dear-inventory "Cin7 Core (DEAR)") [Cin7 Omni](https://flowrunner.ai/integrations/cin7 "Cin7 Omni") [ClickFunnels 2.0](https://flowrunner.ai/integrations/click-funnels-2 "ClickFunnels 2.0") [Clientary](https://flowrunner.ai/integrations/ronin "Clientary") [Clover](https://flowrunner.ai/integrations/clover-pos "Clover") [CNB Exchange Rates](https://flowrunner.ai/integrations/kurzy-cnb "CNB Exchange Rates") [Colete Online](https://flowrunner.ai/integrations/colete-online "Colete Online") [CommerceHQ](https://flowrunner.ai/integrations/commercehq "CommerceHQ") [coreBOS](https://flowrunner.ai/integrations/corebos "coreBOS") [Coupa](https://flowrunner.ai/integrations/coupa "Coupa") [CS-Cart](https://flowrunner.ai/integrations/cs-cart "CS-Cart") [Currency Converter](https://flowrunner.ai/integrations/currency-converter "Currency Converter") [Darujme.cz](https://flowrunner.ai/integrations/darujme-cz "Darujme.cz") [Deskera](https://flowrunner.ai/integrations/deskera "Deskera") [Detrack](https://flowrunner.ai/integrations/detrack "Detrack") [Dext](https://flowrunner.ai/integrations/dext "Dext") [DHL](https://flowrunner.ai/integrations/dhl "DHL") [Digistore24](https://flowrunner.ai/integrations/digistore "Digistore24") [Dokan](https://flowrunner.ai/integrations/dokan "Dokan") [Donorbox](https://flowrunner.ai/integrations/donorbox "Donorbox") [DPD Germany](https://flowrunner.ai/integrations/dpd-germany "DPD Germany") [DPD Romania](https://flowrunner.ai/integrations/dpd-romania "DPD Romania") [DropFunnels](https://flowrunner.ai/integrations/dropfunnels "DropFunnels") [Dynamics 365 Business Central](https://flowrunner.ai/integrations/microsoft-d365-bc "Dynamics 365 Business Central") [e-conomic](https://flowrunner.ai/integrations/e-conomic "e-conomic") [easybill](https://flowrunner.ai/integrations/easybill "easybill") [EasyCargo](https://flowrunner.ai/integrations/easy-cargo "EasyCargo") [Easypay](https://flowrunner.ai/integrations/easypay "Easypay") [EasyPost](https://flowrunner.ai/integrations/easypost "EasyPost") [Easyship](https://flowrunner.ai/integrations/easyship "Easyship") [Ecofleet](https://flowrunner.ai/integrations/ecofleet-cz "Ecofleet") [eDock](https://flowrunner.ai/integrations/edock "eDock") [EenvoudigFactureren](https://flowrunner.ai/integrations/eenvoudigfactureren "EenvoudigFactureren") [elopage](https://flowrunner.ai/integrations/elopage "elopage") [Emporix Commerce](https://flowrunner.ai/integrations/emporix-commerce "Emporix Commerce") [Emporix Orchestration Engine](https://flowrunner.ai/integrations/emporix-oe "Emporix Orchestration Engine") [ERPLY Books](https://flowrunner.ai/integrations/erply-books "ERPLY Books") [ERPNext](https://flowrunner.ai/integrations/erpnext "ERPNext") [Etsy](https://flowrunner.ai/integrations/etsy "Etsy") [EventSquare](https://flowrunner.ai/integrations/eventsquare "EventSquare") [Exact Online](https://flowrunner.ai/integrations/exact-online "Exact Online") [Expedy](https://flowrunner.ai/integrations/expedy-by-makemarket.io "Expedy") [Expensify](https://flowrunner.ai/integrations/expensify "Expensify") [EZOfficeInventory](https://flowrunner.ai/integrations/ezofficeinventory "EZOfficeInventory") [EZRentOut](https://flowrunner.ai/integrations/ezrentout "EZRentOut") [Facebook Catalogs](https://flowrunner.ai/integrations/facebook-catalogs "Facebook Catalogs") [Fakturoid](https://flowrunner.ai/integrations/fakturoid "Fakturoid") [FAPI](https://flowrunner.ai/integrations/fapi "FAPI") [FastBill](https://flowrunner.ai/integrations/fastbill "FastBill") [Fatture in Cloud](https://flowrunner.ai/integrations/fatture-in-cloud "Fatture in Cloud") [Fidoo](https://flowrunner.ai/integrations/fidoo-expense-management "Fidoo") [finaX](https://flowrunner.ai/integrations/finax "finaX") [Fio banka](https://flowrunner.ai/integrations/fio "Fio banka") [Flatfox](https://flowrunner.ai/integrations/flatfox "Flatfox") [Flow Blockchain](https://flowrunner.ai/integrations/flow-blockchain "Flow Blockchain") [Fortnox](https://flowrunner.ai/integrations/fortnox "Fortnox") [FreeAgent](https://flowrunner.ai/integrations/freeagent "FreeAgent") [FreeFinance](https://flowrunner.ai/integrations/free-finance "FreeFinance") [FreshBooks](https://flowrunner.ai/integrations/freshbooks "FreshBooks") [FunnelCockpit](https://flowrunner.ai/integrations/funnelcockpit "FunnelCockpit") [GetMyInvoices](https://flowrunner.ai/integrations/getmyinvoices "GetMyInvoices") [Global Exchange Rates](https://flowrunner.ai/integrations/global-exchange-rates "Global Exchange Rates") [GoCardless](https://flowrunner.ai/integrations/gocardless "GoCardless") [GoodBarber](https://flowrunner.ai/integrations/goodbarber "GoodBarber") [Google Merchant Center](https://flowrunner.ai/integrations/google-shopping "Google Merchant Center") [GoPay](https://flowrunner.ai/integrations/gopay "GoPay") [GP webpay](https://flowrunner.ai/integrations/webpay "GP webpay") [Gumroad](https://flowrunner.ai/integrations/gumroad "Gumroad") [HelloAsso](https://flowrunner.ai/integrations/helloasso "HelloAsso") [HERO Software](https://flowrunner.ai/integrations/hero-software "HERO Software") [HitPay](https://flowrunner.ai/integrations/hitpay "HitPay") [Holded](https://flowrunner.ai/integrations/holded "Holded") [HomeRunner](https://flowrunner.ai/integrations/homerunner "HomeRunner") [Hotmart](https://flowrunner.ai/integrations/hotmart "Hotmart") [iCount](https://flowrunner.ai/integrations/icount "iCount") [iDoklad](https://flowrunner.ai/integrations/idoklad "iDoklad") [Infor M3](https://flowrunner.ai/integrations/infor-m3 "Infor M3") [Instamojo](https://flowrunner.ai/integrations/instamojo "Instamojo") [IntegrityNext](https://flowrunner.ai/integrations/integritynext "IntegrityNext") [Interhyp Gruppe](https://flowrunner.ai/integrations/interhyp-gruppe "Interhyp Gruppe") [Invoice Ninja](https://flowrunner.ai/integrations/invoice-ninja "Invoice Ninja") [Invoiced](https://flowrunner.ai/integrations/invoiced "Invoiced") [Invoicing Plus](https://flowrunner.ai/integrations/invoicing-plus "Invoicing Plus") [iUcto](https://flowrunner.ai/integrations/iucto "iUcto") [Judge.me](https://flowrunner.ai/integrations/judge-me "Judge.me") [KashFlow](https://flowrunner.ai/integrations/kashflow "KashFlow") [Katana](https://flowrunner.ai/integrations/katana-mrp "Katana") [Komercni banka](https://flowrunner.ai/integrations/komercni-banka "Komercni banka") [konfipay](https://flowrunner.ai/integrations/konfipay "konfipay") [Layerise](https://flowrunner.ai/integrations/layerise "Layerise") [Lemon Squeezy](https://flowrunner.ai/integrations/lemon-squeezy "Lemon Squeezy") [Leroy Merlin](https://flowrunner.ai/integrations/leroy-merlin "Leroy Merlin") [lexoffice](https://flowrunner.ai/integrations/lexoffice "lexoffice") [Lightspeed eCom](https://flowrunner.ai/integrations/lightspeed-ecom "Lightspeed eCom") [Lightspeed Retail X-Series](https://flowrunner.ai/integrations/vend "Lightspeed Retail X-Series") [Limble CMMS](https://flowrunner.ai/integrations/limble-cmms "Limble CMMS") [Linx Commerce](https://flowrunner.ai/integrations/linx-commerce "Linx Commerce") [LionWheel](https://flowrunner.ai/integrations/lionwheel-delivery "LionWheel") [Loyverse](https://flowrunner.ai/integrations/loyverse "Loyverse") [Magento 2](https://flowrunner.ai/integrations/magento "Magento 2") [Magic Eden](https://flowrunner.ai/integrations/magic-eden "Magic Eden") [MaintainX](https://flowrunner.ai/integrations/maintainx "MaintainX") [Mamo](https://flowrunner.ai/integrations/mamo "Mamo") [Mangopay](https://flowrunner.ai/integrations/mangopay "Mangopay") [Memberful](https://flowrunner.ai/integrations/memberful "Memberful") [MemberPress](https://flowrunner.ai/integrations/memberpress "MemberPress") [MFR (Mobile Field Report)](https://flowrunner.ai/integrations/mfr-field-service-management "MFR (Mobile Field Report)") [Microsoft Dynamics 365](https://flowrunner.ai/integrations/dynamics-365 "Microsoft Dynamics 365") [Mirakl](https://flowrunner.ai/integrations/mirakl "Mirakl") [Mollie](https://flowrunner.ai/integrations/mollie "Mollie") [Moneybird](https://flowrunner.ai/integrations/moneybird "Moneybird") [Moon Invoice](https://flowrunner.ai/integrations/moon-invoice "Moon Invoice") [Morning](https://flowrunner.ai/integrations/morning "Morning") [MyCarTracks](https://flowrunner.ai/integrations/mycartracks "MyCarTracks") [Navan](https://flowrunner.ai/integrations/navan "Navan") [Neto](https://flowrunner.ai/integrations/neto "Neto") [NetSuite](https://flowrunner.ai/integrations/netsuite "NetSuite") [OANDA Exchange Rates](https://flowrunner.ai/integrations/oanda-exchange-rates "OANDA Exchange Rates") [Oblio](https://flowrunner.ai/integrations/oblio.eu-facturare "Oblio") [Odoo](https://flowrunner.ai/integrations/odoo "Odoo") [Odyssee Field Service](https://flowrunner.ai/integrations/odyssee-field-service "Odyssee Field Service") [OfficeGuy](https://flowrunner.ai/integrations/officeguy "OfficeGuy") [OfficeRnD](https://flowrunner.ai/integrations/officernd "OfficeRnD") [OLX](https://flowrunner.ai/integrations/olx "OLX") [Onfleet](https://flowrunner.ai/integrations/onfleet "Onfleet") [OnlineCheckWriter](https://flowrunner.ai/integrations/onlinecheckwriter "OnlineCheckWriter") [OpenCart](https://flowrunner.ai/integrations/opencart "OpenCart") [openrouteservice](https://flowrunner.ai/integrations/openrouteservice "openrouteservice") [Opn Payments](https://flowrunner.ai/integrations/omise "Opn Payments") [Oracle Fusion Cloud ERP](https://flowrunner.ai/integrations/oracle-fusion-cloud-erp "Oracle Fusion Cloud ERP") [Order Desk](https://flowrunner.ai/integrations/order-desk "Order Desk") [Orderry](https://flowrunner.ai/integrations/orderry "Orderry") [OxaPay](https://flowrunner.ai/integrations/oxapay-crypto-pay-gtw "OxaPay") [Paddle](https://flowrunner.ai/integrations/paddle "Paddle") [Pagar.me](https://flowrunner.ai/integrations/pagar-me "Pagar.me") [PayFunnels](https://flowrunner.ai/integrations/payfunnels "PayFunnels") [PayPal](https://flowrunner.ai/integrations/paypal "PayPal") [Payrexx](https://flowrunner.ai/integrations/payrexx "Payrexx") [Payrexx Platforms](https://flowrunner.ai/integrations/payrexx-platforms "Payrexx Platforms") [Paystack](https://flowrunner.ai/integrations/paystack "Paystack") [Pennylane](https://flowrunner.ai/integrations/pennylane-v2 "Pennylane") [Peppol E-Invoicing](https://flowrunner.ai/integrations/peppol-e-invoicing "Peppol E-Invoicing") [Persat](https://flowrunner.ai/integrations/persat "Persat") [PHP Point Of Sale](https://flowrunner.ai/integrations/php-point-of-sale "PHP Point Of Sale") [Picqer](https://flowrunner.ai/integrations/picqer-debesis "Picqer") [Pigment](https://flowrunner.ai/integrations/pigment "Pigment") [Pixfizz](https://flowrunner.ai/integrations/pixfizz "Pixfizz") [Plentific](https://flowrunner.ai/integrations/plentific "Plentific") [PlentyONE](https://flowrunner.ai/integrations/plentymarkets "PlentyONE") [PowerOffice Go](https://flowrunner.ai/integrations/poweroffice "PowerOffice Go") [PrestaShop](https://flowrunner.ai/integrations/prestashop "PrestaShop") [Printavo](https://flowrunner.ai/integrations/printavo "Printavo") [Printful](https://flowrunner.ai/integrations/printful "Printful") [Printify](https://flowrunner.ai/integrations/printify "Printify") [ProAbono](https://flowrunner.ai/integrations/proabono "ProAbono") [Procountor](https://flowrunner.ai/integrations/procountor "Procountor") [ProfitWell](https://flowrunner.ai/integrations/profitwell "ProfitWell") [Qapla](https://flowrunner.ai/integrations/qapla "Qapla") [Qonto](https://flowrunner.ai/integrations/qonto "Qonto") [Quable PIM](https://flowrunner.ai/integrations/quable-pim "Quable PIM") [Quaderno](https://flowrunner.ai/integrations/quaderno "Quaderno") [Quant Overledger](https://flowrunner.ai/integrations/overledger "Quant Overledger") [QuickBooks Online](https://flowrunner.ai/integrations/quickbooks-online "QuickBooks Online") [QuickFile](https://flowrunner.ai/integrations/quickfile "QuickFile") [Quipu](https://flowrunner.ai/integrations/quipu "Quipu") [Ramp](https://flowrunner.ai/integrations/ramp-service "Ramp") [Razorpay](https://flowrunner.ai/integrations/razorpay "Razorpay") [Rebill](https://flowrunner.ai/integrations/rebill "Rebill") [Recharge](https://flowrunner.ai/integrations/recharge "Recharge") [Recommand](https://flowrunner.ai/integrations/recommand "Recommand") [Recurly](https://flowrunner.ai/integrations/recurly "Recurly") [remberg](https://flowrunner.ai/integrations/remberg-de "remberg") [Remote Retrieval](https://flowrunner.ai/integrations/remote-retrieval "Remote Retrieval") [Rentman](https://flowrunner.ai/integrations/rentman "Rentman") [RepairShopr](https://flowrunner.ai/integrations/repairshopr "RepairShopr") [Revolut Business](https://flowrunner.ai/integrations/revolut-business "Revolut Business") [Sage Accounting](https://flowrunner.ai/integrations/sage-accounting "Sage Accounting") [Sage Intacct](https://flowrunner.ai/integrations/sage-intacct "Sage Intacct") [Salla](https://flowrunner.ai/integrations/salla "Salla") [SamCart](https://flowrunner.ai/integrations/samcart "SamCart") [Sameday](https://flowrunner.ai/integrations/sameday-curier-ro "Sameday") [SAP Business One](https://flowrunner.ai/integrations/sap-business-one "SAP Business One") [SAP S/4HANA](https://flowrunner.ai/integrations/sap-s4hana "SAP S/4HANA") [ScanOrders](https://flowrunner.ai/integrations/scanorders "ScanOrders") [Scarf](https://flowrunner.ai/integrations/scarf-club "Scarf") [Seller Assistant](https://flowrunner.ai/integrations/seller-assistant "Seller Assistant") [Sellercloud](https://flowrunner.ai/integrations/sellercloud "Sellercloud") [SellIntegro CloudPrint](https://flowrunner.ai/integrations/sellintegro-cloud-print "SellIntegro CloudPrint") [Sendcloud](https://flowrunner.ai/integrations/sendcloud "Sendcloud") [Sendle](https://flowrunner.ai/integrations/sendle "Sendle") [SendOwl](https://flowrunner.ai/integrations/sendowl "SendOwl") [ServiceM8](https://flowrunner.ai/integrations/servicem8 "ServiceM8") [ServiceTitan](https://flowrunner.ai/integrations/service-titan "ServiceTitan") [sevdesk](https://flowrunner.ai/integrations/sevdesk "sevdesk") [Seven Senders](https://flowrunner.ai/integrations/sevensenders "Seven Senders") [Ship24](https://flowrunner.ai/integrations/ship24 "Ship24") [ShipBob](https://flowrunner.ai/integrations/shipbob "ShipBob") [shipcloud](https://flowrunner.ai/integrations/shipcloud "shipcloud") [Shipday](https://flowrunner.ai/integrations/shipday "Shipday") [ShipHero](https://flowrunner.ai/integrations/shiphero "ShipHero") [Shippingbo](https://flowrunner.ai/integrations/shippingbo "Shippingbo") [ShippingEasy](https://flowrunner.ai/integrations/shippingeasy "ShippingEasy") [Shippo](https://flowrunner.ai/integrations/shippo "Shippo") [ShipStation](https://flowrunner.ai/integrations/shipstation "ShipStation") [Shop Apotheke](https://flowrunner.ai/integrations/shop-apotheke "Shop Apotheke") [Shopify](https://flowrunner.ai/integrations/shopify "Shopify") [Shoprocket](https://flowrunner.ai/integrations/shoprocket "Shoprocket") [Shuttle](https://flowrunner.ai/integrations/shuttle-payment-links "Shuttle") [SimpleCirc](https://flowrunner.ai/integrations/simplecirc "SimpleCirc") [Simplero](https://flowrunner.ai/integrations/simplero "Simplero") [SimpleShop](https://flowrunner.ai/integrations/simpleshop-cz2 "SimpleShop") [SingleCase](https://flowrunner.ai/integrations/singlecase "SingleCase") [SmartBill](https://flowrunner.ai/integrations/smartbill "SmartBill") [SnelStart](https://flowrunner.ai/integrations/snelstart "SnelStart") [Snipcart](https://flowrunner.ai/integrations/snipcart "Snipcart") [Solana](https://flowrunner.ai/integrations/solana "Solana") [SOS Inventory](https://flowrunner.ai/integrations/sos-inventory "SOS Inventory") [Splitwise](https://flowrunner.ai/integrations/splitwise "Splitwise") [Squad](https://flowrunner.ai/integrations/squad "Squad") [Square](https://flowrunner.ai/integrations/square "Square") [Squarespace](https://flowrunner.ai/integrations/squarespace "Squarespace") [Stamped](https://flowrunner.ai/integrations/stamped "Stamped") [STEL Order](https://flowrunner.ai/integrations/stel-order "STEL Order") [Storeman](https://flowrunner.ai/integrations/storeman "Storeman") [Stripe](https://flowrunner.ai/integrations/stripe "Stripe") [Super Manager](https://flowrunner.ai/integrations/super-manager "Super Manager") [SuperFaktura](https://flowrunner.ai/integrations/superfaktura "SuperFaktura") [SuperHote](https://flowrunner.ai/integrations/superhote "SuperHote") [SureCart](https://flowrunner.ai/integrations/surecart "SureCart") [System Obsługi Najmu](https://flowrunner.ai/integrations/system-obslugi-najmu "System Obsługi Najmu") [Tamio (Plug&Paid)](https://flowrunner.ai/integrations/plug-paid "Tamio (Plug&Paid)") [Thiio](https://flowrunner.ai/integrations/thiio "Thiio") [ThriveCart](https://flowrunner.ai/integrations/thrivecart "ThriveCart") [Tidely](https://flowrunner.ai/integrations/tidely "Tidely") [Tookan](https://flowrunner.ai/integrations/tookan "Tookan") [Tranzila](https://flowrunner.ai/integrations/tranzila "Tranzila") [Tremendous](https://flowrunner.ai/integrations/tremendous "Tremendous") [Tripletex](https://flowrunner.ai/integrations/tripletex "Tripletex") [Trolley](https://flowrunner.ai/integrations/payment-rails "Trolley") [UnionBank](https://flowrunner.ai/integrations/unionbank "UnionBank") [Unleashed Software](https://flowrunner.ai/integrations/unleashed "Unleashed Software") [Upgrade.Chat](https://flowrunner.ai/integrations/upgradechat "Upgrade.Chat") [UPS Quantum View](https://flowrunner.ai/integrations/ups-quantum-view "UPS Quantum View") [Visma eAccounting](https://flowrunner.ai/integrations/visma-eaccounting "Visma eAccounting") [Visma.net ERP](https://flowrunner.ai/integrations/visma-net-erp "Visma.net ERP") [VosFactures](https://flowrunner.ai/integrations/vosfactures "VosFactures") [VTEX](https://flowrunner.ai/integrations/vtex "VTEX") [Vyfakturuj.cz](https://flowrunner.ai/integrations/vyfakturuj-cz "Vyfakturuj.cz") [Wave](https://flowrunner.ai/integrations/wave "Wave") [weclapp](https://flowrunner.ai/integrations/weclapp "weclapp") [WeSupply](https://flowrunner.ai/integrations/wesupply "WeSupply") [wflow](https://flowrunner.ai/integrations/wflow "wflow") [WHMCS](https://flowrunner.ai/integrations/whmcs "WHMCS") [Wise](https://flowrunner.ai/integrations/wise "Wise") [WiziShop](https://flowrunner.ai/integrations/wizishop "WiziShop") [WooCommerce](https://flowrunner.ai/integrations/woocommerce "WooCommerce") [Woosmap](https://flowrunner.ai/integrations/woosmap "Woosmap") [Xero](https://flowrunner.ai/integrations/xero "Xero") [YeshInvoice](https://flowrunner.ai/integrations/yeshinvoice "YeshInvoice") [YNAB](https://flowrunner.ai/integrations/ynab "YNAB") [Yotpo Loyalty](https://flowrunner.ai/integrations/yotpo-loyalty "Yotpo Loyalty") [zistemo](https://flowrunner.ai/integrations/zistemo "zistemo") [Zoho Books](https://flowrunner.ai/integrations/zoho-books "Zoho Books") [Zoho Inventory](https://flowrunner.ai/integrations/zoho-inventory "Zoho Inventory") [Zoho Invoice](https://flowrunner.ai/integrations/zoho-invoice "Zoho Invoice") [Zuora](https://flowrunner.ai/integrations/zuora "Zuora") 326 integrations 15,436 verified actions Verified against each vendor's official API ### The work that never gets easier Your team lives in the gap between systems that should talk and do not. Orders land in the store but not in the ledger. Bills arrive as PDFs that someone keys into the ERP by hand. Bank deposits sit unmatched against invoices until a person opens two tabs and eyeballs the amounts. Month-end turns into a marathon of exports, pivots, and follow-up emails, and the close slips because one number would not tie out. The systems are not the problem. You already run QuickBooks Online or NetSuite, a payment processor like Stripe, a store on Shopify, maybe Acumatica or Sage Intacct on the accounting side and BILL or Ramp for AP. The problem is the manual glue between them, the copy-paste, the re-keying, the second look. And the work that actually carries risk, releasing a payment, posting a journal entry, issuing a refund, is exactly the work you cannot hand to a script that does not know when to stop. - Vendor bills arrive as PDFs that someone keys in by hand. - Bank deposits sit unmatched against invoices for days. - Store orders never make it cleanly into the ledger. - Chargebacks pile up while the evidence window closes. - Month-end close slips because one number will not tie out. ### What you can automate Automations Each of these runs as a flow. The agent does the matching, the enrichment, and the routine data entry on its own. It stops and asks a human at the one point where money moves or a number hits the ledger. Every tool named here is a FlowRunner connector, built and verified against the vendor's official API. 1. ### Invoice validation and three-way match Bills arrive matched and coded. The payment waits for an approver. 1. Trigger A vendor bill lands in BILL, or an invoice PDF hits the AP inbox. 2. Agent The agent reads the invoice, pulls the matching purchase order and receipt from NetSuite or Acumatica, runs the three-way match on quantities and prices, flags variances, and codes the bill to the right GL account and cost center. 3. Result AP stops keying invoices and chasing POs by hand, and every payment that goes out has a named approver behind it in the record. Run 4400 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/billcom "BILL")[](https://flowrunner.ai/integrations/netsuite "NetSuite")[](https://flowrunner.ai/integrations/acumatica "Acumatica")[](https://flowrunner.ai/integrations/quickbooks-online "QuickBooks Online")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Clean matches under your threshold it queues. When the amount is large, the variance is unresolved, or the vendor is new, it stops before scheduling payment and hands the fully-matched bill to an approver who sees the PO, the receipt, and the discrepancy. 2. ### Bank and payment reconciliation The match runs itself. The posting to the ledger waits for a human. 1. Trigger A payout settles in Stripe or PayPal, or a bank feed updates in Xero. 2. Agent The agent pulls the settlement detail, matches each deposit against open invoices and processor fees, groups the batch, and drafts the reconciliation entries with fees split to the right account. 3. Result Reconciliation stops being a two-tab eyeball exercise, and no entry reaches the ledger without a person confirming the numbers tie. Run 4437 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/stripe "Stripe")[](https://flowrunner.ai/integrations/paypal "PayPal")[](https://flowrunner.ai/integrations/xero "Xero")[](https://flowrunner.ai/integrations/quickbooks-online "QuickBooks Online")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Exact matches it stages. Anything that will not tie out, a short payment, an unexplained deposit, a fee that does not reconcile, stops for a controller before the journal entry posts to the ledger. 3. ### Order-to-cash sync Every store order lands in the ledger clean. The odd one out waits. 1. Trigger An order is placed in Shopify, WooCommerce, or BigCommerce, or a subscription renews in Chargebee. 2. Agent The agent creates the sales order or invoice in NetSuite or QuickBooks Online, maps SKUs to the right revenue accounts, applies tax and discounts, records the customer and payment, and reconciles the storefront payout against the deposit. 3. Result Storefront revenue lands in the books the same day it is earned, mapped and taxed correctly, without a nightly export and a manual import. Run 4474 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/shopify "Shopify")[](https://flowrunner.ai/integrations/woocommerce "WooCommerce")[](https://flowrunner.ai/integrations/chargebee "Chargebee")[](https://flowrunner.ai/integrations/netsuite "NetSuite")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Standard orders flow straight through. When the tax treatment is ambiguous, the amount is unusually large, or the customer record does not resolve, it stops for accounting rather than booking revenue on a guess. 4. ### Dispute and chargeback handling The evidence assembles itself. The refund or the fight waits for a call. 1. Trigger A chargeback or dispute opens in Stripe, PayPal, or Paddle. 2. Agent The agent gathers the order, the fulfillment and tracking from ShipBob or ShipStation, the customer history, and the original charge, assembles the evidence package, and drafts the response before the deadline. 3. Result Disputes stop expiring in the queue, and every refund or write-off above the line has a human decision and a timestamp attached. Run 4511 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/stripe "Stripe")[](https://flowrunner.ai/integrations/paddle "Paddle")[](https://flowrunner.ai/integrations/shipbob "ShipBob")[](https://flowrunner.ai/integrations/shipstation "ShipStation")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Small disputes it can resolve on policy. When the amount is above your threshold or the customer is worth keeping, it stops for a manager to decide: submit the evidence and fight, or issue the refund or credit memo and move on. 5. ### Month-end close prep The close package builds itself. The entries post on your sign-off. 1. Trigger The scheduled close date arrives, or a controller kicks off the run. 2. Agent The agent pulls balances and open items across NetSuite, Sage Intacct, or Acumatica, ties subledgers to the GL, reconciles the payment processors and bank feeds, flags unmatched items and accrual candidates, and compiles the close package with every variance called out. 3. Result Close stops being a spreadsheet marathon, and the entries plus the sign-off come out audit-ready the first time. Run 4548 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/netsuite "NetSuite")[](https://flowrunner.ai/integrations/sage-intacct "Sage Intacct")[](https://flowrunner.ai/integrations/acumatica "Acumatica")[](https://flowrunner.ai/integrations/stripe "Stripe")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint It proposes the accruals and adjusting entries and waits. A controller reviews the package, and only then does the agent post the journal entries, with every entry, its backup, and the approver written to the log. ### The pattern that makes it safe to automate Finance automation fails when a script moves money or posts a number that should not have gone through. FlowRunner's answer is the digital andon cord: the agent runs the line and pulls it the instant a step touches the ledger or the bank. It matches, codes, reconciles, and stages the routine work on its own. Scheduling a payment, posting a journal entry, issuing a refund or credit memo above a threshold, releasing a large purchase order, always stops and routes to the approver through the channel they already watch, with the invoice, the match, and the variance attached. And because every one of those decisions is captured with the approver and a timestamp, the record you hand an auditor already exists. That combination, a human on every consequential move and a complete trail behind it, is why finance teams trust it with the ledger and auditors accept what it produces. [The digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord) ### What your team gets Outcomes #### Kill the manual re-keying Invoices, orders, and payments move between your store, processor, and ledger without anyone copy-pasting a number or reconciling by eye. #### Close faster, tie out cleaner Subledgers, bank feeds, and processors are reconciled before close starts, so the package is ready and the entries post on a controller's sign-off. #### A person behind every payment Nothing gets scheduled, posted, or refunded on autopilot above your thresholds. The consequential move always has a named approver. #### Automation that auditors accept Every payment, posting, and refund carries the approver and the timestamp, so the audit trail is a byproduct of the work, not a scramble at year-end. #### Disputes stop expiring Chargeback evidence is assembled and filed before the deadline, and refunds above the line wait for a human call instead of a silent loss. #### Runs on the stack you already have The agent connects to your ERP, accounting system, processors, and stores, every connector verified against the vendor API, so there is no system it cannot reach. ### Built for finance's requirements Controls The controls a controller and an auditor would demand of any automation touching the ledger are native to the platform, not add-ons. #### Complete audit trail Every payment, posting, and refund records the approver and the timestamp alongside the invoice and the match, exportable straight into an audit. The record an auditor asks for already exists. #### RBAC and SSO/SAML Role-based access to the platform itself, so approval authority maps to your segregation-of-duties policy, with SSO and SAML behind your identity provider. #### Bring your own keys Your model providers, your keys. Inference runs on credentials you control, and you pay the provider directly. Financial data never rides on someone else's account. #### Self-hosted option Deploy inside your own infrastructure so invoices, payment detail, and ledger data never leave your environment. > _It doesn't feel like it would probably strip out a significant amount of costs, but it would help drive efficiency and productivity and enable us to scale._ Partner, accounting firm ### The complete finance and operations stack The stack ERP, accounting, payments, e-commerce, and logistics, from NetSuite, Acumatica, Sage Intacct, and QuickBooks Online to Stripe, BILL, Ramp, Shopify, and ShipBob. Every connector is built and verified against the vendor's official API, so an agent calls it the way the vendor's API actually allows. The tools the automations above run on. [BILL](https://flowrunner.ai/integrations/billcom "BILL") [NetSuite](https://flowrunner.ai/integrations/netsuite "NetSuite") [Acumatica](https://flowrunner.ai/integrations/acumatica "Acumatica") [QuickBooks Online](https://flowrunner.ai/integrations/quickbooks-online "QuickBooks Online") [Stripe](https://flowrunner.ai/integrations/stripe "Stripe") [PayPal](https://flowrunner.ai/integrations/paypal "PayPal") [Xero](https://flowrunner.ai/integrations/xero "Xero") [Shopify](https://flowrunner.ai/integrations/shopify "Shopify") [WooCommerce](https://flowrunner.ai/integrations/woocommerce "WooCommerce") [Chargebee](https://flowrunner.ai/integrations/chargebee "Chargebee") [Paddle](https://flowrunner.ai/integrations/paddle "Paddle") [ShipBob](https://flowrunner.ai/integrations/shipbob "ShipBob") [ShipStation](https://flowrunner.ai/integrations/shipstation "ShipStation") [Sage Intacct](https://flowrunner.ai/integrations/sage-intacct "Sage Intacct") [Finance & Accounting 197 integrations](https://flowrunner.ai/integrations/category/finance-accounting) [E-commerce 88 integrations](https://flowrunner.ai/integrations/category/e-commerce) [Logistics & Fulfillment 41 integrations](https://flowrunner.ai/integrations/category/logistics-fulfillment) Every connector verified against each vendor's official API. --- ## FlowRunner for revenue operations Source: https://flowrunner.ai/teams/revops Automate the lead enrichment, CRM hygiene, and handoff work that decides whether pipeline moves or stalls, with an agent that runs the routine and a human who owns every move that touches a named account, a big deal, or the whole database. [Start free](https://app.flowrunner.ai) See the stack The verified stack [1CRM](https://flowrunner.ai/integrations/onecrm "1CRM") [4leads](https://flowrunner.ai/integrations/four-leads "4leads") [Access Charity CRM](https://flowrunner.ai/integrations/access-charity-crm "Access Charity CRM") [Action Network](https://flowrunner.ai/integrations/action-network "Action Network") [ActiveCampaign](https://flowrunner.ai/integrations/activecampaign "ActiveCampaign") [ActiveTrail](https://flowrunner.ai/integrations/activetrail "ActiveTrail") [AcyMailing](https://flowrunner.ai/integrations/acymailing "AcyMailing") [Affinity](https://flowrunner.ai/integrations/affinity "Affinity") [Agendor](https://flowrunner.ai/integrations/agendor "Agendor") [Agile CRM](https://flowrunner.ai/integrations/agile-crm "Agile CRM") [Aimfox](https://flowrunner.ai/integrations/aimfox "Aimfox") [Amwork](https://flowrunner.ai/integrations/amwork "Amwork") [Anabix](https://flowrunner.ai/integrations/anabix "Anabix") [Anymail Finder](https://flowrunner.ai/integrations/anymailfinder "Anymail Finder") [Anysite Web Data](https://flowrunner.ai/integrations/anysite "Anysite Web Data") [Apollo.io](https://flowrunner.ai/integrations/apollo "Apollo.io") [Attentive](https://flowrunner.ai/integrations/attentive "Attentive") [Attio](https://flowrunner.ai/integrations/attio "Attio") [AWeber](https://flowrunner.ai/integrations/aweber "AWeber") [B2Cor CRM](https://flowrunner.ai/integrations/b2cor-crm "B2Cor CRM") [beehiiv](https://flowrunner.ai/integrations/beehiiv "beehiiv") [BetterContact](https://flowrunner.ai/integrations/bettercontact "BetterContact") [Bigin](https://flowrunner.ai/integrations/bigin-by-zoho "Bigin") [BigMailer](https://flowrunner.ai/integrations/bigmailer "BigMailer") [BirdSend](https://flowrunner.ai/integrations/birdsend "BirdSend") [BizMachine](https://flowrunner.ai/integrations/bizmachine "BizMachine") [Boldem](https://flowrunner.ai/integrations/boldem "Boldem") [Bouncer](https://flowrunner.ai/integrations/bouncer "Bouncer") [Brandfetch](https://flowrunner.ai/integrations/brandfetch "Brandfetch") [Braze](https://flowrunner.ai/integrations/braze "Braze") [Brevo](https://flowrunner.ai/integrations/brevo "Brevo") [Campaign Cleaner](https://flowrunner.ai/integrations/campaign-cleaner "Campaign Cleaner") [Campaign Monitor](https://flowrunner.ai/integrations/campaign-monitor "Campaign Monitor") [Capsule CRM](https://flowrunner.ai/integrations/capsule-crm "Capsule CRM") [Captain Data](https://flowrunner.ai/integrations/captain-data "Captain Data") [Clay](https://flowrunner.ai/integrations/clay "Clay") [Clearbit](https://flowrunner.ai/integrations/clearbit "Clearbit") [Clearout](https://flowrunner.ai/integrations/clearout "Clearout") [CleverReach](https://flowrunner.ai/integrations/cleverreach "CleverReach") [Clientjoy](https://flowrunner.ai/integrations/clientjoy "Clientjoy") [Close CRM](https://flowrunner.ai/integrations/closecrm "Close CRM") [Cloze](https://flowrunner.ai/integrations/cloze "Cloze") [CompanyHub](https://flowrunner.ai/integrations/companyhub "CompanyHub") [Constant Contact](https://flowrunner.ai/integrations/constant-contact "Constant Contact") [Contacts+](https://flowrunner.ai/integrations/contactsplus "Contacts+") [Copilot](https://flowrunner.ai/integrations/copilot "Copilot") [Copper](https://flowrunner.ai/integrations/copper "Copper") [Corymbus](https://flowrunner.ai/integrations/corymbus "Corymbus") [Crossbeam](https://flowrunner.ai/integrations/crossbeam "Crossbeam") [CrowdPower](https://flowrunner.ai/integrations/crowdpower "CrowdPower") [Custify](https://flowrunner.ai/integrations/custify "Custify") [Customer.io](https://flowrunner.ai/integrations/customerio "Customer.io") [Datagma](https://flowrunner.ai/integrations/datagma "Datagma") [DataMerge AI](https://flowrunner.ai/integrations/datamerge "DataMerge AI") [Daylite](https://flowrunner.ai/integrations/marketcircle-daylite "Daylite") [DeBounce](https://flowrunner.ai/integrations/debounce "DeBounce") [Digiclose](https://flowrunner.ai/integrations/digiclose "Digiclose") [Disciple.Tools](https://flowrunner.ai/integrations/disciple-tools "Disciple.Tools") [Discko](https://flowrunner.ai/integrations/discko "Discko") [Dotdigital](https://flowrunner.ai/integrations/dotdigital "Dotdigital") [Dr.Tracker](https://flowrunner.ai/integrations/dr-tracker "Dr.Tracker") [Drift](https://flowrunner.ai/integrations/drift "Drift") [Drip](https://flowrunner.ai/integrations/drip "Drip") [Dropcontact](https://flowrunner.ai/integrations/dropcontact "Dropcontact") [Dux-Soup](https://flowrunner.ai/integrations/dux-soup "Dux-Soup") [Dynosend](https://flowrunner.ai/integrations/dynosend "Dynosend") [E-goi](https://flowrunner.ai/integrations/e-goi "E-goi") [Ecomail](https://flowrunner.ai/integrations/ecomail-cz "Ecomail") [Emailable](https://flowrunner.ai/integrations/emailable "Emailable") [Emailkampane](https://flowrunner.ai/integrations/emailkampane "Emailkampane") [EmailListVerify](https://flowrunner.ai/integrations/email-list-verify "EmailListVerify") [EmailOctopus](https://flowrunner.ai/integrations/email-octopus "EmailOctopus") [Emailvalidation.io](https://flowrunner.ai/integrations/emailvalidation "Emailvalidation.io") [Emelia](https://flowrunner.ai/integrations/emelia "Emelia") [Emercury](https://flowrunner.ai/integrations/emercury "Emercury") [Emma](https://flowrunner.ai/integrations/emma "Emma") [Encharge](https://flowrunner.ai/integrations/encharge "Encharge") [Engage](https://flowrunner.ai/integrations/engage "Engage") [EngageBay](https://flowrunner.ai/integrations/engagebay "EngageBay") [Enginemailer](https://flowrunner.ai/integrations/enginemailer "Enginemailer") [EspoCRM](https://flowrunner.ai/integrations/espo-crm "EspoCRM") [eWay-CRM](https://flowrunner.ai/integrations/eway-crm "eWay-CRM") [Exact Spotter](https://flowrunner.ai/integrations/exact-spotter "Exact Spotter") [Explorium AgentSource](https://flowrunner.ai/integrations/explorium-agentsource "Explorium AgentSource") [Facebook Lead Ads](https://flowrunner.ai/integrations/facebook-lead-ads "Facebook Lead Ads") [Findymail](https://flowrunner.ai/integrations/findymail "Findymail") [Flashy](https://flowrunner.ai/integrations/flashyapp "Flashy") [Flexie CRM](https://flowrunner.ai/integrations/flexie-crm "Flexie CRM") [Flexmail](https://flowrunner.ai/integrations/flexmail "Flexmail") [Flodesk](https://flowrunner.ai/integrations/flodesk "Flodesk") [FLOWii](https://flowrunner.ai/integrations/flowii "FLOWii") [FluentCRM](https://flowrunner.ai/integrations/fluentcrm "FluentCRM") [folk](https://flowrunner.ai/integrations/folk "folk") [Follow Up Boss](https://flowrunner.ai/integrations/follow-up-boss "Follow Up Boss") [Force24](https://flowrunner.ai/integrations/force24 "Force24") [ForceManager](https://flowrunner.ai/integrations/force-manager "ForceManager") [Freshworks CRM](https://flowrunner.ai/integrations/freshworks-crm "Freshworks CRM") [FullEnrich](https://flowrunner.ai/integrations/fullenrich "FullEnrich") [GetEmail.io](https://flowrunner.ai/integrations/getemail-io "GetEmail.io") [GetProspect](https://flowrunner.ai/integrations/getprospect "GetProspect") [GetResponse](https://flowrunner.ai/integrations/getresponse "GetResponse") [Gist](https://flowrunner.ai/integrations/gist "Gist") [GoHighLevel](https://flowrunner.ai/integrations/gohighlevel "GoHighLevel") [Gong](https://flowrunner.ai/integrations/gong "Gong") [Google Ads](https://flowrunner.ai/integrations/google-ads "Google Ads") [Google Contacts](https://flowrunner.ai/integrations/google-contacts "Google Contacts") [Happierleads](https://flowrunner.ai/integrations/happierleads "Happierleads") [HeyReach](https://flowrunner.ai/integrations/heyreach "HeyReach") [Hubhus](https://flowrunner.ai/integrations/hubhus "Hubhus") [HubSpot](https://flowrunner.ai/integrations/hubspot "HubSpot") [Hunter.io](https://flowrunner.ai/integrations/hunter "Hunter.io") [Icypeas](https://flowrunner.ai/integrations/icypeas "Icypeas") [Initiative CRM](https://flowrunner.ai/integrations/initiative-crm "Initiative CRM") [Insightly](https://flowrunner.ai/integrations/insightly "Insightly") [Instantly](https://flowrunner.ai/integrations/instantly "Instantly") [Iterable](https://flowrunner.ai/integrations/iterable "Iterable") [JobNimbus](https://flowrunner.ai/integrations/jobnimbus "JobNimbus") [Kartra](https://flowrunner.ai/integrations/kartra "Kartra") [Keap](https://flowrunner.ai/integrations/keap "Keap") [Kit](https://flowrunner.ai/integrations/kit "Kit") [Klaviyo](https://flowrunner.ai/integrations/klaviyo "Klaviyo") [Klenty](https://flowrunner.ai/integrations/klenty "Klenty") [KlickTipp](https://flowrunner.ai/integrations/klicktipp "KlickTipp") [La Growth Machine](https://flowrunner.ai/integrations/lagrowthmachine "La Growth Machine") [Lead Agent](https://flowrunner.ai/integrations/leadagent "Lead Agent") [LeadSquared](https://flowrunner.ai/integrations/leadsquared "LeadSquared") [Leady](https://flowrunner.ai/integrations/leady "Leady") [Lemlist](https://flowrunner.ai/integrations/lemlist "Lemlist") [Linked API](https://flowrunner.ai/integrations/linked-api "Linked API") [LinkedIn](https://flowrunner.ai/integrations/linkedin "LinkedIn") [LinkupAPI](https://flowrunner.ai/integrations/linkupapi "LinkupAPI") [Livespace](https://flowrunner.ai/integrations/livespace-crm "Livespace") [LivingMetrics](https://flowrunner.ai/integrations/livingmetrics "LivingMetrics") [Loopify](https://flowrunner.ai/integrations/loopify "Loopify") [Loops](https://flowrunner.ai/integrations/loops "Loops") [Lusha](https://flowrunner.ai/integrations/lusha "Lusha") [Magentrix](https://flowrunner.ai/integrations/magentrix "Magentrix") [Magileads](https://flowrunner.ai/integrations/magileads "Magileads") [Mail Komplet](https://flowrunner.ai/integrations/mail-komplet "Mail Komplet") [mail2many](https://flowrunner.ai/integrations/mail2many "mail2many") [MailBluster](https://flowrunner.ai/integrations/mailbluster "MailBluster") [MailboxValidator](https://flowrunner.ai/integrations/mailboxvalidator "MailboxValidator") [Mailcheck](https://flowrunner.ai/integrations/mailcheck "Mailcheck") [Mailchimp Marketing](https://flowrunner.ai/integrations/mailchimp-marketing "Mailchimp Marketing") [Maileon](https://flowrunner.ai/integrations/maileon "Maileon") [MailerCheck](https://flowrunner.ai/integrations/mailercheck "MailerCheck") [Mailercloud](https://flowrunner.ai/integrations/mailercloud "Mailercloud") [MailerLite](https://flowrunner.ai/integrations/mailerlite "MailerLite") [Mailkit](https://flowrunner.ai/integrations/mailkit "Mailkit") [Mailmodo](https://flowrunner.ai/integrations/mailmodo "Mailmodo") [Mailrelay](https://flowrunner.ai/integrations/mailrelay "Mailrelay") [Mails.so](https://flowrunner.ai/integrations/mails-so "Mails.so") [Mailshake](https://flowrunner.ai/integrations/mailshake "Mailshake") [Mailvio](https://flowrunner.ai/integrations/mailvio "Mailvio") [ManyReach](https://flowrunner.ai/integrations/manyreach "ManyReach") [Marketo](https://flowrunner.ai/integrations/marketo "Marketo") [Mautic](https://flowrunner.ai/integrations/mautic "Mautic") [Meetime](https://flowrunner.ai/integrations/meetime "Meetime") [Mesh](https://flowrunner.ai/integrations/clay-earth "Mesh") [MessengerOS](https://flowrunner.ai/integrations/messenger-os "MessengerOS") [Mixmax](https://flowrunner.ai/integrations/mixmax "Mixmax") [Monica](https://flowrunner.ai/integrations/monica "Monica") [Moosend](https://flowrunner.ai/integrations/moosend "Moosend") [Moskit CRM](https://flowrunner.ai/integrations/moskitcrm "Moskit CRM") [Mumara](https://flowrunner.ai/integrations/mumara "Mumara") [Myphoner](https://flowrunner.ai/integrations/myphoner "Myphoner") [NectarCRM](https://flowrunner.ai/integrations/nectar-crm "NectarCRM") [Neon CRM](https://flowrunner.ai/integrations/neoncrm "Neon CRM") [NetHunt CRM](https://flowrunner.ai/integrations/nethunt "NetHunt CRM") [NeverBounce](https://flowrunner.ai/integrations/neverbounce "NeverBounce") [Newsman](https://flowrunner.ai/integrations/newsman "Newsman") [Nimble](https://flowrunner.ai/integrations/nimble "Nimble") [noCRM.io](https://flowrunner.ai/integrations/nocrm-io "noCRM.io") [Nutshell](https://flowrunner.ai/integrations/nutshell "Nutshell") [Omnisend](https://flowrunner.ai/integrations/omnisend "Omnisend") [OneClick for Pipedrive](https://flowrunner.ai/integrations/oneclick-for-pipedrive "OneClick for Pipedrive") [OnePageCRM](https://flowrunner.ai/integrations/onepagecrm "OnePageCRM") [Ontraport](https://flowrunner.ai/integrations/ontraport "Ontraport") [Optimizely Campaign](https://flowrunner.ai/integrations/optimizely-campaign "Optimizely Campaign") [Oracle Eloqua](https://flowrunner.ai/integrations/oracle-eloqua "Oracle Eloqua") [Oracle Fusion Cloud Sales](https://flowrunner.ai/integrations/oracle-fusion-cloud-sales "Oracle Fusion Cloud Sales") [Ortto](https://flowrunner.ai/integrations/ortto "Ortto") [Outreach](https://flowrunner.ai/integrations/outreach "Outreach") [Outseta](https://flowrunner.ai/integrations/outseta "Outseta") [People Data Labs](https://flowrunner.ai/integrations/people-data-labs "People Data Labs") [Phantombuster](https://flowrunner.ai/integrations/phantombuster "Phantombuster") [Pipedrive](https://flowrunner.ai/integrations/pipedrive "Pipedrive") [Pipeliner CRM](https://flowrunner.ai/integrations/pipelinercrm "Pipeliner CRM") [PlanOK GCI](https://flowrunner.ai/integrations/planok-gci "PlanOK GCI") [PlusVibe](https://flowrunner.ai/integrations/plusvibe "PlusVibe") [Powerlink (Fireberry)](https://flowrunner.ai/integrations/powerlink "Powerlink (Fireberry)") [PredictLeads](https://flowrunner.ai/integrations/predict-leads-app "PredictLeads") [Quanda](https://flowrunner.ai/integrations/quanda "Quanda") [Quentn](https://flowrunner.ai/integrations/quentn "Quentn") [QuickEmailVerification](https://flowrunner.ai/integrations/quickemailverification "QuickEmailVerification") [Raklet](https://flowrunner.ai/integrations/raklet "Raklet") [Rav Messer](https://flowrunner.ai/integrations/rav-messer "Rav Messer") [RAYNET CRM](https://flowrunner.ai/integrations/raynet-crm-v2 "RAYNET CRM") [RD Station](https://flowrunner.ai/integrations/rd-station "RD Station") [ReachInbox](https://flowrunner.ai/integrations/reachinbox "ReachInbox") [RealMail](https://flowrunner.ai/integrations/realmail "RealMail") [Reoon Email Verifier](https://flowrunner.ai/integrations/reoon-email-verifier "Reoon Email Verifier") [RepliQ](https://flowrunner.ai/integrations/repliq "RepliQ") [Reply.io](https://flowrunner.ai/integrations/reply-io "Reply.io") [Reverse Contact](https://flowrunner.ai/integrations/reversecontact "Reverse Contact") [Robly](https://flowrunner.ai/integrations/robly "Robly") [RocketReach](https://flowrunner.ai/integrations/rocketreach "RocketReach") [SalesBlink](https://flowrunner.ai/integrations/salesblink "SalesBlink") [Salesflare](https://flowrunner.ai/integrations/salesflare "Salesflare") [Salesforce Essentials](https://flowrunner.ai/integrations/salesforce-essentials "Salesforce Essentials") [Salesforce Marketing Cloud](https://flowrunner.ai/integrations/marketing-cloud "Salesforce Marketing Cloud") [Salesforce Pardot](https://flowrunner.ai/integrations/salesforce-pardot "Salesforce Pardot") [Salesforce Pro](https://flowrunner.ai/integrations/salesforce-pro "Salesforce Pro") [Salesforge](https://flowrunner.ai/integrations/salesforge "Salesforge") [Salesloft](https://flowrunner.ai/integrations/salesloft "Salesloft") [SALESmanago](https://flowrunner.ai/integrations/salesmanago "SALESmanago") [Salesmate](https://flowrunner.ai/integrations/salesmate "Salesmate") [SalesTrigger](https://flowrunner.ai/integrations/sales-trigger "SalesTrigger") [Scalelist](https://flowrunner.ai/integrations/scalelist "Scalelist") [Sellsy](https://flowrunner.ai/integrations/sellsy "Sellsy") [Sender](https://flowrunner.ai/integrations/sender "Sender") [SendFox](https://flowrunner.ai/integrations/sendfox "SendFox") [Sendlane](https://flowrunner.ai/integrations/sendlane "Sendlane") [Sendme](https://flowrunner.ai/integrations/sendme123 "Sendme") [SendPulse](https://flowrunner.ai/integrations/sendpulse "SendPulse") [SendX](https://flowrunner.ai/integrations/sendx "SendX") [SensorPro](https://flowrunner.ai/integrations/sensorpro "SensorPro") [SharpSpring](https://flowrunner.ai/integrations/sharpspring "SharpSpring") [SigParser](https://flowrunner.ai/integrations/sigparser "SigParser") [SlimCRM](https://flowrunner.ai/integrations/slimcrm "SlimCRM") [SlimEmail](https://flowrunner.ai/integrations/slimemail "SlimEmail") [Smaily](https://flowrunner.ai/integrations/smaily "Smaily") [SmartEmailing](https://flowrunner.ai/integrations/smart-emailing "SmartEmailing") [Smartlead](https://flowrunner.ai/integrations/smartleadai "Smartlead") [SmartReach.io](https://flowrunner.ai/integrations/smartreach-io "SmartReach.io") [Smoove](https://flowrunner.ai/integrations/smoove "Smoove") [Snov.io](https://flowrunner.ai/integrations/snovio "Snov.io") [SolarMarket](https://flowrunner.ai/integrations/solarmarket "SolarMarket") [Stood CRM](https://flowrunner.ai/integrations/stood-crm "Stood CRM") [Streak](https://flowrunner.ai/integrations/streak "Streak") [SugarCRM](https://flowrunner.ai/integrations/sugarcrm11 "SugarCRM") [SuiteCRM](https://flowrunner.ai/integrations/suitecrm7 "SuiteCRM") [SuperOffice](https://flowrunner.ai/integrations/superoffice "SuperOffice") [Swift Missive](https://flowrunner.ai/integrations/swiftmissive "Swift Missive") [Swipe One](https://flowrunner.ai/integrations/swipe-one "Swipe One") [Swordfish AI](https://flowrunner.ai/integrations/swordfish "Swordfish AI") [Systeme.io](https://flowrunner.ai/integrations/systeme-io "Systeme.io") [Tarvent](https://flowrunner.ai/integrations/tarvent "Tarvent") [Teamgate](https://flowrunner.ai/integrations/teamgate-crm "Teamgate") [Teamleader](https://flowrunner.ai/integrations/teamleader "Teamleader") [Tomba](https://flowrunner.ai/integrations/tomba "Tomba") [UpLead](https://flowrunner.ai/integrations/uplead "UpLead") [uProc](https://flowrunner.ai/integrations/uproc "uProc") [Upsales](https://flowrunner.ai/integrations/upsales "Upsales") [UseINBOX](https://flowrunner.ai/integrations/useinbox "UseINBOX") [User.com](https://flowrunner.ai/integrations/usercom "User.com") [Userlist](https://flowrunner.ai/integrations/userlist "Userlist") [Uspacy](https://flowrunner.ai/integrations/uspacy "Uspacy") [ValidEmail](https://flowrunner.ai/integrations/validemail-co "ValidEmail") [Valuecase](https://flowrunner.ai/integrations/valuecase "Valuecase") [Vayne](https://flowrunner.ai/integrations/vayne "Vayne") [VBOUT](https://flowrunner.ai/integrations/vbout "VBOUT") [Verifi.Email](https://flowrunner.ai/integrations/verifi-email "Verifi.Email") [Vero](https://flowrunner.ai/integrations/vero "Vero") [Vision6](https://flowrunner.ai/integrations/vision6 "Vision6") [Vitally](https://flowrunner.ai/integrations/vitally "Vitally") [Vtiger CRM](https://flowrunner.ai/integrations/vtiger "Vtiger CRM") [Wappalyzer](https://flowrunner.ai/integrations/wappalyzer "Wappalyzer") [Wealthbox](https://flowrunner.ai/integrations/wealthbox "Wealthbox") [Webmetic](https://flowrunner.ai/integrations/webmetic "Webmetic") [Wild Apricot](https://flowrunner.ai/integrations/wild-apricot "Wild Apricot") [Wishpond](https://flowrunner.ai/integrations/wishpond "Wishpond") [Wiza](https://flowrunner.ai/integrations/wiza "Wiza") [Woodpecker](https://flowrunner.ai/integrations/woodpecker "Woodpecker") [Zagomail](https://flowrunner.ai/integrations/zagomail "Zagomail") [Zendesk Sell](https://flowrunner.ai/integrations/zendesk-sell "Zendesk Sell") [ZeroBounce](https://flowrunner.ai/integrations/zerobounce "ZeroBounce") [Zixflow](https://flowrunner.ai/integrations/zixflow "Zixflow") [Zoho Campaigns](https://flowrunner.ai/integrations/zoho-campaigns "Zoho Campaigns") [Zoho CRM](https://flowrunner.ai/integrations/zoho-crm "Zoho CRM") [ZoomInfo](https://flowrunner.ai/integrations/zoominfo "ZoomInfo") 281 integrations 7,503 verified actions Verified against each vendor's official API ### The plumbing everyone forgets until it breaks You are the reason pipeline is clean, and the reason nobody thinks about it until it is not. A lead comes in from a form, an ad, or a list buy, and it needs enriching, deduping, scoring, and routing before a rep ever sees it, or it sits and rots. The CRM fills with half-typed company names, three records for the same account, and owners who left last quarter. Marketing wants sequences enrolled, sales wants deals created, finance wants the numbers to tie, and all of it flows through you. The tools are not the problem. You already run a CRM, an enrichment provider, a sequencer, and an email platform. The problem is the manual glue between them, and the fact that the moves that actually matter, enrolling a strategic account, blasting a segment, reassigning ownership, merging two records into one, are exactly the moves you cannot hand to a blind sync that does not know when to stop. - Inbound leads sit unenriched and unrouted while they go cold. - The CRM fills with duplicates and half-typed company names. - Owners who left last quarter still sit on live accounts. - Deals live in one system and the numbers never tie to the other. - The move that matters, a segment blast or a merge, is exactly what a blind sync should not touch. ### What you can automate Automations Each of these runs as a flow. The agent does the enrichment, the matching, and the routine syncing on its own. It stops and asks a human at the one point where a wrong move is expensive or irreversible, a strategic account, a mass send, a bulk overwrite, a merge. Every tool named here is a FlowRunner connector, built and verified against the vendor's official API. 1. ### Inbound lead enrichment and routing Every lead arrives enriched, scored, and routed. Named accounts wait for a human. 1. Trigger A lead comes in from a form, a Facebook Lead Ads campaign, or a Google Ads landing page. 2. Agent The agent enriches the record with Clearbit and Hunter.io, fills the missing firmographics and verified email, scores it against your fit rules, matches it to an existing account, and routes it to the right rep in HubSpot on the territory logic you already use. 3. Result Reps stop working raw form fills and open leads that are already enriched, scored, and correctly routed, with your top accounts routed by a person, not a rule. Run 4400 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/hubspot "HubSpot")[](https://flowrunner.ai/integrations/clearbit "Clearbit")[](https://flowrunner.ai/integrations/hunter "Hunter.io")[](https://flowrunner.ai/integrations/facebook-lead-ads "Facebook Lead Ads")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Standard leads route on their own. When the enrichment tags the record as a named or strategic account, the agent stops before it assigns and hands the fully-enriched lead to the ops owner, who confirms the routing and the owner before a rep ever touches it. 2. ### CRM data hygiene and dedup The agent finds and stages the duplicates. The merge waits for a sign-off. 1. Trigger A scheduled hygiene run, or a new record created in the CRM. 2. Agent The agent scans Salesforce for duplicate leads, contacts, and accounts, normalizes company names and domains against Clearbit, verifies stale emails with Hunter.io, and compiles the match set with a proposed surviving record and field-by-field resolution. 3. Result The database stops filling with duplicates and half-typed names, and no record is ever merged out of existence without a named human confirming what survives. Run 4437 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/salesforce-pro "Salesforce Pro")[](https://flowrunner.ai/integrations/clearbit "Clearbit")[](https://flowrunner.ai/integrations/hunter "Hunter.io")[](https://flowrunner.ai/integrations/dropcontact "Dropcontact")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Cosmetic fixes it applies on its own. A record merge destroys history and cannot be cleanly undone, so the agent stages the merge and waits. An ops owner reviews the surviving record and the field choices before the merge runs, and every decision lands in the log. 3. ### Sequence and campaign enrollment Enrollment is instant for the routine list. The segment blast stops for a human. 1. Trigger A lead crosses a scoring threshold, changes stage, or a marketing segment is built. 2. Agent The agent resolves the right audience, checks each contact for suppression, active opportunities, and existing enrollments, dedupes against Outreach and Apollo.io sequences already running, and enrolls the routine list into the correct Mailchimp Marketing campaign or Outreach sequence. 3. Result Routine enrollment is instant and clean of double-sends, and no segment-wide blast or strategic outreach ever goes out without a person signing off on who receives it. Run 4474 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/outreach "Outreach")[](https://flowrunner.ai/integrations/apollo "Apollo.io")[](https://flowrunner.ai/integrations/mailchimp-marketing "Mailchimp Marketing")[](https://flowrunner.ai/integrations/hubspot "HubSpot")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Small, well-scoped enrollments go through. A mass send to a whole segment, or enrollment of a named or strategic account, stops for a human who sees the exact audience, the message, and the suppression check before a single email leaves. 4. ### Deal creation and stage sync Deals are created and kept in sync automatically. The high-value deal waits for a human. 1. Trigger A qualified opportunity is created, or a deal changes stage in the CRM. 2. Agent The agent creates the deal with the right amount, stage, and owner, keeps it in sync across Pipedrive and Salesforce, pulls the latest signals from Gong to update the close date and next step, and writes the activity trail back so both systems agree. 3. Result Deals stop living in one system and missing from the other, the numbers tie, and every high-value deal enters the forecast with a named human behind it. Run 4511 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/pipedrive "Pipedrive")[](https://flowrunner.ai/integrations/salesforce-pro "Salesforce Pro")[](https://flowrunner.ai/integrations/gong "Gong")[](https://flowrunner.ai/integrations/hubspot "HubSpot")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Routine deals sync on their own. When the amount crosses your threshold or the deal is flagged strategic, the agent stops before it creates or advances the deal and routes it to the ops owner, who confirms the amount, stage, and owner before it becomes forecast. 5. ### Lead-to-account matching and sales handoff Leads are matched to accounts and handed off clean. Ownership changes wait for a human. 1. Trigger A new lead is enriched, or an account is created or updated in the CRM. 2. Agent The agent matches the lead to the right account in Salesforce using enriched domain and firmographic data from Clearbit and Apollo.io, resolves the existing owner and territory, attaches the lead to the account, and packages the handoff context for the assigned rep. 3. Result Leads land on the right account with full context, and ownership never changes underneath a rep without a person confirming the reassignment. Run 4548 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/salesforce-pro "Salesforce Pro")[](https://flowrunner.ai/integrations/clearbit "Clearbit")[](https://flowrunner.ai/integrations/apollo "Apollo.io")[](https://flowrunner.ai/integrations/outreach "Outreach")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Matching and attachment happen automatically. Reassigning account ownership changes who gets paid and who works the account, so a bulk owner reassignment or a change on a named account stops for the ops owner, who sees the current owner, the proposed owner, and the reason before it applies. ### The pattern that makes it safe to automate RevOps automation fails when a blind sync does something it should not. FlowRunner's answer is the digital andon cord: the agent runs the line and pulls it the instant a step carries real consequence. It enriches, matches, dedupes, and keeps the routine records in sync on its own. Blasting a segment, merging duplicate records, creating a high-value deal, enrolling a strategic account, or reassigning ownership always stops and routes to a person through the channel they already watch, with everything the agent found attached. The reversible, routine work happens immediately; the move that touches revenue or the whole database waits for a human. That is the difference between automation your team trusts with the CRM and automation it does not. [The digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord) ### What your team gets Outcomes #### Leads that arrive ready to work Every inbound lead is enriched, scored, matched, and routed before a rep sees it, so nothing sits and goes cold in a queue. #### A CRM that stays clean Duplicates get found and staged, names and domains get normalized, and stale records get flagged, on a schedule, without a person combing the database by hand. #### Numbers that tie across systems Deals stay in sync across your CRM and sequencer, so the forecast reads the same in every system your team pulls from. #### No unsupervised mass sends An agent never blasts a segment or enrolls a strategic account on its own. The send that reaches your whole list always has a named human behind it. #### No records merged out of existence A merge or a bulk overwrite destroys history, so it always stops for a person who confirms what survives before it runs. #### No coverage gaps The agent connects to the stack you already run, every connector verified against the vendor API, so there is no tool it cannot reach. ### Built for revenue operations' requirements Controls The controls you would demand of any automation touching the CRM, the pipeline, and your prospect data are native to the platform, not add-ons. #### Complete audit trail Every enrollment, merge, deal change, and owner reassignment, the user or agent that took it, the approver, and the timestamp, recorded and exportable when finance or an auditor asks who changed what. #### RBAC and SSO/SAML Role-based access to the platform itself, so ops owns the flows and reps cannot touch them, with SSO and SAML tied to your identity provider. #### Bring your own keys Your model providers, your keys. Enrichment and inference run on credentials you control, and you pay the provider directly. #### Self-hosted option Deploy inside your own infrastructure so your prospect and customer data never leaves your environment. > _We can't get back to them fast enough. And I fear we're going to lose probably half of them just because they're going to go cold._ CEO, automotive services ### The complete revenue operations stack The stack CRM, sales intelligence, outreach, and email marketing, from HubSpot, Salesforce, and Pipedrive to Apollo.io, Clearbit, Outreach, Gong, and Mailchimp Marketing, plus the ad and social sources that feed them. Every connector is built and verified against the vendor's official API, so an agent calls it the way the vendor's API actually allows. The tools the automations above run on. [HubSpot](https://flowrunner.ai/integrations/hubspot "HubSpot") [Clearbit](https://flowrunner.ai/integrations/clearbit "Clearbit") [Hunter.io](https://flowrunner.ai/integrations/hunter "Hunter.io") [Facebook Lead Ads](https://flowrunner.ai/integrations/facebook-lead-ads "Facebook Lead Ads") [Salesforce Pro](https://flowrunner.ai/integrations/salesforce-pro "Salesforce Pro") [Dropcontact](https://flowrunner.ai/integrations/dropcontact "Dropcontact") [Outreach](https://flowrunner.ai/integrations/outreach "Outreach") [Apollo.io](https://flowrunner.ai/integrations/apollo "Apollo.io") [Mailchimp Marketing](https://flowrunner.ai/integrations/mailchimp-marketing "Mailchimp Marketing") [Pipedrive](https://flowrunner.ai/integrations/pipedrive "Pipedrive") [Gong](https://flowrunner.ai/integrations/gong "Gong") [CRM & Sales 162 integrations](https://flowrunner.ai/integrations/category/crm-sales) [Email & Marketing 143 integrations](https://flowrunner.ai/integrations/category/email-marketing) [Marketing & Social 90 integrations](https://flowrunner.ai/integrations/category/marketing-social) Every connector verified against each vendor's official API. --- ## FlowRunner for security teams Source: https://flowrunner.ai/teams/security Automate the identity, triage, and response work that eats your team's day, with an agent that runs the investigation and a human who owns every action that changes access or cannot be undone. [Start free](https://app.flowrunner.ai) See the stack The verified stack [AttachmentAV](https://flowrunner.ai/integrations/attachmentav "AttachmentAV") [Auth0](https://flowrunner.ai/integrations/auth0 "Auth0") [AWS Certificate Manager](https://flowrunner.ai/integrations/aws-acm "AWS Certificate Manager") [AWS Cognito](https://flowrunner.ai/integrations/aws-cognito "AWS Cognito") [AWS IAM](https://flowrunner.ai/integrations/aws-iam "AWS IAM") [AWS KMS](https://flowrunner.ai/integrations/aws-kms "AWS KMS") [Bitwarden](https://flowrunner.ai/integrations/bitwarden "Bitwarden") [Cloudflare](https://flowrunner.ai/integrations/cloudflare "Cloudflare") [Cortex](https://flowrunner.ai/integrations/cortex "Cortex") [ESPY](https://flowrunner.ai/integrations/espy "ESPY") [Google Workspace Admin](https://flowrunner.ai/integrations/google-workspace-admin "Google Workspace Admin") [Grafana](https://flowrunner.ai/integrations/grafana "Grafana") [Greip](https://flowrunner.ai/integrations/greip "Greip") [IdentityCheck](https://flowrunner.ai/integrations/identitycheck "IdentityCheck") [IP2Proxy](https://flowrunner.ai/integrations/ip2proxy "IP2Proxy") [IPQualityScore](https://flowrunner.ai/integrations/ipqualityscore "IPQualityScore") [KnowBe4](https://flowrunner.ai/integrations/knowbe4 "KnowBe4") [LastPass](https://flowrunner.ai/integrations/lastpass "LastPass") [LDAP](https://flowrunner.ai/integrations/ldap "LDAP") [LogicGate Risk Cloud](https://flowrunner.ai/integrations/riskcloud "LogicGate Risk Cloud") [ManageEngine ADManager Plus](https://flowrunner.ai/integrations/manageengine-admanager "ManageEngine ADManager Plus") [Microsoft Entra ID](https://flowrunner.ai/integrations/entra-id "Microsoft Entra ID") [Microsoft Graph Security](https://flowrunner.ai/integrations/ms-graph-security "Microsoft Graph Security") [Microsoft Intune](https://flowrunner.ai/integrations/intunes "Microsoft Intune") [MISP](https://flowrunner.ai/integrations/misp "MISP") [MyPreferences](https://flowrunner.ai/integrations/mypreferences "MyPreferences") [Okta](https://flowrunner.ai/integrations/okta "Okta") [Onsefy](https://flowrunner.ai/integrations/onsefy "Onsefy") [OOPSpam](https://flowrunner.ai/integrations/oopspam-anti-spam "OOPSpam") [PagerDuty](https://flowrunner.ai/integrations/pagerduty "PagerDuty") [PicDefense](https://flowrunner.ai/integrations/picdefense "PicDefense") [Rhombus](https://flowrunner.ai/integrations/rhombus "Rhombus") [SecurityScorecard](https://flowrunner.ai/integrations/securityscorecard "SecurityScorecard") [Sentry](https://flowrunner.ai/integrations/sentry "Sentry") [Splunk](https://flowrunner.ai/integrations/splunk "Splunk") [Surense](https://flowrunner.ai/integrations/surense "Surense") [Team Password Manager](https://flowrunner.ai/integrations/team-password-manager "Team Password Manager") [TheHive](https://flowrunner.ai/integrations/thehive "TheHive") [Twilio Verify](https://flowrunner.ai/integrations/twilio-verify "Twilio Verify") [UptimeRobot](https://flowrunner.ai/integrations/uptimerobot "UptimeRobot") [urlscan.io](https://flowrunner.ai/integrations/urlscan "urlscan.io") [UserCheck](https://flowrunner.ai/integrations/usercheck "UserCheck") [ViewDNS](https://flowrunner.ai/integrations/view-dns "ViewDNS") [WhoisFreaks](https://flowrunner.ai/integrations/whoisfreaks "WhoisFreaks") [Xama](https://flowrunner.ai/integrations/xama-onboarding "Xama") 45 integrations 1,003 verified actions Verified against each vendor's official API ### The work that never stops Your team lives in a queue that never empties. Phishing reports pile up in a shared mailbox. Access requests wait on someone to check the right group. New hires need provisioning today and leavers needed deprovisioning an hour ago. Every SIEM alert could be nothing or could be the one that matters, and the only way to know is to enrich it by hand across six consoles. The tools are not the problem. You already run Okta or Entra, a SIEM, a case manager, threat intel, and a dozen others. The problem is the manual glue between them, and the fact that the risky work, the deprovision, the domain block, the privilege grant, is exactly the work you cannot hand to a script that does not know when to stop. - Phishing reports pile up in a shared mailbox. - Access requests wait on someone to check the right group. - New hires need provisioning today; leavers needed it an hour ago. - Every alert gets enriched by hand across six consoles. - The risky work is exactly what a blind script should not touch. ### What you can automate Automations Each of these runs as a flow. The agent does the enrichment and the routine steps on its own. It stops and asks a human at the one point where a wrong move is expensive or irreversible. Every tool named here is a FlowRunner connector, built and verified against the vendor's official API. 1. ### Phishing report triage Reports arrive investigated. The purge waits for a human. 1. Trigger An employee reports a suspicious email, or one lands in the phishing mailbox. 2. Agent The agent pulls the URLs and sender, detonates the links with urlscan.io, runs the indicators through your Cortex analyzers, correlates them against your MISP threat intel, and opens a TheHive case with the enriched observables and a provisional verdict. 3. Result Analysts stop triaging raw reports by hand and open cases that are already investigated, with the destructive step gated behind a human. Run 4400 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/urlscan "urlscan.io")[](https://flowrunner.ai/integrations/cortex "Cortex")[](https://flowrunner.ai/integrations/misp "MISP")[](https://flowrunner.ai/integrations/thehive "TheHive")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Obvious spam it closes on its own. When the verdict is malicious and the blast radius is more than a handful of mailboxes, it stops before it purges the message org-wide or blocks the sender domain, and hands the fully-enriched case to the on-call analyst. 2. ### Access provisioning for joiners Standard access is instant. Privileged access always has a name behind it. 1. Trigger A new hire or an access request arrives from HR, your ITSM, or an intake form. 2. Agent The agent checks the requested role against policy, resolves the right groups in Okta or Microsoft Entra, confirms the manager approval, and provisions standard birthright access. 3. Result Routine access is instant and consistent with policy; privileged access always has a named human behind it, captured in the audit trail. Run 4437 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/okta "Okta")[](https://flowrunner.ai/integrations/entra-id "Microsoft Entra ID")[](https://flowrunner.ai/integrations/ms-graph-security "Microsoft Graph Security")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Standard access goes through automatically. Anything privileged, an admin role, a production group, a security group, stops for an approver who sees the requester, the exact entitlement, and the justification before it is granted. 3. ### Offboarding the moment someone leaves Access is cut in one flow run. The destructive delete still waits. 1. Trigger HR marks a departure, or an offboarding ticket fires. 2. Agent The agent immediately runs the reversible containment: revoke active sessions and OAuth grants, disable SSO, and remove group memberships across Okta, Entra, and Google Workspace. 3. Result The dangerous gap between a departure and cut access closes to the length of one flow run, without a destructive delete ever happening unsupervised. Run 4474 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/okta "Okta")[](https://flowrunner.ai/integrations/entra-id "Microsoft Entra ID")[](https://flowrunner.ai/integrations/google-workspace-admin "Google Workspace Admin")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Full account deletion and mailbox or data reassignment are hard to undo, so the agent stages them and waits for IT to confirm rather than deleting on autopilot. 4. ### Alert triage and escalation Analysts get enriched cases, not raw alerts. Only real incidents page a person. 1. Trigger A detection fires in Splunk, or a scheduled search crosses a threshold. 2. Agent The agent enriches the alert, correlates the entities against recent activity and threat intel, sets a provisional severity, and opens or updates the case. 3. Result The queue of raw alerts becomes a short list of enriched, correlated cases. A real incident pages a person; noise does not. Run 4511 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/splunk "Splunk")[](https://flowrunner.ai/integrations/pagerduty "PagerDuty")[](https://flowrunner.ai/integrations/thehive "TheHive")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint It pages the on-call engineer through PagerDuty only when confidence crosses the bar you set, and it never runs a production containment action until a human acknowledges. 5. ### Access reviews and certification The review compiles itself. The revocations wait for a sign-off. 1. Trigger A scheduled quarterly review, or an ad-hoc audit request. 2. Agent The agent pulls entitlements across Okta, Entra, and Google Workspace, flags the anomalies, dormant admins, orphaned accounts, standing privileged access, and compiles the review package. 3. Result Access reviews stop being a spreadsheet marathon, and the sign-off plus the revocations come out audit-ready. Run 4548 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/okta "Okta")[](https://flowrunner.ai/integrations/entra-id "Microsoft Entra ID")[](https://flowrunner.ai/integrations/securityscorecard "SecurityScorecard")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint It proposes the revocations and waits. A reviewer signs off, and only then does the agent apply them, with every decision and approver written to the log. ### The pattern that makes it safe to automate Security automation fails when a script does something it should not. FlowRunner's answer is the digital andon cord: the agent runs the line and pulls it the instant a step carries real consequence. It enriches, correlates, and stages the routine work on its own. Deactivating an account, blocking a domain, escalating a privilege, or erasing data on request always stops and routes to a person through the channel they already watch, with everything the agent found attached. Reversible containment happens immediately; the irreversible action waits for a human. That is the difference between automation your team trusts with production and automation it does not. [The digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord) ### What your team gets Outcomes #### Close the offboarding gap Access is contained the moment someone leaves, in one flow run, instead of sitting in a ticket queue while a former employee still has a live session. #### End alert fatigue Analysts open enriched, correlated cases instead of raw alerts and reports. The routine triage is done before a human ever looks. #### Consistent, policy-driven access Every grant follows the same policy check and approval path, so provisioning stops depending on who picked up the ticket. #### Audit-ready by default Every access change and every decision is logged with the approver and a timestamp, so access reviews and audits pull from a record that already exists. #### No unsupervised destructive actions An agent never deletes an account, blocks a domain, or purges mail on its own. The consequential move always has a named human behind it. #### No coverage gaps The agent connects to the stack you already run, every connector verified against the vendor API, so there is no tool it cannot reach. ### Built for security's requirements Controls The controls your own team would demand of any automation touching identity and access are native to the platform, not add-ons. #### Complete audit trail Every action, the user or agent that took it, the approver, and the timestamp, recorded and exportable for review. #### RBAC and SSO/SAML Role-based access to the platform itself, with SSO and SAML so your identity provider stays the source of truth. #### Bring your own keys Your model providers, your keys. Inference runs on credentials you control, and you pay the provider directly. #### Self-hosted option Deploy inside your own infrastructure so sensitive identity and security data never leaves your environment. > _Where actually AI actually stops itself and says, okay, I don't know, I'm going to pull up here and I'm going to get guidance._ CEO, automotive services ### The complete SOC and identity stack The stack Identity providers, directories, threat intelligence, case management, detonation, and posture scoring, from Okta and Microsoft Entra to MISP, TheHive, Cortex, and urlscan. Every connector is built and verified against the vendor's official API, so an agent calls it the way the vendor's API actually allows. The tools the automations above run on. [urlscan.io](https://flowrunner.ai/integrations/urlscan "urlscan.io") [Cortex](https://flowrunner.ai/integrations/cortex "Cortex") [MISP](https://flowrunner.ai/integrations/misp "MISP") [TheHive](https://flowrunner.ai/integrations/thehive "TheHive") [Okta](https://flowrunner.ai/integrations/okta "Okta") [Microsoft Entra ID](https://flowrunner.ai/integrations/entra-id "Microsoft Entra ID") [Microsoft Graph Security](https://flowrunner.ai/integrations/ms-graph-security "Microsoft Graph Security") [Google Workspace Admin](https://flowrunner.ai/integrations/google-workspace-admin "Google Workspace Admin") [Splunk](https://flowrunner.ai/integrations/splunk "Splunk") [PagerDuty](https://flowrunner.ai/integrations/pagerduty "PagerDuty") [SecurityScorecard](https://flowrunner.ai/integrations/securityscorecard "SecurityScorecard") [Identity & Security 39 integrations](https://flowrunner.ai/integrations/category/identity-security) [Developer & Infrastructure 152 integrations](https://flowrunner.ai/integrations/category/developer-infrastructure) Every connector verified against each vendor's official API. --- ## FlowRunner for support teams Source: https://flowrunner.ai/teams/support Automate the triage, drafting, and follow-up work that buries your queue, with an agent that runs the routine and a human who owns every reply, escalation, and bulk action that reaches a customer. [Start free](https://app.flowrunner.ai) See the stack The verified stack [Aidbase](https://flowrunner.ai/integrations/aidbase "Aidbase") [Aircall](https://flowrunner.ai/integrations/aircall "Aircall") [Atera](https://flowrunner.ai/integrations/atera "Atera") [Bland AI](https://flowrunner.ai/integrations/bland "Bland AI") [Bonjoro](https://flowrunner.ai/integrations/bonjoro "Bonjoro") [Brosix](https://flowrunner.ai/integrations/brosix "Brosix") [Chatfuel](https://flowrunner.ai/integrations/chatfuel "Chatfuel") [Chatwork](https://flowrunner.ai/integrations/chatwork "Chatwork") [Cisco Webex](https://flowrunner.ai/integrations/webex "Cisco Webex") [CloudTalk](https://flowrunner.ai/integrations/cloudtalk "CloudTalk") [Crisp](https://flowrunner.ai/integrations/crisp "Crisp") [Daktela](https://flowrunner.ai/integrations/daktela "Daktela") [Dialpad](https://flowrunner.ai/integrations/dialpad "Dialpad") [Discord](https://flowrunner.ai/integrations/discord "Discord") [easiware](https://flowrunner.ai/integrations/easiware "easiware") [Facebook Messenger](https://flowrunner.ai/integrations/facebook-messenger "Facebook Messenger") [Feishu Group Robot](https://flowrunner.ai/integrations/feishu-group-robot "Feishu Group Robot") [Fleep](https://flowrunner.ai/integrations/fleep "Fleep") [Flock](https://flowrunner.ai/integrations/flock "Flock") [FreeScout](https://flowrunner.ai/integrations/freescout "FreeScout") [Freshchat](https://flowrunner.ai/integrations/freshchat "Freshchat") [Freshdesk](https://flowrunner.ai/integrations/freshdesk "Freshdesk") [Freshservice](https://flowrunner.ai/integrations/freshservice "Freshservice") [Front](https://flowrunner.ai/integrations/front-service "Front") [Google Chat](https://flowrunner.ai/integrations/google-chat "Google Chat") [Google Meet](https://flowrunner.ai/integrations/google-meet "Google Meet") [Groove](https://flowrunner.ai/integrations/groove "Groove") [GroupMe](https://flowrunner.ai/integrations/groupme "GroupMe") [HaloPSA](https://flowrunner.ai/integrations/halopsa "HaloPSA") [Handwrytten](https://flowrunner.ai/integrations/handwrytten "Handwrytten") [HappyFox](https://flowrunner.ai/integrations/happyfox-help-desk "HappyFox") [HappyFox Chat](https://flowrunner.ai/integrations/happyfox-chat "HappyFox Chat") [Heartbeat](https://flowrunner.ai/integrations/heartbeat "Heartbeat") [Help Scout](https://flowrunner.ai/integrations/help-scout "Help Scout") [Helpwise](https://flowrunner.ai/integrations/helpwise "Helpwise") [Intercom](https://flowrunner.ai/integrations/intercom "Intercom") [Jitbit Helpdesk](https://flowrunner.ai/integrations/jitbit "Jitbit Helpdesk") [Jotform](https://flowrunner.ai/integrations/jotform "Jotform") [Landbot](https://flowrunner.ai/integrations/landbot "Landbot") [Lark Group Bot](https://flowrunner.ai/integrations/larksuitegrouprobot "Lark Group Bot") [LetterXpress](https://flowrunner.ai/integrations/letterxpress "LetterXpress") [Lightr](https://flowrunner.ai/integrations/lightr "Lightr") [LINE](https://flowrunner.ai/integrations/line "LINE") [LiveAgent](https://flowrunner.ai/integrations/liveagent "LiveAgent") [LiveChat](https://flowrunner.ai/integrations/livechat "LiveChat") [Matrix](https://flowrunner.ai/integrations/matrix "Matrix") [Mattermost](https://flowrunner.ai/integrations/mattermost "Mattermost") [Microsoft Teams](https://flowrunner.ai/integrations/microsoft-teams "Microsoft Teams") [Missive](https://flowrunner.ai/integrations/missive "Missive") [Olark](https://flowrunner.ai/integrations/olark "Olark") [Phaxio](https://flowrunner.ai/integrations/phaxio "Phaxio") [PingBell](https://flowrunner.ai/integrations/pingbell "PingBell") [Print.one](https://flowrunner.ai/integrations/print-one "Print.one") [Re:amaze](https://flowrunner.ai/integrations/reamaze "Re:amaze") [Respond.io](https://flowrunner.ai/integrations/respond-io "Respond.io") [RingCentral](https://flowrunner.ai/integrations/ringcentral "RingCentral") [Ringover](https://flowrunner.ai/integrations/ringover "Ringover") [Rocket.Chat](https://flowrunner.ai/integrations/rocketchat "Rocket.Chat") [Ryver](https://flowrunner.ai/integrations/ryver "Ryver") [ServiceNow](https://flowrunner.ai/integrations/servicenow "ServiceNow") [Slack](https://flowrunner.ai/integrations/slack "Slack") [SolarWinds Service Desk](https://flowrunner.ai/integrations/solarwinds "SolarWinds Service Desk") [Stannp](https://flowrunner.ai/integrations/stannp2 "Stannp") [SupportBee](https://flowrunner.ai/integrations/supportbee "SupportBee") [SurveyMonkey](https://flowrunner.ai/integrations/surveymonkey "SurveyMonkey") [SyncroMSP](https://flowrunner.ai/integrations/syncromsp "SyncroMSP") [tawk.to](https://flowrunner.ai/integrations/tawkto "tawk.to") [TeamViewer](https://flowrunner.ai/integrations/teamviewer "TeamViewer") [Teamwork Desk](https://flowrunner.ai/integrations/teamwork-desk "Teamwork Desk") [Telegram](https://flowrunner.ai/integrations/telegram "Telegram") [Thanks.io](https://flowrunner.ai/integrations/thanks-io "Thanks.io") [Thankster](https://flowrunner.ai/integrations/thankster "Thankster") [Tiflux](https://flowrunner.ai/integrations/tiflux "Tiflux") [Trengo](https://flowrunner.ai/integrations/trengo "Trengo") [Twist](https://flowrunner.ai/integrations/twist "Twist") [Typeform](https://flowrunner.ai/integrations/typeform "Typeform") [WeChat](https://flowrunner.ai/integrations/wechat "WeChat") [WhatsApp Business](https://flowrunner.ai/integrations/whatsapp "WhatsApp Business") [Whereby](https://flowrunner.ai/integrations/whereby "Whereby") [Zammad](https://flowrunner.ai/integrations/zammad "Zammad") [Zendesk](https://flowrunner.ai/integrations/zendesk "Zendesk") [Zoho Cliq](https://flowrunner.ai/integrations/zoho-cliq "Zoho Cliq") [Zoho Desk](https://flowrunner.ai/integrations/zoho-desk "Zoho Desk") [Zoho SalesIQ](https://flowrunner.ai/integrations/zoho-salesiq "Zoho SalesIQ") [Zoho TeamInbox](https://flowrunner.ai/integrations/zoho-teaminbox "Zoho TeamInbox") [Zoom](https://flowrunner.ai/integrations/zoom "Zoom") [Zulip](https://flowrunner.ai/integrations/zulip "Zulip") 87 integrations 1,603 verified actions Verified against each vendor's official API ### The queue that always wins Your queue never sleeps. Tickets land in Zendesk, Intercom, and Freshdesk at the same time customers are messaging you on WhatsApp Business, in Slack, and through a Typeform embed. Every one needs to be read, understood, tagged, and pointed at the right person before anyone can actually help. By the time a ticket reaches the agent who can solve it, half the day is gone to sorting, not solving. The tools are not the problem. You already run a helpdesk, a shared inbox, live chat, and three chat channels. The problem is the manual glue between them, and the fact that the work that matters most, the reply a customer reads, the escalation that pulls in an engineer, the bulk close that touches a hundred tickets at once, is exactly the work you cannot hand to a bot that does not know when to stop. - Tickets arrive across Zendesk, Intercom, WhatsApp, and Slack all at once. - Half the day goes to reading and tagging, not solving. - The right ticket reaches the right person late, or not at all. - At-risk accounts sit in the same queue as password resets. - The reply a customer reads is exactly what a bot should not send alone. ### What you can automate Automations Each of these runs as a flow. The agent does the triage, the drafting, and the routine steps on its own. It stops and asks a human at the one point where a wrong move reaches a customer or cannot be pulled back. Every tool named here is a FlowRunner connector, built and verified against the vendor's official API. 1. ### Inbound triage and routing Every ticket arrives read, tagged, and pointed at the right queue. 1. Trigger A ticket lands in Zendesk or Freshdesk, or a message arrives over WhatsApp Business or Slack. 2. Agent The agent reads the message, classifies it by topic and intent, detects the customer's language and sentiment, sets a provisional priority, tags it, and routes it to the matching team or queue in the helpdesk it belongs to. 3. Result Agents open tickets that are already read, tagged, and routed, so the day starts with solving instead of sorting. Run 4400 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/zendesk "Zendesk")[](https://flowrunner.ai/integrations/freshdesk "Freshdesk")[](https://flowrunner.ai/integrations/whatsapp "WhatsApp Business")[](https://flowrunner.ai/integrations/slack "Slack")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint Clear, routine tickets it routes on its own. When intent is ambiguous, the sentiment reads as angry, or the classifier is unsure, it stops before assigning and hands the tagged ticket to a lead to place, rather than guessing and burying it in the wrong queue. 2. ### Drafted reply, human sends The agent writes the answer. A person approves it before the customer sees it. 1. Trigger A triaged ticket in Intercom or Help Scout matches a known question or a documented resolution. 2. Agent The agent pulls the ticket history and the relevant knowledge, drafts a complete, on-brand reply in the customer's language, cites the article or macro it drew from, and attaches the draft to the ticket for review. 3. Result Agents answer from a finished draft instead of a blank box, and every reply a customer reads was approved by a person. Run 4437 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/intercom "Intercom")[](https://flowrunner.ai/integrations/help-scout "Help Scout")[](https://flowrunner.ai/integrations/zendesk "Zendesk")[](https://flowrunner.ai/integrations/freshdesk "Freshdesk")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint It never sends. The draft waits in Intercom or Help Scout for the assigned agent, who reads it, edits a line if they want, and clicks send. Nothing customer-facing leaves without a human behind it. 3. ### Escalate the at-risk ticket A real emergency reaches a named person. Noise does not. 1. Trigger A ticket crosses a severity threshold, breaches an SLA, or comes from a flagged account in ServiceNow or Freshservice. 2. Agent The agent assembles the full picture, the account, the history, the SLA clock, the sentiment trend, drafts a concise escalation summary, and prepares to route it to the on-call lead over Microsoft Teams or Slack. 3. Result The tickets that actually need a human's attention surface fast, with the context already gathered, and no one gets paged for a routine question. Run 4474 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/servicenow "ServiceNow")[](https://flowrunner.ai/integrations/freshservice "Freshservice")[](https://flowrunner.ai/integrations/microsoft-teams "Microsoft Teams")[](https://flowrunner.ai/integrations/slack "Slack")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint It stops before paging a person. Escalating to a named engineer or a customer's account owner pulls people out of their work, so the agent presents the summary and waits for a lead to confirm the escalation and the recipient rather than paging on a hunch. 4. ### Collect and route feedback CSAT lands, gets read, and reaches the right team the same day. 1. Trigger A customer submits a survey or feedback form through Typeform, Jotform, or SurveyMonkey after a ticket closes. 2. Agent The agent reads each response, scores the sentiment, links it back to the originating ticket and account, tags the theme, and files it so the pattern is visible instead of buried in a spreadsheet. 3. Result Feedback stops dying in a form export. The good signal informs the team, and the at-risk score reaches a person while there is still time to act. Run 4511 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/typeform "Typeform")[](https://flowrunner.ai/integrations/jotform "Jotform")[](https://flowrunner.ai/integrations/surveymonkey "SurveyMonkey")[](https://flowrunner.ai/integrations/slack "Slack")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint A low score or an angry comment is a save-the-account moment, so before it opens a follow-up ticket and pulls in a CX lead over Slack, it stops for a human to confirm the outreach and who owns it. 5. ### Bulk action across a triaged batch The whole batch is prepared in one flow. The mass change waits for a sign-off. 1. Trigger A known issue is resolved, a queue is misassigned, or a recurring topic needs a macro applied across many tickets at once. 2. Agent The agent identifies every affected ticket across Zendesk and Freshdesk, groups them, drafts the reassignment, tag, or macro to apply, and compiles the batch with a preview of exactly which tickets change and how. 3. Result Cleaning up a queue or pushing a known-issue macro stops being an afternoon of manual clicking, and the mass action never fires without a named human behind it. Run 4548 in progress 1. Trigger fired 2. Agent [](https://flowrunner.ai/integrations/zendesk "Zendesk")[](https://flowrunner.ai/integrations/freshdesk "Freshdesk")[](https://flowrunner.ai/integrations/servicenow "ServiceNow")[](https://flowrunner.ai/integrations/front-service "Front")done 3. Human checkpoint waiting 4. Result pending Paused. A person was notified in Slack and by email, with the full context attached. Human checkpoint A bulk reassignment or a mass close touches a hundred customers at once and is painful to reverse, so the agent stages the batch and stops. A lead reviews the preview and approves, and only then does the agent apply the change, with the batch and the approver written to the log. ### The pattern that makes it safe to automate Support automation fails when a bot sends something it should not, to someone it should not. FlowRunner's answer is the digital andon cord: the agent runs the line and pulls it the instant a step reaches a customer or cannot be undone. It reads, classifies, drafts, and stages the routine work on its own. Sending a customer-facing reply, escalating to a named person, reassigning or closing a batch of tickets, or opening a save-the-account follow-up always stops and routes to a person through the channel they already watch, with everything the agent gathered attached. The routine triage happens instantly; the consequential move waits for a human. That is the difference between automation your customers never feel as a downgrade and automation that erodes the trust you spent years building. [The digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord) ### What your team gets Outcomes #### Open solved-ready tickets Every ticket arrives read, tagged, and routed, so agents spend the day resolving instead of sorting a queue by hand. #### Answer from a draft, not a blank box The agent writes the on-brand reply and cites its source; the agent reads, edits, and sends. Faster answers, still human. #### Catch the at-risk account in time SLA breaches, angry sentiment, and low CSAT surface fast with context attached, so a save reaches a person while it still matters. #### One agent across every channel Helpdesks, shared inboxes, live chat, and messaging apps run through the same flow, so a customer gets the same handling wherever they wrote from. #### No customer-facing move on autopilot An agent never sends a reply, escalates, or mass-changes tickets on its own. The move a customer feels always has a named human behind it. #### No coverage gaps The agent connects to the stack you already run, every connector verified against the vendor API, so there is no channel or helpdesk it cannot reach. ### Built for support's requirements Controls The controls a serious support organization needs before it lets automation touch a customer conversation are native to the platform, not add-ons. #### Complete audit trail Every reply, escalation, and bulk action, the agent or user behind it, the approver, and the timestamp, recorded and exportable when a customer or a manager asks what happened. #### RBAC and SSO/SAML Role-based access to the platform, with SSO and SAML, so a junior agent can draft while only a lead can approve a mass close or a customer credit. #### Bring your own keys Your model providers, your keys. The reasoning that reads and drafts customer messages runs on credentials you control, and you pay the provider directly. #### Self-hosted option Deploy inside your own infrastructure so customer conversations and personal data never leave your environment. > _I'm only getting involved on the exceptions._ Director of Operations, beauty brand ### The complete support and messaging stack The stack Helpdesks, ITSM platforms, shared inboxes, live chat, team chat, and the messaging apps customers actually write from, from Zendesk, Freshdesk, and ServiceNow to Intercom, Front, WhatsApp Business, and Slack. Every connector is built and verified against the vendor's official API, so an agent calls it the way the vendor's API actually allows. The tools the automations above run on. [Zendesk](https://flowrunner.ai/integrations/zendesk "Zendesk") [Freshdesk](https://flowrunner.ai/integrations/freshdesk "Freshdesk") [WhatsApp Business](https://flowrunner.ai/integrations/whatsapp "WhatsApp Business") [Slack](https://flowrunner.ai/integrations/slack "Slack") [Intercom](https://flowrunner.ai/integrations/intercom "Intercom") [Help Scout](https://flowrunner.ai/integrations/help-scout "Help Scout") [ServiceNow](https://flowrunner.ai/integrations/servicenow "ServiceNow") [Freshservice](https://flowrunner.ai/integrations/freshservice "Freshservice") [Microsoft Teams](https://flowrunner.ai/integrations/microsoft-teams "Microsoft Teams") [Typeform](https://flowrunner.ai/integrations/typeform "Typeform") [Jotform](https://flowrunner.ai/integrations/jotform "Jotform") [SurveyMonkey](https://flowrunner.ai/integrations/surveymonkey "SurveyMonkey") [Front](https://flowrunner.ai/integrations/front-service "Front") [Helpdesk & ITSM 27 integrations](https://flowrunner.ai/integrations/category/helpdesk-itsm) [Communication & Messaging 205 integrations](https://flowrunner.ai/integrations/category/communication-messaging) Every connector verified against each vendor's official API. --- ## Terms of Service Source: https://flowrunner.ai/terms Date of Last Revision: February 28, 2026 ### 1\. Introduction and Acceptance of Terms Welcome to FlowRunner, a visual workflow automation platform with AI agent orchestration capabilities operated by Midnight Coders, Inc. ("Company," "we," "us," or "our"), a Texas corporation. These Terms of Service ("Terms") govern your access to and use of the FlowRunner platform, including all associated services, features, content, applications, and APIs (collectively, the "Service") available at flowrunner.ai and any related domains. By accessing or using the Service, creating an account, or clicking "I Agree" (or similar affirmation), you ("Customer," "you," or "your") agree to be bound by these Terms. If you are entering into these Terms on behalf of a company or other legal entity, you represent that you have the authority to bind that entity to these Terms, in which case "Customer," "you," or "your" refers to that entity. **If you do not agree to these Terms, you may not access or use the Service.** #### 1.1 Relationship to Other Agreements If you have entered into a separate Master Service Agreement ("MSA"), Enterprise Agreement, or other written agreement with Midnight Coders, Inc. governing your use of the Service, the terms of that agreement will control to the extent they conflict with these Terms. For all other matters not addressed in your separate agreement, these Terms apply. #### 1.2 Changes to These Terms We may update these Terms from time to time. If we make material changes, we will notify you by email (to the address associated with your account) or by posting a prominent notice on the Service at least thirty (30) days before the changes take effect. Your continued use of the Service after the effective date of any changes constitutes your acceptance of the updated Terms. If you do not agree to the updated Terms, you must stop using the Service before the changes take effect. We will not retroactively apply material changes to disputes or events that occurred before the updated Terms became effective. ### 2\. Definitions The following definitions apply throughout these Terms: **"Account"** means a registered FlowRunner account created by or on behalf of a Customer. **"Affiliate"** means any entity that directly or indirectly controls, is controlled by, or is under common control with a party, where "control" means ownership of more than 50% of the voting securities. **"AI Agent"** means an artificial intelligence component configured within the Service that can execute tasks, make decisions within defined parameters, and interact with other components of a Workflow. **"Authorized User"** means any individual who is authorized by Customer to access and use the Service under Customer's Account. **"BYOK"** or **"Bring Your Own Keys"** means the model under which Customer provides their own API keys, credentials, or access tokens for third-party AI services (such as OpenAI, Anthropic, Google, and others) to be used within the Service. **"Cloud Deployment"** means hosting and operation of the Service by the Company on Company-managed infrastructure. **"Community Edition"** means the free, self-hosted version of the Service with the feature set described on our website. **"Customer Content"** means all data, information, text, files, images, workflows, configurations, AI Agent settings, MCP server configurations, and other materials that Customer or its Authorized Users upload to, create within, transmit through, or store in the Service. **"Documentation"** means the user guides, help articles, API references, and other technical materials we make available for the Service at docs.flowrunner.ai or successor URLs. **"Execution"** means one complete run of a Workflow from start to finish, regardless of the number of individual steps or nodes within the Workflow. **"Human-in-Loop"** means the Service's capability to pause an automated Workflow and solicit input, approval, or decisions from a human via email, Slack, WhatsApp, telephone, or other supported communication channels before resuming Workflow execution. **"MCP"** or **"Model Context Protocol"** means the open protocol standard for tool integration that allows AI Agents to interact with external services and tools. **"Midnight Flow"** means the consulting services division operated by or affiliated with the Company, providing implementation, optimization, and compliance consulting services related to the Service. **"Self-Hosted Deployment"** means Customer's installation and operation of the Service on Customer-managed infrastructure. **"Service Level Agreement"** or **"SLA"** means any uptime or performance commitments set forth in a separate written agreement between Customer and the Company. **"Subscription Plan"** means the specific tier of Service (Growth, Professional, Business, Enterprise, or Community Edition) selected by or assigned to the Customer, together with the features, limitations, and pricing applicable to that tier. **"Third-Party Service"** means any software, platform, API, or service not operated by the Company that Customer integrates with or accesses through the Service, including AI model providers, communication platforms, and enterprise applications. **"Workflow"** means an automated sequence of steps, actions, triggers, conditions, and AI Agent operations created by Customer within the Service. ### 3\. Account Registration and Access #### 3.1 Registration Requirements To create an Account, you must provide a valid email address that you control. We do not accept registrations from disposable or temporary email providers. We reserve the right to reject registrations that we reasonably determine are fraudulent, duplicative, or otherwise not legitimate. You represent and warrant that: (a) all information you provide during registration is truthful, accurate, and complete; (b) you will maintain the accuracy of such information; (c) you are at least 18 years of age; and (d) your use of the Service does not violate any applicable law, regulation, or these Terms. #### 3.2 Account Security You are responsible for maintaining the confidentiality of your Account credentials, including passwords and API keys. You are responsible for all activities that occur under your Account, whether or not authorized by you. You must notify us immediately at [security@flowrunner.ai](mailto:security@flowrunner.ai) if you become aware of any unauthorized use of your Account. We are not liable for any loss or damage arising from your failure to secure your Account credentials. #### 3.3 Authorized Users All Subscription Plans include unlimited Authorized Users. You are responsible for ensuring that all Authorized Users comply with these Terms. Any violation of these Terms by an Authorized User is deemed a violation by you. #### 3.4 Free Plan and Trial Each Account may hold one workspace on the Free Subscription Plan at no charge and with no payment method required. The Free Subscription Plan is limited to one workspace per Account. Attempts to obtain additional Free workspaces through multiple registrations or other means constitute a violation of these Terms. The first workspace created by a new Account is placed on a fourteen (14) day trial of the Professional Subscription Plan (the "Trial"). No payment method is required to start a Trial. One Trial is granted per Account; deleting an Account and re-registering with the same email address does not earn a second Trial. Workflow export and workspace transfer are not available while a workspace is on Trial. If no payment method is on file when the Trial ends, the workspace moves to the Free Subscription Plan and Workflows continue to run within the Free plan limits. No Workflow or Customer Content is paused or deleted at Trial expiry. Selecting any paid Subscription Plan during the Trial ends the Trial immediately and begins billing for the selected plan. Accounts that received a $100 USD trial credit under the previous trial program keep their remaining credit balance, subject to the terms under which it was granted. We reserve the right to modify, suspend, or discontinue the Free Subscription Plan or the Trial program at any time without prior notice. ### 4\. Subscription Plans, Fees, and Payment #### 4.1 Subscription Plans The Service is offered through several Subscription Plans, each with different features, execution limits, and pricing as described on our [pricing page](https://flowrunner.ai/pricing). Current plans include Free ($0/month), Starter ($5 or $15/month depending on execution volume), Growth ($45 to $149/month depending on execution volume), Professional ($299/month), Business ($999/month), and Enterprise (custom pricing). The Community Edition is available for free self-hosted deployment with limited features. All cloud-hosted Subscription Plans include unlimited Workflows and unlimited Authorized Users. Feature availability, execution limits, concurrent execution limits, the period for which execution history is visible in the Service, export capabilities, and support levels vary by Subscription Plan as described on our pricing page. Bulk workspace export and programmatic export are available on the Growth Subscription Plan and above; single-Workflow export is available on every plan except while a workspace is on Trial. #### 4.2 Execution Limits and Enforcement Each Subscription Plan includes a monthly allocation of Executions. An "Execution" means one complete run of a Workflow from start to finish, regardless of the number of steps within the Workflow. **When you reach 100% of your monthly Execution limit, all Workflows will be paused immediately.** Workflows will resume automatically at the start of your next billing cycle or upon upgrading to a higher Subscription Plan. We do not charge overage fees; instead, the Service enforces a hard limit on Executions per billing period. We will provide in-application and email notifications as you approach your Execution limit (at approximately 80%, 90%, and 95% of your allocation). These notifications are provided as a courtesy and their absence does not relieve the Execution limit. All Customer Content, Workflow configurations, and Account data are preserved during any pause due to Execution limit enforcement. #### 4.3 Workflow Running Time Limits Each Workflow may contain any number of steps. However, the Service enforces limits on the cumulative "Running Time" of a Workflow Execution. Running Time is calculated as the sum of processing time consumed by individual action executions within the Workflow, including action blocks, transformer blocks, condition blocks, looping blocks, and grouping blocks. Running Time specifically excludes "Waiting Time," which is any period during which a Workflow is suspended and not actively consuming processing resources. Waiting Time includes, without limitation, time spent waiting for Human-in-Loop responses, external webhook callbacks, scheduled delay timers, or other event-driven triggers. Waiting Time has no upper limit — a Workflow may remain in a waiting state indefinitely without consuming Running Time. Running Time limits vary by Subscription Plan as described on our pricing page and in the Documentation. When a Workflow Execution exceeds its applicable Running Time limit, the Execution will be terminated. The Company may update Running Time limits from time to time; any reduction in limits applicable to existing Subscription Plans will be communicated with at least thirty (30) days' notice. #### 4.4 Fees and Payment Fees for the Service are as stated on our pricing page at the time of your subscription or renewal. All fees are quoted in United States Dollars (USD) and are exclusive of all applicable taxes, which you are responsible for paying. Subscription fees are billed monthly in advance unless you have elected annual billing or have separate invoicing arrangements. Payment is due on the date of invoice. A valid payment method on file is required for any paid Subscription Plan, including when moving between paid plans; the Free Subscription Plan never requires one. We accept payment by major credit card (Visa, MasterCard, American Express, Discover), ACH transfer (Business and Enterprise tiers), and invoice billing (Enterprise tier only). If your payment method fails or your account is past due, we may suspend access to the Service until payment is received. We reserve the right to charge interest on overdue amounts at the lesser of 2.1% per month or the maximum rate permitted by applicable law. #### 4.5 Price Changes We may change our pricing at any time. For existing Customers on active Subscription Plans, price changes will take effect at the start of your next billing cycle following at least thirty (30) days' written notice. If you do not agree to a price change, you may cancel your subscription before the new pricing takes effect. #### 4.6 No Refunds All subscription fees are non-refundable. Because the Company provides a permanent Free Subscription Plan and a fourteen (14) day Trial of the Professional Subscription Plan to every new Account, at no cost and with no payment method required, all payments for a paid Subscription Plan represent a deliberate commitment to the Service following a meaningful opportunity to evaluate it. You are responsible for canceling your subscription, or moving your workspace to the Free Subscription Plan, before your next billing cycle if you do not wish to continue. Enterprise Subscription Plans governed by a separate written agreement may contain different refund or termination-for-convenience terms as specified in that agreement. ### 5\. Permitted Use and Restrictions #### 5.1 License Grant Subject to your compliance with these Terms and payment of applicable fees, we grant you a limited, non-exclusive, non-transferable, non-sublicensable, revocable license to access and use the Service during the term of your subscription solely for your internal business purposes, in accordance with the Documentation and your Subscription Plan. #### 5.2 Acceptable Use You agree to use the Service only for lawful purposes and in compliance with all applicable laws and regulations. Without limiting the foregoing, you agree not to: (a) Use the Service to violate any applicable law, regulation, or third-party right, including without limitation data protection, privacy, export control, or anti-spam laws; (b) Use the Service to process, store, or transmit any data that you do not have the right to process, store, or transmit; (c) Use the Service to send unsolicited or unauthorized communications, advertising, or spam through Human-in-Loop or other communication features; (d) Attempt to gain unauthorized access to the Service, other accounts, computer systems, or networks connected to the Service; (e) Interfere with or disrupt the integrity or performance of the Service or the data contained therein; (f) Reverse engineer, decompile, disassemble, or otherwise attempt to discover the source code of the Service, except to the extent expressly permitted by applicable law; (g) Remove, alter, or obscure any proprietary notices or labels on the Service; (h) Use the Service to develop a competing product or service; (i) Resell, sublicense, or make the Service available to third parties as a managed service, except as expressly authorized in writing by the Company; (j) Use the Service to build Workflows or AI Agents intended to deceive, defraud, or harm any person; (k) Circumvent or attempt to circumvent any Execution limits, rate limits, or other technical restrictions of the Service; (l) Use the Service in a manner that exceeds reasonable request volume or constitutes excessive or abusive usage that degrades the Service for other customers. #### 5.3 Customer Responsibility for Workflows You are solely responsible for the Workflows, AI Agents, and automations you create, configure, and deploy using the Service. This includes responsibility for: (a) The accuracy, legality, and appropriateness of all Customer Content processed by your Workflows; (b) Ensuring your Workflows comply with all applicable laws and regulations, including but not limited to HIPAA, SOX, GDPR, CCPA, and other data protection and industry-specific regulations; (c) Obtaining all necessary consents, authorizations, and permissions before processing any personal data, protected health information (PHI), or other regulated data through the Service; (d) Configuring appropriate Human-in-Loop checkpoints for decisions requiring human judgment, particularly in regulated or high-stakes contexts; (e) Testing and validating your Workflows before deploying them in production environments. ### 6\. AI Agents, BYOK Model, and Third-Party AI Services #### 6.1 Bring Your Own Keys (BYOK) The Service operates on a Bring Your Own Keys (BYOK) model. To use AI Agent features, you must provide your own API keys, credentials, or access tokens for supported third-party AI services (e.g., OpenAI, Anthropic, Google, Cohere, and others). The Company does not provide, manage, or subsidize AI service API keys except where explicitly included in an Enterprise Subscription Plan. You are solely responsible for: (a) Obtaining, maintaining, and securing your third-party AI service API keys; (b) All costs, fees, and charges incurred through your use of third-party AI services via your API keys, including usage fees charged by AI model providers; (c) Compliance with the terms of service, usage policies, and acceptable use policies of the third-party AI services you integrate; (d) Monitoring your usage and spending with third-party AI providers. #### 6.2 No Guarantee of AI Outputs **The Company does not guarantee the accuracy, completeness, reliability, suitability, or safety of any output generated by AI Agents within the Service.** AI model outputs are generated by third-party AI services using your API keys and are influenced by factors entirely outside the Company's control, including but not limited to the AI model's training data, configuration parameters, prompt content, and the AI provider's service performance. You acknowledge and agree that: (a) AI Agent outputs may contain errors, inaccuracies, biases, or inappropriate content; (b) AI Agent outputs should not be relied upon as the sole basis for decisions with legal, financial, medical, or other significant consequences without appropriate human review; (c) The Company has no control over and assumes no responsibility for the content or quality of AI model outputs; (d) You are solely responsible for reviewing, validating, and approving AI Agent outputs before acting on them or incorporating them into downstream processes. #### 6.3 Model Context Protocol (MCP) Integration The Service supports the Model Context Protocol (MCP), allowing you to register and connect your own MCP servers to extend AI Agent capabilities. When you register an MCP server: (a) You are solely responsible for the security, availability, and proper functioning of your MCP servers; (b) You are responsible for all data transmitted to and from your MCP servers through the Service; (c) The Company is not responsible for any actions taken by AI Agents through your MCP server connections; (d) You represent and warrant that your MCP servers comply with all applicable laws and do not infringe upon the rights of any third party. ### 7\. Human-in-Loop Communications The Service includes Human-in-Loop capabilities that allow Workflows and AI Agents to pause execution and solicit input from designated humans via email, Slack, WhatsApp, telephone, or other supported communication channels. #### 7.1 Customer Responsibilities When using Human-in-Loop features, you are responsible for: (a) Designating appropriate human recipients for Human-in-Loop requests, including ensuring those individuals have the authority, qualifications, and training to make the decisions requested of them; (b) Obtaining consent from individuals who will receive Human-in-Loop communications; (c) Ensuring Human-in-Loop communications comply with all applicable laws, including anti-spam legislation (CAN-SPAM, TCPA, GDPR communication requirements, and similar laws); (d) The content and accuracy of any decisions, approvals, or input provided by humans in response to Human-in-Loop requests; (e) Configuring appropriate timeouts and fallback procedures for scenarios where human responses are not received in a timely manner. #### 7.2 Company Disclaimers The Company facilitates the delivery of Human-in-Loop communications but does not control, review, or guarantee: (a) The delivery, receipt, or timeliness of communications sent through third-party channels (email, Slack, WhatsApp, telephone); (b) The accuracy, appropriateness, or completeness of human responses received through Human-in-Loop interactions; (c) The qualifications or authority of the humans designated to receive Human-in-Loop requests; (d) The availability or uptime of third-party communication channels used for Human-in-Loop delivery. ### 8\. Customer Content and Data #### 8.1 Ownership As between the parties, you retain all right, title, and interest in and to your Customer Content. The Company does not claim ownership of Customer Content. #### 8.2 License to Customer Content You grant the Company a limited, non-exclusive, worldwide, royalty-free license to access, use, process, copy, store, transmit, and display Customer Content solely as necessary to provide, maintain, and improve the Service, to provide customer support, and to comply with applicable law. This license terminates when you delete your Customer Content or close your Account, subject to any backup retention periods described below. #### 8.3 Customer Content Responsibilities You are solely responsible for the accuracy, quality, integrity, legality, reliability, and appropriateness of all Customer Content. You represent and warrant that you have all rights, licenses, consents, and permissions necessary to submit Customer Content to the Service and to grant the license set forth in Section 8.2. #### 8.4 Data Retention and Deletion Upon termination or expiration of your subscription, your Customer Content will be retained in a paused state for thirty (30) days, during which time you may reactivate your Account by adding a valid payment method, export your Customer Content, or request deletion. After the thirty-day retention period, we may permanently delete your Customer Content from our active systems. Residual copies may persist in encrypted backups for up to ninety (90) days following deletion from active systems, after which they will be purged. #### 8.5 Data Processing and Security We process Customer Content in accordance with our [Privacy Policy](https://flowrunner.ai/privacy). We implement reasonable administrative, technical, and physical safeguards designed to protect Customer Content from unauthorized access, disclosure, alteration, and destruction. **Important:** The Service is a workflow automation platform, not a data storage or data warehousing service. While the Service necessarily stores Customer Content during Workflow execution and for log retention purposes, long-term storage of sensitive data within the Service is at your own risk and discretion. #### 8.6 Sensitive and Regulated Data If you use the Service to process protected health information ("PHI") as defined under HIPAA, personally identifiable financial information, or other categories of regulated data, you are responsible for: (a) Executing any additional agreements required by law (such as a Business Associate Agreement for HIPAA) with the Company prior to processing such data. Contact [legal@flowrunner.ai](mailto:legal@flowrunner.ai) to initiate a BAA; (b) Configuring the Service appropriately for the type of data being processed, including selecting a Subscription Plan with adequate compliance features (audit trails, RBAC, etc.); (c) Ensuring that your use of the Service complies with all applicable data protection, privacy, and security regulations; (d) Conducting your own risk assessments and due diligence regarding the suitability of the Service for your specific compliance requirements. The availability of compliance features (audit trails, RBAC, SLA tracking, SSO, compliance reporting) varies by Subscription Plan. You are responsible for selecting a Subscription Plan that meets your compliance requirements. #### 8.7 Data Processing Agreement To the extent the Company processes Customer Personal Data subject to the European General Data Protection Regulation, the UK General Data Protection Regulation, the California Consumer Privacy Act, the Texas Data Privacy and Security Act, or other applicable data protection legislation, the [Data Processing Agreement](https://flowrunner.ai/data-processing) is incorporated into and forms part of this Agreement. ### 9\. Third-Party Services and Integrations The Service allows you to integrate with and connect to Third-Party Services, including but not limited to AI model providers, communication platforms (Slack, WhatsApp, email services), enterprise applications (ERP systems, CRM systems, practice management software), and custom APIs. #### 9.1 Customer Responsibility Your use of Third-Party Services through the Service is governed by your separate agreements with those third-party providers. You are solely responsible for: (a) Compliance with the terms of service and acceptable use policies of all Third-Party Services you connect to the Service; (b) The security of credentials, API keys, and access tokens for Third-Party Services stored within or used through the Service; (c) Any data transmitted to or received from Third-Party Services through your Workflows; (d) Any costs, fees, or charges imposed by Third-Party Services as a result of your use through the Service. #### 9.2 No Endorsement or Warranty The availability of integrations with Third-Party Services does not constitute an endorsement, recommendation, or warranty by the Company of those services. The Company makes no representations or warranties regarding the availability, accuracy, security, or reliability of any Third-Party Service. #### 9.3 Third-Party Service Disruptions The Company is not responsible for any disruption, modification, or discontinuation of any Third-Party Service, or for any impact such events may have on your Workflows. You are responsible for monitoring the status of Third-Party Services integrated with your Workflows and for configuring appropriate error handling and fallback procedures. ### 10\. Intellectual Property #### 10.1 Company IP The Service, including its proprietary orchestration engine, source code, object code, algorithms, architecture, user interface, design, Documentation, and all related intellectual property, are and remain the exclusive property of the Company and its licensors. These Terms do not grant you any rights to the Company's intellectual property except the limited license expressly set forth in Section 5.1. #### 10.2 Customer IP As between the parties, you retain all intellectual property rights in your Customer Content, Workflows, and AI Agent configurations. The Company does not acquire any ownership rights in Customer Content by virtue of these Terms. #### 10.3 Feedback If you provide the Company with suggestions, feedback, enhancement requests, recommendations, or other input regarding the Service ("Feedback"), you grant the Company a non-exclusive, perpetual, irrevocable, royalty-free, worldwide license to use, modify, and incorporate that Feedback into the Service and our other products and services without obligation to you. #### 10.4 Usage Data The Company may collect and use aggregated, anonymized, and de-identified data derived from your use of the Service ("Usage Data") for purposes of improving, developing, and operating the Service and our other products and services. Usage Data does not include Customer Content or any data that identifies or could be used to identify you or any individual. ### 11\. Compliance Features and Regulatory Considerations The Service offers built-in compliance features on certain Subscription Plans, including audit trail logging, role-based access control (RBAC), SLA tracking and monitoring, SSO/SAML authentication, and compliance reporting tools. #### 11.1 Shared Responsibility Model Compliance with applicable laws and regulations is a shared responsibility: **Company Responsibility:** We provide the technical capabilities, tools, and infrastructure features described in our Documentation. We maintain security practices appropriate for the Service and cooperate reasonably with Customer compliance efforts. **Customer Responsibility:** You are responsible for configuring and using the compliance features appropriately, determining whether the Service meets your specific regulatory requirements, and ensuring your overall use of the Service complies with applicable laws and regulations. #### 11.2 No Compliance Guarantee **The Company does not guarantee that use of the Service, including its compliance features, will make you compliant with any particular law, regulation, or standard (including, without limitation, HIPAA, SOX, GDPR, CCPA, PCI-DSS, or any other regulatory framework).** The compliance features are tools designed to support your compliance efforts, not substitutes for your own compliance programs, legal counsel, and regulatory analysis. #### 11.3 Audit Trail and Log Retention The period for which execution history and audit trail data are visible in the Service varies by Subscription Plan (24 hours on Free and Starter, 7 days on Growth, 30 days on Professional, 90 days on Business, and unlimited on Enterprise). Data outside the visible window is not deleted immediately. It is retained for the periods stated in Section 6 of the Privacy Policy (7 days on Free and Starter, 30 days on Growth, 90 days on Professional and Business, and as configured on Enterprise) and becomes visible again if the workspace moves to a Subscription Plan whose window covers it. You are responsible for exporting or archiving audit data if your compliance obligations require retention beyond the period included in your Subscription Plan. ### 12\. Deployment Options #### 12.1 Cloud Deployment For Cloud Deployment customers, the Company hosts and operates the Service on infrastructure provided by DigitalOcean and related service providers. The Company is responsible for the operational availability, maintenance, security patching, and infrastructure management of the Cloud Deployment environment. #### 12.2 Self-Hosted Deployment For Self-Hosted Deployment customers (Community Edition or Enterprise self-hosted): (a) You are solely responsible for provisioning, securing, maintaining, and operating the infrastructure on which the Service is installed; (b) You are responsible for installing updates, security patches, and new versions of the Service; (c) The Company's uptime commitments, if any, do not apply to Self-Hosted Deployments; (d) The Company is not responsible for data loss, security breaches, or performance issues arising from your infrastructure, network, or operational practices; (e) Community Edition support is limited to community forums and GitHub issues. Enterprise self-hosted customers receive support as defined in their Enterprise agreement. #### 12.3 Sub-Processors For Cloud Deployment, the Company uses the following sub-processors as of the effective date of these Terms: (a) **DigitalOcean** — Cloud infrastructure hosting and compute services; (b) **MongoDB** — Database services; (c) **Redis** — Caching and session management. We may update our list of sub-processors from time to time. If we add a new sub-processor that processes Customer Content, we will provide at least thirty (30) days' notice by updating the sub-processor list on our website. If you object to a new sub-processor, you may terminate your subscription by providing written notice within thirty (30) days of being notified. ### 13\. Confidentiality Each party ("Receiving Party") agrees that all non-public information disclosed by the other party ("Disclosing Party") that is designated as confidential or that the Receiving Party reasonably should understand to be confidential given the nature of the information and circumstances of disclosure ("Confidential Information") will be kept confidential by the Receiving Party and will not be disclosed to any third party except as necessary to perform obligations under these Terms and only to those who are bound by confidentiality obligations at least as protective as those herein. Confidential Information does not include information that: (a) is or becomes publicly available through no fault of the Receiving Party; (b) was already known to the Receiving Party without restriction prior to disclosure; (c) is independently developed by the Receiving Party without use of or reference to the Disclosing Party's Confidential Information; or (d) is rightfully received from a third party without restriction. Notwithstanding the foregoing, a Receiving Party may disclose Confidential Information to the extent required by law, regulation, or court order, provided that the Receiving Party gives the Disclosing Party prompt written notice (to the extent legally permitted) and cooperates with the Disclosing Party's efforts to obtain protective treatment for the information. The confidentiality obligations under this section will survive for three (3) years following the termination or expiration of these Terms, except that obligations regarding trade secrets will survive for as long as the information remains a trade secret under applicable law. ### 14\. Warranties and Disclaimers #### 14.1 Company Warranties The Company warrants that: (a) the Service will perform materially in accordance with the Documentation during the term of your subscription; (b) the Company will provide the Service in a professional and workmanlike manner consistent with generally accepted industry standards; and (c) the Company has the right and authority to enter into these Terms and grant the licenses contemplated herein. #### 14.2 Disclaimer of Warranties EXCEPT FOR THE EXPRESS WARRANTIES SET FORTH IN SECTION 14.1, THE SERVICE IS PROVIDED "AS IS" AND "AS AVAILABLE" WITHOUT WARRANTY OF ANY KIND, WHETHER EXPRESS, IMPLIED, STATUTORY, OR OTHERWISE. THE COMPANY SPECIFICALLY DISCLAIMS ALL IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE, AND NON-INFRINGEMENT, AND ALL WARRANTIES ARISING FROM COURSE OF DEALING, USAGE, OR TRADE PRACTICE. WITHOUT LIMITING THE FOREGOING, THE COMPANY MAKES NO WARRANTY THAT: (a) THE SERVICE WILL MEET YOUR SPECIFIC REQUIREMENTS OR EXPECTATIONS; (b) THE SERVICE WILL BE UNINTERRUPTED, TIMELY, SECURE, OR ERROR-FREE; (c) THE RESULTS OBTAINED THROUGH USE OF THE SERVICE, INCLUDING AI AGENT OUTPUTS, WILL BE ACCURATE, RELIABLE, OR COMPLETE; (d) ANY ERRORS IN THE SERVICE WILL BE CORRECTED; (e) THE SERVICE WILL BE COMPATIBLE WITH YOUR HARDWARE, SOFTWARE, OR THIRD-PARTY SERVICES; (f) THE SERVICE WILL ENSURE COMPLIANCE WITH ANY PARTICULAR LAW, REGULATION, OR STANDARD. #### 14.3 Customer Warranties You represent and warrant that: (a) you have the legal capacity and authority to enter into these Terms; (b) you will use the Service in compliance with all applicable laws and regulations; (c) you have obtained all necessary consents, authorizations, and permissions to submit Customer Content and process data through the Service; and (d) your use of the Service will not infringe upon or misappropriate the rights of any third party. ### 15\. Limitation of Liability #### 15.1 Exclusion of Consequential Damages TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, IN NO EVENT WILL EITHER PARTY BE LIABLE TO THE OTHER PARTY FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, PUNITIVE, OR EXEMPLARY DAMAGES, INCLUDING BUT NOT LIMITED TO DAMAGES FOR LOSS OF PROFITS, REVENUE, GOODWILL, USE, DATA, OR OTHER INTANGIBLE LOSSES, ARISING OUT OF OR RELATED TO THESE TERMS OR THE USE OF OR INABILITY TO USE THE SERVICE, REGARDLESS OF WHETHER SUCH DAMAGES WERE FORESEEABLE AND WHETHER OR NOT A PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. #### 15.2 Liability Cap TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, EACH PARTY'S TOTAL CUMULATIVE LIABILITY ARISING OUT OF OR RELATED TO THESE TERMS WILL NOT EXCEED THE GREATER OF: (A) THE TOTAL FEES PAID BY CUSTOMER TO THE COMPANY DURING THE TWELVE (12) MONTHS IMMEDIATELY PRECEDING THE EVENT GIVING RISE TO THE CLAIM; OR (B) ONE THOUSAND UNITED STATES DOLLARS ($1,000 USD). #### 15.3 Exceptions The limitations in Sections 15.1 and 15.2 will not apply to: (a) either party's indemnification obligations under Section 16; (b) either party's breach of confidentiality obligations under Section 13; (c) Customer's breach of the license restrictions in Section 5; (d) Customer's payment obligations; or (e) liability that cannot be excluded or limited under applicable law. #### 15.4 Basis of the Bargain The limitations of liability set forth in this Section 15 are a fundamental element of the basis of the bargain between the parties and reflect an allocation of risk that both parties have agreed to. The Service would not be provided without these limitations. Each provision of these Terms that provides for a limitation of liability, disclaimer of warranties, or exclusion of damages is intended to be severable and independent of every other such provision. #### 15.5 Specific Disclaimers of Liability Without limiting the generality of the foregoing, the Company will not be liable for: (a) Any errors, inaccuracies, or harmful content in AI Agent outputs generated using Customer's BYOK API keys; (b) Costs or charges incurred by Customer with third-party AI service providers; (c) Decisions made or actions taken based on AI Agent outputs or Human-in-Loop interactions; (d) Regulatory violations resulting from Customer's Workflow configuration or data processing practices; (e) Failures or disruptions of Third-Party Services integrated by Customer; (f) Data loss or security breaches attributable to Customer's infrastructure (for Self-Hosted Deployments); (g) The accuracy, timeliness, or delivery of Human-in-Loop communications through third-party channels. ### 16\. Indemnification #### 16.1 Indemnification by Company The Company will defend, indemnify, and hold harmless Customer and its officers, directors, employees, and agents from and against any third-party claim, action, or proceeding alleging that the Service, as provided by the Company and used in accordance with these Terms, infringes or misappropriates such third party's intellectual property rights, and will pay any damages finally awarded or settlement amounts agreed to, provided that Customer: (a) promptly notifies the Company in writing of the claim; (b) grants the Company sole control of the defense and settlement; and (c) provides reasonable cooperation at the Company's expense. The Company's indemnification obligations do not apply to claims arising from: (i) Customer Content; (ii) Customer's modifications to the Service; (iii) Customer's use of the Service in violation of these Terms; (iv) combination of the Service with products, services, or data not provided by the Company; or (v) use of a superseded version of the Service when a non-infringing version was made available. #### 16.2 Indemnification by Customer Customer will defend, indemnify, and hold harmless the Company and its officers, directors, employees, and agents from and against any third-party claim, action, or proceeding arising from or related to: (a) Customer Content, including claims that Customer Content infringes or misappropriates a third party's rights; (b) Customer's Workflows, AI Agent configurations, and automations built on the Service; (c) Customer's use of the Service in violation of these Terms or applicable law; (d) Customer's use of Third-Party Services through the Service; (e) data processed through Customer's Workflows, including claims related to privacy, data protection, or regulatory non-compliance; and (f) Customer's use of Human-in-Loop features, including communications sent to third parties through the Service. ### 17\. Term and Termination #### 17.1 Term These Terms are effective as of the date you first access or use the Service and continue until terminated in accordance with this Section 17. #### 17.2 Termination by Customer You may terminate your subscription at any time through your Account settings or by contacting [support@flowrunner.ai](mailto:support@flowrunner.ai). Termination will take effect at the end of your current billing period. You will not receive a refund for the remaining portion of your current billing period, except during the first thirty (30) days as described in Section 4.5. #### 17.3 Termination by Company The Company may terminate or suspend your access to the Service: (a) Immediately upon written notice if you breach any material term of these Terms and fail to cure such breach within fifteen (15) days of receiving written notice; (b) Immediately if you breach Sections 5.2 (Acceptable Use) or 6.1 (BYOK responsibilities) in a manner that poses a security risk or legal liability to the Company or other customers; (c) Immediately upon written notice if you fail to pay fees when due and such failure continues for fifteen (15) days after written notice; (d) Upon thirty (30) days' written notice for any reason or no reason (termination for convenience). #### 17.4 Effect of Termination Upon termination or expiration: (a) All licenses granted to you under these Terms immediately terminate; (b) You must cease all use of the Service; (c) Your Customer Content will be retained in a paused state for thirty (30) days as described in Section 8.4; (d) You remain responsible for all fees incurred prior to the effective date of termination; (e) Sections 2, 6.2, 7.2, 8.1, 8.2, 10, 13, 14, 15, 16, 18, 19, 20, and 21 will survive termination. ### 18\. Governing Law and Dispute Resolution #### 18.1 Governing Law These Terms and any disputes arising out of or related to these Terms or the Service will be governed by and construed in accordance with the laws of the State of Texas, without regard to its conflict of law principles. #### 18.2 Exclusive Venue Any legal action or proceeding arising out of or related to these Terms or the Service will be brought exclusively in the state or federal courts located in Dallas County, Texas. Each party irrevocably consents to the personal jurisdiction and venue of such courts and waives any objection to such jurisdiction or venue, including objections based on inconvenient forum. #### 18.3 Informal Resolution Before initiating any formal dispute resolution proceeding, the parties agree to first attempt to resolve any dispute informally by contacting the other party and describing the dispute in writing. Each party agrees to negotiate in good faith for at least thirty (30) days before pursuing formal resolution. Informal dispute resolution notices should be sent to [legal@flowrunner.ai](mailto:legal@flowrunner.ai). #### 18.4 Injunctive Relief Notwithstanding the foregoing, either party may seek injunctive or other equitable relief in any court of competent jurisdiction to prevent the actual or threatened infringement, misappropriation, or violation of its intellectual property rights, confidential information, or data security without first engaging in informal resolution. #### 18.5 Class Action Waiver TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, EACH PARTY AGREES THAT ANY DISPUTE RESOLUTION PROCEEDING WILL BE CONDUCTED ONLY ON AN INDIVIDUAL BASIS AND NOT IN A CLASS, CONSOLIDATED, OR REPRESENTATIVE ACTION. IF FOR ANY REASON A CLAIM PROCEEDS IN COURT, EACH PARTY WAIVES ANY RIGHT TO A JURY TRIAL. ### 19\. Midnight Flow Consulting Services The Company operates Midnight Flow (midnightflow.ai) as a consulting and implementation services division. Midnight Flow may provide workflow implementation, optimization, training, and compliance consulting services in connection with the Service. #### 19.1 Separate Engagement Consulting services provided by Midnight Flow are governed by separate statements of work, consulting agreements, or service orders ("Consulting Agreements"). These Terms govern your use of the FlowRunner platform only and do not govern consulting engagements unless the applicable Consulting Agreement expressly incorporates these Terms. #### 19.2 Enterprise Tier Consulting Enterprise Subscription Plans may include Midnight Flow consulting services as described in the applicable Enterprise agreement. The scope, deliverables, and limitations of included consulting services will be defined in the Enterprise agreement or an associated statement of work. #### 19.3 Consulting Limitations Midnight Flow consulting services are advisory in nature. The Company does not provide legal, tax, regulatory, or compliance advice. Recommendations provided through Midnight Flow consulting services should be reviewed by your own qualified legal, compliance, and financial advisors before implementation. ### 20\. General Provisions #### 20.1 Entire Agreement These Terms, together with the Privacy Policy, any applicable Order Form, and any applicable separate written agreement between the parties, constitute the entire agreement between you and the Company regarding the Service and supersede all prior and contemporaneous agreements, proposals, and communications, whether oral or written. In the event of a conflict between these Terms and a separate written agreement executed by both parties, the separate written agreement will control with respect to the subject matter of that agreement. #### 20.2 Severability If any provision of these Terms is found to be unenforceable or invalid by a court of competent jurisdiction, that provision will be enforced to the maximum extent permissible, and the remaining provisions of these Terms will remain in full force and effect. #### 20.3 Waiver No waiver of any term or condition of these Terms will be deemed a further or continuing waiver of such term or any other term. The Company's failure to assert any right or provision under these Terms does not constitute a waiver of such right or provision. #### 20.4 Assignment You may not assign or transfer these Terms or any rights or obligations hereunder without the prior written consent of the Company. The Company may assign these Terms in connection with a merger, acquisition, reorganization, or sale of all or substantially all of its assets without your consent. Any attempted assignment in violation of this section will be void. These Terms will be binding upon and inure to the benefit of the parties, their successors, and permitted assigns. #### 20.5 Force Majeure Neither party will be liable for any failure or delay in performing its obligations under these Terms (other than payment obligations) to the extent such failure or delay results from circumstances beyond its reasonable control, including but not limited to acts of God, natural disasters, pandemics, war, terrorism, riots, government actions, power failures, internet or telecommunications failures, or cyberattacks. The affected party will promptly notify the other party and use commercially reasonable efforts to mitigate the impact. #### 20.6 Notices All legal notices required or permitted under these Terms must be in writing and will be deemed given when: (a) delivered personally; (b) sent by nationally recognized overnight courier; (c) sent by certified mail, return receipt requested; or (d) sent by email (with confirmation of receipt). Notices to the Company should be sent to: Midnight Coders, Inc. Attn: Legal Department 539 W. Commerce St, Suite 2023 Dallas, TX 75208 Email: [legal@flowrunner.ai](mailto:legal@flowrunner.ai) Notices to you will be sent to the email address associated with your Account. #### 20.7 Independent Contractors The relationship between the parties is that of independent contractors. Nothing in these Terms creates a partnership, joint venture, employment, franchise, or agency relationship between the parties. #### 20.8 No Third-Party Beneficiaries These Terms do not confer any third-party beneficiary rights. Only the parties to these Terms (and their permitted successors and assigns) have any rights or remedies under these Terms. #### 20.9 Export Compliance You agree to comply with all applicable export and import control laws and regulations, including the U.S. Export Administration Regulations, in your use of the Service. You represent that you are not located in a country subject to a U.S. Government embargo, and that you are not listed on any U.S. Government restricted party list. ### 21\. Contact Information If you have questions about these Terms, please contact us at: Midnight Coders, Inc. 539 W. Commerce St, Suite 2023 Dallas, TX 75208 General inquiries: [support@flowrunner.ai](mailto:support@flowrunner.ai) Legal inquiries: [legal@flowrunner.ai](mailto:legal@flowrunner.ai) Security inquiries: [security@flowrunner.ai](mailto:security@flowrunner.ai) BAA requests (HIPAA): [legal@flowrunner.ai](mailto:legal@flowrunner.ai) --- ## Trust & Security Source: https://flowrunner.ai/trust The facts security reviewers ask for, in one place: encryption, access controls, certifications, data handling, and sub-processors. Everything here is backed by our [Privacy Policy](https://flowrunner.ai/privacy), [Terms of Service](https://flowrunner.ai/terms), and [Data Processing Agreement](https://flowrunner.ai/data-processing). ### Security measures #### Encryption in transit All data between you and the platform is encrypted with TLS 1.2 or higher. #### Encryption at rest Customer content and sensitive account data are encrypted at rest with AES-256 or equivalent. #### BYOK credential handling Your AI provider API keys are encrypted at rest and display masked in the platform UI, with reveal available only to authorized users of your account. #### Internal access controls Internal access to personal data is restricted to authorized personnel on a need-to-know basis, with multi-factor authentication required. #### Infrastructure security Hosted in SOC 2-audited DigitalOcean data centers with firewalls, intrusion detection, network segmentation, and a regular patching schedule. #### Data segregation Customer data is logically segregated to prevent cross-customer access. ### Compliance
SOC 2SOC 2 Type II certification is in progress; customers will be notified upon completion. Cloud infrastructure already runs in SOC 2-audited data centers.
HIPAABusiness Associate Agreements are available for customers processing protected health information. A BAA must be executed before processing PHI. Contact legal@flowrunner.ai.
GDPR and privacy lawOur Data Processing Agreement covers GDPR, UK GDPR, the Swiss FADP, CCPA/CPRA, and the Texas Data Privacy and Security Act, and incorporates the EU Standard Contractual Clauses for international transfers.
Audit trails and platform controlsExecution logs on every plan (24 hours of visible history on Free and Starter, 7 days on Growth), audit trails on Professional (30 days) and Business (90 days), unlimited on Enterprise. Role-based access control from Professional; SSO via SAML 2.0 (Okta, Azure AD, Google Workspace, OneLogin) from Business. See pricing for the full matrix.
Human oversight by designWorkflows pause for human judgment at the decision points you define, and the decision trail lands in the audit log. How human-in-the-loop works.
### Sub-processors Current sub-processors for cloud deployments. We provide at least 30 days' notice before adding or replacing a sub-processor that processes customer content. | Sub-processor | Purpose | | --- | --- | | DigitalOcean | Cloud infrastructure hosting and compute services. | | MongoDB | Database services. | | Redis | Caching and session management. | ### Data handling
BYOK means your data goes directWorkflow data sent to AI providers travels directly to them using your credentials. FlowRunner does not control, monitor, or have visibility into it.
Retention and deletionOn termination, customer content is retained in a paused state for 30 days for reactivation, export, or deletion. Encrypted backup copies are purged within 90 days of deletion from active systems.
Self-hosted optionThe free Community Edition and Enterprise self-hosted run entirely on your infrastructure. FlowRunner has no access to customer content on self-hosted installations.
### Frequently asked questions #### Is FlowRunner SOC 2 certified? FlowRunner is pursuing SOC 2 Type II certification and will update customers upon completion. Cloud infrastructure already runs in SOC 2-audited data centers operated by DigitalOcean. #### Does FlowRunner sign HIPAA Business Associate Agreements? Yes. Customers processing protected health information must execute a BAA before processing PHI through the platform. Contact legal@flowrunner.ai to initiate one. #### How is my data encrypted? All data in transit is encrypted with TLS 1.2 or higher. Customer content and sensitive account data are encrypted at rest using AES-256 or equivalent. BYOK API keys are encrypted at rest and display masked in the platform UI. #### Can FlowRunner see the data my workflows send to AI providers? No. Under the BYOK model, workflow data is transmitted directly to your AI providers using your own credentials. FlowRunner does not control, monitor, or have visibility into data sent to third-party AI providers through your API keys. #### Is FlowRunner GDPR compliant? FlowRunner offers a Data Processing Agreement covering GDPR, UK GDPR, the Swiss FADP, CCPA/CPRA, and the Texas Data Privacy and Security Act, with EU Standard Contractual Clauses for international transfers. The DPA is published at flowrunner.ai/data-processing. #### Where is FlowRunner data hosted? Cloud deployments run on DigitalOcean infrastructure in SOC 2-audited data centers, with MongoDB for database services and Redis for caching and session management. Customer data is logically segregated between customers. #### What happens to my data if I cancel? Customer content is retained in a paused state for 30 days, during which you can reactivate, export, or request deletion. After that it is deleted from active systems; residual copies in encrypted backups are purged within 90 days. #### Can I run FlowRunner on my own infrastructure? Yes. The self-hosted Community Edition is free, and Enterprise self-hosted adds clustering and the full compliance suite. On self-hosted deployments FlowRunner has no access to or visibility into your customer content. Security questionnaire, architecture review, or an IT-specific call: [contact us](https://flowrunner.ai/contact) or email [legal@flowrunner.ai](mailto:legal@flowrunner.ai). --- ## Use Cases Source: https://flowrunner.ai/use-cases Each use case below is a real scenario drawn from prospect conversations. FlowRunner agents handle the routine work and surface only the exceptions that need your judgment. ![Invoice validation - 44 of 47 invoices posted automatically, 3 need review](https://flowrunner.ai/assets/hero/hero-usecase1.webp) #### Invoice validation 44 of 47 invoices posted without a touch. Three flagged: PO mismatch and two new vendors. [View details](https://flowrunner.ai/use-cases/invoice-validation) ![Billback reconciliation - Duplicate detection across distributor and brand payments](https://flowrunner.ai/assets/hero/hero-usecase2.webp) #### Billback reconciliation Caught a $1,840 duplicate across two payment systems before it became a write-off. [View details](https://flowrunner.ai/use-cases/billback-reconciliation) ![3PL receipt tracking - 6 of 8 transfer orders received, one overdue](https://flowrunner.ai/assets/hero/hero-usecase3.webp) #### 3PL receipt tracking Shipment TO-4418 is 48 hours late. Follow-up sent, escalation queued. [View details](https://flowrunner.ai/use-cases/3pl-receipt-tracking) ![Stock-out risk detection - 2.9 days of supply remaining alert](https://flowrunner.ai/assets/hero/hero-usecase4.webp) #### Stock-out risk detection 124 of 127 SKUs covered. Three will stock out this week without a reorder. [View details](https://flowrunner.ai/use-cases/stock-out-risk-detection) ![M&A due diligence - Document collection with gaps highlighted](https://flowrunner.ai/assets/hero/hero-usecase5.webp) #### M&A due diligence Two document categories still missing with 30 days to close. The rest is collected. [View details](https://flowrunner.ai/use-cases/ma-due-diligence) ![Prior authorization - Medical forms with missing attachments flagged](https://flowrunner.ai/assets/hero/hero-usecase6.webp) #### Prior authorization 12 submitted today. Two bounced back for missing physician progress notes. [View details](https://flowrunner.ai/use-cases/prior-authorization) ![Data reconciliation - Finding errors in thousands of records](https://flowrunner.ai/assets/hero/hero-usecase7.webp) #### Data reconciliation 1,247 order records matched. Three discrepancies found and routed for review. [View details](https://flowrunner.ai/use-cases/data-reconciliation) ![Client intake - Document collection with one missing item](https://flowrunner.ai/assets/hero/hero-usecase8.webp) #### Client intake W-9, NDA, and insurance received. Tax returns came in the wrong format. [View details](https://flowrunner.ai/use-cases/client-intake) ![Approval routing - Stuck payment approval with vendor inquiry](https://flowrunner.ai/assets/hero/hero-usecase9.webp) #### Approval routing $48K payment stuck on one approval. Reminder sent. The vendor is calling. [View details](https://flowrunner.ai/use-cases/approval-routing) ![Complaint triage - Damaged shipment with claim decision needed](https://flowrunner.ai/assets/hero/hero-usecase10.webp) #### Complaint triage Damaged shipment investigated. Root cause: cold chain break. $34K claim needs your call. [View details](https://flowrunner.ai/use-cases/complaint-triage) --- ## 3PL Receipt Tracking Source: https://flowrunner.ai/use-cases/3pl-receipt-tracking 6 of 8 transfer orders received. 1 in transit. 1 overdue. All without a single status email. The Problem The Automation What Gets Escalated Real Results ### The Problem You have inventory in 4 warehouses managed by 3 different 3PLs. Every day, transfer orders move between them. And every day, your operations team sends emails asking: "Did you receive PO-2847?" "Where's the shipment from Dallas?" "Why hasn't Chicago confirmed receipt?" The 3PLs respond when they can. Sometimes with ASN data. Sometimes with a screenshot. Sometimes not at all. Meanwhile, your system shows inventory that doesn't exist, missing inventory that does, and customers waiting on orders you can't fulfill. ### The Automation Our 3PL tracking agent connects to warehouse WMS systems, EDI feeds, and carrier APIs. It monitors every transfer order from creation to receipt, updates your ERP automatically, and escalates only when something's actually wrong. 01 #### Monitor Agent watches EDI 856s, portal updates, and carrier tracking across all 3PLs 02 #### Reconcile Matches received quantities to shipped quantities, flags discrepancies 03 #### Update Posts receipts to ERP automatically, updates inventory in real-time 04 #### Escalate Alerts on overdue shipments, quantity mismatches, or damage claims ![3PL receipt tracking dashboard showing transfer order status](https://flowrunner.ai/_astro/hero-usecase3.B9UQs9N3.png) 100% Visibility 0 Status emails 3 hrs Saved daily ### What Gets Escalated - **Overdue receipts** - Shipment expected 2+ days ago, no confirmation received - **Quantity mismatches** - Received quantity differs from ASN by >5% - **Damage notifications** - Carrier or 3PL reports damage, agent initiates claim workflow - **Short shipments** - Partial receipt flagged for follow-up with origin warehouse ### Real Results > "We eliminated the daily 3PL status call. The agent catches discrepancies within hours instead of weeks. Last month it flagged a $23K inventory mismatch we would have never found." \- Director of Operations, omnichannel retailer #### See your transfer orders in real-time Connect your 3PLs in under an hour. Start with one warehouse, scale to all of them. [Start free trial](https://app.flowrunner.ai) --- ## Approval Routing Source: https://flowrunner.ai/use-cases/approval-routing Payment approval stuck 4 days. Vendor called twice. Agent escalated to CFO with full context. The Problem The Automation Smart Routing Rules Real Results Complete Audit Trail ### The Problem A $47,000 vendor payment sits in someone's inbox for 5 days. The vendor calls asking about status. Your AP team doesn't know who has it. The approver was out sick. The backup approver wasn't notified. Now you're paying expedite fees and damaging a relationship. This happens because approval workflows are invisible. You can't see what's pending, what's stuck, or who's the bottleneck. So payments are late, vendors are frustrated, and early-pay discounts expire. ### The Automation Our approval routing agent manages the entire workflow: routing requests to the right approvers, escalating when deadlines approach, and handling exceptions like out-of-office coverage. Every approval has an audit trail. Nothing gets lost. 01 #### Route Agent sends approval requests via email, Slack, or mobile based on amount and type 02 #### Track Monitors response status, detects delays, checks approver availability 03 #### Escalate Routes to backup approver after timeout, escalates to manager for urgent items 04 #### Complete Posts approved transactions to ERP, notifies AP team, updates vendor ![Approval routing dashboard showing stuck payment and escalation](https://flowrunner.ai/_astro/hero-usecase9.Bb2BEp6Z.png) 68% Faster approvals 0 Lost approvals 2.3 days Avg. cycle time ### Smart Routing Rules - **Amount-based** - Under $5K: Manager approval. Over $5K: Director + Finance - **Category routing** - Marketing spend to CMO, CapEx to CFO, HR to CHRO - **Vendor-specific** - New vendors require Procurement approval regardless of amount - **Time-based** - Escalate if no response in 24 hours, escalate again at 48 hours - **Out-of-office** - Auto-route to designated backup when primary is unavailable ### Real Results > "We eliminated the 'where's my approval?' emails completely. The agent knows who's out of office, routes to backups automatically, and escalates before vendors start calling. Our average approval time went from 7 days to 2.3 days." \- VP Finance, 300-person professional services firm ### Complete Audit Trail Every approval is tracked: - Who requested and when - Who approved and timestamp - Approval method (email click, mobile app, Slack) - Comments and attachments - Escalation history - Final posting to ERP Ready for auditors. Ready for compliance. Ready for your peace of mind. --- ## Billback Reconciliation Source: https://flowrunner.ai/use-cases/billback-reconciliation Catch duplicate payments and pricing mismatches across distributor and brand systems before they cost you thousands. The Problem The Automation Discrepancies Caught Real Results Why Traditional Tools Fail ### The Problem You sell through distributors. They charge you promotional allowances, marketing fees, and price protection adjustments. The problem? You're paying invoices from distributors while also receiving chargebacks from brands - and nobody's checking if they match. The result is predictable: duplicate payments, missed deductions, and pricing errors that drain 2-5% of your gross margin. Your finance team spends 12+ hours weekly reconciling spreadsheets, but they're always behind. By the time they catch an error, the window to dispute has closed. ### The Automation Our billback reconciliation agent connects to your ERP, distributor portals, and brand chargeback systems. It matches every transaction across systems, flags discrepancies in real-time, and routes exceptions to your team with full context. 01 #### Connect Integrate with ERP, distributor EDI feeds, and brand chargeback portals 02 #### Match AI correlates invoices, payments, and chargebacks across all systems 03 #### Flag Identifies duplicates, price mismatches, and missing deductions automatically 04 #### Recover Routes disputes to the right person with documentation ready to submit ![Billback reconciliation dashboard showing duplicate payment detection](https://flowrunner.ai/_astro/hero-usecase2.BByICEQt.png) $47K Recovered monthly 12 hrs Saved weekly 99.2% Accuracy rate ### Common Discrepancies Caught - **Duplicate payments** - Same invoice paid to distributor and deducted by brand - **Rate mismatches** - Promotional allowance on invoice differs from agreed rate - **Timing issues** - Chargebacks applied to wrong fiscal period - **Missing deductions** - Eligible marketing funds not claimed from brands - **Quantity variances** - Units billed don't match units sold per POS data ### Real Results > "We found $47,000 in duplicate payments in the first month. The agent catches discrepancies within 24 hours instead of 60 days. We're actually recovering money now instead of just trying to stop the bleeding." \- CFO, $50M consumer goods distributor ### Why Traditional Tools Fail Excel-based reconciliation breaks down because: - Data arrives in different formats from 10+ sources - Product codes don't match across distributor and brand systems - Timing lags mean you're always reconciling stale data - Rules are too complex for basic matching algorithms FlowRunner's AI handles the complexity: fuzzy matching on product descriptions, understanding date ranges for promotional periods, and learning your specific distributor-brand relationships. #### Find your duplicate payments We'll analyze your last 90 days of transactions and show you exactly what's been missed. [Get a free reconciliation audit](https://flowrunner.ai/contact) --- ## Client Intake Source: https://flowrunner.ai/use-cases/client-intake Document collection 87% complete. 7 items received, 1 missing. Onboard clients in days, not weeks. The Problem The Automation What Gets Validated Real Results Works With Your Tools ### The Problem New clients need to submit 8-15 documents: contracts, tax forms, ID verification, banking details, compliance questionnaires. You send a checklist. They send back partial responses. You email asking for the missing items. They respond with 3 of 5. You email again. Two weeks later, the client is still not onboarded. They've lost enthusiasm. Your team has spent 6 hours on email ping-pong. And you still don't have a complete file. ### The Automation Our client intake agent manages the entire onboarding workflow: sending personalized requests, tracking document receipt, validating completeness, and escalating only when human judgment is needed. Clients know exactly what's required. You know exactly what's missing. 01 #### Request Agent sends branded intake forms with clear instructions and examples 02 #### Collect Monitors email, portal uploads, and e-signature platforms for documents 03 #### Validate Checks for completeness, extracts data, validates formats and signatures 04 #### Escalate Routes complete intakes to onboarding team, flags exceptions for review ![Client intake dashboard showing document collection progress](https://flowrunner.ai/_astro/hero-usecase8.wSWAekU-.png) 5 days Avg. onboarding 73% Faster intake 0 Follow-up emails ### What Gets Validated - **Document completeness** - All required pages present, no blank required fields - **Signature verification** - Documents signed by authorized parties - **Data extraction** - Key information pulled and entered into your CRM/ERP - **Format compliance** - Files in correct format, not corrupted, readable - **Expiration dates** - IDs, certifications, insurance docs not expired ### Real Results > "Our onboarding time dropped from 3 weeks to 5 days. The agent sends friendly reminders automatically, so our team stopped chasing documents. Clients actually comment on how smooth our process is now." \- Head of Client Operations, wealth management firm ### Works With Your Tools Integrates with your existing stack: - CRM (Salesforce, HubSpot, Pipedrive) - E-signature (DocuSign, Adobe Sign, HelloSign) - File storage (SharePoint, Google Drive, Dropbox) - Forms (Typeform, JotForm, custom portals) - Communication (email, Slack, SMS) --- ## Complaint Triage Source: https://flowrunner.ai/use-cases/complaint-triage Damaged shipment flagged. Claim decision needed. Routed to claims team with photos and order history. The Problem The Automation Intelligent Classification Real Results Beyond Routing ### The Problem Customer complaints arrive by email, phone, chat, and social media. Someone has to read each one, figure out what it's about, decide how urgent it is, and route it to the right team. That takes time. Meanwhile, the customer is waiting. The result? High-priority issues sit in general queues. Simple refunds get escalated to managers. Product defects aren't aggregated to identify patterns. And your team spends 40% of their time on routing instead of resolution. ### The Automation Our complaint triage agent reads incoming complaints from any channel, classifies the issue type, assesses urgency, and routes to the right team with full context. High-priority issues get immediate attention. Routine requests get automated responses. Nothing falls through cracks. 01 #### Capture Monitors email, chat, social, and voice transcripts for incoming complaints 02 #### Classify AI identifies issue type: shipping damage, product defect, billing error, etc. 03 #### Prioritize Assesses urgency based on customer value, issue severity, and regulatory risk 04 #### Route Sends to appropriate team with context, customer history, and recommended action ![Complaint triage dashboard showing damaged shipment claim](https://flowrunner.ai/_astro/hero-usecase10.DJDsMMVL.png) 4.2 hrs Avg. response time 94% Routed correctly 23 hrs Faster resolution ### Intelligent Classification - **Shipping issues** - Damaged, lost, or delayed shipments: Logistics team - **Product defects** - Quality complaints, safety issues: QA + Product team - **Billing disputes** - Overcharges, refund requests: Finance team - **Service failures** - Support experience, account issues: Customer Success - **Legal/Regulatory** - Safety reports, compliance issues: Legal immediately ### Real Results > "The agent routes 94% of complaints correctly without human touch. Our response time dropped from 24 hours to 4 hours. Most importantly, we caught a product defect pattern in week 2 that would have taken us months to identify manually." \- VP Customer Experience, DTC e-commerce brand ### Beyond Routing FlowRunner doesn't just route - it helps resolve: - **Auto-responses** - Immediate acknowledgment with expected resolution time - **Context assembly** - Pulls order history, previous interactions, customer value - **Suggested actions** - Recommends refund amount, replacement product, or escalation - **Pattern detection** - Identifies emerging issues before they become crises - **SLA tracking** - Monitors resolution times, escalates breaches automatically #### Respond faster to every complaint Route your last 100 complaints through FlowRunner and see the difference in response time and accuracy. [Get a free analysis](https://flowrunner.ai/contact) --- ## Data Reconciliation Source: https://flowrunner.ai/use-cases/data-reconciliation 47,000 records processed. 23 discrepancies found. 0.05% error rate - caught before month-end close. The Problem The Automation Errors Caught Real Results Integrations ### The Problem You have the same data in three systems: your ERP, your CRM, and your data warehouse. And every month, they don't match. Your finance team exports everything to Excel, writes VLOOKUP formulas, and spends a week hunting for the $0.02 difference that's preventing close. The real cost isn't the time - it's the errors you miss. A customer marked active in Salesforce but inactive in your billing system. An invoice posted twice. A payment applied to the wrong account. By the time you find them, the damage is done. ### The Automation Our reconciliation agent connects to any system with an API or database access. It compares records across systems using fuzzy matching, learns your business rules, and flags only the exceptions that need human judgment. 01 #### Extract Pulls data from source systems on schedule or on-demand 02 #### Match AI correlates records across systems using keys, fuzzy logic, and business rules 03 #### Validate Checks for missing records, field mismatches, and out-of-range values 04 #### Report Generates exception reports with suggested resolutions and routes to owners ![Data reconciliation dashboard showing error detection across records](https://flowrunner.ai/_astro/hero-usecase7.BgzscgJQ.png) 99.9% Accuracy 40 hrs Saved monthly 3 days Faster close ### Types of Errors Caught - **Missing records** - Transaction in System A, no corresponding record in System B - **Field mismatches** - Same record, different values (amounts, dates, statuses) - **Orphaned data** - Child records without parent records - **Duplicate detection** - Same transaction recorded multiple times - **Referential integrity** - Invalid foreign keys, inactive customers with open orders ### Real Results > "We reconciled 50,000 customer records across three systems in 2 hours. The agent found 147 discrepancies our manual process missed for months. Month-end close went from 8 days to 5." \- Controller, B2B software company ### Any Data, Any System FlowRunner reconciles across: - Databases (PostgreSQL, MySQL, SQL Server, Oracle) - Cloud apps (Salesforce, NetSuite, HubSpot, Stripe) - Files (CSV, Excel, XML, JSON) - APIs (REST, GraphQL, SOAP) - Data warehouses (Snowflake, BigQuery, Redshift) #### Find your data errors Run a free reconciliation analysis on your critical data sets. See what you've been missing. [Request analysis](https://flowrunner.ai/contact) --- ## Invoice Validation Source: https://flowrunner.ai/use-cases/invoice-validation 47 invoices arrived this morning. 44 posted automatically. 3 need your decision. The Problem The Automation What Gets Escalated Real Results Implementation ### The Problem Your AP team spends 4+ hours every morning processing invoices. Most are routine: same vendors, same amounts, same approvals. But 6% have exceptions: missing PO numbers, price mismatches, new vendors requiring setup. Those exceptions stall everything. The result? Invoices sit in queues. Vendors call asking about payment status. Early payment discounts expire. And your team is stuck doing data entry instead of managing vendor relationships. ### The Automation Our invoice validation agent connects to your ERP, email, and vendor portals. It reads incoming invoices, matches them to POs, checks for duplicates, and validates amounts against contracted rates. 01 #### Intake Agent monitors email, portals, and shared folders for new invoices 02 #### Validation Extracts data, matches POs, checks pricing, flags exceptions 03 #### Escalation Posts clean invoices automatically. Routes exceptions to AP team with context ![Invoice validation dashboard showing 44 of 47 invoices processed automatically](https://flowrunner.ai/_astro/hero-usecase1.7IBTBhlX.png) 94% Auto-processed 3 min Avg. review time $12K Monthly savings ### What Gets Escalated - **Missing PO numbers** - Agent flags invoice, suggests matching PO from vendor history - **Price variances >5%** - Escalates with contract terms and approval workflow - **New vendors** - Routes to vendor setup process with banking details - **Duplicate detection** - Catches invoices already paid or in process - **Tax discrepancies** - Flags sales tax calculation errors ### Real Results > "We went from 4 hours of invoice processing every morning to 20 minutes of exception review. The agent catches duplicates we used to miss and pays early enough to capture discounts we were leaving on the table." \- VP Finance, 200-person manufacturing company ### Implementation **Week 1:** Connect to your ERP and email. Train the agent on your invoice formats and approval rules. **Week 2:** Run in shadow mode: agent processes but doesn't post, you verify accuracy. **Week 3:** Go live with auto-posting for invoices above confidence threshold. --- ## M&A Due Diligence Source: https://flowrunner.ai/use-cases/ma-due-diligence Document collection at 73% complete. 47 items received, 12 pending, 6 flagged for review. The Problem The Automation Document Intelligence Real Results Built for Deal Complexity ### The Problem Every M&A deal requires hundreds of documents: financial statements, contracts, IP filings, employee records, regulatory filings. You send a request list. They send back a Dropbox link. You spend days sorting, renaming, and cross-referencing against your checklist. Then the questions start: "Did we get the 2022 audit?" "Where's the employment agreement for the CTO?" "Is this the signed version or the draft?" Your junior associates spend 60% of their time on document logistics instead of analysis. ### The Automation Our due diligence agent manages the entire document workflow: sending requests, tracking receipts, organizing files, and flagging gaps. It reads documents as they arrive, extracts metadata, and updates your checklist automatically. 01 #### Request Agent sends structured request lists to target company with clear instructions 02 #### Collect Monitors email, portals, and file shares for incoming documents 03 #### Organize Reads documents, extracts metadata, files in correct folder structure 04 #### Track Updates checklist, flags missing items, routes questions to target company ![M&A due diligence dashboard showing document collection progress](https://flowrunner.ai/_astro/hero-usecase5.BhITAvRc.png) 60% Faster close 200+Doc types tracked $400/hr Associate time saved ### Document Intelligence The agent doesn't just file documents - it understands them: - **Version detection** - Identifies drafts vs. signed copies, flags when executed version is missing - **Date validation** - Confirms document covers required period (e.g., "fiscal 2023") - **Party matching** - Verifies contracts reference correct legal entities - **Red flag detection** - Highlights unusual clauses, missing signatures, or expired agreements - **Cross-reference** - Links related documents (e.g., amendment to original agreement) ### Real Results > "We cut document collection time from 3 weeks to 8 days. The agent caught a missing consent that would have delayed closing by 2 weeks. Our associates actually had time to read the documents instead of just organizing them." \- Managing Director, middle-market private equity firm ### Built for Deal Complexity FlowRunner handles the messiness of real deals: - Multiple document sources (email, VDR, file shares, physical scans) - Inconsistent naming and formatting - Partial submissions requiring follow-up - Confidentiality restrictions and access controls - Multiple workstreams (legal, financial, commercial, technical) --- ## Prior Authorization Source: https://flowrunner.ai/use-cases/prior-authorization Medical forms with 3 missing attachments flagged. Complete submission in 24 hours instead of 8 days. The Problem The Automation Missing Items Caught Real Results Compliance ### The Problem Every specialty medication requires prior authorization. Your team fills out the same forms repeatedly, attaches clinical documentation, and submits to insurance. Then the rejections come back: "Missing lab results." "Need 90-day medication history." "Form incomplete on page 3." Each rejection adds 3-5 days. Patients wait in pain. Providers call asking for status. And your staff spends 40% of their time on rework instead of patient care. ### The Automation Our prior authorization agent manages the entire workflow: extracting patient data from your EHR, pre-populating forms, checking for required attachments, and tracking submissions through approval. It catches missing information before submission, not after rejection. 01 #### Extract Pulls patient demographics, diagnosis codes, and medication history from EHR 02 #### Validate Checks payer requirements, flags missing clinical documentation 03 #### Submit Completes forms, attaches documents, submits via payer portal or fax 04 #### Track Monitors status, escalates delays, routes approvals back to clinical team ![Prior authorization dashboard showing form status and missing attachments](https://flowrunner.ai/_astro/hero-usecase6.BLHbIsTz.png) 68% Faster approval 94% First-pass rate 12 hrs Per case saved ### Common Missing Items Caught - **Clinical documentation** - Progress notes, lab results, imaging reports required by payer - **Step therapy proof** - Documentation of failed preferred alternatives - **Specialty provider credentials** - Verification prescribing physician is in-network - **Quantity limit justification** - Medical necessity for doses exceeding standard limits - **Diagnosis specificity** - ICD-10 codes with required specificity for approval ### Real Results > "Our first-pass approval rate went from 34% to 94%. The agent catches missing attachments before we submit, so we stopped getting rejections for incomplete paperwork. Patients get their medications 5 days faster on average." \- Director of Patient Access, specialty pharmacy ### HIPAA-Ready Compliance Built for healthcare from day one: - End-to-end encryption for all PHI - Complete audit trail of every action - Role-based access controls - BAAs available with all major payers - On-premise deployment option #### Accelerate patient access Process your first 50 prior authorizations free. See the time savings and approval rate improvement firsthand. [Start free trial](https://app.flowrunner.ai) --- ## Stock-Out Risk Detection Source: https://flowrunner.ai/use-cases/stock-out-risk-detection 2.9 days of supply remaining. Reorder now or lose $47K in sales. The Problem The Automation Risk Levels What the Agent Considers Real Results ### The Problem You find out about stock-outs when customers complain. Or when your Amazon seller rating drops. Or when the replenishment order you placed last week won't arrive for 10 more days. Your ERP has reorder points, but they're static. It doesn't account for velocity changes, supplier delays, or the promotion your marketing team just launched. By the time the system flags low inventory, it's already too late. ### The Automation Our stock-out risk agent monitors inventory levels, sales velocity, supplier lead times, and incoming shipments across all channels. It calculates days-of-supply in real-time and escalates when you have enough time to act - not when it's already a crisis. 01 #### Monitor Tracks inventory levels, sales velocity, and incoming shipments across all SKUs 02 #### Predict Calculates days-of-supply based on current velocity and confirmed inbound inventory 03 #### Alert Flags at-risk SKUs with time remaining and recommended reorder quantity 04 #### Route Sends alerts to purchasing team with supplier contact and last order details ![Stock-out risk alert showing low inventory warning](https://flowrunner.ai/_astro/hero-usecase4.Xbxt0Skh.png) 73% Fewer stock-outs $2.1M Sales protected 4 days Earlier warning ### Risk Levels - **Critical (0-3 days)** - Immediate action required. Expedite shipment or transfer from another location. - **High (4-7 days)** - Place emergency order. Consider air freight for high-value SKUs. - **Medium (8-14 days)** - Standard reorder. Monitor velocity for sudden spikes. - **Low (15+ days)** - Normal replenishment cycle. No action needed. ### What the Agent Considers Unlike static reorder points, FlowRunner factors in: - 7-day, 30-day, and 90-day velocity trends - Seasonal patterns and promotional calendars - Supplier lead time variability - Inbound shipments already in transit - Safety stock requirements by SKU - Multi-channel sales (DTC, wholesale, marketplace) ### Real Results > "We went from reactive to proactive. The agent caught a velocity spike on our top SKU 5 days before we would have stocked out. That single alert saved us $127,000 in lost sales during peak season." \- VP Supply Chain, consumer electronics brand #### Protect your best sellers Start monitoring your top 100 SKUs free for 30 days. See which products are at risk before it's too late. [Start free trial](https://app.flowrunner.ai) --- - [Concepts](https://flowrunner.ai/concepts.md): The vocabulary of Orchestration as a Service. Canonical definitions of human-in-the-loop, agents, connectors, and the rest of the FlowRunner category. --- ## Agent Directory Source: https://flowrunner.ai/concepts/agent-directory Updated September 8, 2026 The Agent Directory is a curated library of pre-built, deployable AI agents authored by FlowRunner, partners, and the community, designed for reuse across customer environments. **TL;DR** - The Agent Directory is where pre-built AI agents live, ready to deploy. - Every agent in the directory inherits the same guarantee: it knows when to stop and ask a human for help. Installing an agent does not mean inheriting someone else’s risk. - A directory entry is a complete agent, not a template. You install it and use it; you do not assemble it. - Today, FlowRunner authors the directory. The forward direction is partner-authored and community-authored agents alongside. - The reuse pattern is intentional. An agent designed once should be deployable into many customer environments with configuration rather than re-construction. ### What it means The Agent Directory is the catalog of agents available to FlowRunner customers. A directory entry has three properties that distinguish it from a template or a sample: - **It is deployable.** A directory entry can be installed into a customer environment and start doing its job. No assembly required. - **It is parameterized.** The same agent serves many customers because it is configurable. The customer supplies the integrations, the data sources, and the policies; the agent supplies the capability. - **It is governed.** Every agent in the directory operates under the same [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) the platform enforces on agents built in the [Agent Factory](https://flowrunner.ai/concepts/agent-factory). The governance does not depend on who authored the agent. The category we are defining ([Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service)) treats agents as a unit of value that should be produced once and reused many times. The directory is the place that reuse happens. The vision is that the directory grows along three axes: - **FlowRunner-authored agents** for common patterns we know our customers need. This is what exists today. - **Partner-authored agents** built by consultants, system integrators, and specialist vendors. The Midnight Flow consulting arm and FlowRunner partners are the natural authors here. - **Community-authored agents** built by FlowRunner customers and made available to others. The forward-state agents (partner and community) are how the directory becomes a leverage layer for the whole ecosystem rather than a starting point library FlowRunner ships. ### What it is not The Agent Directory is not a template library. Templates require assembly. Directory entries do not. The distinction matters because the customer’s question shifts: not “how do I build this” but “which one do I deploy.” The Agent Directory is not the integrations catalog. The integrations catalog lists what FlowRunner can talk to (Salesforce, QuickBooks, Acumatica, and so on). The directory lists agents that do specific jobs across those integrations. The Agent Directory is not a marketplace, in the sense of a transactional storefront with buy buttons. The framing is _directory_ deliberately. The point is discoverability and reuse, not commerce. Pricing and packaging belong to FlowRunner’s tiers, not to individual directory entries. The Agent Directory is not the same as the [Agent Factory](https://flowrunner.ai/concepts/agent-factory). The factory is where agents get built. The directory is where they are kept for reuse. The two are related but separate: not every agent built in the factory has to be published to the directory, and not every agent in the directory has to be built in the factory. ### How FlowRunner implements it Today, the directory contains FlowRunner-authored agents covering common operations patterns: vendor invoice intake, expense exception routing, document extraction, sales handoff coordination, and similar jobs. A customer installs an agent from the directory, supplies the connectors and policies relevant to their environment, and runs it. The technical mechanics: - **Parameterization.** Every directory entry exposes the configuration points the customer needs to provide: which CRM, which accounting system, which Slack channel, what threshold for human review. - **Composability.** A directory agent can be invoked from another agent built in the [Agent Factory](https://flowrunner.ai/concepts/agent-factory), the same way any agent can call any other agent. - **Versioning.** Directory entries are versioned. A customer’s deployment of an agent stays on the version they installed; upgrades are explicit. - **Governance.** Directory agents run under the [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) the same as any other agent in the orchestration environment. The forward direction (partner and community authoring) is what makes the directory a leverage layer. The architecture supports it today; the authoring program and the publication pipeline mature as the customer base and partner network grow. ### Where the term comes from The choice of _directory_ over _marketplace_ or _library_ is positioning work. _Library_ implies passive collection. The directory is more active than that. Entries are deployable, not just downloadable. _Marketplace_ implies transactions and storefronts. The directory does not lean on commerce as the organizing principle. Pricing is a tier-level decision in FlowRunner, not a per-entry one. _Directory_ sits in the middle: a curated, discoverable, organized set of capabilities that customers and partners and (eventually) community members contribute to and draw from. The closest analog is an enterprise software catalog or an enterprise app store, but stripped of the transactional and customer-acquisition framing. The Agent Directory is one of three pillars of [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service). The [Agent Factory](https://flowrunner.ai/concepts/agent-factory) is where agents come from. The directory is where they live. The [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) governs how they behave once they are deployed. ### Frequently asked questions #### Is the Agent Directory the same as a template library? No. Templates require assembly; directory entries do not. The Agent Directory is a curated library of pre-built, deployable AI agents authored by FlowRunner, partners, and the community, designed for reuse across customer environments. You install an entry, supply your connectors and policies, and run it. #### What is the difference between the Agent Directory and the Agent Factory? The factory is where agents get built; the directory is where they are kept for reuse. Not every agent built in the factory has to be published to the directory, and not every directory entry has to have been built in the factory. Both run under the same governance. #### Who authors the agents in the directory? FlowRunner authors the directory today. The forward direction is partner-authored and community-authored agents alongside, from consultants, system integrators, specialist vendors, and customers. Governance does not depend on the author: every entry runs under the same platform-enforced Code of Conduct. #### Does installing an agent from the directory cost extra? No. Pricing is a tier-level decision in FlowRunner, not a per-entry one, so the directory is not a storefront with buy buttons. Entries are versioned: your deployment stays on the version you installed, and upgrades are explicit rather than pushed to you. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Agent Factory Source: https://flowrunner.ai/concepts/agent-factory Updated September 8, 2026 The Agent Factory is FlowRunner's visual and conversational interface for assembling AI agents from connectors, MCP services, other agents, and existing flows, combined with business logic and built without writing code. **TL;DR** - The Agent Factory is where new AI agents get built in FlowRunner. - The agents the factory produces know when to stop and ask a human for help. The pause-and-escalate pattern that makes agents production-ready is built in, not bolted on. Builders get it by default. - Two modes, same factory: a visual editor for assembling agents step by step, and a Conversational AI for describing the agent in plain language. The chat sits on top of the visual editor and can hand off to manual fine-tuning at any point. - Agents compose from four kinds of building blocks: connectors, MCP services, other agents, and existing flows. Business logic wires them together. An agent assembled in the factory can include triggers, actions, conditions, and variables, in addition to the agentic capabilities the platform adds on top. - The factory’s job is to make a deployable, production-ready agent cheap enough to build that operations teams stop treating agent creation as a special project. ### What it means Building an agent used to mean writing Python, wiring tools to a framework, and testing in a notebook. That changed fast. Most business software vendors now ship some kind of agent builder, and the visual-builder experience is table stakes. What separates platforms today is what the agents that come out of them can actually do. The Agent Factory’s answer is that the agents you build here know when to stop and ask a human for help. Every agent the factory produces has the same escalation pattern built in: pause at uncertainty, route the decision to a human through the right channel, resume on response. The builder does not have to remember to add it; it is there by default. That is what “production-ready” means in the FlowRunner sense, and it is what most other agent builders leave to the builder to figure out. The factory assembles agents from four kinds of building blocks: - **Connectors.** Reusable bridges between FlowRunner and third-party services (Salesforce, QuickBooks, Slack, Acumatica, and many more). Each connector exposes the actions a third-party system offers in a form an agent can call. New connectors enter the catalog as integration needs surface; a connector to a new system typically takes 30 minutes or less to produce. - **MCP services.** Model Context Protocol servers. The MCP standard is how the AI industry is converging on a portable way for AI systems to call tools and access data. Agents in the factory consume MCP services directly, the same way they consume native connectors. - **Other agents.** An agent assembled in the factory can call other agents as tools. A specialist agent (extract data from a PDF) becomes a building block for a coordinator agent (handle the whole vendor invoice intake process). - **Existing flows.** A complete FlowRunner workflow can be invoked from inside an agent as if it were a function. A flow that handles vendor onboarding becomes one step in a larger procurement agent’s job. Business logic wires these together. Where to branch. When to call a human. What to do on error. Agents in FlowRunner can be substantial structures with triggers, actions, conditions, and variables in addition to the agentic capabilities the platform adds on top. The result is a deployable agent that can be published to the [Agent Directory](https://flowrunner.ai/concepts/agent-directory) or used directly. Two modes sit on top of the same factory. The visual editor is for builders who prefer to drag and drop. The Conversational AI is for builders who prefer to describe the agent in plain language; the chat sits on top of the visual editor, so a conversation can hand off to manual fine-tuning at any point. The agent produced is the same regardless of which doorway you walked in through, and the two modes are complementary rather than exclusive. ### What it is not The Agent Factory is not a chatbot builder. Chatbot builders produce conversational interfaces as the deliverable; the chatbot is what the customer ships. The Agent Factory uses a conversational interface (the Conversational AI) as one way to build agents, but what gets built is an agent that takes actions in business systems, not a chatbot. The chat is how you tell the factory what to build, not what the factory builds for you. The Agent Factory is not a pure workflow builder either, though there is real overlap. An agent assembled in the factory can include triggers, actions, conditions, and variables the same way a traditional workflow does. The factory composes those workflow primitives with agentic capabilities: the agent makes decisions, calls tools, asks for help when it needs to, and can be called by other agents in turn. The unit is an agent that contains workflow logic, not a workflow that contains nothing else. The Agent Factory is not a code generator. It does not produce Python or TypeScript for the builder to maintain. The agent is a first-class FlowRunner entity, deployed and run by the platform. The Agent Factory is not a thin wrapper over LangChain or another agent framework. The agent the factory produces is native to FlowRunner’s orchestration model, including [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) as a callable tool and the [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) governance the platform enforces. ### How FlowRunner implements it The factory operates in two layers. The first layer is connectors. Each connector is a discrete, reusable bridge to an external service. Connector production is fast because the factory’s connector-creation tooling does much of the integration work that engineers would otherwise do by hand: reading the third-party API surface, mapping its actions, wiring authentication. The current measured production time for a new connector is 30 minutes or less, and the error rate on the AI-generated integrations the factory produces is 0.2 percent. The second layer is agents. An agent assembled in the factory bundles connectors, MCP services, other agents, and existing flows together with business logic. The bundle becomes a single deployable unit. The unit composes. An agent built today can be a building block in another agent tomorrow. A flow that handles a specific business process can be invoked from any agent that needs that process as a step. This composability is what makes the factory more than a faster way to build one agent. It is a way to grow a library of capabilities that build on each other. Agents the factory produces are published to the [Agent Directory](https://flowrunner.ai/concepts/agent-directory) for reuse, or deployed directly into a customer’s environment. ### Where the term comes from The factory metaphor is deliberate. Factories produce instances of a designed product, repeatably, at a cost that scales. The naming signals that agent creation in FlowRunner is high-volume and routine, not an artisanal one-off. The naming also reflects what makes the FlowRunner approach different from the agent builders most business platforms now offer. Single-vendor agent builders (the ones inside CRM, ERP, or marketing platforms) tend to produce agents that work inside that vendor’s walls. The Agent Factory produces agents that work across systems, compose with each other, and run under the same governance regardless of which vendor’s data they touch. The unit (a deployable agent) is the same; the orchestration environment it lives in is different. The Agent Factory is one of three pillars of [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service). It is where agents enter the orchestration environment. The other two pillars, the [Agent Directory](https://flowrunner.ai/concepts/agent-directory) and the [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct), handle what happens to agents once they exist. ### Frequently asked questions #### Is the Agent Factory the same as a chatbot builder? No. Chatbot builders produce a conversational interface as the deliverable. The Agent Factory is FlowRunner's visual and conversational interface for assembling AI agents from connectors, MCP services, other agents, and existing flows, combined with business logic and built without writing code. The chat is the input, not the output. #### Do I need to write code to build an agent in the Agent Factory? No. Two modes sit on the same factory: a visual editor for assembling agents step by step, and a Conversational AI for describing the agent in plain language. The chat sits on top of the visual editor, so a conversation can hand off to manual fine-tuning at any point. #### What can an agent be built from? Four kinds of building blocks: connectors, MCP services, other agents, and existing flows. Business logic wires them together with triggers, actions, conditions, and variables. Because agents and flows are themselves building blocks, an agent built today becomes a component in a larger agent tomorrow. #### Is the Agent Factory a wrapper over LangChain? No. Agents built here are native FlowRunner entities, deployed and run by the platform, with human-in-the-loop available as a callable tool and Code of Conduct governance enforced by the runtime. The factory does not generate Python or TypeScript for you to maintain. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Automation exceptions Source: https://flowrunner.ai/concepts/automation-exceptions Updated September 8, 2026 Automation exceptions are the items in an automated process that need human judgment when the rules do not cleanly apply, the data is incomplete, or the cost of being wrong is too high to automate. **TL;DR** - Automation exceptions are the items in an automated process that need human judgment. - Three triggers: the rules do not cleanly apply, the data is incomplete, or the cost of being wrong is too high to automate. - Exception is the word operations leaders naturally use. In FlowRunner’s prospect interviews, it surfaced unprompted in 5 of 15 conversations as the most resonant single word in the buyer’s vocabulary. - Exceptions are why agents stop and ask a human for help. They are the focal point of the orchestration platform, not an edge case. ### What it means Most automation tools imagine processes that follow rules. Build the rules, automate the process, done. The world does not work that way. Every process has items that the rules do not cleanly cover. Those items are exceptions. The three triggers cover the common cases: - **The rules do not cleanly apply.** The invoice has a discount line the policy does not anticipate. The vendor sent two different W-9 forms with conflicting numbers. The contract has a clause the standard template never accounts for. Rules can be written for the common case; exceptions are the items where the rules and the case do not match. - **The data is incomplete.** The vendor record is missing a tax ID. The customer’s address has a typo. The line item description has nothing in it. The system has fields where information should be; the case has fields where information is missing. - **The cost of being wrong is too high to automate.** The decision is technically within the rules but the consequences of a wrong answer outweigh the value of the speed. Wires above a certain threshold. Refunds above a certain limit. Anything customer-facing on a top-10 account. In a well-orchestrated process, the rules handle the items they can handle. Exceptions are routed to humans. The humans only see the items that need their judgment, not the whole flow of work. ### What it is not An exception is not an error. Errors are when something goes wrong (the API returned a 500, the database timed out, the file was malformed). Exceptions are when something needs judgment (the rule did not cleanly apply). Errors are a system problem to fix. Exceptions are a business decision to make. An exception is not an edge case. Edge cases are rare and unusual; exceptions are routine. A finance team can expect 5 to 15 percent of invoices to require human judgment in any given month. That is not an edge case; that is the baseline. An exception is not an exception in the programming sense. Programming exceptions are control flow constructs (try/catch/throw) for handling errors. The qualifier “automation” disambiguates: we mean exceptions in the workflow and business process sense, not the language-feature sense. An exception is not a failure. A process that produces exceptions is a process that is working correctly. The exceptions are the items the system flagged for human attention, which is what it was supposed to do. A process that produces no exceptions and runs through everything automatically is not better; it is suspicious. ### How FlowRunner implements it Exception detection happens at runtime, not at design time. The agent or workflow runs the process. At each step, the platform checks the conditions that mark an item as an exception: a rule mismatch, missing data, a threshold breach. When an exception is detected, the platform invokes [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) with the context the human needs to decide quickly. Specifics: - **Detection triggers are configurable per workflow.** What counts as an exception depends on the business. The platform supports the common patterns (confidence thresholds, rule mismatches, missing required fields, monetary limits) and lets the workflow designer add others. - **Routing is multi-channel.** Exceptions go to humans through the channel they actually use: Slack, email, WhatsApp, phone. The orchestration layer picks the channel based on who is being asked. - **Context is assembled.** The human receiving the exception sees the data, the reasoning that led to the flag, and the choices available. The handoff is a scannable case with options, not a vague “have a look at this.” - **Resolution is captured.** What the human decided is recorded in the audit trail. The workflow resumes from where it paused. The decision becomes input to subsequent steps. Exceptions are also the data that improves the system over time. The patterns that humans resolve consistently in the same way become candidates for new rules. The patterns that humans struggle with become candidates for process redesign. The audit trail makes both kinds of pattern visible. ### Where the term comes from The word is the buyer’s word. In 15 prospect calls FlowRunner ran during early sales, _exception_ surfaced unprompted in 5 of them. It is the single word operations leaders most naturally use when describing the work that does not fit their automated processes today. The qualifier “automation” was added in this concept page to disambiguate from the programming meaning. In conversation, operations leaders just say “exceptions” and the context makes the meaning clear. The longer form is for written content, where the search and AI systems benefit from the disambiguator. Exceptions are central to how the category we are defining ([Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service)) treats human and AI work. Rules handle the items they can handle. Exceptions are routed through [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop). The combination is what produces the system prospects described as a [digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord): automation that knows when to stop. ### Frequently asked questions #### Is an automation exception the same as an error? No. Errors are when something goes wrong: an API returned a 500, a database timed out, a file was malformed. Automation exceptions are the items in an automated process that need human judgment when the rules do not cleanly apply, the data is incomplete, or the cost of being wrong is too high to automate. #### Are automation exceptions rare? No. Exceptions are routine, not edge cases. A finance team can expect 5 to 15 percent of invoices to require human judgment in any given month. A process that produces no exceptions and runs everything through automatically is not better than one that flags them; it is suspicious. #### How does FlowRunner decide what counts as an exception? Detection happens at runtime and the triggers are configurable per workflow. The platform supports confidence thresholds, rule mismatches, missing required fields, and monetary limits, and the workflow designer can add others. When one fires, the platform invokes human-in-the-loop with the context the human needs. #### What happens to an exception after a human resolves it? The decision is recorded in the audit trail, the workflow resumes from where it paused, and the decision becomes input to the next steps. Over time the patterns humans resolve the same way become candidates for new rules, and the ones they struggle with point at process redesign. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## BYOK (Bring Your Own Keys) Source: https://flowrunner.ai/concepts/byok Updated September 8, 2026 BYOK is a model where the customer supplies their own AI provider API credentials, pays providers directly for usage, and FlowRunner orchestrates calls across providers without taking a markup on inference. **TL;DR** - BYOK is short for Bring Your Own Keys. The customer supplies their own AI provider API credentials to FlowRunner. - The customer pays AI providers (Anthropic, OpenAI, and others) directly for inference. FlowRunner does not take a markup. - The same workflow can call different providers for different steps. An invoice agent can use Claude for extraction and a different model for classification within the same run. - The provider relationship belongs to the customer. Switching providers, negotiating with them, or auditing usage happens directly between the customer and the provider, not through FlowRunner. ### What it means When an AI agent does work, it calls an AI provider to do the inference. Someone has to pay the provider for that inference. Most AI platforms include the inference cost in their pricing, either by reselling provider capacity at a markup or by bundling a fixed inference allowance into the tier. BYOK is a different model. The customer maintains their own account with the AI provider. The customer supplies the API credentials to FlowRunner. FlowRunner uses those credentials to make calls on the customer’s behalf. The inference cost goes to the provider’s bill, not to FlowRunner’s. Three properties follow from the model: - **No markup on inference.** FlowRunner charges for the orchestration platform (tier-based pricing for workflows, executions, audit retention, and so on). FlowRunner does not charge for the AI tokens the agents consume. The customer pays the provider what the provider charges. - **Provider flexibility.** A FlowRunner workflow can use any provider the customer has credentials for. A single agent can use one provider for one step and a different provider for another step, picked based on which model is best for that specific job. - **Cost transparency.** The customer’s AI bill comes from the provider, line-itemed by usage. The customer can see exactly which workflows consumed which models for which work. Optimizing AI cost becomes a matter of looking at the provider’s invoice, not reverse-engineering a platform’s bundled pricing. ### What it is not BYOK is not a discount on AI. FlowRunner does not sell AI capacity; the customer buys it from providers directly. The platform fee is for orchestration, not inference. BYOK is not a billing trick. The arrangement is operational: the agents make actual API calls using the customer’s keys, the provider records the usage against the customer’s account, the provider bills the customer. There is no obfuscation layer. BYOK is not provider lock-in by FlowRunner. The customer’s relationship is with the AI provider. Switching providers, adding new providers, or canceling FlowRunner does not affect the underlying provider relationship. BYOK is not the only deployment model. Customers who prefer FlowRunner to handle provider billing have that option in some tiers. BYOK is the default model because it is cleaner for cost accounting and compliance. ### How FlowRunner implements it The mechanics: - **Credential management.** The customer supplies API keys for each provider they want the platform to use. The platform stores them securely and uses them to make inference calls on the customer’s behalf. - **Provider selection per workflow.** Each agent or workflow specifies which provider to call for which step. The selection can be based on cost, capability, latency, or compliance. - **Multi-provider orchestration.** A single agent run can call multiple providers. The platform handles the provider switching, response normalization, and any error handling needed. - **Usage records on the platform side.** The platform records which workflows called which providers, for which work, in the audit trail. The provider’s invoice is the source of truth on cost; the platform’s records are the source of truth on what work was being done at the time. The combination matters for buyers in regulated industries. Compliance and finance can look at the provider’s invoice for cost. They can look at the platform’s audit trail for what the AI was doing when the cost was incurred. The two records reconcile against each other. ### Where the term comes from BYOK is an industry term that started in SaaS, originally for security and customer-managed encryption keys. The pattern was simple: the customer holds the keys to their own data; the SaaS vendor cannot access the data without the customer’s keys. The AI provider version of BYOK borrows the same pattern. The customer holds the API keys; the AI vendor (FlowRunner, in this case) uses the customer’s keys to call providers on the customer’s behalf. The customer retains the provider relationship, the billing relationship, and the ability to audit usage at the provider level. What FlowRunner adds to the industry pattern is the orchestration angle. Because the platform supports multiple providers under the same set of customer keys, an agent or workflow can pick the best provider for each step. This is harder when the platform is reselling provider capacity, because the platform’s incentives push toward whichever provider it has the best margin on. With BYOK, the platform has no margin to defend. It calls the provider the workflow says to call. BYOK is one of several deployment characteristics that distinguish [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service) from AI-platform-plus-billing models. The orchestration platform’s job is to coordinate, govern, and supervise the AI work. The AI providers’ job is to do the inference. BYOK keeps the two roles separate. ### Frequently asked questions #### Does BYOK cost more than a platform that bundles AI usage? No. FlowRunner takes no markup on inference, so the AI bill comes from the provider at the provider's own rate. BYOK is a model where the customer supplies their own AI provider API credentials, pays providers directly for usage, and FlowRunner orchestrates calls across providers without taking a markup on inference. #### Is BYOK the same as bring your own encryption key? No, though the name is borrowed. Encryption BYOK means the customer holds the keys to their own data so the vendor cannot read it. AI BYOK means the customer holds the API credentials for their AI providers, keeps the provider relationship, and pays for inference directly. #### Can one workflow use more than one AI provider? Yes. A single agent run can call different providers for different steps, picked on cost, capability, latency, or compliance. An invoice agent can use one model for extraction and another for classification in the same run. FlowRunner handles provider switching and response normalization. #### Which FlowRunner plans include BYOK? All of them, starting with the Free plan at $0. AI agents with BYOK, all integrations, human-in-the-loop, and unlimited users and workflows are on every tier. Plans differ on executions, concurrency, and log history. Your AI spend is separate and goes straight to the provider. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Code of Conduct Source: https://flowrunner.ai/concepts/code-of-conduct Updated September 8, 2026 The Code of Conduct is the platform-enforced governance layer that defines how AI agents communicate, negotiate conflicts, escalate to humans, and produce audit trails when working alongside other agents. **TL;DR** - The Code of Conduct is the rule layer that AI agents follow when they operate inside FlowRunner. - It is the layer that guarantees agents stop and ask a human for help at the right moments, regardless of who authored the agent or how it was configured. Without that guarantee, the “agents that ask” promise is per-agent goodwill, not platform infrastructure. - Four functions: 1. How agents communicate with each other 2. How they resolve conflicts 3. When they escalate to humans 4. What audit trails they leave behind - Audit trails and human escalation are shipped capabilities today. Inter-agent negotiation and conflict resolution are the forward direction as the multi-agent world matures. - The point is that governance is infrastructure, not paperwork. The platform handles the rules so the customer does not have to remember to apply them. ### What it means When a single agent runs alone, governance is mostly about logging. What did the agent do? With what data? On whose behalf? When? When many agents run together, governance gets harder. Two agents try to act on the same record. An agent makes a decision that depends on data another agent owns. A compliance auditor asks who decided what and when, and the answer has to assemble across the entire orchestration environment. The Code of Conduct is the layer that handles the harder version. It defines, and the platform enforces, four kinds of agent behavior: - **Communication.** How agents share state, hand off context, and call each other. The Code of Conduct defines the protocols so the calls are consistent regardless of which team or partner authored the agent. - **Conflict negotiation.** What happens when two agents have contradictory instructions or try to act on the same data. The protocol determines who yields, who escalates, and what the resolution path looks like. - **Escalation to humans.** When an agent must stop and ask. The Code of Conduct enforces escalation triggers consistently, regardless of which agent encounters the condition. [Human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) is the execution pattern; the Code of Conduct is what guarantees the pattern is invoked when it should be. - **Audit trails.** What gets recorded so the chain of decisions can be reconstructed. Who decided what. With what input. Under whose authority. The audit trail covers the orchestration as a whole, not just individual agents in isolation. The word _enforce_ is doing important work. The Code of Conduct is not a policy document the customer signs and hopes their agents follow. It is platform infrastructure. The orchestration runtime applies the rules whether the agent author wrote them in or not. ### What it is not The Code of Conduct is not RBAC. Role-based access control governs which _humans_ can do which things. The Code of Conduct governs which _agents_ can do which things, and how they interact with each other. The two are complementary, not the same. The Code of Conduct is not audit logs. Logs are a recording surface. Governance is an enforcement surface. The platform records what happened (logs) and also constrains what was allowed to happen (Code of Conduct). Recording without enforcement is post-hoc forensics; enforcement without recording is unaccountable. The Code of Conduct is not a configuration screen for individual agent permissions. Per-agent permissions exist as a layer, but they are not the Code of Conduct. The Code of Conduct sets the rules every agent in the environment follows by default, regardless of who authored the agent or which customer deployed it. The Code of Conduct is not a written policy or compliance document. It is enforced code. The customer’s policies and the regulator’s requirements inform what the Code of Conduct enforces, but the Code of Conduct itself is the runtime enforcement layer, not the paperwork. ### How FlowRunner implements it What is shipped today: - **Audit trails.** Every agent action, every human handoff, every decision routed through [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) is recorded with the data, the actor, and the timestamp the auditor needs. SLA tracking and audit-retention windows are available at the appropriate FlowRunner tiers. - **Human escalation enforcement.** The platform guarantees that the escalation conditions configured for an agent are honored. An agent cannot bypass the escalation rule by accident or by design. - **Role-based access control.** Permissions over which humans can author, deploy, and operate which agents. What is the forward direction: - **Agent negotiation protocols.** As more agents enter the orchestration environment and start needing to work together, the Code of Conduct will define the protocols by which they communicate and resolve disagreements. The architecture supports this work; the protocols mature as the multi-agent world materializes. - **Cross-agent communication standards.** A consistent way for agents from different sources (built in the [Agent Factory](https://flowrunner.ai/concepts/agent-factory), installed from the [Agent Directory](https://flowrunner.ai/concepts/agent-directory)) to interoperate without each customer wiring it themselves. - **Conflict resolution.** Platform-level rules for handling the edge cases that emerge when many agents operate on the same data. The Code of Conduct also intersects with [Flow conversation](https://flowrunner.ai/concepts/flow-conversation). When a human gives a running flow new instructions through a conversational channel, the Code of Conduct governs which operators are allowed to give which instructions to which flows, and what kinds of behavior changes are in bounds. ### Where the term comes from Codes of conduct are the older idea this name borrows from. Professional codes (medical, legal, engineering) define the conduct expected of practitioners, with enforcement mechanisms when conduct falls short. The metaphor maps to AI agents cleanly enough: there should be a defined standard of behavior; the platform should enforce it; falling short should have consequences. The naming is also a positioning bet. Most orchestration discussions today center on capability (“what can the agents do?”). The Code of Conduct frames the more important question as one of regulation (“what should the agents do, and how do we make sure they do it?”). That frame is closer to how operations leaders and compliance officers actually think about adding AI to their environments. The Code of Conduct is one of three pillars of [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service). The [Agent Factory](https://flowrunner.ai/concepts/agent-factory) is where agents come from. The [Agent Directory](https://flowrunner.ai/concepts/agent-directory) is where they live for reuse. The Code of Conduct is what keeps them from becoming a mess once there are many of them. ### Frequently asked questions #### Is the Code of Conduct the same as RBAC? No. RBAC governs which humans can do which things. The Code of Conduct is the platform-enforced governance layer that defines how AI agents communicate, negotiate conflicts, escalate to humans, and produce audit trails when working alongside other agents. FlowRunner ships both, and they are complementary. #### Is the Code of Conduct a written policy document? No. It is enforced code, not paperwork. The orchestration runtime applies the rules whether the agent author wrote them in or not, so an agent cannot bypass an escalation condition by accident or by design. Your policies inform what it enforces; they are not what it is. #### What parts of the Code of Conduct exist today? Audit trails, human escalation enforcement, and role-based access control ship today. Inter-agent negotiation protocols, cross-agent communication standards, and conflict resolution are the forward direction as multi-agent environments mature. The architecture supports that work; the protocols firm up as the need arrives. #### Which FlowRunner plan do I need for audit trails and compliance features? Professional at $299 per month adds 30-day audit trails and RBAC. Business at $999 adds 90-day audit trails, SLA tracking, SSO and SAML, and compliance reporting. Enterprise is custom, with unlimited audit retention and a self-hosted option. Escalation enforcement itself runs on every plan. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Connectors Source: https://flowrunner.ai/concepts/connectors Updated September 8, 2026 Connectors are reusable bridges between FlowRunner and third-party services that any agent or workflow can compose into a capability. **TL;DR** - Connectors are the bridges between FlowRunner and every third-party service your agents and workflows talk to. - A connector is built once and used everywhere. The same QuickBooks connector serves the invoice agent, the expense agent, and the reporting agent, without separate work for each. - New connectors enter the catalog at 30 minutes or less of measured production time. Error rate on those AI-generated integrations is 0.2 percent. - Connectors carry the data the agents use to decide. They are the substrate that lets the agents that know when to stop and ask a human for help actually know what is happening. ### What it means An agent that handles invoices needs to read invoices from email or a document service, look up vendor records in the accounting system, post journal entries, and send Slack messages when something needs review. None of those actions are part of the AI model or the agent’s logic. They are calls to external services. The thing that lets the agent make those calls is a connector. A connector does three things: - **Exposes the actions a third-party service offers.** The QuickBooks connector exposes “create invoice,” “look up vendor,” “post journal entry,” and the rest of what QuickBooks lets you do through its API. - **Hides the integration mechanics.** The agent does not handle authentication, retry logic, rate limits, schema variation between versions, or the dozens of small things that make calling a real service different from calling a documented API. The connector handles those. - **Stays reusable across the orchestration environment.** The same connector is called by any agent, any workflow, any flow that needs to talk to that service. There is one connector per service, not one per use case. The third property is the one that matters most for orchestration. As the number of agents and workflows in an environment grows, the cost of per-use-case integration work grows linearly with it unless connectors are reused. With reuse, the integration work happens once per service, not once per use case. ### What it is not A connector is not an API. The API is what the third-party service publishes. The connector is the FlowRunner-side bridge that consumes the API and exposes its actions in a form agents and workflows can call. A connector is not an adapter pattern. Adapters in software engineering translate between two interfaces at runtime. Connectors are persistent, reusable, catalog-level entities that the platform manages, not lightweight runtime translators. A connector is not what other workflow tools call an “integration.” Most workflow automation tools (Zapier, Make) bundle the trigger, the action, the data shape, and the use case into one integration unit. Connectors separate the bridge to the service (one connector) from the use cases that consume it (many agents and workflows). A connector is not a single action. A connector exposes the set of actions a third-party service offers; an agent picks the actions it needs from the connector and uses them. ### How FlowRunner implements it The connector layer sits underneath the Agent Factory. Both layers are part of the same factory architecture, but they serve different roles: connectors are the bridges, agents are the capabilities composed from them. Connector production is fast because the factory’s connector-creation tooling does most of the integration work that engineers would otherwise do by hand: reading the third-party API surface, mapping its actions, wiring authentication, handling pagination and error cases. The current measured production time for a new connector is 30 minutes or less, and the error rate on the AI-generated integrations the factory produces is 0.2 percent. The catalog covers the services FlowRunner’s customers ask for: Salesforce, QuickBooks, NetSuite, Acumatica, Slack, HubSpot, Stripe, ShipBob, Parseur, and many more. New connectors enter the catalog as integration needs surface; a service that does not have a connector today can have one by the end of the day, not the end of the quarter. Connectors compose with the rest of the orchestration model. A connector is one of the four kinds of building blocks the [Agent Factory](https://flowrunner.ai/concepts/agent-factory) assembles agents from. An agent assembled from connectors operates under the [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) governance the platform enforces. ### Where the term comes from _Connector_ is a standard term in software engineering. The choice to use the standard term rather than coining a new one is deliberate. The category we are defining ([Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service)) is built on terms operations leaders and technical buyers recognize without translation. What FlowRunner adds to the standard idea is the production-speed expectation and the reuse contract. A connector built in 30 minutes is qualitatively different from a connector built in a month. The first one fits into a workflow where new integration needs are routine. The second one is a special project. The factory’s job is to keep the first model viable as the customer’s environment grows. ### Frequently asked questions #### Is a FlowRunner connector the same as a Zapier integration? No. Zapier and Make bundle the trigger, the action, the data shape, and the use case into one integration unit. Connectors are reusable bridges between FlowRunner and third-party services that any agent or workflow can compose into a capability. One connector per service, many use cases on top. #### What is the difference between a connector and an API? The API is what the third-party service publishes. The connector is the FlowRunner-side bridge that consumes that API and exposes its actions in a form agents and workflows can call. The connector handles authentication, retries, rate limits, and schema variation so the agent does not have to. #### What happens if FlowRunner does not have a connector for a service I use? It gets built. New connectors enter the catalog at a measured production time of 30 minutes or less, with a 0.2 percent error rate on the AI-generated integrations the factory produces. A service without a connector today can have one by the end of the day. #### Do FlowRunner plans limit which connectors I can use? No. All integrations are on every plan, including Free at $0. Executions, concurrency, and log history are the only meters that change between tiers. The same connector serves every agent and workflow that needs the service, so there is no per-use-case integration charge. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Digital andon cord Source: https://flowrunner.ai/concepts/digital-andon-cord Updated September 8, 2026 The digital andon cord is a metaphor for AI that stops the line when it hits uncertainty, borrowed from the Toyota Production System pull cord that signals a line needs human judgment. **TL;DR** - The metaphor came from a prospect call. A CEO described what he wanted from AI as “a digital andon cord.” - In Toyota’s factories, any worker on the assembly line can pull a cord to stop production when something looks wrong. Stopping the line is a feature, not a failure. - In AI workflows, the same principle applies: agents that stop and ask a human for help when they hit uncertainty produce higher quality than agents that confidently guess. - The metaphor captures, in three words, what AI ought to do when it is not sure: pause, signal, wait for judgment. ### What it means In a Toyota factory, every workstation has a cord (originally a light, then a cord) within arm’s reach. The cord is called the andon. Any worker who notices a defect, an irregularity, or anything they cannot resolve on their own pulls the cord. Pulling the cord lights up a board, alerts the team leader, and stops the line. The first time you hear this it sounds like a productivity disaster. Cars stop being built. Workers stop working. But the Toyota Production System produced some of the highest quality and lowest defect rates in industrial history, and the andon cord is one of the reasons. The line stopping is exactly what you want. The cost of catching a defect at the workstation is small. The cost of catching it after fifty more cars have been built on top of it is huge. A digital andon cord is the same idea applied to AI workflows. The agent runs. When the agent hits uncertainty (the data is incomplete, the rule does not cleanly apply, the cost of being wrong is too high), it pulls the cord. The workflow pauses. The right human is notified. They make the call. The workflow resumes. What the metaphor captures that the industry has not is the _desirability_ of stopping. AI that never stops is presented as a strength: full autonomy, no humans in the way. In practice, AI that never stops produces confident wrong answers at scale. AI that knows when to stop produces accurate ones. ### What it is not The digital andon cord is not error handling. Errors are exceptions the developer anticipated and wrote code for. The andon cord is for the situations the developer did not anticipate (or could not have). The digital andon cord is not an interruption. Interruptions are unwanted breaks in flow. The andon cord is a deliberate signal that the workflow needs attention. It is wanted, not unwanted. The digital andon cord is not a sign of weak AI. The opposite. AI that can identify the boundary of its own competence is more mature than AI that cannot. Knowing what you do not know is harder than answering confidently. The digital andon cord is not the same as routing every decision through a human. Toyota workers do not pull the cord for every car. They pull it when something is wrong. The digital andon cord is the _exception_ pattern, not the default. ### How FlowRunner implements it In FlowRunner, the digital andon cord is the [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) capability. AI agents pull the cord by invoking a human-in-the-loop step at runtime. The step routes the decision to a person through the channel they use, delivers full context, captures the response, and feeds it back into the workflow. What triggers the cord pull is configurable. Common triggers: - **Confidence threshold.** The AI is below a configured confidence level on a classification or extraction. - **Rule boundary.** A rule fires that says “this kind of case always needs a human” (every wire over a threshold, every customer with a contract exception, every refund over a limit). - **Missing data.** The agent does not have enough information to decide. Asking the human is faster and safer than guessing. - **Cost of error.** The downstream cost of being wrong is high enough that even moderate uncertainty justifies the pause. What makes the implementation production-ready is the same set of properties that distinguishes [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) from a basic approval gate: the pause is autonomous, the routing is multi-channel, the context is preserved, the resumption is seamless. ### Where the term comes from A CEO in one of our prospect calls said: “Sounds like a digital andon cord.” He was responding to a demo of FlowRunner’s human-in-the-loop capability. He had spent years running operations at scale and recognized the pattern immediately. The phrase has stuck because it does what good metaphors do: it makes a new idea instantly familiar by mapping it onto a well-understood older one. The Toyota Production System framing also does work the marketing version cannot. It signals to operations leaders that stopping the line is a feature of the system, not a confession of weak AI. Operations leaders who came up through manufacturing or supply chain often know the andon cord story already and do not need the framing explained. For everyone else, the short version is this: the most reliable systems are the ones that know when to stop. The category FlowRunner is defining ([Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service)) treats that knowing-when-to-stop capability as core platform infrastructure, not an afterthought. ### Frequently asked questions #### Is a digital andon cord the same as human-in-the-loop? They describe the same capability from different angles. The digital andon cord is a metaphor for AI that stops the line when it hits uncertainty, borrowed from the Toyota Production System pull cord that signals a line needs human judgment. Human-in-the-loop is the execution pattern that implements it. #### Does stopping the line mean the AI is weak? The opposite. AI that can identify the boundary of its own competence is more mature than AI that cannot. Toyota's line-stopping practice produced some of the lowest defect rates in industrial history. AI that never stops produces confident wrong answers at scale; AI that knows when to stop produces accurate ones. #### Does the cord get pulled on every workflow run? No. Toyota workers do not pull the cord for every car, and agents do not pull it for every run. It is the exception pattern, not the default. Triggers are configurable: a confidence threshold, a rule boundary, missing data, or a downstream cost of error high enough to justify the pause. #### Is the digital andon cord just error handling? No. Errors are the cases a developer anticipated and wrote code for. The cord is for the situations nobody anticipated, or could not have. It is also not an interruption: it is a deliberate signal that the workflow needs attention, wanted rather than unwanted. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Flow conversation Source: https://flowrunner.ai/concepts/flow-conversation Updated September 8, 2026 Flow conversation is a real-time, bidirectional communication capability that lets humans query, instruct, and guide active FlowRunner workflows without pausing execution or losing state. **TL;DR** - Flow conversation is how a human talks to an AI workflow that is already running. - Three kinds of communication: queries (where are you, what are you doing), instructions (change priority, add this constraint), and guidance (here is new context the flow should incorporate). - The flow does not stop. State is preserved. The conversation happens alongside execution, not instead of it. - When a buyer asks “can I check on my AI workflows in real time,” or “can I give a running agent new instructions,” or “is there a way to talk to my AI workers without interrupting them,” flow conversation is the FlowRunner answer. ### What it means Most workflow tools give operators two options: build the workflow ahead of time, then watch a dashboard after it runs. The dashboard is passive. It shows what happened. It does not let the operator interact with the workflow while it is happening. Flow conversation is the third option. The operator can communicate with the running flow at any point. The flow keeps running. The conversation is bidirectional. Three kinds of communication compose: - **Queries.** The operator asks the flow about itself. _Where are you in the process? What records have you handled? Which exceptions are pending? Why did you route that last item to manual review?_ The flow answers with its state, its decisions, and its reasoning. No screenshots, no spreadsheet pulls, no waiting for a dashboard refresh. - **Instructions.** The operator gives the flow a new constraint or change. _Skip the Acme account until I tell you otherwise. Stop sending anything to legal until Tuesday. Treat invoices over $50,000 as exceptions for the next 48 hours._ The flow incorporates the instruction into its execution without restart. - **Guidance.** The operator adds context the flow could not have known at build time. _A new policy went into effect this morning, here is the document. The CFO wants to see anything over $10K personally for the next two weeks. The Acme team is renegotiating their contract, hold any disputes._ The flow factors the guidance into the decisions it makes from that point on. In each case the flow keeps running. There is no pause, no restart, no replay. The conversation happens in the same execution context as the work itself. ### What it is not Flow conversation is not [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop). HITL is request-response at exception points. The agent stops, the human decides, the agent resumes. Flow conversation does not require a stop. The human initiates, the flow keeps running, the dialogue happens alongside. Flow conversation is not a dashboard. Dashboards are passive. They show. Flow conversation is active. It listens, answers, and acts on instructions. Flow conversation is not a chatbot. Chatbots are wrappers around language models that answer questions. Flow conversation is the actual workflow on the other end of the message, including the state of every step it has run, the records it has handled, and the decisions it has made. Flow conversation is not ChatOps. ChatOps is operations control via chat (deploying through Slack, triggering scripts from messages). It treats chat as a control surface for static tools. Flow conversation is a control and dialogue surface for stateful running workflows. Flow conversation is not voice-based AI. The medium is conversational, but the substance is the workflow’s state and behavior, not a virtual assistant. ### How FlowRunner implements it Flow conversation is a runtime capability of every FlowRunner workflow. The mechanics: - **Message channel.** Operators communicate with a running flow through the same channels FlowRunner uses for [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop): Slack, email, WhatsApp, phone. The same identity, the same workspace, the same conversation thread. - **State-aware responses.** When the operator queries the flow, the flow answers from its own state. It knows where it is in the process, which records it has handled, which exceptions are pending. The answer is what is actually true at that moment, not what a dashboard reflects after the next refresh. - **Instruction injection.** When the operator gives the flow a new constraint or guidance, the flow incorporates it into the next decisions it makes. State is preserved. Other work in progress is not affected. - **Governed by the Code of Conduct.** The [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) defines which operators can instruct which flows and what kinds of instructions are allowed. Conversation is governed, not free-form. Audit trails record every query, instruction, and guidance message alongside the flow’s response and the resulting behavior change. The pattern composes with the rest of the orchestration model. A flow can be in the middle of executing, paused for a human-in-the-loop decision on one branch, and simultaneously in a flow conversation with a different operator giving it guidance on a different branch. The platform tracks the whole thing. ### Where the term comes from The capability does not have an established industry name. The closest existing terms each miss something important: - _ChatOps_ describes ops tooling, not workflow dialogue. - _Conversational AI_ describes chatbots, where the conversation is the product. - _Workflow monitoring_ describes passive observation, not bidirectional control. - _Human-in-the-loop_ describes pause-for-decision moments, not ongoing dialogue. - _Agentic interface_ is too vague; it could mean any UI for an AI agent. We are using _flow conversation_ because the FlowRunner unit of work is a flow, and conversation is the right word for a bidirectional, multi-turn, real-time exchange. The term is distinctive enough to be ownable and clear enough to be understood on first read. The capability sits inside the broader category of [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service). Most orchestration platforms treat the operator as an observer. FlowRunner treats the operator as a participant. Flow conversation is the participation surface. ### Frequently asked questions #### Is flow conversation the same as human-in-the-loop? No. Human-in-the-loop is request-response at pause points: the agent stops, the human decides, the agent resumes. Flow conversation is a real-time, bidirectional communication capability that lets humans query, instruct, and guide active FlowRunner workflows without pausing execution or losing state. The human starts it, and nothing stops. #### What can I actually say to a running workflow? Three kinds of message. Queries ask the flow about its own state, decisions, and reasoning. Instructions add a constraint, such as holding one account or treating invoices over $50,000 as exceptions for 48 hours. Guidance supplies new context the flow could not have known at build time. #### Is flow conversation the same as ChatOps? No. ChatOps treats chat as a control surface for static tools: deploying through Slack, triggering scripts from messages. Flow conversation is a control and dialogue surface for stateful running workflows. The thing answering is the workflow itself, with its own execution state, not a bot in front of it. #### Can anyone give instructions to any running flow? No. The Code of Conduct defines which operators can instruct which flows and what kinds of instruction are in bounds. Conversation is governed, not free-form. Every query, instruction, and guidance message is recorded in the audit trail alongside the flow's response and the resulting behavior change. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Human-in-the-loop Source: https://flowrunner.ai/concepts/human-in-the-loop Updated September 8, 2026 Human-in-the-loop is an execution pattern where AI agents pause autonomously, assemble the relevant context and the decision choices available, route to a human via their preferred channel, and resume the moment the human responds. **TL;DR** - Human-in-the-loop is how AI agents stop and ask a human for help when they hit a decision they should not make on their own. - Four moves define the pattern: pause autonomously, assemble context and decision choices, route to a human via the right channel, resume on response. - The pattern is not an approval gate stapled onto an automation. It is a runtime capability the agent uses as a callable tool. - In FlowRunner, agents invoke human-in-the-loop the way a function calls another function. The assembled context and pre-structured choices make the decision fast. ### What it means Most automation tools think about human review as a checkpoint. Build the workflow. Add an approval step. The approval step blocks until a human clicks yes or no. Every workflow run hits the same gate whether the underlying decision is hard or trivial. Human-in-the-loop is different. It is a runtime pattern the AI agent uses on its own initiative. The agent runs as far as it can run on its own. When it hits a decision it should not make, it pauses execution, contacts a human through the channel that human actually uses, and resumes the moment the human answers. The pattern has four specific moves: 1. **Pause autonomously.** The agent decides when to stop, based on what it sees. Sometimes that is a confidence threshold (the AI is below 90% sure). Sometimes it is a rule (any wire over $10,000 needs review). Sometimes it is missing data (the invoice references a vendor the system has never seen). The agent stops itself; nobody had to anticipate every case at build time. 2. **Assemble the context and the decision choices.** The agent does the editorial work before it ever contacts the human. It pulls the data the human will need to understand the situation, summarizes the reasoning that led to the pause, and structures the available choices so the human picks rather than composes. The handoff is a scannable case with options, not a vague “have a look at this.” 3. **Route to a human via the right channel.** Different decisions need different humans, and different humans live in different tools. Some respond fastest in Slack. Some answer email. Some need a phone call. The orchestration layer picks the channel based on who is being asked. 4. **Resume the moment the human responds.** The decision flows back into the agent’s execution. The agent continues from where it paused. Other steps in the workflow do not have to be replayed. State is preserved. Each of those four moves is the difference between human-in-the-loop and the cheap version. Skip any one and you get something less useful. ### What it is not Human-in-the-loop is not an approval gate. Approval gates are static checkpoints in a workflow. Every run hits the same gate. The human sees a name and a yes/no button. They have no context for the decision they are being asked to make. Human-in-the-loop is not Slack-based confirmation. Routing a “click yes to continue” message into Slack is a delivery mechanism. The hard parts are deciding when to pause and what context to send. Human-in-the-loop is not the ML training term. In machine learning, “human in the loop” refers to humans labeling training data or correcting model outputs to improve future performance. In production workflows, human-in-the-loop is a runtime pattern for handling decisions the AI should not make autonomously. The terms share a name and almost nothing else. Human-in-the-loop is not the same as keeping a person in charge. Plenty of companies keep humans in charge of everything by refusing to automate. That is just manual work. Human-in-the-loop means automating what can be automated and routing only the judgment to humans. ### How FlowRunner implements it In FlowRunner, human-in-the-loop is a first-class capability AI agents invoke as a callable tool. The agent decides at runtime that it has hit a decision it should not make alone. It calls the human-in-the-loop tool. The platform handles routing, context delivery, response capture, and resumption. A few specifics that matter: - **Channel routing.** The platform supports email, Slack, WhatsApp, and phone as decision channels. The agent or the workflow designer specifies which to use for which kind of decision. - **Context delivery.** The platform delivers a complete, scannable case to the human: the data, the reasoning that led to the pause, the options available, the cost of being wrong. The human can decide in seconds rather than reverse-engineering the situation. - **Subflows as callable tools.** A complex human review can itself be a subflow that the agent invokes. The subflow can include conditional routing (different humans for different conditions), escalation rules (auto-escalate if no response in 15 minutes), and structured response capture. - **State preservation.** The agent’s execution state is preserved across the pause. When the human responds, the agent resumes exactly where it stopped, with the human’s decision available as input to the next step. The pattern composes. An agent can invoke a human-in-the-loop step. The step can invoke another agent. That agent can invoke another human. The orchestration layer tracks the whole chain and produces a single audit trail. ### Where the term comes from The phrase “human in the loop” originated in control systems and military command, then spread to machine learning. Both of those uses describe humans positioned somewhere inside a system’s feedback cycle. FlowRunner uses the term in its production workflow sense: AI agents executing automated work pausing for human judgment at the points that need it. A prospect in one of our calls described what they wanted as a [digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord), which is the same idea in different language. We use both terms. The pattern is the differentiator. The category we are defining, [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service), treats human-in-the-loop as a core platform capability rather than a feature. The work AI handles is the work AI should handle. The work humans handle is the work humans should handle. The platform routes between them. Human-in-the-loop covers the request-response moments. The related capability for ongoing dialogue with a running workflow is [Flow conversation](https://flowrunner.ai/concepts/flow-conversation). HITL is what an agent invokes when it needs a decision before it can continue. Flow conversation is what a human invokes when they want to check on, instruct, or steer a workflow that has not stopped. ### Frequently asked questions #### Is human-in-the-loop the same as an approval gate? No. An approval gate is a static checkpoint every run hits, with a yes/no button and no context. Human-in-the-loop is an execution pattern where AI agents pause autonomously, assemble the relevant context and the decision choices available, route to a human via their preferred channel, and resume the moment the human responds. #### Does human-in-the-loop mean the same thing in AI workflows as it does in machine learning? No. In machine learning, human-in-the-loop means people labeling training data or correcting model outputs to improve future performance. In production workflows it is a runtime pattern: the agent pauses on a live decision it should not make alone and routes it to a person. Same name, different job. #### Which channels can an agent use to reach a human? Email, Slack, WhatsApp, and phone. The agent or the workflow designer picks the channel per decision type, so the person answering gets the request where they already work. Escalation rules can reroute if nobody responds in time, and the decision flows back into the same execution. #### Do I need a paid FlowRunner plan to use human-in-the-loop? No. Human-in-the-loop is on every plan, including the Free plan at $0. So are AI agents, BYOK, all integrations, and unlimited users and workflows. Plans differ on executions, concurrency, and log history, not on capability. New accounts start with a 14-day Professional trial, no credit card. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Orchestration as a Service Source: https://flowrunner.ai/concepts/orchestration-as-a-service Updated September 8, 2026 Orchestration as a Service is a platform category that coordinates, governs, and supervises multi-agent environments while keeping humans in control of the decisions that require judgment. **TL;DR** - Orchestration as a Service is the platform layer that sits above AI agents and makes them work together. - The category exists because every company is about to run many agents from different vendors with no shared governance. - It covers three jobs: coordinating agents, governing their behavior, and ensuring they stop and ask a human for help at the decisions that should not be automated. - FlowRunner is the first platform purpose-built for this category. Agent builders compete below it; we coordinate them. ### What it means Software vendors are racing to ship AI agents. Salesforce has them. Microsoft has them. OpenAI ships new ones every quarter. Niche startups ship agents for narrow jobs. Inside a year or two, the typical mid-market company will have agents from five to ten different vendors operating across its sales, finance, ops, and support workflows. Nobody has thought through what happens next. Two agents try to update the same record. A human asks one agent for an answer that depends on data another agent owns. A compliance auditor asks who decided what and when, and the answer is scattered across vendor logs in five different formats. That is the coordination problem Orchestration as a Service solves. The category has three responsibilities: - **Coordinate.** Route work to the right agent. Resolve conflicts when two agents try to act on the same data. Hand off context cleanly between agents so the next one can pick up where the last one left off. - **Govern.** Define rules the agents follow when they talk to each other. Produce the audit trails compliance and security teams need. Enforce limits on what each agent is allowed to do. - **Supervise.** Pause when an agent hits a decision that should not be automated. Route that decision to a human through the channel they use. Resume cleanly once the human has answered. The third responsibility is the one most easy to skip and the most expensive to skip. Agents that plow ahead through uncertainty produce confident wrong answers. Agents that know when to stop and ask a human for help produce trust. ### What it is not OaaS is not just an agent builder. Pure agent-building tools like LangChain, CrewAI, and OpenAI’s agent SDK focus on producing capable individual agents and stop there. OaaS includes agent building (FlowRunner does this through the Agent Factory) but adds the layers that make many agents work together: coordination, governance, and supervision. The Agent Factory is the entry point; the orchestration that surrounds it is the destination. OaaS is not workflow automation. Tools like Zapier and Make connect SaaS apps by trigger and action. They do not coordinate AI agents that operate over time, ask questions, and need supervision. OaaS is not iPaaS. Enterprise integration platforms like MuleSoft and Boomi move data between systems. They are not designed around agents as first-class actors that make decisions and need governance. OaaS is not RPA. Robotic process automation drives keystrokes through legacy UIs. AI agents work at the data and intent layer, not the screen. The way to think about it: agent builders ship the workers. OaaS is the floor where the workers coordinate, the supervisor that watches them, and the auditor that records what happened. ### How FlowRunner implements it FlowRunner is the first OaaS platform. The product is organized around three pillars that map directly to the three responsibilities of the category. - The [Agent Factory](https://flowrunner.ai/concepts/agent-factory) is FlowRunner’s visual and conversational interface for assembling AI agents from connectors, MCP services, other agents, and existing flows. It is how new agents enter the orchestration environment without an engineering team. - The Agent Directory is a curated library of pre-built agents that are ready to deploy and reuse. - The [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) is the platform-enforced governance layer that defines how agents communicate, escalate, and leave audit trails. [Human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) is the supervision pattern that runs through all three. Agents do not just produce outputs and hope. They pause autonomously when they hit uncertainty, assemble the context and decision choices, route to a human through the right channel, and resume the moment the human responds. [Flow conversation](https://flowrunner.ai/concepts/flow-conversation) is the complementary capability for ongoing dialogue with running workflows. Operators can query a flow about its state, give it new instructions, or steer it with fresh context without pausing execution. Where human-in-the-loop handles the agent-initiated pause for a decision, flow conversation handles the human-initiated check-in or instruction. The result is a platform a COO can buy with confidence today (it eliminates the manual processing eating their team’s time) and that becomes more valuable every quarter (as the company adds more agents that need coordination). ### Where the term comes from Orchestration as a Service is the category FlowRunner is naming. Other vendors describe what they do as workflow automation, AI agent platforms, or intelligent automation. None of those names captures the coordination, governance, and supervision job that becomes critical when a company runs many agents. The naming choice is deliberate. _Orchestration_ is the right verb because the job is to make many actors play in time, not to produce any single output. _As a Service_ signals that it is a platform commitment, not a feature. The platform takes responsibility for the orchestration; the customer does not have to build it. A category name is a bet that the industry will need a word for this job before too long. FlowRunner is the bet. ### Frequently asked questions #### What is the difference between Orchestration as a Service and workflow automation? Orchestration as a Service is a platform category that coordinates, governs, and supervises multi-agent environments while keeping humans in control of the decisions that require judgment. Workflow automation connects SaaS apps by trigger and action. It does not coordinate agents that operate over time, ask questions, and need supervision. #### Is Orchestration as a Service the same as iPaaS? No. iPaaS platforms like MuleSoft and Boomi move data between systems. Orchestration as a Service treats AI agents as first-class actors that make decisions, need governance, and escalate to humans. iPaaS is designed around data in motion; OaaS is designed around agents in operation. #### Why would a company need an orchestration layer if it already has AI agents? Because the agents come from different vendors and do not coordinate with each other. Two agents update the same record. One needs context another owns. An auditor asks who decided what, and the answer sits in five log formats. The orchestration layer is where coordination, governance, and supervision live. #### What does FlowRunner charge for Orchestration as a Service? Tiers run from Free at $0 to Enterprise at custom pricing, with Starter at $5 or $15, Growth at $45, $89, or $149, Professional at $299, and Business at $999. Every tier includes unlimited users and workflows, AI agents with BYOK, all integrations, and human-in-the-loop. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- - [Compare FlowRunner](https://flowrunner.ai/compare.md): Head-to-head comparisons of FlowRunner against the platforms operations leaders are also evaluating. Honest positioning on where each tool wins. --- ## FlowRunner vs Make Source: https://flowrunner.ai/compare/flowrunner-vs-make FlowRunner vs Make: FlowRunner does everything Make does and adds native AI-agent orchestration with human oversight. Where FlowRunner wins, where Make still helps, how to choose. 2026-07-18 Updated September 7, 2026 12 min ![Shared foundation, different ceilings: Make's linear scenario beside FlowRunner, which runs that same workflow and adds the agentic layer above it, agent-invoked human review, long waits, and runtime tool choice.](https://flowrunner.ai/images/compare/make-shared-foundation.webp) **TL;DR** - Start with what these two share: the cloud, connector-based scenarios Make is known for, connecting apps, chaining modules from a trigger through a set of actions. FlowRunner builds all of that. FlowRunner is [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service), so it also runs the layer above: coordinating AI agents, governing multi-agent work, and keeping a human in control of the decisions that need judgment. Read them side by side and the shape is plain: Make does the linear scenario; FlowRunner does that and the agentic orchestration on top. - This page is about AI-agent automation with human oversight. For that specific work the fit is not 50/50: FlowRunner is built for it, and Make is being stretched to reach it. We will still show you exactly where Make is the better call, because sometimes it is. - FlowRunner is stronger where this reader lives: human-in-the-loop as a native agent tool on every tier, run-based billing that does not multiply with steps, waits up to a year, HIPAA with a BAA at mid-tier prices, and agent-tool metadata on the whole catalog. - Make is stronger on raw connector count, per-module error handling, and community depth. ### Who this comparison is for You are building AI-agent automation where the agent has to do more than fire a linear scenario and finish. It needs to stop and pull in a person when the rules do not cleanly apply. It may run for a while. It may touch data an auditor will ask about later. You might be coming from Make or starting fresh; either way you are deciding which platform is built for that work. The overlap is large: the linear, cloud scenario building Make is known for, FlowRunner does too. What FlowRunner adds on top is the agentic layer, agents that pause for a human, long-running orchestration, and built-in compliance. Where that added layer matters, FlowRunner is not merely also an option; it is the platform built for it. This page is about exactly where Make’s ceiling is and where FlowRunner keeps going: the agent that hits an [automation exception](https://flowrunner.ai/concepts/automation-exceptions) it should not decide alone and has to hold the run open until a human answers. ### What FlowRunner is good at FlowRunner is built around one commitment: agents that know when to stop and ask a human for help. Everything else in the platform exists to make that commitment reliable at production scale. The mechanism is the part that linear tools do not have. In FlowRunner, an AI agent invokes one of your own Flows as a callable tool. When the agent calls that flow-tool, it suspends while the flow runs, and the flow can hold on an external callback until a human replies through email, Slack, WhatsApp, phone, or a Form, then resume mid-run. This is [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) as an execution pattern: the agent pauses on its own judgment, assembles the context and the choices available, routes to a human on their preferred channel, and resumes the moment they answer. Some teams call it a [digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord), after the Toyota factory pull cord any worker can pull to stop the line when something needs a human eye. ![Human review built into the run versus a workaround assembled around it. Make's Enterprise-only closed-beta approach stores state externally and reprocesses through another scenario, while FlowRunner's agent calls a human-review flow, suspends, and the human answers in-channel (email, Slack, WhatsApp, phone, or Form) before the same run resumes.](https://flowrunner.ai/images/compare/make-human-in-the-loop-architecture.webp) FlowRunner is strong on: - **Human-in-the-loop as an agent tool, native on every tier.** The agent decides at runtime to pull in a person, calls a human-review flow as a tool, and suspends until they reply in their own channel: email, Slack, WhatsApp, phone, or a Form. It can loop over a batch, ask a follow-up question, and escalate if no one answers. This is native on every plan, including Free, not gated behind an Enterprise plan. - **Run-based billing.** One complete workflow run, start to finish, is one execution. A 10-step run is one execution, not ten, and there is no per-step or per-operation multiplier. That removes the step-multiplication math that per-operation billing forces on builders. - **Long waits and long runtimes.** A run can suspend for up to 30 days on Growth and up to a year on Professional, Business, and Enterprise. Cumulative runtime runs to 1 hour on Growth, 4 hours on Professional, 12 hours on Business, and unlimited on Enterprise. - **Compliance at mid-tier prices.** Audit trails and RBAC start at the Professional tier ($299/month). HIPAA support with a BAA is available. SSO/SAML and 90-day audit retention arrive at Business ($999/month). - **A verified, agent-ready catalog with depth.** 1,800+ verified integrations and 56,000+ callable actions, each carrying the structured metadata an LLM needs to call it as a tool, all built and verified against each vendor’s official API. The catalog includes full-CRUD databases, vector stores, and deep ERPs, plus 75 integrations with no first-party Make equivalent, including X, Redis, Oracle, DynamoDB, and the major vector stores. - **[BYOK](https://flowrunner.ai/concepts/byok) with no markup.** BYOK is a model where the customer supplies their own AI provider API credentials, pays providers directly for usage, and FlowRunner orchestrates calls across providers without taking a markup on inference. For a buyer who already has an Anthropic, Google, or Azure agreement, that means your own rates, your own data-processing terms, and a provider your security team already approved, rather than a new bundled inference vendor to review. That pause-for-a-human moment is the load-bearing part of FlowRunner’s architecture. It is why agentic work that needs human judgment is native here rather than assembled from workarounds. ### What Make is good at Make (formerly Integromat) is a cloud scenario builder with a large connector directory. Make advertises 3,000+ integration apps, with roughly 3,543 listings live today. Make is strong on: - **Connector directory size.** The 3,000+ figure is real as a directory count. If you need to touch a long tail of services, Make probably lists something for it. - **AI features on the low paid tiers.** Make’s Core plan starts at $12/month (as of July 2026), and AI features reach down to the low paid tiers, with an effective AI entry point around $9 to $12/month on annual billing. Make does not hold its AI back for higher plans. FlowRunner includes AI agents on its Free plan and $5 Starter, so the entry price itself is not a Make advantage, but Make’s low tiers are a real place to build. - **Mature error directives.** Make gives each module explicit error handling: Ignore, Resume, Commit, Rollback, and Break, plus an Incomplete Executions retry queue that parks a failed run for later reprocessing. - **Community and templates.** A large community, an active forum, and a broad template gallery mean most problems have a documented answer. - **Track record.** Make has been in market for years under two names and has handled a lot of production traffic. It is a known quantity. Make is built for linear, cloud-based scenario work. ### Where FlowRunner is stronger For the agentic, human-oversight work this page is about, these are the differences that decide the platform: - **Wait and runtime ceilings.** This is the single largest architectural gap. A Make scenario run is capped at 40 minutes on paid plans and 10 minutes on the free plan, and the Sleep module tops out at 300 seconds, which is 5 minutes. If a human does not answer within 5 minutes, a native Make wait cannot hold the run, and you fall back to webhooks and reprocessing scenarios. FlowRunner suspends a run for up to a year and runs for up to 12 hours or unlimited. For any workflow that waits on a person, this difference decides the platform. - **Human-in-the-loop native on every tier, as an agent tool.** Make’s Human in the Loop app is available on the Enterprise plan only and is currently in closed beta, available only to invited customers. FlowRunner’s human-in-the-loop is native starting at Growth ($45/month), and it is architecturally different: the agent invokes a Flow as a callable tool and suspends until a human replies, rather than routing through a linear approval step. The pause condition is decided at runtime based on what the agent sees, and the context delivered to the human is assembled, not pasted. - **Run-based billing.** Make bills per credit (formerly operations), where one module run over a batch of items burns many credits. A 3-module scenario over 10 items is roughly 31 operations. FlowRunner bills per execution, where the whole run is one unit. At volume, the difference between per-step and per-run billing is large and, more importantly, predictable. You are not re-budgeting every time you add a step to an agent. - **Compliance at mid-tier prices.** As of July 2026, Make does not sign a BAA on any plan, and its SSO, audit, and RBAC controls live on the Enterprise plan. FlowRunner ships HIPAA support with a BAA, plus audit trails and RBAC, starting at Professional ($299/month). For a regulated builder, this moves the compliant configuration down from Enterprise pricing to mid-tier. - **Agent-tool metadata on the whole catalog.** Every one of FlowRunner’s 56,000+ actions carries typed parameters, descriptions, and sample output, so an agent can call it as a tool directly. Make and other workflow-first tools treat apps as steps first; FlowRunner treats every action as a tool an agent can reach for. FlowRunner also has 75 integrations with no first-party Make equivalent, including X, Redis, Oracle, DynamoDB, and the major vector stores. - **BYOK with no markup on inference.** You supply your own AI provider keys and pay providers directly. FlowRunner does not resell inference. Your orchestration fee stays separate from your model spend. ![Wait and runtime ceilings from five minutes to one year. Make's native wait tops out at the 5-minute Sleep module and a 40-minute paid run cap; FlowRunner suspends up to a year and runs up to 12 hours or unlimited by tier.](https://flowrunner.ai/images/compare/make-wait-runtime-ceilings.webp) ### Where Make is stronger This is the section most comparison pages skip. Make wins on several dimensions that matter, and if these are your priorities, Make is the right call. - **Raw connector directory.** Make advertises 3,000+ apps against FlowRunner’s 1,800+ verified integrations, and even after discounting the community-built, internal, discontinued, and long-tail regional connectors, Make’s verified first-party count (roughly 2,534 by an independent audit) is far larger. But the few hundred mainstream apps that most teams actually build on are the ones FlowRunner already covers, in depth (roughly 18 callable actions each). Where Make still wins outright is the long tail: if a specific regional or niche service is your single deciding factor, Make probably lists it and FlowRunner may not. - **Per-module error directives.** Make’s Ignore, Resume, Commit, Rollback, and Break directives, plus the Incomplete Executions retry queue, are more granular out of the box than FlowRunner’s Handle Error block and Repeat plus Wait pattern, which cover the common retry and rollback cases with less per-module wiring. - **Community size and templates.** Make’s community, forum, and template gallery are larger than FlowRunner’s today, a real head start for self-service problem-solving. FlowRunner’s answer is a growing catalog of install-and-configure packages, but Make’s ecosystem is larger right now. - **Longer track record.** Make has been in production for years under two names, and for a buyer who weights operational maturity heavily, that road matters. The tradeoff is that Make’s design center was set before agentic, human-in-the-loop work existed, which is the shape FlowRunner was built for. If any of those outweigh the differences above, Make is the right call, and choosing it is defensible. ![One scenario, two meters: Make's credits (formerly operations) multiply per module over each item, roughly 31 credits for a 3-module run over 10 items, while FlowRunner counts one execution per whole run. Your data branches; your bill should not.](https://flowrunner.ai/images/compare/make-billing-credits-vs-executions.webp) ### Feature comparison | Dimension | FlowRunner | Make | | --- | --- | --- | | Primary buyer | Builder plus operations team | Builder, cloud scenario author | | Billing unit | Execution (one whole run) | Credit / operation (per module run) | | Cost of a 10-step run x 1,000/mo | 1,000 executions | Many thousands of credits | | Entry price | Free $0 (100 executions); Starter $5/mo (300) or $15 (3,000); Growth $45/mo (12,000) | Free $0 (1,000 credits); $12/mo Core (as of Jul 2026) | | AI entry point | Free plan (BYOK, no markup) | ~$9 to $12/mo (annual) | | Unlimited users | Yes | Yes | | Max single-run time | 1h Growth, 4h Pro, 12h Business, unlimited Enterprise | 40 min paid / 10 min free | | Max native wait | 30 days Growth, up to 1 year other tiers | Sleep module 300s (5 min) | | Human-in-the-loop | Native on every tier, agent-invoked callable tool | Enterprise-only, closed beta | | HITL channels | Email, Slack, WhatsApp, phone, Form | Webhook / Slack workarounds; beta app | | AI agents | Native ([Agent Factory](https://flowrunner.ai/concepts/agent-factory)) | Yes (paid plans) | | MCP | Consumes external MCP servers; exposes catalog as MCP server (shipping ~Aug 2026) | Native server and client | | Integration directory | 1,800+ verified integrations, 56,000+ actions | 3,000+ listed apps (~3,543 live) | | Agent-tool metadata on actions | On every action | Not the model | | Error handling | Handle Error block + Repeat/Wait retry | Per-module directives + retry queue | | HIPAA / BAA | HIPAA with BAA | No BAA on any plan (as of Jul 2026) | | Audit trails / RBAC | Professional ($299) | Enterprise | | SSO/SAML | Business ($999) | Enterprise | | SOC 2 | Security program designed to meet common audit requirements | Type II | | Self-hosting | Enterprise on-prem; Community Edition coming | No (cloud-only) | | BYOK for AI | Yes, no markup on inference | Model-provider connectors | ### Decision framework ![A compliance decision gate for PHI workflows: if a signed BAA is required, Make is unavailable and FlowRunner is supported. HIPAA and BAA, audit trails, RBAC, and SSO compared, with FlowRunner reaching audit and RBAC at Professional versus Make's Enterprise tier.](https://flowrunner.ai/images/compare/make-compliance-gate.webp) **Pick FlowRunner if:** - Your agents need to stop and pull in a human mid-run, and the human might take hours or days to answer. - You are building AI agents where human-in-the-loop is the point, and you do not want it gated to an Enterprise closed beta. - Predictable run-based billing matters more than a low sticker price, because your workflows run many times a month at many steps each. - You handle regulated data and need HIPAA with a BAA, audit trails, and RBAC without paying Enterprise prices for them. - You want every action in the catalog to be callable by an agent as a tool, with the metadata to make that reliable. **Pick Make if:** - Your scenarios are short, run in the cloud, and finish in well under 40 minutes with no mid-run human wait longer than 5 minutes. - Raw connector directory breadth is your single most important criterion and you need a listing for a long tail of services today. - You want the granular per-module error directives (Ignore, Resume, Commit, Rollback, Break) and the Incomplete Executions retry queue as your reliability model. - You do not handle protected health information and do not need a signed BAA. If the answer is honestly mixed, run a real workflow on both before committing. The right platform is the one that fits the shape of your work, not the one with the larger directory. ### Migration considerations ![Migration map from a Make scenario to a FlowRunner flow: triggers, actions, filters, and routers carry over; Make's error directives and Incomplete Executions map to Handle Error and Repeat plus Wait, and a native Human Review Flow is added. The logic carries over; the execution model changes.](https://flowrunner.ai/images/compare/make-migration-map.webp) For builders already on Make and weighing a switch, the practical questions: - **What maps mechanically?** Triggers, actions, and basic scenario structure map directly. A Make module usually has a FlowRunner action equivalent, and Make’s router and filter logic maps to FlowRunner’s Condition and Value Router blocks. - **What gets re-expressed?** Make’s per-module error directives (Ignore, Resume, Commit, Rollback, Break) become FlowRunner’s Handle Error block and the Repeat plus Wait retry pattern. The Incomplete Executions retry queue maps to that same retry pattern. The logic carries over; the shape changes. - **What gets better in the process?** Any scenario that fought the 40-minute run cap or the 5-minute Sleep ceiling stops being a workaround. Long waits and long runtimes become native, so the “split it into two scenarios that hand off” pattern disappears. - **Who runs the migration?** The Midnight Flow consulting team runs migrations as part of FlowRunner Enterprise onboarding. Other tiers self-migrate with the platform’s import tools. - **How long does it take?** A typical mid-market deployment with 20 to 50 scenarios migrates over 4 to 8 weeks with consulting support, or longer self-served, depending on how many scenarios relied on Make-specific error directives. If you would like a free migration consultation, reach out via the [Contact page](https://flowrunner.ai/contact) and we will scope the engagement before any commitment. ### Frequently asked questions #### Is FlowRunner cheaper than Make? Yes on the sticker, and usually per run as well. Make's Core plan starts at $12/month (as of July 2026) and its AI entry point sits around $9 to $12/month on annual billing. FlowRunner's Free plan is $0 for 100 executions a month with AI agents, every integration, and human-in-the-loop included, and Starter is $5/month for 300 executions or $15 for 3,000. The real comparison is the billing unit. Make bills per credit (formerly operations), where every module run burns a credit and a single module run over a batch of items burns many, so a 10-module scenario costs about ten credits per run and Core's 10,000 credits are roughly 1,000 runs, around $0.012 a run. FlowRunner bills per execution, where one complete run from start to finish is one execution regardless of how many steps it has: Starter at $15 is $0.005 a run, and Growth at $45 for 12,000 executions is under $0.004. Make is cheap to start; FlowRunner is cheaper to start and stays flat as scenarios grow. #### What is Make's execution time limit? A single Make scenario run is capped at 40 minutes on paid plans and 10 minutes on the free plan, and each individual module is capped at roughly 40 seconds (community-documented via Make's own timeout error). Make's Sleep module allows a maximum delay of 300 seconds, which is 5 minutes. That ceiling is the practical reason Make is a poor fit for long-running or human-in-the-loop work: if a person does not respond within 5 minutes, a native Sleep-based wait cannot hold the run. FlowRunner suspends a run for up to a year and runs for up to 12 hours or unlimited depending on tier. #### Does Make have human in the loop? Yes, but with two constraints. Make's Human in the Loop app is available on the Enterprise plan only and is currently in closed beta, available only to invited customers (per Make's own app listing). Outside that, teams build approvals with webhooks, Slack messages, and the Incomplete Executions retry queue. FlowRunner's human-in-the-loop is native on every plan, including Free, the AI agent invokes a Flow as a callable tool and suspends until a human replies through email, Slack, WhatsApp, phone, or a Form, and the wait can last up to a year. #### Is Make HIPAA compliant? As of July 2026, Make does not sign a Business Associate Agreement on any plan, and healthcare teams should treat it as not HIPAA compliant for workflows that touch protected health information (per Make's community support responses and third-party compliance reviews). Make is SOC 2 Type II certified, which covers general security posture but is not the same as a signed BAA. FlowRunner offers HIPAA support with a BAA and ships audit trails and RBAC starting at the Professional tier ($299/month). #### What is a good cheaper alternative to Make for AI agents with human oversight? If your work is short, cloud-only, and per-operation, Make fits that shape well. If your work is long-running, needs a human to weigh in mid-run, and touches regulated data, Make's ceilings (40-minute run, 5-minute native wait, Enterprise-gated human-in-the-loop and compliance) push the total cost up through the Enterprise tier. FlowRunner is built agent-first for exactly that case: per-execution billing, waits up to a year, human-in-the-loop native on every tier, and HIPAA plus audit plus RBAC at mid-tier prices. #### Does FlowRunner have as many integrations as Make? No, not on raw directory count. Make advertises 3,000+ apps (roughly 3,543 live listings). A verified app-by-app audit shows a large share of that directory is community-built, internal utilities, discontinued apps, or long-tail regional and vertical connectors. FlowRunner covers 1,800+ verified integrations and 56,000+ callable actions, built and verified against each vendor's official API, with agent-tool metadata on every action so an LLM can call it directly. FlowRunner also has 75 integrations with no first-party Make equivalent, including X, Redis, Oracle, DynamoDB, and the major vector stores. The honest summary: Make has more listings; FlowRunner has verified, mainstream, agent-ready coverage plus depth Make's directory does not. #### Can I migrate from Make to FlowRunner? Yes. Triggers, actions, and basic scenario structure map over. Make's per-module error directives (Ignore, Resume, Commit, Rollback, Break) are re-expressed with FlowRunner's Handle Error block and Repeat plus Wait retry pattern. The Midnight Flow consulting team runs the migration as part of Enterprise onboarding, and other tiers can self-migrate. A typical mid-market deployment with 20 to 50 scenarios migrates over 4 to 8 weeks with consulting support. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) > _I like that aspect of the human in the loop when it's needed. That's ultimately what we need to get to._ VP Operations, CPG brand --- ## FlowRunner vs n8n Source: https://flowrunner.ai/compare/flowrunner-vs-n8n FlowRunner vs n8n: FlowRunner covers the same automation no-code, with human oversight and built-in compliance; n8n adds developer code and free self-host. How to choose. 2026-06-07 Updated September 7, 2026 10 min ![Who can ship the automation? n8n's process-owner idea goes through an engineering queue, code, and self-hosted infrastructure to reach production, while FlowRunner's goes through a visual or conversational builder with governance so the operations team ships it directly.](https://flowrunner.ai/images/compare/n8n-who-can-ship.webp) **TL;DR** - Start with the automation both run: connector-based, trigger-to-action flows, now with AI agents and human-in-the-loop on top. FlowRunner runs all of it, and the operations team that owns the process ships it without an engineering bottleneck or compliance to bolt on. FlowRunner is [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service), built so non-developers ship production agents. What n8n adds is a more permissive code surface (a full JavaScript or Python runtime with npm and network access, beyond FlowRunner’s sandboxed Custom Cloud Code) and free unlimited self-hosting today. - This page is about AI-agent automation with human oversight. For that specific work the fit is not 50/50: FlowRunner is built for it, and n8n is reached through developer ownership of code and infrastructure. We will still show you exactly where n8n is the better call, because sometimes it is. - FlowRunner is stronger where this reader lives: operations-team accessibility for non-developers, human-in-the-loop as a native agent tool on every tier, compliance with a BAA at mid-tier prices, and agent-tool metadata on every action across a verified catalog. - n8n is stronger on free unlimited self-hosting, a more permissive code runtime (JavaScript or Python with npm and network access, where FlowRunner runs sandboxed Custom Cloud Code), its GitHub community and template library, fair-code control, and a handful of protocol-level nodes (message brokers, SSH, FTP, Git) FlowRunner does not yet cover. ### Who this comparison is for You are building AI-agent automation where the agent has to do more than fire a linear sequence and finish. It needs to stop and pull in a person when the rules do not cleanly apply. It may touch data an auditor will ask about later. And it needs to get built and maintained by the team that owns the process, whether or not that team writes code. You might be a COO, VP of operations, or operations director at a 50-to-500-person company, coming from n8n or starting fresh, deciding which platform is built for that work. The overlap is large: the connector-based, trigger-to-action automation n8n runs, FlowRunner runs too, with AI agents and human-in-the-loop on top, and the operations team ships it without an engineering bottleneck or compliance to bolt on. What n8n adds is a more permissive code surface (a full JavaScript or Python runtime with npm and network access, beyond FlowRunner’s sandboxed Custom Cloud Code) and free unlimited self-hosting today. This page is about where those facts leave you: the agent that hits an [automation exception](https://flowrunner.ai/concepts/automation-exceptions) it should not decide alone and has to hold the run open until a human answers. ![Developer control or operations accessibility: two different design centers. n8n is strongest when a developer team owns the platform, with free self-hosting, a full JavaScript and Python runtime with npm packages and network access, protocol nodes, and a large community. FlowRunner is strongest when the process owner needs to ship production automation, with no-code accessibility, a visual and conversational builder, managed-cloud governance, an agent-ready catalog under human oversight, and sandboxed Custom Cloud Code usable as a step or an agent tool. The distinction is not code versus no-code, it is a more permissive code runtime on n8n against operations accessibility plus sandboxed code on FlowRunner.](https://flowrunner.ai/images/compare/n8n-developer-vs-operations.webp) ### What FlowRunner is good at FlowRunner is built around one commitment: agents that know when to stop and ask a human for help, built by the team that owns the process rather than by an engineering group they have to queue behind. Everything else in the platform exists to make that commitment reliable at production scale. The mechanism is the part that a wire-in approval node does not have. In FlowRunner, an AI agent invokes one of your own Flows as a callable tool. When the agent calls that flow-tool, it suspends while the flow runs, and the flow can hold on an external callback until a human replies through email, Slack, WhatsApp, phone, or a Form, then resume mid-run. This is [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) as an execution pattern: the agent pauses on its own judgment, assembles the context and the choices available, routes to a human on their preferred channel, and resumes the moment they answer. Some teams call it a [digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord), after the Toyota factory pull cord any worker can pull to stop the line when something needs a human eye. ![Human-in-the-loop, step versus tool. n8n's Send and Wait for Response is a capable node placed at design time; FlowRunner's agent detects uncertainty, calls a human-review flow at runtime, suspends, and the human responds in-channel before the same agent resumes.](https://flowrunner.ai/images/compare/n8n-human-in-the-loop-architecture.webp) FlowRunner is strong on: - **Operations-team accessibility.** The platform is designed so the people who own the process build the automation. The visual editor and the Conversational AI mode let an operations director ship a working, production agent without a JavaScript primer and without owning the infrastructure it runs on. Describe the agent you want in plain language and the Conversational AI assembles it, then hands off to manual fine-tuning at any point. - **Human-in-the-loop as an agent tool.** The agent decides at runtime to pull in a person, calls a human-review flow as a tool, and suspends until they reply in their own channel: email, Slack, WhatsApp, phone, or a Form. It can loop over a batch, ask a follow-up question, and escalate if no one answers. The pause is decided by what the agent sees, not fixed at design time. It is native on every plan, including Free. - **Compliance with a BAA at mid-tier prices.** Audit trails and RBAC start at the Professional tier ($299/month), and HIPAA support with a BAA is available. SSO/SAML and 90-day audit retention arrive at Business ($999/month). For a regulated mid-market buyer, the compliant configuration lives at mid-tier rather than behind an Enterprise negotiation. - **A verified, agent-ready catalog with depth.** 1,800+ verified integrations and 56,000+ callable actions (as of September 2026, and growing), roughly 26 actions per integration, each carrying the structured metadata an LLM needs to call it as a tool, all built and verified against each vendor’s official API. - **Long waits and long runtimes.** A run can suspend for up to 30 days on Growth and up to a year on Professional, Business, and Enterprise. Cumulative runtime runs to 1 hour on Growth, 4 hours on Professional, 12 hours on Business, and unlimited on Enterprise. - **[BYOK](https://flowrunner.ai/concepts/byok) with no markup.** BYOK is a model where the customer supplies their own AI provider API credentials, pays providers directly for usage, and FlowRunner orchestrates calls across providers without taking a markup on inference. For a buyer who already has an Anthropic, Google, or Azure agreement, that means your own rates, your own data-processing terms, and a provider your security team already approved. That pause-for-a-human moment, invoked by an agent an operations team can build, is the load-bearing part of FlowRunner’s architecture. It is why agentic work that needs human judgment is native here rather than assembled from workarounds or handed to an engineering team. ### What n8n is good at n8n has built one of the widely used workflow automation platforms in the market, with a developer-first design and a mature self-hosted deployment model. The credentials it cites are large: approximately 192,000 GitHub stars, 200,000+ community members, and a funding history that runs from a $180 million Series C in October 2025 (Accel-led, $2.5 billion valuation) through a strategic investment from SAP in May 2026 at a $5.2 billion valuation, for $254 million in total funding. The community is large and the documentation is thorough. n8n is strong on: - **Free unlimited self-hosting.** n8n’s self-hosted Community edition runs unlimited executions for free under its fair-code license, and self-hosting is the flagship deployment model rather than an afterthought. FlowRunner Cloud does not match a $0 unlimited self-host on sticker. Two things sit alongside that free number: self-hosting carries the operating cost of running the infrastructure yourself, which industry estimates put into six figures a year for a maintained production deployment once you count servers, upgrades, and DevOps time, and n8n Cloud signs no BAA, so a regulated team on the free self-host path owns all of its own compliance. - **A large, code-extensible node ecosystem.** n8n’s directory advertises ~1,953 listings, though that number counts triggers and actions separately and is dominated by community-maintained npm nodes, and n8n Cloud exposes a curated subset; for anything unlisted, its generic HTTP node lets a developer wire up any REST API by hand. That breadth-via-community-and-code model is a genuine fit for a developer team that will extend the platform itself. On a like-for-like, first-party basis FlowRunner is now clearly ahead: n8n’s roughly 330 built-in core app nodes against FlowRunner’s 1,800+ verified integrations, covered in the sections below. - **A more permissive code surface.** n8n’s Code node runs JavaScript or Python in a full runtime, with npm packages and network access, at any point in a workflow. FlowRunner also runs code: its Custom Cloud Code block executes JavaScript as a step, and can be handed to an AI agent as a tool it calls on its own. The difference is the sandbox: FlowRunner’s runs JavaScript only, with no npm or direct network (I/O goes through the HTTP Request and database blocks), and caps execution at ten seconds. For a developer team that wants an unrestricted runtime with any package, n8n’s node is more permissive today, though FlowRunner is adding custom actions with arbitrary JavaScript, shipping in August 2026. - **A large developer community and template library.** With approximately 192,000 GitHub stars, 200,000+ community members, a several-year head start, and a homepage that lists 9,500+ community templates, n8n has a bigger community and a richer template ecosystem than FlowRunner does today. For a developer starting cold, that head start on ready-made examples is real. FlowRunner’s answer is a growing catalog of install-and-configure packages, but n8n’s community library is larger right now. If your team is developer-owned and wants to host and extend the platform with code, those strengths line up with how you work. ![Same billing unit, different operating model. Both platforms bill one whole workflow run as one execution; the difference is who runs the stack, n8n's free self-host means you own the infrastructure, while FlowRunner's managed cloud builds in audit trails, RBAC, and a BAA at mid-tier.](https://flowrunner.ai/images/compare/n8n-billing-execution-model.webp) ### Where FlowRunner is stronger For the agentic, human-oversight work this page is about, built and maintained by an operations team, these are the differences that decide the platform: - **Operations team can build and maintain.** This is the largest practical difference for the buyer this page is written for. n8n’s design assumes developers will own the platform: the visual editor is forgiving for simple flows, but the platform expects code knowledge for anything substantial, and self-hosting expects infrastructure ownership. FlowRunner is built so the operations team that owns the process owns the automation, no engineering bottleneck and no translation between business logic and JavaScript. The Conversational AI builder, which assembles an agent from a plain-language description, has no n8n equivalent today. - **Compliance with a BAA at mid-tier prices.** FlowRunner ships audit trails and RBAC, with a BAA available, starting at Professional ($299/month). n8n has no Cloud BAA at any price, and gates SSO/SAML, RBAC, and audit streaming to its custom-priced Enterprise tier, with the Community edition offering essentially no multi-user governance. A regulated team on n8n ends up either negotiating Enterprise or self-hosting and owning all of its own compliance. For a 100-person healthcare or financial services company, that difference can drive the decision on its own. - **Human-in-the-loop as an agent tool.** n8n’s human-in-the-loop is its Send and Wait for Response node: the workflow reaches the node, sends a message across a channel, and waits for an approval, free-text, or form reply. It is a capable step, and it is placed in the workflow at design time. FlowRunner works differently. The agent itself decides at runtime to pull in a person, calls a human-review flow as a callable tool, and suspends until they reply. The pause condition is decided by what the agent sees rather than fixed in the wiring, the context delivered to the human is assembled rather than pasted, and the flow can loop, ask a follow-up, and escalate. That is the difference between a wait step you place in advance and a human the agent decides to call at runtime. - **Wait ceilings without silent eviction.** Both platforms can suspend a run for a person who takes days to answer. n8n’s Wait node carries a data-pruning caveat: the default 14-day execution-data max age can evict a longer-waiting run, a failure mode that is easy to miss. FlowRunner suspends a run for up to a year on Professional and above as a documented ceiling, with no equivalent pruning risk. For work that waits on a person who might take a week, that difference in reliability matters. - **Depth per integration, with agent-tool metadata.** FlowRunner’s 2,100+ integrations span 56,000+ actions, roughly 26 per integration, and every action carries typed parameters, descriptions, and sample output so an agent can call it as a tool directly. Full-CRUD databases and deep ERPs are in the catalog, and every action is agent-ready by default rather than being a workflow node an agent has to be wrapped around. n8n’s directory lists more entries overall, but they are workflow nodes without that metadata; on verified first-party integrations FlowRunner is ahead (1,800+ against n8n’s ~330 built-in core app nodes), and every FlowRunner action ships the structured metadata an agent needs to invoke it directly. - **Composability across agents.** An agent assembled in the [Agent Factory](https://flowrunner.ai/concepts/agent-factory) can call other agents and existing flows as tools. The unit of composition is an agent, not a workflow node, and the [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) provides platform-enforced governance for how agents communicate, escalate to humans, and produce audit trails across many agents from many authors. n8n’s composability stays at the workflow level, with no equivalent inter-agent governance layer today. ![Wait reliability without silent eviction. n8n's default 14-day execution-data pruning can evict a longer-waiting run; FlowRunner suspends up to a year on Professional and above as a documented ceiling, with runtime from 1 hour on Growth to unlimited on Enterprise.](https://flowrunner.ai/images/compare/n8n-wait-reliability.webp) ### Where n8n is stronger This is the section most comparison pages skip. n8n wins on real dimensions, and if these are your priorities, n8n is the right call. - **Free unlimited self-hosting, available today.** n8n’s self-hosted Community edition runs unlimited executions for free, and its self-hosting is mature and flagship, which FlowRunner does not match today (its Enterprise on-prem option exists, but the free Community Edition is still coming). The free number comes with the operating cost of running it yourself, into six figures a year by industry estimates for a maintained production deployment, and no Cloud BAA, so it does not clear a regulated use case on its own. But for a team with the DevOps capacity and no compliance requirement, free unlimited self-hosting is a genuine advantage n8n has and FlowRunner does not. - **Message brokers, protocol nodes, and the community long tail.** This is where n8n’s node set genuinely reaches further. n8n has built-in nodes for the Kafka, RabbitMQ, and MQTT message brokers and for core SSH, FTP, and Git, which FlowRunner does not yet cover (SFTP is on the roadmap). For a genuinely obscure service, n8n’s community-node long tail or its generic HTTP node may reach it when FlowRunner’s verified catalog does not. This is not a raw-count story: on verified first-party app integrations FlowRunner is ahead (1,800+ against n8n’s ~330 built-in core app nodes), and n8n’s advertised ~1,953 counts triggers and actions separately and is community-dominated. It is about those specific protocol nodes and the community tail. If one of them is your single deciding factor, n8n probably has it and FlowRunner may not. - **A more permissive code runtime.** n8n’s Code node runs JavaScript or Python in a full runtime with npm packages and network access. FlowRunner runs code too: its Custom Cloud Code block executes sandboxed JavaScript and can be handed to an agent as a tool, with custom actions arriving in August 2026. But n8n’s runtime is more permissive today: Python, external packages, and network calls from the code itself. For a team that wants unbounded code with any library and owns the engineers to use it, n8n is the more permissive platform. - **A large developer community and template ecosystem.** n8n’s community, its 9,500+ community templates, and its several-year head start give a developer starting cold a large library of examples and answers. FlowRunner’s community is much smaller today. For a team that values that established ecosystem, n8n’s is larger right now. - **Fair-code licensing and developer control.** Some buyers will not deploy proprietary platforms, or want to fork and extend the code, or need to self-host without a commercial contract. n8n’s fair-code model (its Sustainable Use License restricts reselling n8n commercially, but permits self-hosting freely) is the cleaner answer for those teams. FlowRunner offers a free Community Edition for self-hosting, but it is proprietary and still coming. If any of those outweigh the differences above, n8n is the right call, and choosing it is defensible. ![Directory size versus agent-ready depth. n8n's ~1,953 site listings are not like-for-like: they count triggers and actions separately, lean on community npm nodes, and n8n Cloud exposes a subset. FlowRunner's 400+ verified first-party integrations expose 8,000+ callable actions at roughly 23 each, with agent-tool metadata on every action, and 228 of 246 tracked n8n app-nodes are already built.](https://flowrunner.ai/images/compare/n8n-directory-size-vs-depth.webp) ### Feature comparison | Dimension | FlowRunner | n8n | | --- | --- | --- | | Primary buyer | Operations team plus builder | Developer / platform team | | Who builds and maintains | Operations team, no code required | Developer team, code and infrastructure ownership | | Billing unit | Execution (one whole run) | Execution (one whole run); Community self-host unlimited free | | Visual builder | Yes (operations-oriented) | Yes (developer-oriented) | | Conversational AI builder | Yes | No | | Custom code | Custom Cloud Code (sandboxed JS, agent-tool); custom actions ~Aug 2026 | Code node: JavaScript/Python, npm, network | | AI agents | Native ([Agent Factory](https://flowrunner.ai/concepts/agent-factory)) | AI Agent node (LangChain) | | Human-in-the-loop | Native every tier; agent-invoked callable tool; can loop, branch, escalate | Send and Wait for Response node (design-time step) | | HITL reviewer experience | Reply in-channel: email, Slack, WhatsApp, phone, Form | Slack, Gmail, Email, Teams, Telegram, Discord, WhatsApp, Chat | | Max native wait | 30 days Growth, up to 1 year other tiers | Wait node; 14-day execution-data pruning can evict longer waits | | MCP | Consumes external MCP servers; exposes catalog as MCP server (shipping ~Aug 2026) | Native (server trigger + client) | | Integration directory | 1,800+ verified first-party, 56,000+ actions (~26 each) | ~330 built-in core app nodes (site lists ~1,953, community-heavy) | | Mainstream app coverage | Parity: 228 of 246 tracked n8n app-nodes built, plus 40+ n8n lacks | Broadest node list overall | | Agent-tool metadata on actions | On every action | Not the model | | Cloud entry price | Free $0 (100 executions); Starter $5/mo (300) or $15 (3,000); Growth $45/mo (12,000) | €20/mo Starter | | Free evaluation | Permanent Free plan (100 executions/mo, every integration and human-in-the-loop) plus 14 days of Professional on signup, no card | Free Starter and Pro; unlimited free self-host | | HIPAA / BAA | HIPAA with BAA available | No Cloud BAA (self-host + own BAAs only) | | Audit trails / RBAC | Professional ($299) | Enterprise (custom) | | SSO/SAML | Business ($999) | Enterprise (custom); Business €667 self-host has SSO | | Self-hosting | Enterprise on-prem; Community Edition coming | Fair-code, mature, free unlimited executions | | BYOK for AI | Yes, no markup on inference | No native BYOK model | | Inter-agent governance | [Code of Conduct](https://flowrunner.ai/concepts/code-of-conduct) | No equivalent | | SOC 2 | Security program designed to meet common audit requirements | Yes | ### Decision framework ![A compliance decision gate for PHI workflows: if a signed BAA is required, n8n Cloud is unavailable and FlowRunner is supported. HIPAA and BAA, audit trails, RBAC, SSO, and self-hosting compared, with FlowRunner reaching audit and RBAC at Professional while n8n gates them to Enterprise, and n8n offering free unlimited self-hosting.](https://flowrunner.ai/images/compare/n8n-compliance-gate.webp) **Pick FlowRunner if:** - Your agents need to stop and pull in a human mid-run, and the human might take hours or days to answer, and you want that pause decided by the agent at runtime rather than wired in as a fixed step. - The operations team that owns the process needs to build and maintain the automation without an engineering bottleneck or a JavaScript primer. - You handle regulated data and need HIPAA with a BAA, audit trails, and RBAC at mid-tier prices, which n8n Cloud cannot provide at any price. - You want every action in the catalog callable by an agent as a tool, with the depth and the metadata to make that reliable. - The buyer in your organization is a COO, VP of operations, or operations director, not a CTO. **Pick n8n if:** - You have a developer team that owns automation as part of its job and wants to write and control the code. - Free unlimited self-hosting is your priority, you have the DevOps capacity to run it, and you do not need a Cloud BAA. - You need the broadest node list, including message brokers and core SSH, FTP, or Git, or a specific long-tail or community-contributed node today. - An unrestricted code runtime is non-negotiable: Python, npm packages, or network calls from the code itself, beyond FlowRunner’s sandboxed Custom Cloud Code. - Fair-code licensing matters to your procurement or technical leadership, or you plan to fork and extend the platform. If the answer is honestly mixed, run a real workflow on both before committing. The right platform is the one that fits how your team actually works, not the one with the larger node directory. ### Migration considerations ![Migration map from an n8n workflow to a FlowRunner flow: triggers, actions, IF/filter, and switch nodes carry over; the Send and Wait node becomes a native Human Review Flow, and code nodes are re-expressed as a Custom Cloud Code block, a custom action, or logic in flows. The logic carries over; the ownership model changes.](https://flowrunner.ai/images/compare/n8n-migration-map.webp) For teams already on n8n and weighing a switch, the practical questions: - **What maps mechanically?** Triggers, actions, and basic flow structure usually map directly. Most n8n app nodes have a FlowRunner action equivalent, and the step structure carries over. - **What gets re-expressed?** JavaScript in n8n custom nodes does not always translate node-for-node. Logic that lived in code becomes business logic in FlowRunner flows, a Custom Cloud Code block, or a custom action. The Send and Wait for Response node becomes an agent-invoked human-in-the-loop Flow that the agent calls at runtime and that reaches more channels. - **What gets better in the process?** The team’s relationship to the automation changes: the operations team owns it instead of engineering. A human step wired in at design time becomes a pause the agent decides on at runtime, and a compliant configuration reaches down to mid-tier instead of Enterprise. - **What stays the same?** Your integrations with third-party systems mostly stay the same. We connect to the same Salesforce, the same QuickBooks, the same Slack workspace. - **Who runs the migration, and how long?** The Midnight Flow consulting team runs migrations as part of FlowRunner Enterprise onboarding, and other tiers self-migrate with the platform’s import tools. A typical mid-market deployment with 20 to 50 active workflows migrates over 4 to 8 weeks with consulting support, or longer self-served. If you would like a free migration consultation, reach out via the [Contact page](https://flowrunner.ai/contact) and we will scope the engagement before any commitment. ### Frequently asked questions #### Is FlowRunner cheaper than n8n? It depends on what you need, and on both platforms the billing unit is the same. n8n and FlowRunner both bill per execution, where one full workflow run counts as one unit regardless of step count, so neither has the per-step task math that Zapier or Make impose. n8n's self-hosted Community edition runs unlimited executions for free, which FlowRunner Cloud does not match on sticker. That free path carries the operating cost of running the infrastructure yourself: servers, upgrades, and the DevOps time to keep it healthy, which industry estimates put into six figures a year for a maintained production deployment. On managed cloud, FlowRunner starts lower: a permanent Free plan at 100 executions a month, then Starter at $5/month for 300 executions or $15 for 3,000, with every integration, AI agents, and human-in-the-loop included at every step, where n8n Cloud's cheapest plan is Starter at €20/month. FlowRunner Growth ($45/month, 12,000 executions) sits between n8n's Starter and its Pro plan (€50/month). The gap widens the moment compliance matters. FlowRunner ships audit trails and RBAC, with a BAA available, starting at Professional ($299/month). n8n has no Cloud BAA at any price, and gates SSO, RBAC, and audit streaming to its custom-priced Enterprise tier, so a regulated team ends up either on Enterprise or self-hosting and owning all of its own compliance. #### Can I migrate from n8n to FlowRunner? Yes. Triggers, actions, and basic flow structure usually map directly. JavaScript in n8n custom nodes does not translate node-for-node, so logic that lived in code becomes business logic in FlowRunner flows or is re-expressed with FlowRunner's blocks. The Midnight Flow consulting team runs migrations as part of FlowRunner Enterprise onboarding, and other tiers self-migrate with the platform's import tools. Reach out via the Contact page for a free migration consultation. #### Does FlowRunner have as many integrations as n8n? Yes, and on verified first-party integrations FlowRunner has more. FlowRunner ships 1,800+ verified first-party integrations against n8n's roughly 330 built-in core app nodes. n8n's site advertises around 1,953, but that number counts trigger and action listings separately, is dominated by community-maintained npm nodes, and n8n Cloud exposes only a curated subset, so it is not a like-for-like figure. An adversarial app-by-app audit found FlowRunner has built 228 of the 246 tracked n8n app-nodes, and FlowRunner covers 40+ integrations n8n has no built-in node for, including Databricks, Snowflake, Azure data services, and a full security and ITSM stack. Where n8n's built-in set genuinely reaches further is protocol nodes: message brokers (Kafka, RabbitMQ, MQTT) and core SSH, FTP, and Git, which FlowRunner does not yet cover. Every FlowRunner action also ships agent-tool metadata, so an agent can call it as a tool directly. #### Can I self-host FlowRunner like I can n8n? Both platforms self-host, and n8n's self-hosting is more mature today: it is the flagship deployment model, free, unlimited on executions, and available now. FlowRunner offers a self-hosted Enterprise option with full feature parity including audit trails, SLA tracking, RBAC, and SSO, and a free Community Edition for single-instance deployments is coming. One difference to weigh: self-hosting either platform means you own the infrastructure and its operating cost, and n8n Cloud signs no BAA, so a regulated team that self-hosts for free also owns all of its own compliance. #### Does FlowRunner work without writing code? Yes. Both the visual editor and the Conversational AI mode let non-developers build agents and flows and ship them to production without owning code or infrastructure. n8n works without code for simple flows, and it exposes custom JavaScript and Python anywhere for teams that want that flexibility, but the platform's design assumes code fluency for anything substantial. That is the core difference in design center: FlowRunner is built so the operations team that owns the process ships the automation; n8n is built for developers who want to write and own the code. #### Is n8n open source and FlowRunner is not? n8n uses a fair-code license (its Sustainable Use License restricts reselling or embedding n8n commercially); FlowRunner is proprietary. FlowRunner offers a free self-hosted Community Edition that covers the open-core use cases, though it is coming rather than available today, and it is not fair-code. The choice between fair-code and proprietary matters most to teams who plan to fork or extensively modify the platform, or who need to self-host without a commercial contract. For those teams n8n's model is the cleaner answer. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) > _I like that aspect of the human in the loop when it's needed. That's ultimately what we need to get to._ VP Operations, CPG brand --- ## FlowRunner vs Power Automate Source: https://flowrunner.ai/compare/flowrunner-vs-power-automate FlowRunner vs Power Automate: for cloud AI-agent automation FlowRunner does more, without the premium-connector paywall or Microsoft lock-in. Where each wins, and how to choose. 2026-07-18 Updated September 7, 2026 13 min ![Premium connectors or the whole catalog. Power Automate includes standard connectors like Outlook, Teams, SharePoint, and Excel but gates HTTP, SQL Server, Salesforce, ServiceNow, Oracle, and AWS or Azure services behind its Premium plan at $15 per user per month. FlowRunner includes its whole catalog of 2,100+ integrations and 56,000+ callable actions on every plan, including Free, with no premium-connector gate.](https://flowrunner.ai/images/compare/power-automate-connector-gate.webp) **TL;DR** - Start with the cloud automation both run: connectors, triggers, actions, approvals. FlowRunner runs all of it, and it runs the layer above, coordinating AI agents and keeping a human in control of the decisions that need judgment, without a per-connector paywall or a Microsoft-estate dependency. FlowRunner is [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service). Two things Power Automate does that FlowRunner does not: desktop RPA, screen-level automation of legacy Windows apps, and the very broadest inherited compliance (FedRAMP, government clouds). - This page is about AI-agent automation with human oversight in a Microsoft-independent stack. For that specific work the fit is not 50/50: FlowRunner is built for it, and Power Automate is being stretched to reach it. We will still show you exactly where Power Automate is the better call, because sometimes it is. - FlowRunner is stronger where this reader lives: per-execution billing that does not multiply with steps or retries, no premium-connector paywall, multi-channel agent-invoked human-in-the-loop that waits up to a year, and BYOK across dozens of AI providers with no markup on inference. - Power Automate is stronger on inherited Microsoft compliance breadth, desktop RPA, native Microsoft 365 and Dynamics depth, enterprise ubiquity, and its Copilot toolchain. ### Who this comparison is for You are a builder who has Power Automate available, probably because it came bundled with Microsoft 365, and you are deciding whether to keep building agent automation on it or move that work somewhere designed for it. The agents you are building need to do more than fire a linear flow and finish. They need to call out to non-Microsoft systems. They need to wait on a person who might not answer today. They may run against a limit you did not know existed until a run failed at scale. The overlap is large: the cloud, connector-based automation Power Automate runs, FlowRunner runs too, and it adds the agentic layer, agents that pause for a human and long-running orchestration, without the premium-connector paywall or the Microsoft-estate lock-in. Two things Power Automate does that FlowRunner does not: desktop RPA that drives legacy Windows apps by their screen, and the very broadest inherited compliance (FedRAMP, government clouds). This page is about where those facts leave you: the agent that has to reach outside Microsoft, hold a run open until a human weighs in, and do it without a per-user premium license gating the connectors that matter. ![One run or 40,000 requests. Power Automate meters every action, retry, HTTP call, and pagination request as a Power Platform request against a ceiling of 40,000 per user per 24 hours, or 6,000 on free and seeded licenses. FlowRunner counts one complete workflow run as one execution, so retries and pagination do not multiply the count, with Growth starting at $45 per month. The workflow may look the same; the billing math does not.](https://flowrunner.ai/images/compare/power-automate-billing-model.webp) ### What FlowRunner is good at FlowRunner is built around one commitment: agents that know when to stop and ask a human for help. Everything else in the platform exists to make that commitment reliable at production scale, and to do it without tying your automation to a single vendor’s ecosystem. The mechanism is the part that linear, estate-bound tools do not have. In FlowRunner, an AI agent invokes one of your own Flows as a callable tool. When the agent calls that flow-tool, it suspends while the flow runs, and the flow can hold on an external callback until a human replies through email, Slack, WhatsApp, phone, or a Form, then resume mid-run. This is [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) as an execution pattern: the agent pauses on its own judgment, assembles the context and the choices available, routes to a human on their preferred channel, and resumes the moment they answer. Some teams call it a [digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord), after the Toyota factory pull cord any worker can pull to stop the line when something needs a human eye. ![Human-in-the-loop compared, an approval step versus a callable tool. Power Automate places an approval at design time inside a cloud flow, routes it to Outlook, Teams, or the action center, stores history in Dataverse, and caps the wait at the 30-day flow run limit. In FlowRunner the agent detects uncertainty at runtime, calls a Human Review Flow as a tool, suspends the run, reaches the person through email, Slack, WhatsApp, phone, or a Form, and resumes mid-run on every tier, able to loop, ask a follow-up, and escalate.](https://flowrunner.ai/images/compare/power-automate-human-in-the-loop.webp) FlowRunner is strong on: - **No premium-connector paywall.** FlowRunner’s whole catalog is available at every tier. The [connectors](https://flowrunner.ai/concepts/connectors) you need most (HTTP, databases, Salesforce, ServiceNow, and every non-Microsoft SaaS) are not gated behind a per-user upgrade. Connectors are reusable bridges between FlowRunner and third-party services that any agent or workflow can compose into a capability, and none of them cost extra to reach. - **Run-based billing.** One complete workflow run, start to finish, is one execution. A 10-step run is one execution, not ten, and retries and pagination do not multiply the count. That removes the per-request math that Power Platform request metering forces on builders. - **Long waits and long runtimes.** A run can suspend for up to 30 days on Growth and up to a year on Professional, Business, and Enterprise. Cumulative runtime runs to 1 hour on Growth, 4 hours on Professional, 12 hours on Business, and unlimited on Enterprise. A pending human decision does not expire at 30 days. - **[BYOK](https://flowrunner.ai/concepts/byok) across dozens of providers with no markup.** BYOK is a model where the customer supplies their own AI provider API credentials, pays providers directly for usage, and FlowRunner orchestrates calls across providers without taking a markup on inference. You are not confined to Azure OpenAI, and you are not buying AI capacity through a Copilot Credits meter. For a buyer who already has an Anthropic, Google, or Azure agreement, that means your own rates, your own data-processing terms, and a provider your security team already approved. - **Compliance at mid-tier prices.** Audit trails and RBAC start at the Professional tier ($299/month). HIPAA support with a BAA is available. SSO/SAML and 90-day audit retention arrive at Business ($999/month), without a premium-connector gate and without standardizing on the Microsoft estate. - **A verified, agent-ready catalog.** 1,800+ verified integrations and 56,000+ callable actions (as of September 2026, and growing), roughly 26 actions per integration, each carrying the structured metadata an LLM needs to call it as a tool, all built and verified against each vendor’s official API reference. Agents built in the [Agent Factory](https://flowrunner.ai/concepts/agent-factory) compose those connectors, other agents, and existing flows without writing code. That pause-for-a-human moment is the load-bearing part of FlowRunner’s architecture. It is why agentic work that needs human judgment is native here rather than assembled from workarounds, and it reaches the systems and the people your agents need without a premium license or a Microsoft standardization. ### What Power Automate is good at Power Automate is Microsoft’s automation platform. Inside the Microsoft estate it runs deep: a certified connector library, desktop RPA that FlowRunner does not have, and compliance inherited from the Microsoft cloud including government clouds. Power Automate is strong on: - **Inherited Microsoft compliance breadth.** A HIPAA Business Associate Agreement is available by default through the Microsoft Online Services DPA, with no special negotiation, and Power Automate cloud is in scope. Beyond that, Microsoft carries SOC 2, ISO 27001, HITRUST, and FedRAMP attestations including the government clouds (GCC, GCC High, DoD). Microsoft Purview provides audit with 90-day retention, and Entra ID provides SSO. This is a compliance surface no independent competitor, FlowRunner included, matches on breadth. - **Genuine desktop RPA.** Power Automate desktop flows provide attended and unattended robotic process automation that drives legacy and modern desktop and web applications by their actual user interface. If you need to automate an old Windows application that has no API, Power Automate can click through its screens. FlowRunner does not do this at all. - **Native Microsoft 365, Dynamics, and Azure depth.** Inside SharePoint, Outlook, Teams, Excel, Dataverse, and Dynamics, Power Automate’s integrations are built by the same company that builds the apps. - **Enterprise ubiquity and zero net-new procurement.** Power Automate is seeded into most Microsoft 365 subscriptions. For a Microsoft-standardized organization, adopting it means no new vendor, no new security review, and no new line item. - **AI Builder’s prebuilt models.** Both platforms have AI agents, natural-language flow builders, and MCP, so those are not the difference. Where Power Automate’s AI goes further for a Microsoft shop is AI Builder: prebuilt, no-code models for document, receipt, and invoice extraction, prediction, and classification, wired into the Microsoft stack. FlowRunner orchestrates any model you bring through BYOK, but does not ship prebuilt ML models like these. - **Process mining.** Power Automate Process Mining analyzes how work actually flows through your systems to find automation candidates, a capability aimed at large process-heavy organizations. Power Automate is built for automation inside the Microsoft estate. ![Two AI operating models. Power Automate's AI runs on the Microsoft stack, Copilot, AI Builder, and Copilot Studio over Azure OpenAI, with agents and MCP metered through Copilot Studio at $200 for 25,000 Copilot Credits per month. FlowRunner uses BYOK across dozens of providers including Anthropic, Google, Azure, and OpenAI, with no inference markup, your own contracts and rates, and one orchestration layer across every provider.](https://flowrunner.ai/images/compare/power-automate-byok.webp) ### Where FlowRunner is stronger For the agentic, human-oversight, Microsoft-independent work this page is about, these are the differences that decide the platform: - **No premium-connector paywall.** This is the sharpest practical gap. Power Automate’s free and Microsoft 365-seeded entitlement includes standard connectors only. The genuinely useful ones (HTTP, SQL Server, Salesforce, Oracle, ServiceNow, AWS and Azure services, and every custom connector) are premium connectors that require the Premium plan at $15/user/month. Because it is per user, the cost climbs with every person who runs a flow touching those systems. FlowRunner’s entire catalog is available on every plan, including Free. The connector you need is never a line-item upgrade. - **Per-execution billing instead of per-request metering.** Power Automate meters Power Platform requests, and Microsoft’s own documentation is explicit that every action counts, from initializing a variable to a compose step, and that retries and pagination requests count too. A Premium user gets 40,000 requests per user per 24 hours; the free and Office 365-seeded ceiling is 6,000 per user per day. A busy multi-step flow burns through that faster than teams expect. FlowRunner bills one execution per whole run regardless of step count, so the cost is predictable and the metering does not shape how you design. - **Human-in-the-loop across every channel, waiting up to a year.** Power Automate Approvals routes to Outlook actionable email, a Teams adaptive card, or the action center, it depends on Dataverse, and a pending approval dies at the 30-day flow run cap. FlowRunner’s human-in-the-loop reaches people through email, Slack, WhatsApp, phone, or a Form, the agent invokes it as a callable tool rather than a linear approval step, and the wait can last up to a year. For any workflow where the human might take days or weeks, that difference decides the platform. - **The 30-day run ceiling does not exist.** A Power Automate cloud flow run maxes out at 30 days, pending approvals included, and synchronous HTTP requests time out at 120 seconds. FlowRunner suspends a run for up to a year and runs up to 12 hours or unlimited depending on tier, and a flow can begin with any block, not only a trigger. Long-running orchestration is native, not a workaround built from child flows. - **BYOK across dozens of AI providers, no markup.** Power Automate’s AI is Microsoft-stack-centric, powered by Azure OpenAI, with agents and MCP living in Copilot Studio behind a Copilot Credits meter. FlowRunner lets you bring your own keys across dozens of providers, pay them directly at your negotiated rates under your own DPA, and orchestrate across them without an inference markup. For a buyer with an existing Anthropic, Google, or Azure agreement, that is their processor, their contract, their rate. - **Compliance without the Microsoft lock-in or the connector gate.** FlowRunner ships HIPAA support with a BAA, plus audit trails and RBAC, starting at Professional ($299/month), with SSO/SAML at Business ($999/month). Power Automate has wider compliance breadth, but reaching a compliant, connector-complete configuration on Power Automate means the Premium plan and standardizing on the Microsoft estate. FlowRunner reaches a compliant configuration at mid-tier prices without either. ![Time ceilings for waits and long-running orchestration. A single Power Automate cloud flow run is capped at 30 days including pending approvals, which time out at that cap, and synchronous HTTP requests time out at 120 seconds. FlowRunner suspends a run for up to 30 days on Growth and up to a year on Professional, Business, and Enterprise, with cumulative runtime of 1 hour on Growth, 4 hours on Professional, 12 hours on Business, and unlimited on Enterprise.](https://flowrunner.ai/images/compare/power-automate-time-ceilings.webp) ### Where Power Automate is stronger This is the section most comparison pages skip. Power Automate wins on real, decisive dimensions, and if these are your priorities, Power Automate is the right call. - **Compliance breadth you cannot get independently.** Power Automate inherits Microsoft’s full estate: HIPAA BAA by default, SOC 2, ISO 27001, HITRUST, and FedRAMP including government clouds, plus Purview audit and Entra ID SSO. FlowRunner offers HIPAA support with a BAA and mid-tier audit and RBAC, but it does not carry FedRAMP or HITRUST, and it cannot serve GCC High or DoD environments. For a federal agency, a defense contractor, or an organization that has standardized on Purview for audit, Power Automate wins outright. We will not pretend that gap is small. - **Real desktop RPA.** This is the clearest line. Power Automate desktop flows automate legacy applications by driving their user interface, attended or unattended. FlowRunner does not do UI automation or screen scraping at all. If part of your process is clicking through a Windows application that has no API, only one of these two platforms can do it, and it is not FlowRunner. - **Native Microsoft depth.** For automation that lives inside SharePoint, Outlook, Teams, Excel, Dataverse, and Dynamics, Power Automate’s first-party integrations go deeper than any third party, FlowRunner included, can match. Same-vendor integration is a genuine advantage. - **Zero net-new procurement.** Power Automate is already in most Microsoft 365 subscriptions. For a Microsoft-standardized organization, that means no new vendor to onboard, no new security review, and no new contract. FlowRunner is a net-new purchase and a net-new security review. - **Process mining and a compounding Microsoft AI investment.** Power Automate Process Mining analyzes how work flows through your systems to surface automation candidates, which FlowRunner does not offer, and AI Builder’s prebuilt models plus a broad Copilot investment compound inside the Microsoft estate. If your organization is standardizing on Copilot, Power Automate is where that investment pays off. If any of those outweigh the differences above, Power Automate is your platform, and choosing it is defensible. ![Desktop RPA is the clearest line between the platforms. Power Automate desktop flows drive legacy Windows applications by their user interface, attended or unattended, so a process with no API can still be automated by clicking through its screens. FlowRunner is API and agent orchestration only, with no UI automation and no screen scraping, so it has no desktop RPA equivalent. If part of the process is clicking through a Windows application by its interface, only Power Automate can do it.](https://flowrunner.ai/images/compare/power-automate-desktop-rpa.webp) ### Feature comparison | Dimension | FlowRunner | Microsoft Power Automate | | --- | --- | --- | | Primary buyer | Builder plus operations team | Microsoft-standardized enterprise | | Billing unit | Execution (one whole run) | Power Platform request (per action, retry, pagination) | | Metering ceiling | Per-tier execution count, no per-action metering | 40,000 requests/user/24h Premium; 6,000 free/seeded | | Entry price for full connectors | Free $0, whole catalog included; paid plans from $5/mo Starter | $15/user/mo Premium (as of Jul 2026) | | Connector gating | No gating; all connectors every tier | Standard free; premium + custom behind Premium | | Unlimited users | Yes, every plan including Free | No (per-user and per-bot licensing) | | Max single-run duration | 1h Growth, 4h Pro, 12h Business, unlimited Enterprise | 30 days incl. pending approvals | | Max human wait | 30 days Growth, up to 1 year other tiers | 30 days (approval dies at run cap) | | Human-in-the-loop | Agent-invoked callable tool, every tier | Approvals (Outlook / Teams / action center; Dataverse-bound) | | HITL channels | Email, Slack, WhatsApp, phone, Form | Outlook email, Teams, action center | | AI agents | Native ([Agent Factory](https://flowrunner.ai/concepts/agent-factory)), BYOK across providers | Copilot + AI Builder + Copilot Studio (Azure OpenAI) | | AI provider model | BYOK, dozens of providers, no inference markup | Microsoft-stack-centric; Copilot Credits meter | | MCP | Consumes external MCP servers; exposes catalog as MCP server (shipping ~Aug 2026) | Via Copilot Studio (generative orchestration) | | Desktop RPA / UI automation | No (API and agent orchestration only) | Yes, attended and unattended | | Connector library | 1,800+ verified integrations, 56,000+ actions | 1,000+ certified (Microsoft cites up to 1,400) | | HIPAA / BAA | HIPAA support with BAA | Yes, by default via Microsoft Online Services DPA | | FedRAMP / HITRUST / gov clouds | No | Yes (GCC, GCC High, DoD) | | Audit / SSO | Audit + RBAC at Professional ($299); SSO/SAML at Business ($999) | Purview audit (90-day); Entra ID SSO | | Self-hosting | Enterprise on-prem; Community Edition coming | No (on-prem gateway is a data bridge only) | ### Decision framework ![Compliance is a platform gate decided before the build. If a workflow processes regulated data and needs FedRAMP, a government cloud, or the broadest Microsoft compliance, Power Automate wins on compliance breadth. If it needs HIPAA with a BAA, audit trails, and RBAC without a Microsoft-estate lock-in, FlowRunner reaches a compliant cloud configuration at mid-tier prices. A capability table compares HIPAA and BAA, FedRAMP and government clouds, audit and RBAC, SSO and SAML, connector-complete setup, and self-hosting across both platforms.](https://flowrunner.ai/images/compare/power-automate-compliance-gate.webp) **Pick FlowRunner if:** - The connectors you actually need are the ones Power Automate paywalls (HTTP, SQL, Salesforce, ServiceNow, non-Microsoft SaaS), and you do not want a per-user Premium license to reach them. - Your agents need to stop and pull in a human mid-run through Slack, WhatsApp, phone, or a Form, and the human might take days or weeks to answer. - You want per-execution billing you can predict, instead of counting every action, retry, and pagination call against a 40,000-request daily ceiling. - You want BYOK across dozens of AI providers at your own rates and under your own DPA, not Azure OpenAI through a Copilot Credits meter. - You need HIPAA with a BAA, audit trails, and RBAC at mid-tier prices without standardizing your whole automation stack on Microsoft. **Pick Power Automate if:** - Your automation lives inside Microsoft 365, Dynamics, and Azure, and you want first-party depth from the vendor that builds those apps. - You need desktop RPA to drive a legacy Windows or web application by its user interface, attended or unattended. FlowRunner cannot do this. - You operate in a FedRAMP, HITRUST, or government-cloud environment (GCC, GCC High, DoD) where Microsoft’s inherited compliance is a hard requirement. - Power Automate is already seeded into your Microsoft 365 subscription and avoiding a net-new vendor and security review is the deciding factor. - Your organization is investing in Copilot broadly and you want automation that compounds with that stack, including Process Mining. If the answer is honestly mixed, run one real workflow on both before committing. The right platform is the one that fits the shape of your work and your stack, not the one that came bundled. ### Migration considerations For builders already on Power Automate and weighing a switch, the practical questions: ![Migration map from a Power Automate cloud flow to a FlowRunner flow. Triggers, actions, conditions, loops, approvals, and the Scope plus run-after try/catch pattern carry over to FlowRunner's trigger, action, condition, repeat, Human Review Flow, and Handle Error blocks. Approvals become agent-invoked human-in-the-loop, the per-action request math disappears, and runs can wait far longer. Power Automate desktop flows have no direct equivalent and must stay or be replaced with an API path. The cloud-flow logic carries over; the operating model changes.](https://flowrunner.ai/images/compare/power-automate-migration-map.webp) - **What maps mechanically?** Cloud flow logic carries over. Triggers, actions, conditions, and loops map directly. Power Automate’s try/catch pattern (Scopes plus “Configure run after”) re-expresses as FlowRunner’s Condition, Repeat, and Handle Error blocks. Approvals become FlowRunner’s agent-invoked human-in-the-loop across more channels, and without the 30-day death. - **What does not map, honestly?** Desktop RPA. FlowRunner does not do UI automation or screen scraping, so any Power Automate desktop flow that drives a legacy Windows application by its interface has no FlowRunner equivalent. If that RPA is load-bearing for your process, keep it on Power Automate or replace it with an API path where one exists. - **What gets better in the process?** Any flow that fought the 40,000-request daily ceiling stops counting individual actions. Any human-in-the-loop step that risked timing out at 30 days can now wait up to a year. Any connector you were paying $15/user/month to reach is included. - **Who runs the migration?** The Midnight Flow consulting team runs cloud migrations as part of FlowRunner Enterprise onboarding. Other tiers self-migrate with the platform’s import tools. - **How long does it take?** A typical mid-market cloud deployment migrates over 4 to 8 weeks with consulting support, longer if a meaningful share of the estate depends on desktop RPA that has to be re-architected around APIs. If you would like a free migration consultation, reach out via the [Contact page](https://flowrunner.ai/contact) and we will scope the engagement, including the desktop-RPA question, before any commitment. ### Frequently asked questions #### Is FlowRunner a good Power Automate alternative? It depends on whether your automation lives inside Microsoft. If your work is Microsoft 365, Dynamics, and Azure end to end, and you need desktop RPA to drive legacy Windows apps, Power Automate fits that case: it is seeded into your existing subscriptions and inherits Microsoft's compliance estate. FlowRunner is the better alternative when your automation is Microsoft-independent, the useful connectors you need (HTTP, SQL, Salesforce, ServiceNow, non-Microsoft SaaS) are what Power Automate paywalls behind its Premium plan, you want per-execution billing instead of per-request metering, and you are building AI agents that pause for a human across email, Slack, WhatsApp, phone, or a Form rather than only Outlook and Teams. #### How much do Power Automate premium connectors cost? Power Automate's free and Microsoft 365-seeded entitlement includes standard connectors only. The genuinely useful connectors (HTTP, SQL Server, Salesforce, Oracle, ServiceNow, AWS and Azure services, and all custom connectors) are premium connectors that require the Power Automate Premium plan at $15/user/month (as of July 2026). That is per user, so the cost scales with how many people run flows that touch those connectors. FlowRunner does not gate its connectors by tier. The whole catalog of 2,100+ integrations is available on every plan, including the Free plan at $0 and Starter at $5/month. #### Does Power Automate have human in the loop? Yes. Power Automate has Approvals with five approval types, and a flow can pause on a 'Start and wait for an approval' action. Two constraints matter for agent work. First, approvals route to Outlook actionable email, a Teams adaptive card, or the action center, so there is no native Slack, WhatsApp, phone, or web-form channel, and the feature depends on Dataverse. Second, a pending approval dies at the 30-day flow run cap. FlowRunner's human-in-the-loop is architecturally different: the AI agent invokes a Flow as a callable tool and suspends until a human replies through email, Slack, WhatsApp, phone, or a Form, and the wait can last up to a year, on every tier. #### What is Power Automate's flow run time limit? A single Power Automate cloud flow run has a maximum duration of 30 days, and Microsoft's own documentation states this includes flows with pending steps like approvals, which then time out after 30 days. Synchronous HTTP requests time out at 120 seconds, and the minimum scheduled recurrence is 60 seconds. The 30-day ceiling is the practical reason a long-running human-in-the-loop flow can fail: if the approver takes longer than 30 days, the run and the pending approval both expire. FlowRunner suspends a run for up to a year on Professional, Business, and Enterprise, and up to 30 days on Growth. #### Does Power Automate have AI agents? Yes. Power Automate has Copilot (powered by Azure OpenAI) for building and describing flows, AI Builder for prebuilt models, and full agents through Copilot Studio, which also carries its Model Context Protocol support. The difference is model breadth and cost. Power Automate's AI is Microsoft-stack-centric, and agents plus MCP live in Copilot Studio at $200 for 25,000 Copilot Credits per month. FlowRunner uses BYOK: you supply your own keys across dozens of AI providers, pay providers directly at your negotiated rates, and FlowRunner takes no markup on inference. Note that Power Automate's bundled 5,000 AI Builder credits per month are being removed in November 2026, so new customers buy Copilot Credits. #### Is Power Automate HIPAA compliant? Yes. This is a real Power Automate strength. A HIPAA Business Associate Agreement is available by default through the Microsoft Online Services DPA with no special negotiation, Power Automate cloud is in scope, and Microsoft also holds SOC 2, ISO 27001, HITRUST, and FedRAMP attestations including government clouds. That inherited compliance breadth is wider than FlowRunner's. FlowRunner offers HIPAA support with a BAA and ships audit trails and RBAC starting at the Professional tier ($299/month) without the Microsoft lock-in and without gating the useful connectors behind a premium plan. The honest framing is not 'more compliant.' It is a compliant configuration at mid-tier prices without tying your automation to the Microsoft estate. #### Can I migrate from Power Automate to FlowRunner? Yes, with one honest exception. Cloud flow logic maps over: triggers, actions, conditions, loops, and the try/catch pattern (Power Automate's Scopes plus 'Configure run after') re-express as FlowRunner's Condition, Repeat, and Handle Error blocks. Approvals become FlowRunner's agent-invoked human-in-the-loop across more channels. The exception is desktop RPA. FlowRunner does not do UI automation or screen scraping, so Power Automate desktop flows that drive legacy Windows applications do not have a FlowRunner equivalent and stay where they are. The Midnight Flow consulting team runs cloud migrations as part of Enterprise onboarding; other tiers self-migrate. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) > _I like that aspect of the human in the loop when it's needed. That's ultimately what we need to get to._ VP Operations, CPG brand --- ## FlowRunner vs Zapier Source: https://flowrunner.ai/compare/flowrunner-vs-zapier FlowRunner vs Zapier: FlowRunner does everything Zapier does and adds native AI-agent orchestration with human oversight. Where FlowRunner wins, where Zapier still helps, how to choose. 2026-07-18 Updated September 7, 2026 13 min ![Choosing for the shape of the work: a linear Zapier trigger-to-action chain beside a FlowRunner agent that calls tools, loops, waits, and pulls in human judgment at runtime.](https://flowrunner.ai/images/compare/zapier-shape-of-work.webp) **TL;DR** - Start with what these two share: the linear, app-to-app automation Zapier is known for, connecting apps, moving data from a trigger through a chain of actions, ready-made building blocks. FlowRunner does all of that. FlowRunner is [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service), so it also runs the layer above the linear flow: coordinating AI agents, governing multi-agent work, and keeping a human in control of the decisions that need judgment. Read them side by side and the shape is plain: Zapier does the linear slice; FlowRunner does that slice and the agentic orchestration on top. - This page is about AI-agent automation with human oversight. For that specific work the fit is not 50/50: FlowRunner is built for it, and Zapier is being stretched to reach it. We will still show you exactly where Zapier is the better call, because sometimes it is. - FlowRunner is stronger where this reader lives: human-in-the-loop as a native agent tool on every tier, run-based billing that does not multiply with steps, waits up to a year, HIPAA with a BAA at mid-tier prices, and real depth per integration. - Zapier is stronger on raw connector breadth, its ready-made recipe ecosystem, and market familiarity. ### Who this comparison is for You are building AI-agent automation where the agent has to do more than fire a linear sequence and finish. It needs to stop and pull in a person when the rules do not cleanly apply. It may run many steps thousands of times a month. It may touch data an auditor will ask about later. You might be coming from Zapier or starting fresh; either way you are deciding which platform is built for that work. The overlap is large: the linear, app-to-app automation Zapier is known for, FlowRunner does too. What FlowRunner adds on top is the agentic layer, agents that pause for a human, long-running orchestration, and built-in compliance. Where that added layer matters, FlowRunner is not merely also an option; it is the platform built for it. This page is about exactly where Zapier’s ceiling is and where FlowRunner keeps going: the agent that hits an [automation exception](https://flowrunner.ai/concepts/automation-exceptions) it should not decide alone and has to hold the run open until a human answers. ### What FlowRunner is good at FlowRunner is built around one commitment: agents that know when to stop and ask a human for help. Everything else in the platform exists to make that commitment reliable at production scale. The mechanism is the part that linear tools do not have. In FlowRunner, an AI agent invokes one of your own Flows as a callable tool. When the agent calls that flow-tool, it suspends while the flow runs, and the flow can hold on an external callback until a human replies through email, Slack, WhatsApp, phone, or a Form, then resume mid-run. This is [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) as an execution pattern: the agent pauses on its own judgment, assembles the context and the choices available, routes to a human on their preferred channel, and resumes the moment they answer. Some teams call it a [digital andon cord](https://flowrunner.ai/concepts/digital-andon-cord), after the Toyota factory pull cord any worker can pull to stop the line when something needs a human eye. ![Human approval is a step, human judgment is a tool. Zapier's fixed Request Approval step (design-time, single, not inside a loop, reviewer opens Zapier) beside FlowRunner's agent-invoked human review flow that reaches email, Slack, WhatsApp, phone, or a form and resumes the run mid-execution.](https://flowrunner.ai/images/compare/zapier-human-in-the-loop-architecture.webp) FlowRunner is strong on: - **Run-based billing.** One complete workflow run, start to finish, is one execution. A 10-step run is one execution, not ten, and there is no per-step or per-tool-call multiplier. That removes the task math that per-task billing forces on builders. - **Long waits and long runtimes.** A run can suspend for up to 30 days on Growth and up to a year on Professional, Business, and Enterprise. Cumulative runtime runs to 1 hour on Growth, 4 hours on Professional, 12 hours on Business, and unlimited on Enterprise. - **Compliance at mid-tier prices.** Audit trails and RBAC start at the Professional tier ($299/month). HIPAA support with a BAA is available. SSO/SAML and 90-day audit retention arrive at Business ($999/month). - **A verified, agent-ready catalog with depth.** 1,800+ verified integrations and 56,000+ callable actions (as of September 2026, and growing), roughly 26 actions per integration, each carrying the structured metadata an LLM needs to call it as a tool, all built and verified against each vendor’s official API. - **[BYOK](https://flowrunner.ai/concepts/byok) with no markup.** BYOK is a model where the customer supplies their own AI provider API credentials, pays providers directly for usage, and FlowRunner orchestrates calls across providers without taking a markup on inference. For a buyer who already has an Anthropic, Google, or Azure agreement, that means your own rates, your own data-processing terms, and a provider your security team already approved, rather than a new bundled inference vendor to review. That pause-for-a-human moment is the load-bearing part of FlowRunner’s architecture. It is why agentic work that needs human judgment is native here rather than assembled from workarounds. ### What Zapier is good at Zapier is the most widely used tool in the automation category. The step-by-step Zap editor, the largest connector directory anywhere, and the surrounding ecosystem mean almost any two apps you want to connect already have a documented recipe. Zapier is strong on: - **Connector breadth, decisively.** Zapier advertises 9,000+ apps, far more than FlowRunner or any direct competitor lists. If your deciding factor is whether a specific long-tail service is supported, Zapier almost certainly has it. - **The getting-started ecosystem.** Thousands of pre-built recipes, a tutorial for practically every app pairing, and the Copilot natural-language builder give a newcomer a large library to start from. - **A well-known free plan.** Zapier’s free plan runs 100 tasks a month with no time limit, and its paid Professional plan starts at $29.99/month monthly (around $19.99 on annual billing, as of July 2026, 750 tasks base). It is a familiar on-ramp. FlowRunner’s Free plan is the same $0 with no time limit, on 100 whole runs rather than 100 action steps, and its paid plans start at $5, so the on-ramp is matched rather than conceded. - **Market familiarity.** Zapier is SOC 2 Type II certified, supports GDPR and CCPA, and is a known quantity in most procurement conversations. That is not a product capability, but it is real, and it lowers buying friction. For linear automation that fits that shape, few tools match Zapier’s breadth and on-ramp. ### Where FlowRunner is stronger For the agentic, human-oversight work this page is about, these are the differences that decide the platform: - **Run-based billing.** Zapier bills per task, where each successful action step is one task and an MCP tool call is two. A 10-step Zap run 1,000 times a month is roughly 10,000 tasks. FlowRunner bills per execution, where the whole run is one unit regardless of step count. At volume, the difference between per-step and per-run billing is large and, more importantly, predictable. You are not re-budgeting every time you add a step to an agent. - **Compliance at mid-tier prices.** As of July 2026, Zapier signs no BAA on any plan and states that PHI is not supported on the platform, which is a hard blocker for healthcare work. FlowRunner ships HIPAA support with a BAA, plus audit trails and RBAC, starting at Professional ($299/month). For a regulated builder, this moves the compliant configuration down to mid-tier instead of ruling the platform out entirely. - **Wait and runtime ceilings.** Zapier’s Delay action holds a run for a maximum of 30 days, and every Zap must start with a trigger. FlowRunner suspends a run for up to a year, runs for up to 12 hours or unlimited depending on tier, and a flow can begin with any block, not only a trigger. For work that waits on a person who might take days to answer, that ceiling difference decides the platform. - **Depth per integration, with agent-tool metadata.** Zapier’s 9,000+ apps span roughly 30,000 actions, about 3.3 actions per app. FlowRunner’s 2,100+ integrations span 56,000+ actions, roughly 26 per integration, and every action carries typed parameters, descriptions, and sample output so an agent can call it as a tool directly. The catalog includes full-CRUD databases, vector stores, and deep ERPs where Zapier’s per-app depth is shallow. Zapier lists more apps; FlowRunner reaches deeper into each one and ships every action agent-ready. - **Unlimited users and self-hosting.** FlowRunner includes unlimited users on every plan, including Free, where Zapier’s lower tiers are seat-limited (its Team plan is commonly cited at 25 seats). FlowRunner also offers an Enterprise on-prem deployment with a Community Edition coming, where Zapier is cloud-only with a US-based data path. - **Human-in-the-loop as an agent tool.** Zapier’s Request Approval is a capable feature: the reviewer can approve, decline, or edit the submitted data, with reminders and an audit trail. But it is a fixed step you place after the agent, single-shot, unable to sit inside a loop, and the reviewer has to log into Zapier to answer. FlowRunner works differently. The agent itself decides at runtime to pull in a person, calls a human-review flow as a tool, and suspends until they reply in their own channel: email, Slack, WhatsApp, phone, or a Form. It can loop over a batch, ask a follow-up question, and escalate if no one answers. That is the difference between an approval step you place in advance and a human the agent decides to call at runtime. ![Same workflow, different multiplication: per task versus per execution. A 10-step workflow run 1,000 times a month costs roughly 10,000 Zapier tasks but 1,000 FlowRunner executions.](https://flowrunner.ai/images/compare/zapier-billing-tasks-vs-executions.webp) ### Where Zapier is stronger This is the section most comparison pages skip. Zapier wins on real dimensions, and if these are your priorities, Zapier is the right call. - **Raw connector breadth.** Zapier advertises 9,000+ apps against FlowRunner’s 1,800+ verified integrations. But that headline number is mostly long tail: the few hundred apps the vast majority of teams actually use are the ones FlowRunner already covers, in depth (roughly 18 callable actions each versus Zapier’s ~3.3). Where Zapier still wins outright is the genuinely obscure or niche service; if a specific long-tail app is your single deciding factor, Zapier probably lists it and FlowRunner may not. - **The getting-started ecosystem.** Zapier’s library of pre-built recipes, its tutorial for nearly every app pairing, and its Copilot builder get a builder starting cold to a first automation quickly, faster than FlowRunner does today. FlowRunner’s answer is a growing catalog of install-and-configure packages, but Zapier’s recipe ecosystem is larger right now, and for a newcomer that head start is real. - **Market familiarity.** Zapier is one of the most recognized names in automation and clears procurement conversations easily. For a buyer who weights that familiarity and the broad surrounding surface (Interfaces, Tables, Chatbots), that recognition can matter. - **SSO from a mid-tier plan.** Zapier includes SAML SSO on its Team plan ($103.50/month monthly, around $69 annual). If SSO is a hard requirement, you reach it without an Enterprise negotiation, whereas FlowRunner’s SSO/SAML arrives at Business ($999/month). If any of those outweigh the differences above, Zapier is the right call, and choosing it is defensible. ![Breadth versus depth: two catalog strategies. Zapier's 9,000-plus apps at roughly 3.3 actions each beside FlowRunner's 350-plus verified integrations at roughly 23 actions each.](https://flowrunner.ai/images/compare/zapier-breadth-vs-depth.webp) ### Feature comparison | Dimension | FlowRunner | Zapier | | --- | --- | --- | | Primary buyer | Builder plus operations team | Builder, linear automation author | | Billing unit | Execution (one whole run) | Task (per successful action step; MCP call = 2 tasks) | | Cost of a 10-step run x 1,000/mo | 1,000 executions | ~10,000 tasks | | Free to start | Free plan: 100 executions/mo (whole runs), permanent, no card, every integration and human-in-the-loop included; plus 14 days of Professional on signup | 100 tasks/mo (action steps), permanent | | Paid entry price | $5/mo Starter (300 executions); $15 (3,000); $45/mo Growth (12,000) | $29.99/mo Professional (~$19.99 annual, 750 tasks) | | AI entry point | Free plan (BYOK, no markup) | AI Agents (billed by activities) + Copilot | | Unlimited users | Yes, every plan including Free | Seat-limited on lower tiers (Team commonly cited at 25) | | Max native wait | 30 days Growth, up to 1 year other tiers | Delay action 30 days | | Max single-run time | 1h Growth, 4h Pro, 12h Business, unlimited Enterprise | Not a fixed run cap; per-Zap 100-step limit | | Human-in-the-loop | Native every tier; agent-invoked callable tool; can loop, branch, escalate | Request Approval (Pro+); self-approve on Pro; single-shot; first responder only; not in loops | | HITL reviewer experience | Reply in-channel: email, Slack, WhatsApp, phone, Form | Email/Slack notify only; reviewer logs into Zapier to answer | | Start block | Any block (trigger, action, condition, loop) | Trigger required | | AI agents | Native ([Agent Factory](https://flowrunner.ai/concepts/agent-factory)) | Yes (native Agents + Copilot builder) | | MCP | Consumes external MCP servers; exposes catalog as MCP server (shipping ~Aug 2026) | Native server (~30,000 actions; 2 tasks/call) | | Integration directory | 1,800+ verified integrations, 56,000+ actions (~26/integration) | 9,000+ apps (~30,000 actions, ~3.3/app) | | Agent-tool metadata on actions | On every action | Not the model | | Retry | Handle Error block + Repeat/Wait retry | Autoreplay, Pro+, up to 5 tries (5m/30m/1h/3h/6h) | | HIPAA / BAA | HIPAA with BAA | No BAA on any plan; not HIPAA-eligible (as of Jul 2026) | | Audit trails / RBAC | Professional ($299) | Enterprise | | SSO/SAML | Business ($999) | Team plan | | SOC 2 | Security program designed to meet common audit requirements | Type II | | Self-hosting | Enterprise on-prem; Community Edition coming | No (cloud-only, US data path) | | BYOK for AI | Yes, no markup on inference | Bundled inference (AI by Zapier) | ### Decision framework ![For PHI workflows this is an eligibility question: a compliance decision gate. If a signed BAA is required, Zapier is unavailable and FlowRunner is supported. HIPAA and BAA, audit trails, RBAC, and self-hosting compared, with FlowRunner reaching audit and RBAC at Professional versus Zapier's Enterprise tier.](https://flowrunner.ai/images/compare/zapier-compliance-gate.webp) **Pick FlowRunner if:** - Your agents need to stop and pull in a human mid-run, and the human might take hours or days to answer, not seconds. - You are building AI agents where human-in-the-loop is the point, and you do not want it limited to a self-approve-only step that cannot run inside a loop. - Predictable run-based billing matters more than a low sticker price, because your workflows run many steps many times a month and per-task math is hard to budget. - You handle regulated data and need HIPAA with a BAA, audit trails, and RBAC, which Zapier cannot provide at any price. - You want every action in the catalog callable by an agent as a tool, with the depth (full-CRUD databases, vector stores, deep ERPs) and the metadata to make that reliable. **Pick Zapier if:** - Your automations are short, linear, run in the cloud, and never need a human to weigh in mid-run for longer than a quick self-approval. - Raw connector breadth is your single most important criterion, and you need a listing for a specific long-tail service today. - You want a fast first build, and Zapier’s library of pre-built recipes and per-pairing tutorials is the on-ramp you value. - You need SAML SSO from a mid-tier plan and do not handle protected health information. If the answer is honestly mixed, run a real workflow on both before committing. The right platform is the one that fits the shape of your work, not the one with the larger app directory. ### Migration considerations ![Migration map from Zapier to FlowRunner: the logic carries over, the execution model changes. Zapier constructs (trigger, action, filter, paths, autoreplay, request approval) map to FlowRunner blocks (trigger, action, condition, value router, handle error plus repeat and wait, human-in-the-loop flow).](https://flowrunner.ai/images/compare/zapier-migration-map.webp) For builders already on Zapier and weighing a switch, the practical questions: - **What maps mechanically?** A Zap’s trigger becomes a FlowRunner trigger, and each action step becomes a FlowRunner action. Most Zapier app actions have a FlowRunner action equivalent, and the linear step structure carries over directly. - **What gets re-expressed?** Zapier’s Filters and Paths map to FlowRunner’s Condition and Value Router blocks. Autoreplay-style retries map to FlowRunner’s Handle Error block plus a Repeat and Wait retry pattern. The Request Approval step becomes an agent-invoked human-in-the-loop Flow that can wait far longer and reach more channels. The logic carries over; the shape gets more capable. - **What gets better in the process?** Any automation that fought the per-task budget stops multiplying cost by step count, because the whole run bills as one execution. Any approval that was stuck as a self-approve-only step, or that could not live inside a loop, becomes a native agent-invoked pause that reaches the right person on their channel. - **Who runs the migration?** The Midnight Flow consulting team runs migrations as part of FlowRunner Enterprise onboarding. Other tiers self-migrate with the platform’s import tools. - **How long does it take?** A typical mid-market deployment with 20 to 50 Zaps migrates over 4 to 8 weeks with consulting support, or longer self-served, depending on how many Zaps relied on Zapier-specific behaviors. If you would like a free migration consultation, reach out via the [Contact page](https://flowrunner.ai/contact) and we will scope the engagement before any commitment. ### Frequently asked questions #### Is FlowRunner cheaper than Zapier? It depends on volume, and the billing unit is where the real difference lives. Take a concrete case: a five-step flow that runs ten times a week. That is 40 runs a month, which Zapier counts as 200 tasks (five per run) and FlowRunner counts as 40 executions. Two hundred tasks is double Zapier's 100-task free plan, so that flow does not run on Zapier for free, and the cheapest paid option is Professional (from $29.99/month monthly, around $19.99 annual, as of July 2026, 750 tasks). On FlowRunner the same flow uses 40 of the Free plan's 100 executions, so it runs at $0, with every integration, AI agents, and human-in-the-loop included. Outgrow Free and Starter is $5/month for 300 executions or $15 for 3,000; Growth is $45/month for 12,000. The two bills have different shapes. Zapier meters every step of every run and an MCP tool call counts as two tasks, so the number climbs as you add steps or frequency. FlowRunner's price is flat within each plan. Push that flow to a few hundred runs a month, on the order of a thousand tasks, and Zapier's task bill passes FlowRunner's $15 Starter and keeps climbing while FlowRunner does not move. Zapier's free plan fits a one- or two-step Zap at low frequency; FlowRunner's Free plan covers whole runs, and its paid plans stay cheaper and more predictable the moment real volume or step count enters the picture. #### How does Zapier's AI agent and human-in-the-loop compare to FlowRunner? Both platforms have AI agents and both have human-in-the-loop, so the honest comparison is about architecture, not existence. Zapier offers native AI Agents (billed by activities) and a Copilot natural-language builder, plus a Human in the Loop feature called Request Approval that pauses a Zap for a reviewer, who can approve, decline, or edit the submitted data. Request Approval is a linear step: it is available on Professional and higher, a reviewer on the Professional plan can only send approvals to themselves, and the action cannot sit inside a Looping step. It is also single-shot, the reviewer answers by logging into Zapier (email and Slack only notify), and when several reviewers are configured the first to respond completes the review, so there is no dual-control or quorum approval. FlowRunner's human-in-the-loop is architecturally different. The AI agent invokes one of your own Flows as a callable tool, the agent suspends while that flow runs, and the flow holds on an external callback until a person replies through email, Slack, WhatsApp, phone, or a Form, then resumes mid-run. The pause is decided at runtime by what the agent sees, and it is native on every plan, including the Free plan. #### Is Zapier HIPAA compliant? No. As of July 2026, Zapier does not sign a Business Associate Agreement on any plan and states that regulated healthcare data including protected health information is not supported on the platform (per Zapier's own data-privacy FAQ, reported by HIPAA Journal). Covered entities can use Zapier for administrative tasks that do not touch PHI, but not for PHI workflows. Zapier is SOC 2 Type II certified, which covers general security posture but is not the same as a signed BAA. FlowRunner offers HIPAA support with a BAA and ships audit trails and RBAC starting at the Professional tier ($299/month). #### What counts as a task in Zapier pricing? A Zapier task is each successful action step in a Zap. Triggers never count, and filters and Paths are free, but every action that runs successfully consumes one task, so a multi-step Zap burns one task per action each time it runs. An MCP tool call costs two tasks and draws from the same quota your Zaps use. This is the multiplication that makes multi-step, high-volume automations hard to budget on Zapier. FlowRunner bills per execution instead, where the whole run is a single unit no matter how many steps or tool calls it contains, which removes the per-step math. #### What is a good cheaper alternative to Zapier for AI agents with human oversight? If your automations are short, linear, cloud-only, and you value the largest connector directory and a fast first build, Zapier covers that well. If your agents need to stop and pull in a human mid-run, run many steps at high volume, or touch regulated data, Zapier's per-task metering, its Professional-tier self-approve-only human-in-the-loop, and its lack of a BAA push the total cost and risk up. FlowRunner is built agent-first for that case: per-execution billing, human-in-the-loop as a callable agent tool on every tier, waits up to a year, and HIPAA plus audit plus RBAC at mid-tier prices. #### Does FlowRunner have as many integrations as Zapier? No, not on raw directory count. Zapier advertises 9,000+ apps, far more than FlowRunner's 1,800+ verified integrations. The difference is depth. Zapier's 9,000+ apps span roughly 30,000 actions, which averages about 3.3 actions per app. FlowRunner exposes 56,000+ callable actions across 2,100+ integrations, roughly 26 actions per integration, and every action carries the structured metadata an agent needs to call it as a tool. FlowRunner's integrations are built and verified against each vendor's official API, and the catalog includes full-CRUD databases, vector stores, and deep ERPs where Zapier's per-app depth is shallow. The honest summary: Zapier lists more apps; FlowRunner goes deeper per integration and ships every action as an agent-callable tool. #### Can I migrate from Zapier to FlowRunner? Yes. Zaps map cleanly in structure: a Zap's trigger becomes a FlowRunner trigger, and each action step becomes a FlowRunner action. Zapier's Filters and Paths map to FlowRunner's Condition and Value Router blocks, and Autoreplay-style retries map to the Handle Error block plus a Repeat and Wait retry pattern. Anything that fought Zapier's task budget or its self-approve-only approval step gets re-expressed as a single billed execution and a native agent-invoked human-in-the-loop. The Midnight Flow consulting team runs migrations as part of Enterprise onboarding, and other tiers self-migrate. A typical mid-market deployment with 20 to 50 Zaps migrates over 4 to 8 weeks with consulting support. By [Mark Piller](https://flowrunner.ai/about#mark-piller), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) > _I like that aspect of the human in the loop when it's needed. That's ultimately what we need to get to._ VP Operations, CPG brand --- ## Accounting Automation Source: https://flowrunner.ai/solutions/accounting-automation End-to-end accounting automation across Acumatica, QuickBooks, NetSuite, Bill.com, Stripe, and Slack, with human-in-the-loop review on the exceptions that actually matter. ![Bold concept hero: three stacked pills labeled AP, AR, CLOSE joined by a single sage green connector line interrupted by one amber alert dot, beside large mono-caps text reading ACCOUNTING AUTOMATION.](https://flowrunner.ai/images/solutions/accounting-automation-hero.webp) Most pages a finance leader will read after searching “accounting automation” describe a single-step tool (OCR for invoices, expense management, payment processing) and call it automation. The honest read, which most of those pages will not say, is that automating one step out of seven does not eliminate the work. It just moves the stitching tax to someone else on the finance team. The pattern that actually shortens close and frees senior finance hours is end-to-end orchestration across the systems already in place: the ERP, the AP and AR tools, the bank and payments stack, and the Slack or email channel where approvals happen. The work that has to disappear is the work between those systems, not inside any one of them. ### What CFOs mean when they say accounting automation The CFO question underneath the search is not “what software exists in this category.” It is closer to “how do I handle the growing volume of bills, payments, and reconciliations without hiring proportionally.” Finance leaders describe the same upstream friction across conversations. Treat the following as a pattern from those conversations, not as measured benchmarks. - Manual reconciliation between systems that should agree but do not. Pulling totals out of an operational system, dumping to CSV, and loading into the GL. - Duplicate payment risk where a company and its distributor might both pay the same bill, and the duplicate only surfaces during reconciliation. - Approvals that sit in Slack threads and email chains until someone remembers, with no audit trail of who said yes and on what basis. - Senior finance staff entering bills when someone is out, because there is no system bridge that survives a vacation. Single-purpose tools chip at each of these individually. They do not address the seam between them, which is where the time goes. ### Where the piecemeal stack breaks down Most finance teams already run a parsing tool, an AP product like Bill.com or Ramp, an ERP like QuickBooks Online or NetSuite or Acumatica, a payment stack including Stripe, and Slack for approvals. The bills still take too long. Three structural gaps explain why: - **Data movement between systems that should agree.** Each tool is correct in isolation; the spreadsheet that reconciles them is where the actual close happens. - **Approval workflows where reviewers rubber-stamp without context.** A request for sign-off without the PO, the vendor history, and the comparable invoice is a request to guess. People guess fast. - **Exception handling that depends on someone remembering.** Billbacks that look off, freight allocations that do not balance, accruals that need a judgment call. There is no system that owns these; they live in heads. A close checklist tracks that these were done. A reconciliation workspace helps a reviewer approve them. Neither produces them. The production is manual. ### What FlowRunner does differently The orchestration layer is the category that owns the seam between finance tools. It listens for what the ERP, the AP product, and the payment stack emit. It gathers context the individual tool did not have. It pulls a named human in at the moments that need judgment. FlowRunner is built for that layer. Operationally, three things change: - **Workflows connect the tools already in place** rather than replacing them. The ERP stays the ERP. The AP product stays the AP product. The orchestration layer sits above them, coordinating. - **Human in the loop, by design.** When a payment does not match an invoice, when a billback looks off, when an approval threshold trips, the workflow pauses and asks a named reviewer in Slack with the full context attached. Resume happens after the reviewer responds. - **Audit trail per execution.** Each run produces a step-by-step record: which steps fired, which data they touched, which person approved any paused step. Designed to meet common audit requirements; not a substitute for a Trust page or certification. ### Accounting workflows finance teams orchestrate on FlowRunner These are documented patterns finance teams adapt to their own setup. Each is a workflow guide with the integrations, the trigger, and the human review points spelled out. - [Automate AP in Acumatica end-to-end](https://flowrunner.ai/workflows/automate-with-acumatica), from invoice intake through matching, Slack approval, and ERP posting. - Use Parseur and Acumatica together for [automating invoice data entry from documents into the ERP](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack), with duplicate detection on the front end. - For teams running on QuickBooks Online, handle [invoice processing from email to payment in QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) without manual entry. - Run [automated payment reconciliation across Stripe and QuickBooks](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack), with mismatches flagged for a reviewer instead of dumped to a spreadsheet. - [Route QuickBooks approvals into Slack](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) with the context needed to validate rather than rubber-stamp. If the close itself (task coordination, journal entry approvals, consolidation) is also a bottleneck, the adjacent surface is [close management software](https://flowrunner.ai/solutions/close-management-software), which describes where dedicated close platforms fit and where they do not. ### How to decide what to automate first A practical sequence that works for finance teams entering this category: - Start with the work that is repetitive, rule-based, and currently consuming senior finance time. A controller entering AP when the clerk is out is the canonical example. - Prioritize processes where an exception today becomes a cash flow or compliance problem next month. Duplicate payments, missed billbacks, unsigned approvals. - Skip low-volume work that is annoying but rarely runs. It feels worth automating; the ROI is poor. - For a scoring approach, see the longer write-up on [how to decide what’s worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating). A reasonable scope for a first workflow is one process across two or three systems, in production within a few weeks. Bring a real workflow to evaluation, not a hypothetical: an AP path, a reconciliation step, a close-week task that always runs late. The FlowRunner team maps it end to end with the human review points marked, then ships it. --- ## Close Management Software Source: https://flowrunner.ai/solutions/close-management-software FlowRunner orchestrates the reconciliation, exception handling, and approvals upstream of your close platform, across Stripe, QuickBooks, Acumatica, and Slack. ![Bold concept hero: a calendar grid with one filled and checkmarked cell beside large mono-caps text reading MONTH-END CLOSE D+0, with a single amber alert dot in one mid-grid cell.](https://flowrunner.ai/images/solutions/close-management-software-hero.webp) Finance leaders searching for close management software are almost always trying to compress one number: the days between period end and a clean trial balance. The category itself, though, packages a narrower job than the search query implies. Most close platforms coordinate the close once the data is in. They do not address the upstream reconciliation and exception handling that finance teams report as the actual reason the close takes as long as it does. This page is not a sales pitch for replacing your close platform. It is an honest description of what dedicated close management software does well, where the bottleneck actually sits for most mid-market finance teams, and how an orchestration layer above the ERP handles the part of the close that the close platform was never designed to own. ### What close management software is, and what it does not solve Close management software is a category of tools that coordinates the month-end close. The established shape of the category includes a few distinct sub-segments: - **Dedicated close platforms** (FloQast, Numeric, BlackLine) focus on close task checklists, journal entry management, reconciliation workspaces, and review workflows. - **Consolidation suites** (OneStream, Workiva, CCH Tagetik) focus on multi-entity consolidation, intercompany eliminations, and statutory reporting. - **ERP-native close modules** in NetSuite, Sage Intacct, and Acumatica handle period-end close inside the system of record. These are mature, capable platforms for the problems they were built for. The honest gap they share is the assumption underneath: that the data feeding the close is already clean. Reconciliation workspaces help reviewers approve reconciliations, but they do not produce the reconciliations. Journal entry trackers help approve entries, but they do not pull the data from the source systems the entries are based on. The manual work happens upstream, before the close platform ever sees it. ### Where finance teams actually lose time during the close Conversations with CFOs and VPs of Finance during early FlowRunner customer discovery surface the same upstream pain points. Treat these as patterns observed in those conversations rather than as measured benchmarks: - **Manual reconciliation between systems that should agree but do not.** Billing platform totals against the GL. Distributor billbacks against AR. Freight and royalty allocations done in a pivot table because no system owns them end to end. Senior finance time gets consumed pulling data from one system, dumping it to CSV, and loading it into a second database. - **Catching exceptions and duplicate payments before they become cash flow problems.** CFOs name duplicate payment risk to distributors and vendors as a specific close-time concern, particularly where both parties may issue a payment for the same item and the duplicate only surfaces in reconciliation. - **Chasing approvals across email and Slack.** Journal entry approvals, accrual sign-offs, and unusual transaction reviews sit in threads. Items fall through the cracks and surface as last-day fire drills. - **Repetitive data dumps from one system to another.** Senior finance staff spend close-week hours moving data between systems that have no native bridge. None of these are problems a close checklist solves. The close checklist tracks that the work happened. The work itself happens somewhere else. ### How FlowRunner fits with the close management software you already have FlowRunner is not a replacement for a close platform, a consolidation suite, or an ERP close module. It is an orchestration layer above those systems. The category here is orchestration as a service: a layer above the systems of record that listens for what they emit, gathers context the close platform did not have, and pulls a human in at moments that need judgment. FlowRunner is built for that layer. What that looks like operationally: - **Reconciliation runs before the close window opens.** Mismatches surface as exceptions days earlier than they would in a Friday-of-close-week pivot table. - **Exceptions route to a named reviewer in Slack with full context.** The Stripe charge, the matching QuickBooks invoice, the discrepancy, and the recommended action arrive in one message. The reviewer resolves; the flow resumes. - **Approvals capture the approver, the timestamp, and the input.** Every execution produces a step-by-step record that is designed to meet common audit requirements. - **Human in the loop, by design.** The agent does not guess on judgment calls. Where the data is ambiguous, the flow pauses and asks. This is the differentiator from auto-resolve tools that silently apply rules and leave the trail to reconstruct after the fact. The integrations layer covers the close-adjacent systems where the upstream work actually originates. The current set is published at [flowrunner.ai/integrations](https://flowrunner.ai/integrations) and includes Stripe, QuickBooks Online, Acumatica, NetSuite, Sage Intacct, Bill.com, Ramp, and Slack. ### Close workflows finance teams orchestrate with FlowRunner These are documented workflow patterns finance teams adapt to their own close process: - [Automate revenue reconciliation before the close](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) by matching every Stripe payment against open QuickBooks Online invoices and surfacing mismatches before the close window opens. - [Route financial approvals through Slack with an audit trail](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) so journal entry, accrual, and unusual-transaction approvals capture the named approver and timestamp instead of disappearing into a thread. - [Coordinate Acumatica AP with a human-in-the-loop review step](https://flowrunner.ai/workflows/automate-with-acumatica) for invoices over threshold, so exceptions get caught before they become close-week reconciliation work. - [Eliminate manual invoice entry into the ERP](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) with duplicate detection that reduces close-time exception volume. - [Process invoices from email to payment](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) so the data feeding the books is clean before the close starts rather than reconciled after. ### When a dedicated close platform is the right tool, and when orchestration is The two are complementary. The honest scoping line: - **Choose a dedicated close platform** when the bottleneck is task coordination across a large team, journal entry tracking, statutory consolidation across multiple entities, or close-process maturity assessment. FloQast, Numeric, and OneStream are strong fits for those needs. - **Choose orchestration** when the bottleneck is upstream: reconciliation between systems, exception handling, approval routing across Slack and email, and repetitive data movement that consumes senior finance time. Many teams will eventually run both. FlowRunner does not ship a close checklist UI, a journal entry tracker, or a consolidation engine, and it is not a replacement for those capabilities when they are the actual need. ### What to evaluate when choosing close management software A practical sequence for finance leaders comparing options: - **Locate the bottleneck.** Is the constraint task tracking, reconciliation, approval chasing, or consolidation? The answer points to a different tier of tool. - **Estimate the exception ratio.** What share of close-time work is exception handling versus structured task completion? High exception load suggests orchestration; low exception load suggests a close platform. - **Check the integration fit.** Does the tool integrate with the ERP, billing system, and communication stack already in place, or does it assume a migration? Tools that assume migration tend to deliver slower. - **Inspect the audit trail.** For steps involving judgment (accrual approvals, unusual transactions, exception resolutions), what does the trail capture, and who is named on each decision? - **Ask about the human-in-the-loop pattern.** Does the tool pause and ask on ambiguous data, or does it auto-resolve and leave the trail to reconstruct? For audit-defensibility, the former matters more than the marketing headlines suggest. --- ## Journal Entry Automation Source: https://flowrunner.ai/solutions/journal-entry-automation Automate recurring and calculated journal entries into Acumatica, QuickBooks, NetSuite, and Sage Intacct, with variance-flagged entries paused in Slack. ![Bold concept hero: a stack of ledger lines with four marked posted in sage green and one held by an amber pause glyph, beside large mono-caps text reading POST OR PAUSE.](https://flowrunner.ai/images/solutions/journal-entry-automation-hero.webp) Most journal entry automation marketing treats the post as the work. The post is the easy part. The work is the calculation that produces the entry: pulling freight charges out of a carrier statement, allocating them across SKUs in a pivot table, accruing for a vendor that has not invoiced yet, reclassing a Stripe payout across revenue accounts. That work sits in spreadsheets one person owns. When that person is out, senior finance staff enter bills and rebuild the pivot table from scratch. Automating the post without owning the calculation upstream gets you a faster way to push the same hand-built numbers into the GL. This page is about the other path. FlowRunner sits between source systems and the GL, builds the entry from live data on a schedule, and pauses anything that looks wrong before it lands in the ledger. ### Why journal entry automation stalls inside most finance teams The honest baseline most mid-market finance teams operate from is not “we have no JE automation.” It is “JE automation handles the simple cases and the hard ones still sit in spreadsheets.” Three patterns show up across CFO conversations: - **Recurring entries** run on a schedule but pull from static templates. When the inputs need to change, someone updates the template, which is a manual step that drifts. - **Calculated entries** (freight allocations, accruals, intercompany, reclasses) sit in pivot tables. One person owns them. The CFO has told us in conversations that freight cost allocation in particular is still built in pivot tables and uploaded to the ERP manually, which is the realistic baseline this page replaces. - **ERP-native JE tools** post on schedule but offer little control over the calculation inputs or the exception routing. You can schedule a recurring JE; you cannot easily say “but pause this one if freight is more than ten percent off last month’s average and ping the controller.” What the CFO actually wants is the inverse of what most automation tools deliver: full control over the calculation, scheduled posting, and a pause point for the entries that need a human read. ### What FlowRunner automates in the journal entry lifecycle The orchestration covers the entry types where the manual work actually lives: - **Recurring entries** triggered on a schedule, with values pulled from the system of record at run time rather than a static template. - **Calculated entries** built from live source data (Stripe payouts, AP bills, freight invoices, payroll exports) and posted into the ERP. - **Reversing entries** handled automatically in the following period, with the reversal traced to the original posting. - **Adjusting entries** surfaced for human review with the supporting data attached, then posted on approval. FlowRunner posts these entries into Acumatica, QuickBooks Online, NetSuite, and Sage Intacct through their respective APIs. The integrations layer that handles the connections, the source-system reads, and the messaging is published at [flowrunner.ai/integrations](https://flowrunner.ai/integrations). ### Human-in-the-loop review for entries that need judgment A variance threshold is the simplest version of “pause this if it looks wrong.” A freight accrual that comes in at three times last period gets held in Slack with the calculation, the source data, and the proposed posting in one message. The reviewer approves, edits, or kicks it back without leaving the message. The entry posts only after the human input is captured. The seam this page is really about is the seam between the source systems where the calculation has to be built and the GL where the entry has to land. ERPs own posting. Spreadsheets own calculation. Email and Slack own the review. Nothing owns the path between them, which is where the work piles up. An orchestration layer is the category that owns that seam: a system above the systems of record that pulls in source data, runs the calculation, asks a human when the answer is ambiguous, and writes the entry to the ledger. FlowRunner is built for that layer. ### Audit trail, not a black box Each posted journal entry has a run log capturing the trigger, the source data snapshot, the calculation that produced the number, and the reviewer action if a human was in the loop. Finance leadership can see what posted and why without filing a ticket with IT or paying a consultant to reconstruct the chain. The logic lives in the workflow editor where finance owns it, not in a script someone else has to maintain. ### Where journal entry automation connects to the rest of close Journal entries are downstream of work the close depends on. Cleaner upstream means fewer adjusting entries at the end of the month: - [Reconcile Stripe payments against QuickBooks invoices](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) before the close window opens, so revenue-side mismatches surface as exceptions days earlier and reduce the volume of adjusting entries. - [Automate Acumatica bill posting](https://flowrunner.ai/workflows/automate-with-acumatica) so AP bills feed accrual JEs from a real record rather than a spreadsheet. - [Sync Stripe revenue into QuickBooks](https://flowrunner.ai/workflows/connect-quickbooks-online-with-stripe) so revenue events post to the ledger as they happen. - [Run the invoice-to-QuickBooks workflow](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) so the AP bills that become accrual JEs are captured at intake with duplicate detection. For a wider view of how this fits with task tracking, consolidation, and dedicated close platforms, see the [close management software](https://flowrunner.ai/solutions/close-management-software) page. ### How finance teams roll this out A pragmatic sequence that has held up in practice: - **Start with one entry.** Pick a recurring or calculated entry currently owned by a spreadsheet. Freight allocation, a specific accrual, a standard reclass. One person knows it cold. - **Add a variance threshold and a Slack reviewer before turning posting on.** Run the workflow in pause-on-everything mode for a cycle so the reviewer sees what the agent would have posted and how it lines up with what they would have done manually. - **Turn posting on for the entries that match.** Keep the pause for the entries that drift. - **Expand after one full close cycle.** Add the next entry type only after the first one has run end to end through a month-end. The audit trail is what makes that expansion safe. FlowRunner does not replace your ERP’s general ledger. It feeds it cleaner, and it gives the controller a place to stand when an entry needs a second pair of eyes. --- ## Marketing Workflows Source: https://flowrunner.ai/solutions/marketing-workflows Run lead routing, scoring, attribution, and lifecycle logic across HubSpot, Salesforce, Slack, and the rest of the marketing stack without manually patching every handoff. ![Bold concept hero: a horizontal chain of five tool tiles linked by sage arrows with one amber break marker on a single connector, beside large mono-caps text reading CROSS-TOOL MARKETING HANDOFFS.](https://flowrunner.ai/images/solutions/marketing-workflows-hero.webp) Marketing workflows almost never fail inside HubSpot, Marketo, or Mailchimp. They fail at the handoffs between them. A lead lands in a form, gets scored in one tool, needs to route through another, and the rep sees a Slack ping with half the context. The native automation in each platform covers the work that happens inside that platform’s boundary; the seams between them are where marketing ops actually spends its week. This page is for the marketing ops or demand gen lead who already runs a multi-tool stack and is tired of patching cross-tool handoffs every time a campaign rolls out. It describes the workflows that span systems, where they break, and how an orchestration layer that runs above the stack handles them without becoming yet another silo. ### The workflows marketing ops actually owns Across a typical mid-market marketing stack, the cross-tool sequences look like this: - **Inbound lead intake and routing.** Form submission lands in HubSpot, needs enrichment, deserves a routing rule that respects round-robin and territory, and should result in an SDR seeing it within minutes, not the next morning. - **Lead scoring across signals.** Score lives in one tool, behavior lives in another, fit data lives in a third. The composite score that drives MQL handoff is rarely native to any single platform. - **MQL-to-SQL handoff.** The handshake between marketing and sales is the most consequential workflow in the building and almost always the most patched together. - **Campaign attribution.** Pulling weekly attribution into a format the rest of the business will trust usually means a manual export, a spreadsheet, and a Monday morning that should have been spent on something else. - **Lifecycle email logic that depends on cross-tool signals.** A trigger that depends on product usage from one tool, lead score from a second, and account stage from a third. Each tool can express the trigger it sees; none of them can express the combined rule. ### Why native automation in each tool stops short HubSpot, Marketo, and Mailchimp each automate inside their own boundary, and they do that part well. Segment and Mixpanel emit useful signals downstream but they do not orchestrate the action across the stack. Slack notifications without context become noise, get muted, and leads sit. The gap is not a missing feature in any one tool. The gap is a coordination layer above the tools. The honest read on this category, which most marketing automation guides will not say, is that the answer is not “consolidate onto fewer platforms.” Mid-market marketing teams will keep their stack mixed for the same reason they got there: each tool earned its slot by being best at one thing. The work is making them act as one system without forcing a migration. An orchestration layer is the category that owns that work; FlowRunner is one example of one, built specifically for the seams between marketing tools where the handoffs actually live. ### How FlowRunner runs cross-tool marketing workflows FlowRunner triggers on real events from the tools in the stack rather than scheduled polling. A form submission in HubSpot, a status change in Salesforce, a behavior event from a product analytics tool, an email reply landing in a shared mailbox. Conditional logic combines fields from multiple systems before deciding what to do next. Routing, scoring updates, lifecycle steps, and notifications all happen from the same flow with the same context. The judgment calls stay with a human. Ambiguous lead ownership when two reps could legitimately claim it, a bounced enrichment lookup that needs a second look, an account-level handoff to sales that should not be auto-routed without eyes on it. FlowRunner pauses on those, asks the named owner in Slack with the full context attached, and resumes when they decide. Every run produces a step-by-step audit trail so attribution and routing decisions are reviewable later, not assumed. ### Workflow patterns marketing ops can ship The reference patterns sit in the workflow library: - [Qualifying inbound leads without manual research](https://flowrunner.ai/workflows/orchestrate-google-forms-and-hubspot-and-slack) from form to enrichment to owner assignment, with the Slack notification carrying the context the rep needs to act. - [Connect HubSpot with Slack](https://flowrunner.ai/workflows/connect-hubspot-with-slack) so MQL handoff messages reach the right rep with the right fields, not a generic “new lead” ping. - [Campaign attribution delivered automatically](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack) into a Slack channel each week from Salesforce data, with no manual export step in the middle. ### Deciding what to put on FlowRunner first Start with the workflow that breaks most often or the one where a broken handoff costs the most pipeline. The decision framework in [how to know what’s worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) is a useful filter for prioritizing the first build. Keep the human in the loop on the steps that require judgment; let FlowRunner carry the routing, enrichment, notification, and audit trail around them. The fastest result usually comes from picking one cross-tool seam, closing it cleanly, and using the trust that builds to take on the next. --- ## Procurement Automation Source: https://flowrunner.ai/solutions/procurement-automation Run PO approvals, vendor onboarding, contract renewals, and matching as orchestrated flows across the ERP, contract, and AP systems procurement already uses. ![Bold concept hero: three system silhouettes with a sage green ribbon connecting them across the top and one amber chevron rising from the ribbon, beside large mono-caps text reading PO TO PAY ORCHESTRATED END TO END.](https://flowrunner.ai/images/solutions/procurement-automation-hero.webp) Procurement automation is sold as a software category and bought as a coordination problem. The vendor pitch is a single platform that absorbs the procurement function. The reality on the buyer side is that PO approvals, vendor onboarding, contract renewals, and three-way matching already happen across the ERP, a contract repository, a sourcing tool, an AP system, and a few Slack channels. The gap procurement leaders feel is not the absence of a procurement tool. It is the inconsistency of execution between the tools already in place. This page describes that coordination problem and how FlowRunner runs procurement processes across the procurement stack instead of asking the team to consolidate into a new one. It is written for procurement operators who run the process, separate from finance buyers picking a procure-to-pay platform. ### What procurement automation actually means for a procurement team Procurement automation, used precisely, is the structured execution of four recurring jobs across whichever systems the company runs: - PO approval routing with thresholds, category logic, and escalation - Vendor onboarding with documentation tracking (W-9, COI, banking details, compliance attestations) - Contract lifecycle, especially renewal surfacing before the auto-renew clause fires - Three-way matching across PO, receipt, and invoice when the data lives in different systems The value is not in digitizing forms. Forms have been digital for fifteen years. The value is in connecting the sourcing tool, the ERP, the contract repository, and the finance approvals into one flow that runs the same way every time, with one audit trail across the whole motion. ### The procurement processes most worth automating Procurement teams that run this exercise honestly almost always surface the same set of candidates. The candidates differ in volume from team to team; the list is consistent: - **PO approval routing** with consistent thresholds, category-owner review, and escalation when an approver is out. The [PO approval workflow](https://flowrunner.ai/blog/po-approval-workflow) failure modes are the recurring source of audit findings in this category. - **Vendor onboarding** with structured document collection and validation against the ERP’s existing vendor list before any record is created. - **Contract renewal surfacing** so renewals reach the contract owner weeks before auto-renew, with renewal context (spend, dispute history, scorecard) attached. - **Three-way matching** across PO, receipt, and invoice when those three records live in three different systems and reconciliation is currently a spreadsheet. - **Supplier scorecard data collection** pulled from operational events (invoice timing, dispute counts, on-time delivery) instead of compiled in an end-of-quarter pivot table. A team that automates one of these well usually finds the next two adjacent. ### Where procurement automation usually breaks The standard advice on procurement automation is to pick a procurement application. The honest version, which most posts on this topic will not say, is that picking one solves a slice of the problem and leaves the handoffs intact. The breakage pattern is consistent: - **Approvals stall** because routing rules live in one person’s head, not in a system. The audit trail loses the step where the request was nudged through informally. - **Vendor records drift** between the sourcing tool, the ERP, and the contract repository. A vendor name spelled three different ways becomes three vendor records. - **Contract renewals get tracked in a spreadsheet** that nobody opens until after the auto-renew has already fired. - **Matching exceptions get emailed around** with no link back to the PO record. Months later, an auditor asks who reconciled the freight line on PO 4421 and the answer requires inbox archaeology. The honest baseline is that most mid-market procurement teams have these processes documented somewhere. The documentation describes the rule. The execution diverges, quietly, across buyers, business units, and the time of the month. ### How FlowRunner runs procurement processes across the stack FlowRunner treats procurement automation as a coordination problem across the systems already in place. It does not become the procurement system of record. It coordinates the ones that already are. The shape of that coordination: 1. **Approval routing** runs with structured logic, including amount thresholds, category, cost center, and vendor risk. Slack or email is the human interface; the routing engine, the audit capture, and the system-of-record updates run inside the flow. 2. **Vendor record alignment** happens by triggering on changes in one system and updating the others, with a human checkpoint when fields conflict. A banking detail change in the ERP triggers a verification step before the AP system honors it. 3. **Contract renewals** surface on a defined cadence and route to the contract owner with renewal context attached: spend over the term, dispute history, scorecard data. The trigger is the date, not someone remembering to look. 4. **Three-way matching** pulls PO, receipt, and invoice data into one comparison step and flags exceptions to a named reviewer. Matched documents flow through; exceptions pause for judgment with the full extract attached. 5. **An audit trail** captures every decision, including who approved, when, and on what data. The trail is the artifact that turns the workflow into a control rather than a habit. The category that owns this work is orchestration as a service: a layer above the procurement tools that listens for what they emit, gathers context the procurement tool did not have on its own, and pulls a human in only at moments that require judgment. FlowRunner is built for that layer. It complements the procurement application or the ERP procurement module already in place rather than replacing either one. The distinction is intentional and worth keeping straight when scoping the work. ### Example procurement workflows already running on FlowRunner These are documented patterns procurement and finance teams adapt to their own stack: - [Automating vendor invoices into Acumatica with duplicate checks](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack), where matched invoices flow through and duplicates pause for a named reviewer with the suspected match attached. - [Processing vendor documents into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack), with vendor lookup and outlier checks happening before a bill is created. - [Invoice processing from email to payment](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) through QuickBooks Online, where the inbox is the trigger and the audit trail spans the parser, the ERP, and the Slack approval. - [Everything you can automate with Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) as a broader reference for the downstream finance leg of the procurement-to-pay motion. Each pattern is the AP-side mirror or the downstream leg of a procurement decision. They are useful as concrete proof of the shape an orchestrated procurement workflow takes when the system of record is the ERP and the coordination layer runs above it. ### How to decide what to automate first The most common mistake on a first procurement-automation initiative is to start with the most visible process rather than the most leveraged one. Three sequencing heuristics that hold up: - **Start with the process that has the highest volume of recurring exceptions.** Not the most visible one, the one that generates the most repeat work for the team. Recurring exceptions are where the orchestration earns its keep. - **Look for steps where the same data is rekeyed across two or more systems.** Rekeying is a tax. Each instance is small; the aggregate is what the team feels. - **Prioritize processes where a missed step has financial or compliance consequences.** Contract renewals, vendor compliance documents, and matching are the candidates with the sharpest downside if execution lapses. The framework on [how to know what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) handles this as a financial calculation rather than a gut call. For teams choosing between an end-to-end procurement application and an orchestration layer, the [FlowRunner vs Procurify scope comparison](https://flowrunner.ai/blog/flowrunner-vs-procurify-automation-in-procurement) walks the trade-off explicitly. The two questions usually need to be answered in that order. ### How this fits with the systems you already run The integrations procurement teams ask about most: - **ERP and accounting:** [NetSuite](https://flowrunner.ai/integrations/netsuite), [Acumatica](https://flowrunner.ai/integrations/acumatica), [QuickBooks Online](https://flowrunner.ai/integrations/quickbooks-online). Vendor validation and three-way match patterns run on top of the existing data model rather than against a parallel one. - **Contracts and signatures:** [DocuSign](https://flowrunner.ai/integrations/docusign) for supplier agreements, with signed PDFs attached to the vendor or PO record on completion. - **AP:** [Bill.com](https://flowrunner.ai/integrations/billcom) for accounts payable, when the procurement flow continues into the payment leg. - **Internal coordination:** [Slack](https://flowrunner.ai/integrations/slack) for approval requests, exception escalations, and renewal alerts. The human-in-the-loop step lives here for most teams. - **Document intake:** [Parseur](https://flowrunner.ai/integrations/parseur) for parsing supplier-submitted PDFs and forms when intake comes through email. The flow is built once around the team’s actual approval matrix and actual systems. It does not require ripping out the procurement application, the ERP, or the contract repository to install a new one. ### Getting started The fastest evaluation path is a demo built against the team’s real procurement stack rather than a generic walkthrough. The conversation usually covers three things: which process to automate first, where the human checkpoints have to stay for control reasons, and what the audit trail needs to capture per execution. The first build typically targets the highest-volume exception path, because that is the one that produces the cleanest near-term result, and the rest of the procurement-automation roadmap is sequenced from there. --- ## Property Management Automation Software Source: https://flowrunner.ai/solutions/property-management-automation-software FlowRunner is the orchestration layer between your property management platform and your accounting, banking, and tenant comms, with humans on exceptions. ![Bold concept hero: an apartment-block glyph with three sage green connector lines fanning up to coin, bank, and chat nodes, one amber checkpoint dot on the middle line, beside mono-caps text reading PROPERTY SYSTEMS CONNECTED.](https://flowrunner.ai/images/solutions/property-management-automation-software-hero.webp) Search for property management automation software and the results assume one thing: that you do not have a property management platform yet. Most operators running multifamily, single-family, or mixed portfolios already do. AppFolio, Buildium, or Yardi is the system of record for units, leases, tenants, and the rent ledger. The work that still eats your team’s week is not inside that platform. It is in the seams between it and everything else. This page is not a pitch to replace your property management platform. It describes where the manual work actually survives after you have one, and how an orchestration layer above your stack handles the cross-system processes the PM platform was never built to own. ### What property management automation software does not cover A property management platform automates the parts of operations that live inside its own four walls: rent posting, lease records, maintenance ticket intake, owner statements, the tenant portal. These are mature products and they do that job well. The honest limit is the boundary of the system itself. The processes that still run on manual effort are the ones that cross that boundary: - **Rent and payment reconciliation against the books.** Payments land in a processor and a bank account; the GL lives in QuickBooks or your accounting suite. Matching deposits to the rent roll and chasing the variances is manual. - **Vendor and maintenance invoice intake.** A plumber emails a PDF. Someone reads it, codes it to a property and a GL account, checks it against the work order, and keys it into AP. - **Owner and tenant communication that depends on a threshold or an event.** A delinquency crosses a line, a lease is 60 days from expiry, an owner draw needs approval. The trigger sits in one system; the notification and the decision sit somewhere else. - **Document collection at move-in and renewal.** Signed leases, insurance certificates, and W-9s for vendors get gathered across email and e-signature, then re-entered by hand. None of these are a PM platform’s job. The PM platform records that the lease exists. Pulling the renewal context together, routing it for a decision, and writing the result back across systems is work that happens between systems. ### How FlowRunner fits with the property management platform you already run FlowRunner is not a property management system. It is the layer above AppFolio, Buildium, or Yardi that connects them to accounting, banking, document, and communication tools, and runs the cross-system processes as one flow. This is the seam the search term hides. A property management platform owns the units and the ledger. Your accounting suite owns the books. Your payment processor owns the money movement. Your inbox owns the vendor invoices and the tenant threads. No single one of those systems was built to coordinate the other three, and the coordination is exactly where an operations team becomes the manual glue. That coordination work is a category of its own: orchestration as a service, a layer that listens for what each system emits, gathers the context any one of them is missing, and pulls a person in only when a payment, an exception, or an owner-facing decision needs judgment. FlowRunner is built for that layer. What that looks like in practice: - **Reconciliation runs continuously, not at month-end.** Payments get matched against the rent roll as they post, and variances surface as exceptions while they are still cheap to resolve. - **Exceptions route to a named person with full context.** The vendor invoice, the matching work order, the property code, and the discrepancy arrive in one Slack message. The reviewer decides; the flow resumes and writes back. - **Human in the loop, by design.** The agent does not guess on an ambiguous charge or an unusual owner draw. Where the call needs judgment, it pauses and asks. That is the difference from auto-post rules that apply silently and leave the trail to reconstruct later. - **Every execution leaves an audit trail.** Who approved the invoice, when, against what data. Property accounting is owner money, and the trail is what makes a workflow a control rather than a habit. ### Integrations property operators connect FlowRunner runs on top of the systems a property operation already uses. The full set is published at [flowrunner.ai/integrations](https://flowrunner.ai/integrations); the ones property teams reach for most: - **[QuickBooks Online](https://flowrunner.ai/integrations/quickbooks-online)** for the books, where rent, fees, and vendor bills reconcile against the ledger. - **[Stripe](https://flowrunner.ai/integrations/stripe)** for rent and fee collection, matched back to the rent roll as payments post. - **[Bill.com](https://flowrunner.ai/integrations/billcom)** for vendor and maintenance AP when the invoice flow continues into payment. - **[DocuSign](https://flowrunner.ai/integrations/docusign)** for leases, renewals, and vendor agreements, with signed PDFs attached to the right record on completion. - **[Parseur](https://flowrunner.ai/integrations/parseur)** for reading vendor invoices and tenant documents that arrive as email attachments. - **[Slack](https://flowrunner.ai/integrations/slack)** for the human checkpoint: approvals, exception escalations, and threshold alerts. ### How it works 1. **A trigger fires** when a payment posts, a vendor invoice hits the inbox, or a lease crosses a date threshold in your PM platform or accounting system. 2. **An agent gathers context and acts**, matching the payment to the rent roll, coding the invoice to a property and GL account, or assembling the renewal packet across systems. 3. **A human resolves the exception** in Slack when the data is ambiguous or the decision involves owner money, then the flow resumes and writes the result back. ### What you get - Cross-system reconciliation that surfaces variances as they happen instead of during a month-end scramble. - Vendor and maintenance invoices coded and routed without manual rekeying, with exceptions flagged to a named reviewer. - Threshold and event-driven communication that fires on the data, not on someone remembering to check. - An audit trail per execution that captures the approver, the timestamp, and the underlying data, suited to the owner-money standard property accounting runs to. ### Property workflows that map directly onto a portfolio These are documented FlowRunner patterns built for other finance and operations teams. The integrations and the shape transfer cleanly to property work: - [Reconcile incoming payments against the books](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) by matching every Stripe payment to its open invoice and surfacing variances to a reviewer, the same pattern rent reconciliation needs. - [Route approvals through Slack with an audit trail](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) so owner draws, refunds, and unusual charges capture the named approver instead of disappearing into a thread. - [Process invoices from inbox to payment](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack), the exact shape a maintenance-invoice flow takes from the vendor’s emailed PDF through to AP. - [Automate document intake with duplicate detection](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) for vendor invoices, so the same bill submitted twice pauses for a human rather than paying twice. ### When a PM platform is the right tool, and when orchestration is The two are complementary, and it is worth keeping the line straight. Choose a property management platform when the need is the system of record itself: unit and lease tracking, the tenant portal, maintenance ticketing, owner statements. AppFolio, Buildium, and Yardi are built for that and FlowRunner does not try to be. Choose orchestration when the bottleneck is the work between systems: reconciliation against the books, vendor invoice intake, document collection, and event-driven communication that the PM platform records but does not run end to end. For a side-by-side on where each tool’s scope begins and ends, the [FlowRunner versus AppFolio](https://flowrunner.ai/blog/flowrunner-vs-appfolio-property-management-automation) and [FlowRunner versus Buildium](https://flowrunner.ai/blog/flowrunner-vs-buildium-property-management-automation) breakdowns walk the boundary explicitly. The [property management automation guide](https://flowrunner.ai/blog/property-management-automation-guide) covers which processes are worth automating first, and the framework on [how to know what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) turns that into a calculation rather than a gut call. Before committing to a build, [what makes automations hold up in production](https://flowrunner.ai/blog/the-truth-about-building-automations) is worth reading, because a property workflow that handles owner money has to survive the messy cases, not just the clean demo. --- ## Purchase Order Automation Software Source: https://flowrunner.ai/solutions/purchase-order-automation-software Automate PO routing, vendor document handling, and three-way matching across the ERP and procurement stack you already run. Humans review exceptions. ![Bold concept hero: three document silhouettes labeled PO, GR, INV connected by a single sage tick, beside large mono-caps text reading PO MATCHED BOOKED, with one amber exception badge.](https://flowrunner.ai/images/solutions/purchase-order-automation-software-hero.webp) Purchase order automation software is two different products under one search term, and the version a procurement team needs depends almost entirely on what they already run. For teams without a system of record, it means a standalone application that owns the catalog, the requisition, the PO, and the approval ledger in one place. For teams already running NetSuite, Acumatica, SAP, or Coupa, it means something else: automation that sits around the system of record they already trust, handling the work between it and the rest of the stack. The honest read on this category, which most PO automation reviews will not say plainly, is that the structural choice (replace versus orchestrate) decides whether the rollout works far more reliably than any feature checklist. This page is written for the orchestrate side of that choice. ### What procurement teams actually mean by PO automation The two interpretations rarely get separated in marketing copy, so the buyer ends up comparing tools that solve different problems against each other: - **Standalone PO software** consolidates requisition, catalog, PO creation, and approval inside one application. It is the right answer when no system of record exists yet, or when the existing ERP’s procurement module is too thin to live in. - **Automation around the existing stack** keeps the ERP as the PO master and adds coordination across the systems that surround it: shared mailboxes that receive vendor documents, approval channels in Slack or email, the AP tool that processes invoices, and the contract repository that holds the agreements. Most mid-market procurement teams have already chosen a system of record. The work hiding inside the PO automation question is therefore rarely “which standalone PO tool to buy” and almost always “how to route approvals, handle vendor documents, perform three-way matching, and triage exceptions across the systems already in place.” That is the work this page describes. ### Where manual PO processes break down The repeating failure pattern across mid-market procurement teams is consistent enough to name: - Approval routing handled in email or chat with no consistent record of who approved what against which policy. The PO record shows Approved; the audit cannot reconstruct who signed off or on what basis. - Vendor invoices and receipts landing in a shared inbox, requiring manual entry into the ERP because nothing wires the inbox to the vendor master. - Three-way matching done by AP staff comparing PDFs to ERP records line by line, with line-item discrepancies caught (or missed) by hand. - Price-variance, partial-receipt, and missing-PO exceptions sitting in queues with no named owner and no defined escalation path. - Recurring and renewal POs surfacing late because no system watches for them between cycles. Each of these can be patched in isolation. None of them can be solved by buying another standalone tool that adds its own database to reconcile against the ERP. ### What to look for in purchase order automation software A short list of capabilities that matter once the orchestrate-versus-replace question is settled: - **Works with the systems already in place.** Connects to the ERP as the PO master rather than re-hosting the vendor record in a separate database. - **Conditional approval routing.** Routes by amount, category, vendor, and cost center, not just a linear threshold ladder. The [PO approval workflow explainer](https://flowrunner.ai/blog/po-approval-workflow) covers the design questions in detail. - **Vendor document intake.** Accepts PDF invoices and receipts from shared inboxes, structures the data, and validates it against the open PO before any record is posted. - **Three-way matching.** Compares the PO, the goods receipt, and the vendor invoice and flags variance with the context needed to resolve it. The pattern is defined consistently across ERP documentation, including [SAP’s standard three-way match reference](https://help.sap.com/docs/SAP_ERP/). - **Human-in-the-loop on exceptions.** Pauses the flow and asks a named reviewer on ambiguity rather than auto-approving or auto-rejecting. - **Per-execution audit trail.** Captures every approval and match decision against the PO record with the user, timestamp, and input attached. ### How FlowRunner approaches PO automation FlowRunner is not a standalone PO application. It is the orchestration layer between the systems that already hold pieces of the procurement process. The ERP holds the PO master. The shared inbox receives vendor documents. The contract repository holds the agreement. Slack holds the approval. FlowRunner runs above them and produces one consistent audit trail across the lot. Three reference patterns show the shape: - [Automating the full bill lifecycle in Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) coordinates AP intake, exception escalation, and posting with a human in the loop on the cases that need judgment. - [Vendor documents into Acumatica with duplicate checks](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) parses inbound PDFs, validates against the open PO, and routes price variance and duplicate-bill exceptions through Slack to a named reviewer before any record is written. - [Vendor invoice processing into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) runs the same shape against NetSuite as the system of record, with the duplicate check and exception escalation handled before the bill is created. - [Email-to-QuickBooks invoice automation](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) shows the same pattern against QuickBooks Online for teams whose vendor master lives there. The orchestration layer above procurement, AP, and the ERP is the category that owns this work. FlowRunner is built for that layer. The differentiator on PO automation is not a better form for the buyer to fill out. It is what happens between the request and the posted bill, and whether exceptions get resolved by a human with the right context attached or auto-resolved with the trail to reconstruct later. ### The integrations procurement teams actually ask about - **ERP and system of record:** [NetSuite](https://flowrunner.ai/integrations/netsuite), [Acumatica](https://flowrunner.ai/integrations/acumatica), [QuickBooks Online](https://flowrunner.ai/integrations/quickbooks-online). FlowRunner reads the open PO, vendor master, and chart of accounts; writes the bill, the posting, and the approval evidence back. - **Document capture:** [Parseur](https://flowrunner.ai/integrations/parseur) for parsing inbound PDF invoices, receipts, and statements out of a shared mailbox into structured fields. - **Approval and exception channel:** [Slack](https://flowrunner.ai/integrations/slack) for routing approvals to named reviewers and handling exceptions with the full PO context attached to the message. - **Contract repository:** DocuSign for supplier agreements and renewal references. The PO flow gets built once against the team’s real approval matrix and real systems. Replacing the ERP or the AP tool is not part of the install. ### Building the business case The honest framing on ROI: - **Time recovered** from manual approval chasing, PDF data entry, and side-by-side three-way matching is the most reliable savings. The work to remove is the part that has no judgment in it. - **Reduction in maverick spend** comes from consistent policy enforcement on approvals across categories, not from blocking spend after the fact. - **Faster cycle time** from requisition to PO to paid invoice compounds because exceptions surface days earlier, not because the happy-path is shorter. - **Audit defensibility** comes from the per-PO trail of approvals, document matches, and exception resolutions living in one reconstructible place rather than across inboxes. The honest baseline matters here: most procurement teams have some controls today. PO automation tightens them, removes the manual chase work, and produces an auditable record. It does not invent governance from nothing, and any vendor pitch that promises a percentage cycle-time reduction without measuring the current cycle should be treated as marketing rather than evidence. The framework for [how to calculate what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) handles the math as a financial question rather than a gut call. ### Getting started A realistic sequence for evaluating an orchestrate-versus-replace PO automation approach: 1. **Map the current PO lifecycle** across the systems already in place. Where does the requisition originate, who approves at which threshold, which inbox receives the invoice, which tool runs the match, where does the bill post? 2. **Identify the two or three highest-volume exception types.** Price variance over X percent. Receipt-PO mismatch on quantity. Duplicate-bill risk on the same vendor. Those are the patterns where automation produces the cleanest near-term result. 3. **Pilot one approval path or one document type end to end** before expanding. The intake-to-match path against one ERP is usually the right first scope. 4. **Book a demo against your real stack.** The conversation worth having is the one anchored on your current ERP, your real approval matrix, and a sample of recent PO exceptions, not on a generic PO automation pitch. FlowRunner is designed to support common audit requirements: every PO execution keeps a record with timestamps, approvers, matched documents, and exception resolutions attached. That trail is the artifact procurement and finance leaders typically ask for first when PO automation shows up in a controls review. --- ## Purchase Request Software Source: https://flowrunner.ai/solutions/purchase-request-software Capture requests with the data finance needs, route them on policy, and hand approved requests to your ERP without rekeying. Humans review exceptions. ![Bold concept hero: a request slip glyph with a sage arrow forking into two routing branches and one amber checkpoint diamond, beside large mono-caps text reading REQUEST ROUTED ON POLICY.](https://flowrunner.ai/images/solutions/purchase-request-software-hero.webp) A purchase order is a commitment. A purchase request is the moment before the commitment, when someone asks to spend and the organization decides whether they can. That gap is where purchase request software earns its keep, because policy is cheap to enforce on a request and expensive to unwind on a PO that has already gone to a vendor. Get the intake right and the rest of procure-to-pay inherits clean, approved, correctly coded inputs. Get it wrong and every downstream step spends time fixing what the front door let through. This page is written for the procurement operator who owns that front door, separate from the finance buyer choosing a full procure-to-pay suite. It describes what purchase request software is meant to do, where the work actually gets lost, and how an orchestration layer above your existing systems handles requisition intake without asking your team to move into another application. For the commitment side of the process, the [purchase order automation page](https://flowrunner.ai/solutions/purchase-order-automation-software) covers the PO, three-way matching, and the invoice that follows. ### What purchase request software is supposed to do Purchase request software governs intake and requisition: the structured capture of an ask before any order exists. Treat the following as the standard shape of the category across procurement tools and ERP requisition modules, not as a proprietary claim: - **Capture the request with the fields finance actually needs.** Vendor, line items, GL coding, business justification, requested delivery date. A free-text email captures none of these consistently. - **Route on policy, not by forwarding.** Purchase request workflows commonly route on amount thresholds combined with department, GL account, or category, with delegate handling when an approver is out. The routing rule is the product, not the form. - **Give the requester visibility without a ping.** The person who submitted the request should see where it sits without messaging the buyer to ask. - **Hand the approved request off cleanly.** An approved request becomes a PO or a pending bill in the system of record, ideally without anyone rekeying the line items. Here is what most purchase request software reviews will not say plainly: the form was never the hard part. Digitizing a requisition form is a solved problem and has been for years. The work that decides whether the tool actually helps is the routing logic underneath the form and the handoff on the other side of approval. ### Where the back-and-forth actually lives The honest before-state is not chaos. Most teams have a requisition process. It runs on email, a shared spreadsheet, and an approval matrix that lives partly in a document and partly in the buyer’s memory. That process works just well enough to avoid a crisis, which is exactly why it persists. The cost is cumulative, and it concentrates in three places: - **Reconstruction.** Requests arrive over email, chat, and tickets, so the buyer rebuilds the same fields (vendor, coding, justification) by hand every time before anything can move. - **Routing by memory.** When the approval matrix lives in someone’s head, escalations stall when an approver is out and out-of-policy spend slips through because no rule caught it. - **The buyer as help desk.** Requesters chase status, which pulls the buyer into answering “where is my request” instead of doing sourcing work. Approval routing, status visibility, and ERP handoff are the parts of the flow where time disappears when the process runs on email and spreadsheets. None of these is a single catastrophic failure. It is twenty small ones a month, and the buyer absorbs the aggregate. ### How FlowRunner approaches purchase requests FlowRunner is not a standalone requisition application that re-hosts your vendor list in a new database. It is an orchestration layer that runs the intake-to-handoff path across the systems you already trust. A structured intake captures the fields finance requires, with conditional logic for capex, software, or contracted spend. Routing rules then run on amount thresholds, department, GL account, or vendor attributes, and escalate to a delegate when an approver is unavailable. The request is the front door of procure-to-pay, and no requisition form owns the layer behind it: the routing decisions, the status updates, the clean write into the ERP, and the audit record that ties them together. That layer is a category of its own, orchestration as a service, a system above the systems of record that listens for an inbound request, gathers the coding and policy context the form did not carry, and pulls a human in only when judgment is required. FlowRunner is built for that layer. Status posts back to the requester in the channel they already use, so the buyer stops being the status desk. Approved requests are pushed into the ERP or AP system as a PO or pending bill, with the original request preserved as the audit trail. Out-of-policy and ambiguous requests do not get auto-approved; the flow pauses and asks a named reviewer in [Slack](https://flowrunner.ai/integrations/slack), with the full request context attached, then resumes once the human decides. The routing design questions this raises are covered in the [PO approval workflow explainer](https://flowrunner.ai/blog/po-approval-workflow), which applies directly to request-stage approvals. ### How it works 1. **Trigger:** a request arrives through an intake form, a shared mailbox, or a chat submission, and the flow structures it into the fields finance requires. 2. **Agent action:** routing rules evaluate amount, department, GL account, and vendor attributes, post status back to the requester, and on approval write a PO or pending bill into the ERP without rekeying. 3. **Human in the loop:** out-of-policy, over-threshold, or ambiguous requests pause and route to a named approver with full context, and the request record captures who approved, when, and on what version. ### The systems a request flows through Intake touches fewer systems than matching does, so the handoff is the part worth getting right: - **ERP and accounting:** [NetSuite](https://flowrunner.ai/integrations/netsuite), [Acumatica](https://flowrunner.ai/integrations/acumatica), and [QuickBooks Online](https://flowrunner.ai/integrations/quickbooks-online), where an approved request becomes a PO or pending bill. The reference patterns for [automating the bill lifecycle in Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) and [processing documents into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) show the downstream leg the approved request feeds. - **Approval and status channel:** [Slack](https://flowrunner.ai/integrations/slack), where approvers act and requesters see status without logging into another tool. - **Vendor document intake:** [Parseur](https://flowrunner.ai/integrations/parseur), so the invoice that follows an approved request can be parsed from email and posted, with a human review point on exceptions. The [email-to-payment workflow](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) and the [Acumatica intake pattern with duplicate checks](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) show that path end to end. - **Contracts:** [DocuSign](https://flowrunner.ai/integrations/docusign), so a request tied to a master agreement or renewal surfaces the existing contract rather than starting a duplicate. ### What you get - Requests captured once, in the structure finance needs, instead of reconstructed from email by the buyer. - Routing that runs on policy every time, with escalation when an approver is out, so out-of-policy spend gets caught at intake rather than at invoice. - Approved requests written into the ERP without rekeying, which removes the coding errors that otherwise surface during three-way matching. - A request record that preserves who approved what, when, and on which version, as the audit trail finance asks for first in a controls review. ### What to evaluate before you pick a tool Adoption is what makes a request tool work, so weigh these before the feature checklist: - **Routing depth.** Can rules combine amount thresholds with department, category, vendor, and delegate logic, or is it a single linear threshold ladder? - **The handoff.** Does the tool write the approved request into the ERP and AP systems you already pay for, or does it ask you to make a new app the system of record? - **The requester path.** Is there a clear way to submit and a clear view of status? A tool nobody adopts produces no governance, regardless of its routing engine. - **Exception handling.** How does it treat rejections, partial approvals, and requests that need a human conversation before they can proceed? Pausing for a named reviewer beats auto-approving and reconstructing the trail later. Request intake is the front of procure-to-pay, and its value compounds when it connects cleanly to PO creation, receiving, and matching. The broader coordination across those steps is covered on the [procurement automation page](https://flowrunner.ai/solutions/procurement-automation), and the [framework for deciding what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) handles the sequencing as a financial question rather than a gut call. The fastest way to evaluate the intake-to-handoff path is against your real approval matrix and a sample of recent requests, not a generic walkthrough. --- ## Supplier Portal Software Source: https://flowrunner.ai/solutions/supplier-portal-software Coordinate supplier onboarding, document collection, and vendor validation across the ERP, e-signature, and AP systems you already run. Humans review exceptions. ![Bold concept hero: a stack of three vendor-card silhouettes with a large APPROVED stamp on the top card, beside large mono-caps text reading VENDOR ONBOARDED ERP-READY, with one amber exception badge in the lower corner.](https://flowrunner.ai/images/solutions/supplier-portal-software-hero.webp) Most teams searching for supplier portal software do not actually want another portal. They want clean vendor data in the ERP, an approval routing that matches their real approval matrix, and a way to stop paying vendors whose COI expired three weeks ago. A portal product is one way to get there. It is not the only way, and for teams already running NetSuite, Acumatica, or QuickBooks Online, it often is not the fastest way. The honest read on this category, which most supplier-portal comparison posts will not say, is that a portal solves the intake-form problem and then creates a new silo procurement has to reconcile against the ERP, the AP system, and the contract repository. Coordination across those systems is the actual work. FlowRunner is built for that coordination layer: a system that sits above the ERP and the AP system, collects supplier documents, validates them against the records that already exist, routes approvals through the channels the team already uses, and pulls a human in when anything looks ambiguous. ### What procurement teams actually need from a supplier portal The real job behind “supplier portal” is a sequence of tasks that almost never lives in a single tool today: - Collect supplier documents (W-9, COI, banking details, certifications, signed agreements) - Validate those documents against the company’s standards and against the ERP’s existing vendor list - Route approvals through the company’s actual approval matrix (category owner, controller, sometimes legal) - Push a clean record into the ERP only after validation and approval - Keep tracking the vendor after onboarding (renewals, document expirations, banking changes) Most teams we have spoken to coordinate this through shared inboxes, spreadsheets, and Slack threads, with the final record landing in NetSuite, Acumatica, SAP, or QuickBooks Online. The portal category exists because that coordination is painful. The portal category is also incomplete, because the documents and the vendor master record end up in two different systems and procurement still has to reconcile them. This page describes a coordination approach for teams who already run an ERP and want supplier onboarding to flow through it cleanly. It is not a description of a packaged portal product. ### Where standalone portals break down Portal products tend to be strong at the front door and weak everywhere else. Common gaps: - **The portal stores documents; the ERP stores the vendor master.** Banking details, tax IDs, and remit-to addresses entered in the portal have to be re-entered in NetSuite or Acumatica, or synced through a connector that breaks quietly. - **Approval routing rarely matches the actual approval matrix.** Portals offer linear approval chains; real procurement approval is conditional (category over $50k routes to the CFO, indirect spend routes to the category owner, anything touching healthcare data routes to compliance). Teams fall back to email. - **Duplicate vendor checks need ERP access.** A portal alone cannot tell you that “Acme Industries LLC” is already in NetSuite as “Acme Industries, Inc.” with a different tax ID. The duplicate check has to query the ERP. - **Expirations get tracked in the portal but AP keeps paying.** A COI expires, the portal flags it, and Bill.com keeps cutting checks because nothing wired the two together. The pattern is consistent: the portal solves intake, then the same coordination problem appears one layer downstream. ### How FlowRunner coordinates supplier onboarding across existing systems FlowRunner treats supplier onboarding as a flow that crosses the systems already in place, not a destination application. A typical pattern: 1. **Intake.** A supplier-facing form (or shared mailbox routed through Parseur) collects documents and structured fields. The trigger is the submission. 2. **Validation.** Before any record gets created, FlowRunner checks the submitted vendor against the ERP. The pattern of [validating vendor data before it hits NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) is documented end to end; the same pattern runs against [Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and against [QuickBooks Online when AP is the destination](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack). Duplicate name, mismatched tax ID, mismatched remit-to: these get caught here. 3. **Approval routing.** Internal approvals fan out through Slack, matched to the company’s approval matrix. Signed agreements route through DocuSign. The agent is aware of both and waits for the right combination before continuing. 4. **ERP write.** Only after validation and approval does FlowRunner create or update the vendor record in the ERP, with the document trail attached as references. 5. **Exception handling.** Anything ambiguous (a probable duplicate, a tax ID that does not validate, an expired COI on day one) routes to a named procurement reviewer in Slack. The human resolves; the flow resumes from where it paused. The agent does not create a bad record and clean it up later. The orchestration layer above the ERP and the AP system is the answer here; FlowRunner is one example of one. The differentiator is not a fancier intake form. It is what happens between intake and the ERP write, and how exceptions get handled by a human rather than auto-resolved. ### Ongoing vendor management, not just onboarding The same coordination pattern handles the parts of vendor management that portal products typically punt on: - **Document expiration tracking.** COI and certification expirations trigger renewal requests on a schedule, with the AP system held back from paying out-of-compliance vendors until the document refreshes. - **Banking detail changes.** Banking updates for existing vendors are a well-known fraud vector. FlowRunner routes those changes through a verification step (callback to a known contact, second-factor sign-off) before the ERP record updates. The verification step is not optional in the flow. - **Scorecard data collection.** Invoice timing, dispute counts, on-time delivery, and price variance get pulled from the AP and ERP systems on a schedule, [checking vendor history before creating a bill](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) as part of the pattern. - **Contract renewal surfacing.** Upcoming renewals get flagged to the category owner in Slack with the vendor’s history attached, well before the auto-renew clause fires. ### How this fits with the systems you already run The integrations procurement teams ask about most: - **ERP:** [NetSuite](https://flowrunner.ai/integrations/netsuite), [Acumatica](https://flowrunner.ai/integrations/acumatica), [QuickBooks Online](https://flowrunner.ai/integrations/quickbooks-online), SAP. Vendor validation patterns are documented per ERP and run on top of the ERP’s existing data model. - **Procurement and AP:** Coupa for sourcing and contract management, Bill.com for AP, NetSuite or Acumatica for the vendor master. FlowRunner coordinates between them rather than replacing any of them. - **Document signing:** [DocuSign](https://flowrunner.ai/integrations/docusign) for supplier agreements, with signed PDFs attached to the vendor record on completion. - **Internal coordination:** [Slack](https://flowrunner.ai/integrations/slack) for approval requests, exception escalations, renewal alerts, and the human-in-the-loop step. - **Intake:** [Parseur](https://flowrunner.ai/integrations/parseur) for parsing supplier-submitted PDFs and forms when intake comes through email. The flow is built once around the team’s actual approval matrix and actual systems. It does not require ripping out an existing tool to install another one. ### When a dedicated portal product still makes sense The honest scoping line: large enterprises with thousands of active suppliers and a dedicated supplier-experience function may want a branded supplier-facing destination. A coordination approach does not replicate that. FlowRunner is the better fit when the priority is clean vendor data in the ERP, approval consistency, exception handling, and ongoing vendor management rather than a supplier-facing brand experience. Mid-market teams running NetSuite, Acumatica, SAP, or QuickBooks Online and using Slack for internal coordination usually get further faster by coordinating across those systems than by adding a portal product on top of them. FlowRunner is a coordination layer, not a packaged portal product, and it is not a replacement for Coupa or SAP Ariba. Those systems own sourcing, contract management, and category strategy. The supplier-onboarding flow sits alongside them or in front of them, depending on the team. ### What it looks like to evaluate FlowRunner for supplier onboarding A demo conversation typically covers: - **What we walk through:** a real onboarding flow built against the team’s actual ERP and approval matrix, including a probable-duplicate exception and a banking-change verification. - **What the team brings:** current onboarding steps, the real approval rules (with the conditional branches), the systems supplier data lands in today, and a sample of past onboarding events to ground the flow. - **What gets built first:** usually the intake-to-validation path, because that produces the cleanest near-term result. Renewals, banking-change verification, and scorecard collection get added once the intake path is running. FlowRunner is designed to support common audit requirements: every onboarding event keeps a document trail with timestamps, approvers, and the version of the vendor record before and after each change. That trail is the artifact procurement and finance leaders typically ask for first when supplier onboarding shows up in an audit scope. --- ## Supplier Relationship Management Software Source: https://flowrunner.ai/solutions/supplier-relationship-management-software Operationalize SRM across the ERP, contract, and AP systems procurement already runs. FlowRunner is the orchestration layer, not a second system of record. ![Diagram contrasting a tangled multi-system supplier data architecture against a single ERP coordinated by an orchestration layer.](https://flowrunner.ai/images/solutions/supplier-relationship-management-software-hero.webp) The SRM category has a structural problem that most buyer’s guides do not name: every dedicated supplier relationship management tool becomes a second system of record alongside the ERP, the contract repository, and the AP system. The team installs it to consolidate supplier visibility, and ends up reconciling supplier data across one more application than they started with. Procurement asks for one view of the supplier; the stack now has four. The honest read on this category, which most SRM comparison posts will not say, is that the gap is not “we need an SRM tool.” The gap is that supplier data already lives in the ERP, contract data already lives in DocuSign or a contract repository, and approval routing already happens in Slack and email. None of those systems talk to each other in the way procurement actually works. Adding a fifth system on top does not close that gap. Coordinating across the four already in place does. ### What procurement teams actually need from supplier relationship management software Procurement leaders consistently ask for four things from an SRM investment: a clean supplier record, controlled onboarding, visible approvals, and an audit trail across the systems where supplier data already lives. The conventional answer is to install an SRM platform that becomes the new master record. That answer triggers a second migration: data moves out of the ERP, into the SRM, and then back into the ERP through a sync connector that becomes its own maintenance surface. The alternative is to keep the ERP (NetSuite, Acumatica, SAP) as the master and use an orchestration layer to enforce SRM behavior across it. The supplier record stays where finance and AP already trust it. The SRM behaviors (onboarding controls, approval routing, document tracking, exception escalation) run as flows over the top of the systems already in place. ### Where standalone SRM platforms create friction for procurement The friction pattern is consistent across the category. Duplicate supplier records appear between the SRM tool and the ERP the moment the sync connector misses an update, and procurement spends hours each month reconciling which version is correct. Approval routing configured inside the SRM rarely matches how finance actually approves vendors in the ERP, so teams end up running two approval paths in parallel: the one the SRM enforces and the one finance actually trusts. Contract renewal data sits in a separate repository the SRM cannot read without another integration, which means renewal dates surface to procurement late, after the auto-renew clause has already fired. The honest baseline is that many mid-market teams still run supplier onboarding through shared inboxes and spreadsheets, not because dedicated SRM tools do not exist, but because adopting one means rebuilding processes the ERP already supports. The cost of the rebuild outweighs the value of the new system. The shared-inbox baseline survives because nothing has been faster. ### How FlowRunner operationalizes SRM across existing systems FlowRunner treats supplier relationship management as a coordination problem rather than a destination application. The ERP holds the vendor master. The contract repository holds the agreement. The AP system processes the bill. The messaging channel holds the approval. FlowRunner runs above them. Vendor onboarding starts from a form submission or an email that arrives in a shared mailbox. The flow validates the submission’s tax ID, checks for duplicates against the ERP’s existing vendor list, and routes approvals through the channels procurement and finance already use. Only after validation and approval does the vendor record land in the ERP. The pattern is documented end to end against NetSuite in the [validate vendor data against NetSuite before posting](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) workflow, and runs the same way for [vendor validation and duplicate checks in Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and for [automating vendor invoice processing end-to-end across mailbox, Parseur, QuickBooks Online, and Slack](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack). Supplier data integrity works the same way at the change level. When a vendor record changes (a new banking detail, an updated remit-to, a refreshed COI), FlowRunner validates the change against the ERP before posting, with a verification step on the changes that carry fraud risk. Approval routing for POs and vendor onboarding runs through structured flows with a human in the loop at defined checkpoints, and the full document trail is preserved per event. Contract and renewal visibility comes from reading contract metadata out of the existing repository (DocuSign, a contract management tool) and surfacing renewal dates to procurement through Slack or email before the renewal window closes. The orchestration layer is the category that owns this work; FlowRunner is one example of one, built for the layer between the ERP, the contract system, and the channels procurement already runs in. ### What this looks like in practice The reference patterns sit in the workflow library rather than in marketing copy. The vendor validation pattern against NetSuite shows how a vendor submission gets checked for duplicates and tax-ID mismatches before any record is written. The Acumatica pattern shows the same shape against a different ERP, with the AP-side duplicate check running before bill creation. The end-to-end mailbox-to-QuickBooks pattern shows what vendor invoice processing looks like when the orchestration layer handles intake, vendor lookup, exception escalation, and posting as one flow. Each pattern shows orchestration across the procurement stack rather than coordination through a separate SRM database. The integrations procurement teams ask about most: [NetSuite](https://flowrunner.ai/integrations/netsuite) and [Acumatica](https://flowrunner.ai/integrations/acumatica) for the vendor master, [DocuSign](https://flowrunner.ai/integrations/docusign) for supplier agreements, and [Slack](https://flowrunner.ai/integrations/slack) for the approval and exception channels. The flow is built once around the team’s actual approval matrix and actual systems. ### Where FlowRunner fits, and where a dedicated SRM platform may still make sense The honest scoping line: FlowRunner is the right fit when the ERP and the contract systems are already in place and the gap is process orchestration, approvals, and supplier data integrity across them. A dedicated SRM platform is still the better answer when the priority is supplier-facing capability rather than internal coordination. Organizations that need a branded supplier portal as a destination, structured supplier scorecards built into the tool, or formal sourcing event management with bid intake and scoring will get more out of a dedicated SRM product than out of an orchestration layer. FlowRunner can also complement a dedicated SRM rather than replace it. When the SRM holds supplier scorecards and the sourcing flow, FlowRunner handles the coordination between the SRM, the ERP, and the approval channels: the SRM still owns supplier engagement, and the orchestration layer keeps the supplier data consistent with the system of record. The two categories solve different problems. ### Next step The fastest way to evaluate this approach is a demo built against the team’s real onboarding flow and the systems already in place: the ERP, the approval channel, and one or two of the contract or AP tools the team runs today. The conversation is anchored on those systems rather than on a generic SRM pitch, because that is where the actual coordination work lives. --- - [Blog](https://flowrunner.ai/blog.md): Insights on AI automation, operations efficiency, and human-in-the-loop workflows. --- ## Automate Data Extraction: What It Actually Means for Finance Teams Source: https://flowrunner.ai/blog/automate-data-extraction Articles May 27, 2026 Updated May 28, 2026 8 min read Extraction turns unstructured documents into structured fields. The value lives in what surrounds extraction: routing, validation, exception handling, and ERP posting. ![A bald cartoon man at an inspection station between a pile of raw documents and a row of structured field cubbies, examining one document with an amber-tinted magnifying glass while others pass through untouched.](https://flowrunner.ai/images/blog/automate-data-extraction-hero.webp) Automated data extraction is the bridge step that turns unstructured documents into structured records, and on its own it does not eliminate a single hour of senior finance time. The leverage is on either side of the bridge: what the extraction triggers off, what validates the output, what catches the cases the model got wrong, and what posts the clean records to the system of record. Finance leaders evaluating extraction tools tend to read the vendor pitch and assume the bridge is the whole road. It is not. The bridge is one stage in a six-stage workflow, and the other five stages decide whether the automation is a control or a liability. This article is for the CFO or VP of Finance trying to figure out where extraction fits in the AP and reconciliation stack before committing budget to a tool. The argument is structural: extraction by itself produces a list of fields, the workflow around extraction produces a useful business outcome, and the standard accuracy conversation skips the part that actually matters. ### What automated data extraction actually means In plain finance terms, automated data extraction pulls fields out of documents and converts them into structured records. The fields look like this: vendor name, invoice number, invoice date, line items with descriptions and amounts, subtotal, tax, freight, total, payment terms, due date. The inputs are usually emails with PDF attachments, scanned paper forwarded from the field, statements pushed from vendor portals, or images shared by an AP coordinator from a phone. The outputs are structured records routed somewhere a system can act on them: an ERP (Acumatica, NetSuite, Sage, Dynamics), an accounting platform (QuickBooks Online, Xero), a database or data warehouse, or in the lighter-weight cases a spreadsheet. The shape of that destination matters more than the extraction itself. A row in a QuickBooks bill has different required fields than a row in NetSuite’s vendor bill, which differs from a row in a freight cost allocation model that lands in an Excel workbook. The destination defines the schema; the schema defines what the extraction needs to produce. Finance leaders are searching for this now because document volume is climbing without proportional headcount, and the document mix is widening. A growing distributor base means new vendor formats each quarter. A new freight provider means a new invoice layout. A new payment processor pushes statements in its own template. The historical answer was a junior staffer rekeying. That answer stops working when document volume crosses some threshold and the rekeying drifts from “manageable chore” into “the reason we can’t close the month on time.” ### Where extraction fits in the workflow Extraction is one stage in a six-stage chain. The chain is the unit of value, not the stage: - **Capture.** Documents arrive: email attachments, portal pushes, scans, drive-folder drops. - **Extract.** Field-level extraction with a confidence score per field. - **Validate.** Totals, tax math, PO match, vendor match, dollar-threshold check. - **Route.** Clean records continue; low-confidence or flagged records pause and ask. - **Post.** Structured records land in the ERP, accounting system, or warehouse. - **Reconcile.** The downstream system’s truth gets compared against the source. Extraction alone produces a list of fields. The list does not know whether the line items sum to the invoice total. It does not know whether this vendor has been paid for this invoice number already. It does not know whether the dollar amount crosses an approval threshold. The other five stages are where those questions get answered. A useful extraction workflow does not stop at the bridge; it owns the chain. The structuring step that follows extraction is where the fields get reshaped into the schema the downstream system actually wants. A vendor invoice extraction produces fields; the bill record in the ERP needs those fields mapped to specific accounts, departments, and projects. That mapping is its own discipline. See the deeper write-up on [structuring extracted data for downstream systems](https://flowrunner.ai/blog/transform-data-block-the-power-tool-for-structuring-automation-data) for how that stage works in practice. ### Why extraction without exception handling falls short The honest read on automated data extraction, which most “what is data extraction” guides won’t say, is that the accuracy conversation everyone has is the wrong conversation. Extraction vendors compete on accuracy percentages. Finance leaders evaluate vendors on accuracy percentages. The percentage tells you almost nothing about whether the workflow is safe to operate. Two reasons. First, accuracy varies by vendor format. A new distributor invoice template that the model hasn’t seen drops the accuracy on that vendor’s invoices from whatever the marketing pages claim down to something materially worse, and you do not find out until the bill is wrong. Second, OCR confidence varies field by field within a single document. The vendor name might come back at high confidence, the invoice number at moderate confidence, and one line-item amount at low confidence. A workflow that treats the document as one accept-or-reject decision wastes the per-field signal that would actually let you act on it. Without a pause-and-ask step, errors flow straight into the ERP and surface later as duplicate payments or reconciliation breaks. Finance leaders consistently raise duplicate-payment risk as top-of-mind when extraction is wired directly into payment systems without validation steps. The cost of catching a bad bill three weeks after it posted, after the payment has gone out and the vendor has cashed it, is materially higher than the cost of pausing on a low-confidence line item at the moment of extraction. The honest baseline at most mid-market finance teams: a junior staffer rekeys or spot-checks the extraction output, fixes the obvious mistakes, and posts the cleaned records. That is not no-process. That is informal review. The contrast worth drawing is between informal review and structured review, not between informal review and no review at all. Human in the loop is the design pattern that makes extraction safe to automate. Confident records post automatically. Low-confidence records get flagged with the source document attached, the field that triggered the flag highlighted, and the reviewer pinged where they already work. That is the pattern in the workflow that [routes extraction exceptions to a human reviewer in Slack](https://flowrunner.ai/workflows/connect-parseur-with-slack). The reviewer sees the field, the document, and the suggested value; one click decides whether the value passes through or gets corrected. The category-level point this article is making is the one most extraction guides skip. Extraction is a bridge step. Capture, validate, route, post, and reconcile sit on either side of it. The work between the extraction tool and the system of record, and the human judgment that has to live somewhere in that work, is not the failure case for the automation. It is the work. An orchestration layer is what owns that seam: a system above the extraction tool, the ERP, and the inbox, that listens for documents arriving, gathers the validation context the extraction tool doesn’t have, and pulls a human in at the moments that need judgment. FlowRunner is built for that layer. ### What the workflow looks like in practice Take an AP example. An invoice arrives in a shared AP inbox. The orchestration picks it up, sends it to the extraction tool (Parseur, in most of the patterns published on this site), receives back a structured field set with per-field confidence scores. The vendor gets matched against the vendor master in the ERP. The total gets validated against the line-item sum. The invoice number gets checked against open and recently-posted bills for that vendor (the duplicate-payment guard). If a PO is referenced, the line items get matched against the PO. If every check passes and every field is above confidence threshold, the bill posts to the ERP. The audit trail gets written: source document, extracted fields, validation results, the timestamp, the workflow run ID. If any check fails, the workflow pauses. The reviewer gets a Slack message with the document attached, the offending field highlighted, and a one-click approve or correct affordance. On approval, the workflow resumes and posts. On correction, the corrected value is what gets posted, and the override is logged. That shape is implemented across ERPs that finance leaders evaluating extraction actually use: - [Automated invoice extraction into Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) with the Slack exception path for distributors and direct vendors. - [Vendor document extraction into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) for the same pattern in NetSuite environments. - [Extract invoice data from email into QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) for mid-market companies on QuickBooks Online. - The capability summary in [everything you can automate with Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) for a broader view of where extraction fits among the other Acumatica automation patterns. For documents going to a workbook rather than an ERP (freight allocation tables, reconciliation worksheets, distributor billback summaries), the same pattern routes to Excel instead of an accounting system; see the [PDF data extraction to Excel](https://flowrunner.ai/workflows/pdf-data-extraction-to-excel) pattern for the workbook variant. The audit trail in this pattern is not something you build separately. It accumulates as a byproduct of the orchestration. Every document captured, every field extracted, every validation result, every override, every approval is logged with timestamp and actor. When an audit asks “who approved this bill on what basis,” the answer is in the workflow run record, not in someone’s email archive. ### What to evaluate before committing to automated extraction Five questions that hold up across every finance-leader conversation on this topic: - **Volume and variance.** How many documents per month, and how many distinct vendor formats? Volume justifies the investment; variance determines how often the extraction tool has to learn a new layout and how often format drift will trigger exceptions. A low-variance, high-volume workflow (one freight provider, ten thousand invoices a month) is the easy case. A high-variance workflow (a thousand vendors, hundred-document months) needs more reviewer capacity per document and a longer training curve. - **Where the extracted data has to land.** ERP, accounting system, data warehouse, spreadsheet, or all of the above. The destination shapes the schema. Multi-destination workflows need a structuring step in between that maps fields to each target. - **Who owns the exceptions and how they get notified.** Slack is the dominant answer for mid-market companies because that is where the reviewer already lives. Email is workable but slower. The choice matters because reviewer latency is the rate-limiting factor on the workflow’s throughput. - **Which validations matter.** Total math, PO match, vendor match, dollar threshold, duplicate check, compliance-sensitive accounts. Each is a separate validation rule. The honest scoping question for a finance leader is which subset is actually material to your business, not which subset the vendor’s marketing page brags about. - **What gets logged and where the log lives.** The audit trail is the deliverable. If the extraction tool’s logs do not include the validation context, the routing decisions, and the human approvals, the audit story is incomplete. The orchestration layer is where the complete log lives. The honest scoping question that closes most evaluations: which document type, posted into which system, with which exception path. Pick one. Prove the audit trail in a real month-end. Expand to the next document type from there. ### From extraction to a finance workflow that scales The end state worth working toward is not a higher accuracy number on the extraction vendor’s dashboard. It is fewer senior finance hours on rekeying and faster catch of mismatches before they cost money. Extraction is the starting point. Routing, validation, exception handling, and ERP posting are what turn the starting point into a finance workflow that scales. The CFOs who succeed with automated data extraction are not the ones who picked the highest-accuracy tool. They are the ones who defined the validation rules, owned the exception list, and made sure the audit trail wrote itself as a byproduct of the workflow. The accuracy number is a feature of the bridge. The exception list is the asset. ### Quick answers **What does it mean to automate data extraction in a finance context?** Pulling fields like vendor name, invoice number, line items, totals, and dates out of documents (PDFs, emails, scans) and converting them into structured records a downstream system can act on. The records land in an ERP, accounting system, database, or spreadsheet. **Does automated extraction eliminate the need for a human reviewer?** No. Vendor formats drift, OCR confidence varies, and line items do not always sum cleanly. The viable pattern is confident records post automatically and low-confidence records get flagged to a reviewer with the source document attached. The reviewer is the control that makes extraction safe to scale. **What gets extracted today, and where does the data land?** AP invoices, vendor statements, distributor billbacks, freight invoices, and bank confirmations are the most common starting points. Destinations are typically QuickBooks, Acumatica, NetSuite, or a data warehouse. The destination shapes the schema; the schema shapes everything that comes before it. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Automate Invoice Processing: The Step Vendors Skip Source: https://flowrunner.ai/blog/automate-invoice-processing Articles May 28, 2026 Updated June 19, 2026 9 min read Most invoice automation sells capture accuracy. The step that protects cash is validation: catching duplicates, billbacks, and missing POs after capture. ![An editorial-cartoon inspection checkpoint: a conveyor belt of invoices passing through a gate labeled validation, most flowing to a green POSTED tray while one is lifted into a separate amber REVIEW tray.](https://flowrunner.ai/images/blog/automate-invoice-processing-hero.webp) The word “automate” in “automate invoice processing” hides a switch, and most vendors sell you the wrong side of it. The side they demo is capture: an invoice arrives, the software reads it, and the fields land in your system without anyone typing. That part is solved, commoditized, and roughly the same across every tool that sells it. The side that decides whether your books stay clean is validation, and it is the side the demo skips. This article is about that switch, and about why the number a vendor leads with (extraction accuracy) is the least decision-relevant number in the entire evaluation. If you are weighing whether to automate invoice processing, this is the distinction to hold onto: getting the data off the page is not the hard problem. Knowing whether the invoice is real, unique, and owed is the hard problem. This piece sits one level below the broader [accounts payable automation question](https://flowrunner.ai/blog/how-to-automate-accounts-payable); it is specifically about the path an invoice takes from arrival to a posted bill, and where that path quietly breaks. ### The three layers, and the one that matters Automated invoice processing is three layers stacked, not one feature. Naming them separately is the first useful thing to do, because vendor pitches blur them constantly. | Layer | What it does | Maturity | Where the risk lives | | --- | --- | --- | --- | | **Capture** | Pull the invoice out of email, PDF, or a vendor portal and extract structured fields (vendor, invoice number, date, line items, totals) | Commoditized. Many tools do this well. | Low. A wrong field is visible and correctable. | | **Validation** | Check the extracted data against the ERP, the vendor master, prior invoices, and open POs before the bill is posted | Uneven. This is where implementations fall short. | High. A missed duplicate or billback error is money out the door. | | **Posting** | Create the bill in the ERP with the right coding and route it for approval | Straightforward once validation passes | Low, conditional on validation being right first. | Read the maturity column top to bottom. Capture is a solved category problem. Posting is mechanical once the data is trusted. The entire difficulty, and the entire reason to care which tool you pick, lives in the middle row. A finance team that buys on capture accuracy and inherits a weak validation layer has automated the easy 80% and left the part that protects cash to chance. Here is what most posts on this topic will not say plainly: extraction accuracy is the wrong headline number. A tool can read an invoice with perfect fidelity and still wave through a bill that should never have been paid. The invoice is captured correctly and is a duplicate. The line items are extracted correctly and exceed what the contract allows. The total is read correctly and there is no purchase order behind it. None of those are capture failures. Every one of them is a validation failure, and validation is the layer the marketing number does not measure. ![A horizontal three-stage flow diagram](https://flowrunner.ai/images/blog/automate-invoice-processing-1.webp) ### Why validation is hard, and why it is not a parsing problem Capture happens inside one document. You hand the parser an invoice and it hands you fields. Everything it needs is on the page. Validation is the opposite. To answer “should this bill be paid,” the system has to look away from the document and reach into other systems. Is this invoice number already in the ledger against this vendor? That answer lives in the ERP. Is the claimed rate within the contract? That lives in the vendor agreement. Does a purchase order exist, and do the line items match it? That lives in the procurement record. Is this vendor even in the master file yet? Another lookup, another system. That reach is the whole reason validation is hard, and it is also the reason a better parser does not fix it. You can improve extraction accuracy from 96% to 99% and not move the duplicate-payment risk at all, because the duplicate check was never about reading the page better. It was about cross-referencing the page against four other systems at the moment of capture. Validation is a coordination problem dressed up as a document problem. This is the seam where a parsing tool, an ERP module, and an approval app each own a fragment and none owns the whole. The parser knows the fields. The ERP knows the ledger. The contract repository knows the terms. The approval tool knows who signs. The invoice has to be checked against all of them before it posts, and nothing in that list was built to coordinate the other three. A validation layer that spans systems is an orchestration problem, and an orchestration layer is the category that owns it. We call this category [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service): a platform category that coordinates, governs, and supervises multi-agent environments while keeping humans in control of the decisions that require judgment. That layer sits above the systems of record. It listens when a new invoice arrives. It gathers the ledger history, the contract terms, and the PO the parser never had. It pulls a human in only when the cross-check cannot resolve on its own, in the [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) pattern that pauses, assembles context, routes the decision, and resumes. [FlowRunner](https://flowrunner.ai) is built for that layer. It does not replace your parser or your ERP. It is the thing that makes their separate facts agree before a bill becomes a payment. ### The exception path is the actual product Once you accept that validation is the point, the design question stops being “how accurate is capture” and becomes “what happens when validation says no.” That path, the exception path, is the actual product you are buying. Everything that clears automatically is invisible by design. The bills that pause are where the finance team’s day happens. The validated finance buyers we talk to frame their own work this way without prompting. One described his AP day plainly: when someone is out, he might enter a bill into the system himself. That is the tell. When a senior finance person is the backup data-entry clerk, the routine work still carries just enough judgment per invoice to block fully unattended automation, and the automation that exists is doing capture and nothing past it. The judgment did not disappear. It got pushed onto whoever was available. A workable exception path has a specific shape, and it is not “email it back to AP.” When an invoice cannot be auto-resolved, the system routes it to a channel (Slack or Teams) with three things attached: the original document, the extracted fields, and the reason the system paused. The reviewer responds in the thread. The response is logged as part of the audit trail. The [email-to-payment workflow we publish for QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) shows that shape end to end, with the same pattern published for [vendor invoices into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) and [parsed documents into Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack). The mechanics are nearly identical across ERPs; only the field mapping changes. The difference between that and a forwarded email is the difference between a control and a hope. A forwarded email loses the context, loses the audit trail, and lands in an inbox that already has 200 other things in it. A routed exception with full context attached is a decision a human can make in seconds, with a record of what they decided and why. ![A Slack-style exception notification card](https://flowrunner.ai/images/blog/automate-invoice-processing-2.webp) #### The exceptions worth designing for Not every [automation exception](https://flowrunner.ai/concepts/automation-exceptions) is equal. The four that move cash, and therefore the four a validation layer has to handle before anything else, are concrete: - **Duplicate invoices.** Match on vendor identity, amount, invoice number, and date before the bill reaches an approver. The risk finance leaders name here is specific: when both the company and a distributor can initiate payment against the same item, double-paying is easy and the check belongs at capture, not at month-end reconciliation when the cash is already gone. - **Billback and contract mismatches.** Distributor billbacks are a running back-and-forth between what the vendor charged and what the contract allows. Compare the claimed amount against the terms before payment goes out, not after the AR team raises a credit memo three months later. - **Missing or mismatched POs.** Three-way match against invoice, purchase order, and receipt is standard AP control practice. When the records exist, the bill clears against them. When they do not, it pauses for a human to find the missing record or override with a logged reason. The automation question is whether your tool enforces that check before approval or pretends the bill cleared. - **Unknown vendors and out-of-tolerance amounts.** Vendor not in the master file. Total off the PO by more than a set tolerance. The flag fires, the bill pauses, the reviewer gets the full picture. ### Evaluating automated invoice processing software If you are shopping for the “best automated invoice processing software,” the honest framing is that the category sells four different shapes of tool, and they are strong in different layers. Sort them by where they live, not by their feature lists. - **Document parsing tools** (Parseur, Rossum, Docparser) are strong on capture and stop there by design. They get clean fields out of messy documents. What happens to those fields next is your problem to solve, which makes them an excellent capture layer and not an invoice-processing solution on their own. - **Native ERP modules** handle capture and basic validation inside the system of record. NetSuite, for example, documents [Bill Capture as part of its accounts payable features](https://www.netsuite.com/portal/products/erp/financial-management/accounts-payable.shtml); Acumatica and Sage Intacct ship comparable AP-automation capabilities (check current vendor documentation for module names, which change between releases). These are credible for capture and in-ERP validation. They tend to be rigid about routing exceptions outside the ERP, which is exactly where many exceptions need to go (to a sales rep for a billback, to a buyer for a PO, to an account manager for a contract term). - **Standalone AP platforms** (BILL, Tipalti, Ramp) bundle payment execution with processing and are the right answer when AP is the whole job. Their validation logic lives inside their product, which is a strength when the data you need is inside their product and a constraint when your check depends on data outside it (a contract in your document store, a PO in your ERP, a vendor risk flag in your CRM). - **Orchestration approaches** connect these pieces rather than replacing them. The capture tool, the ERP, the contract repository, and the approval channel all participate in one workflow where the validation logic is explicit and exceptions route with full context. This is the shape that fits when validation depends on data the AP product cannot see. There is no universally best tool, and any listicle that crowns one is selling placement. The best tool is the one whose strongest layer matches your weakest one. A team with clean POs and a tight vendor master needs capture and not much else. A team drowning in billbacks and missing POs needs validation that reaches across systems, and capture is the part they should worry about least. ![A calm, near-monochromatic finance dashboard panel titled "This month"](https://flowrunner.ai/images/blog/automate-invoice-processing-3.webp) The number on that dashboard worth watching is not “auto-cleared.” It is “held for review,” and specifically how fast those held items resolve. A high auto-clear rate with a stagnant review queue is not a win; it is a backlog wearing a green metric. ### Where to start, by document source and ERP The starting point is not a tool. It is a single invoice flow chosen for volume and predictability. Pick the vendor invoice stream with the highest monthly count and the most consistent format, automate that, and let the harder document types wait. The pattern holds regardless of the pieces. Document parsing into a structured ERP record is a viable first step, demonstrated by FlowRunner-orchestrated workflows pulling email invoices through a parser into [QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack), [NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack), or [Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack). The [Acumatica capability catalog](https://flowrunner.ai/workflows/automate-with-acumatica) shows invoice handling as one part of a broader footprint inside one ERP. Treat each of these as a pattern to adapt, not a finished product to copy: the capture tool can change, the ERP can change, the channel can change. The capture-validate-route-post structure is what stays fixed. What this is not, and the brief is explicit on it, is a case for removing finance review. Automating invoice processing does not eliminate AP staff or finance judgment; it relocates them onto the exceptions, which is where they were always most valuable. The ninety-something percent of bills that flow clean become invisible. The handful that need a human get a human, fast, with the full picture attached. ### Deciding whether it is worth it The ROI conversation for invoice automation is quieter and more honest than the vendor slides suggest. For a lean mid-market finance team, automation rarely strips out significant headcount cost. What it changes is what senior finance time is spent on, and it removes the failure mode where an exception sits in someone’s inbox for days because no one owned it. Better questions than “how much will this save”: - How many invoices per month, and what is the realistic projection over the next eighteen months? Flat volume means the ROI is mostly exception detection. Growing volume means it is absorbing growth without adding a clerk. - How often does AP currently catch a duplicate before it pays? If the answer is “at reconciliation, if at all,” the validation layer is what you are actually buying. - What is the cycle time from invoice received to bill posted, and how much of it is waiting on a human who was never told it was their turn? - How much senior finance time goes to bill entry when someone is out? That number is the cost of capture-only automation, and it is usually larger than it looks. The broader framework for that decision, applied to any process and not just invoices, is laid out in [how to know what’s worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating). So the number to evaluate a vendor on is not extraction accuracy. It is the catch rate on the exceptions that move cash, and the cycle time on resolving them once caught. A tool that reads invoices flawlessly and lets a duplicate through has automated the part that was never going to hurt you. Buy the layer that protects the money, not the layer that reads the page. ### Quick answers #### What does it mean to automate invoice processing? It means moving an invoice from arrival to a posted bill across three layers: capture (extract the fields), validation (check the data against your ERP, vendor master, prior invoices, and POs), and posting (write the bill with coding and approval). Capture is the commoditized part. Validation is where automation either protects cash or quietly stops being a control. #### Is OCR enough to automate invoice processing? No. OCR and parsing get structured data out of the document, which is the solved part of the problem. Extraction can be perfectly accurate while the invoice is still a duplicate, a billback that exceeds contract terms, or missing a matching purchase order. Those are validation failures, not capture failures, and they are where the money is lost. #### What is the hardest part of invoice automation to get right? The exception path. Deciding what happens when an invoice cannot be auto-resolved, routing it to the right person with the original document and the reason it paused attached, and logging the response as part of the audit trail. A system without a designed exception queue turns into another inbox. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Automated Bookkeeping Software: What It Actually Automates, and Where Your Time Still Goes Source: https://flowrunner.ai/blog/automated-bookkeeping-software Articles May 28, 2026 Updated June 19, 2026 9 min read Automated bookkeeping software is a stack of capabilities, not one product. A CFO's framework for evaluating it against your real chores, not feature lists. ![A sage-green sorting machine feeding a stack of invoices and statements into a rust-colored filing cabinet drawer labeled THE BOOKS, with one document kicked sideways onto an amber REVIEW tray.](https://flowrunner.ai/images/blog/automated-bookkeeping-software-hero.webp) Automated bookkeeping software is the wrong phrase for what finance leaders are actually shopping for, and the wrong phrase costs them the evaluation. “Software” implies one product you buy, install, and finish. What growing finance teams need is a set of capabilities that handle the flow of work into and out of the books, and almost none of that flow happens inside the book itself. The ledger was never the bottleneck. The bottleneck is everything that has to happen before a clean number lands in it. That distinction is the whole game when you evaluate this category. Most of the roundups that rank for “automated bookkeeping software” hand you a ranked list of products and call the work done. They are selling you a ledger, or a point tool that bolts onto one, and calling it automation. The honest read, which the roundups will not give you, is that picking the product is the easy decision. The hard decision is figuring out where senior finance time actually leaks, and matching a capability to each leak. ### What automated bookkeeping software actually automates in 2026 The category is not one thing. It spans three layers that get marketed under the same banner, and conflating them is how finance teams buy the wrong tool. - **Dedicated bookkeeping platforms** (Zeni, Digits, Pilot and similar) bundle categorization, reporting, and in some cases a human bookkeeping service into a single product. They aim to be the system you log into. - **Ledger-native automation** is the rules and AI features already inside QuickBooks, Xero, NetSuite, and Sage Intacct: bank feed rules, recurring transactions, anomaly surfacing. You already pay for these. - **Connective automation layers** sit around whatever ledger you run and move work between it and the other systems that feed it: inboxes, payment processors, ERPs, expense tools, distributor portals. Underneath the marketing, every tool in all three layers is really claiming to automate some subset of five distinct capabilities. Evaluate them separately, because no single product is equally strong at all five. ![A horizontal flow diagram of five labeled stages reading left to right "Capture", "Categorize", "Reconcile", "Route for approval", "Handle exception"](https://flowrunner.ai/images/blog/automated-bookkeeping-software-1.webp) | Capability | What it does | Where the exceptions land | | --- | --- | --- | | **Document capture** | Pulls bills, receipts, and statements out of email, PDFs, and portals into structured data | A line item the parser misreads; a vendor format it has never seen | | **Transaction categorization** | Assigns the right account and class to each transaction | A transaction that fits two categories; a new vendor with no history | | **Reconciliation** | Matches transactions across the ledger, the bank, and the processor | The payment short by a fee; the payout that does not propagate; the billback that does not match | | **Approval routing** | Sends transactions to the right person to sign off | The approver who rubber-stamps without context; the threshold nobody enforces | | **Exception handling** | Routes the cases that do not fit the rule to a human with the context to resolve them | This is the capability the others fall back on, and the one most tools treat as an afterthought | Look down the right-hand column. “Automated” in practice means the tool handles the matched, repetitive cases and leaves the exceptions for a person. That is fine. It is correct, even, because the exceptions are the cases that need judgment. The trap is buying a tool on the strength of the first four columns and discovering that the fifth, the one where your senior finance time actually goes, is a notification with no context attached. ### Evaluate against your chores, not the vendor’s feature list Start the evaluation from the wrong end on purpose. Not “what can this tool do,” but “where in my close do senior finance people still touch transactions by hand.” Those touch points are your chores, and they map cleanly onto the five capabilities above. The patterns are consistent enough across finance leaders to predict, and they show up in [conversations with finance teams scaling without proportional hiring](https://midnightflow.ai/portfolio/scale-your-cfo-practice-without-hiring/): - **Manual reconciliation between systems that should agree but do not.** The canonical version is exporting from an operations or distributor system, pivoting it in Excel, then loading it into the ledger. It is a reconciliation problem dressed up as a data-entry chore. - **Duplicate payment risk on distributor billbacks.** When a company and its distributor can both issue payment against the same charge, a double pay is easy and nobody catches it for weeks. This is an exception-handling problem, not an edge case to wave off. A tool that cannot flag a probable duplicate before it posts is not solving the chore that keeps the CFO up at night. ![A Slack-style exception notification card titled "Payment paused: possible duplicate"](https://flowrunner.ai/images/blog/automated-bookkeeping-software-2.webp) - **Freight and allocation entries done in pivot tables.** A stable rule, rebuilt by hand every month, then uploaded to the ledger. A categorization-and-journal problem where the rule almost never changes once documented. - **AP approvals that get rubber-stamped.** Approvers sign off because the bill is just paperwork they want off their desk, not because they validated it. An approval-routing problem that fails quietly: it looks like governance and functions like a chore. Write down your three worst versions of these. Then take that list to every vendor and ask three questions the feature page will not answer for you: 1. **How do exceptions surface, and to whom?** Is it a notification with a dollar amount, or a request with the purchase order, the vendor history, and the budget code attached? The difference decides whether your controller resolves the exception in thirty seconds or chases context across four tabs. 2. **Who owns the audit trail?** Can you see and export which steps ran, which data they used, and who approved each paused step? Or does that record live in a vendor’s logs, or worse, in a consultant’s black box you cannot see into? 3. **What happens when a source system changes a field?** Real finance stacks shift. If the answer is “the integration breaks silently,” you have bought a maintenance liability, not an automation. ### Why dedicated bookkeeping platforms rarely finish the job alone Dedicated platforms are good at what they are built for. The high-volume matched flow, the categorization, the clean monthly reporting: a modern bookkeeping platform handles that core well, and for a company whose finance complexity stops at the edge of one tool’s data model, a dedicated platform may be the whole answer. Most growing companies are not that company. Real bookkeeping work crosses systems the platform does not own. Bank feeds, payment processors, the ERP running alongside the books at mid-size companies, expense tools, document inboxes, distributor portals. A dedicated platform automates the matched transactions inside its own data model with confidence, and that confidence ends precisely at the boundary of what it can see. The cross-system exception, the approval that needs context from somewhere else, the reconciliation against a source the platform does not connect to: those fall back to a human, because the platform was never designed to reach across the boundary. So the honest baseline most finance teams actually live with is not one tool. It is a primary ledger, plus a couple of point tools, plus a residue of manual reconciliation stitching them together. The residue is the work. And the residue is exactly what no single bookkeeping platform was built to own, because owning it means operating in the space between systems rather than inside one. ### Where connective automation fits alongside your ledger This is the seam most category roundups skip, because it does not fit on a product-comparison grid. There is a layer of work that lives in the space between the books and everything that updates them. Document arrives in an inbox and has to become a bill in the ledger. A Stripe payout has to be matched against open invoices in a second system. A vendor document has to land in NetSuite or Acumatica without anyone retyping it. None of that work happens inside any one product. It happens between them, and the systems on either side of the gap do not know the other exists. That gap is a category, not a missing feature. We call this category [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service): a platform category that coordinates, governs, and supervises multi-agent environments while keeping humans in control of the decisions that require judgment. An orchestration layer is what owns the gap in finance specifically. It sits above the systems of record and listens for what they emit. It gathers the context none of them share on their own. It routes the [automation exception](https://flowrunner.ai/concepts/automation-exceptions) to a named person at the moment judgment is required, and it writes a structured record of who decided what across the whole flow. The ledger holds the books. The orchestration layer runs the work around the books. FlowRunner is built for that layer, which is why it sits alongside QuickBooks, NetSuite, Acumatica, and Sage Intacct rather than competing with any of them at the ledger. Concrete is better than abstract here, so the patterns finance teams run today: - Invoice email parsed into a [QuickBooks bill with duplicate detection before it posts](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack). - [Stripe payouts reconciled against open QuickBooks invoices](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack), with mismatches paused for a reviewer. - [Vendor documents pulled into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) without manual entry. ![A calm, near-monochromatic finance dashboard panel titled "This month"](https://flowrunner.ai/images/blog/automated-bookkeeping-software-3.webp) The shape is the same in every case. Automation handles the matched flow. The human reviews the flagged exceptions. The audit trail stays intact across both. That is what [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) means in finance: not an approval button bolted onto an automation, but the system knowing when to stop the line and ask a named person for help, then resuming with the answer recorded. For the deeper read on which workflows around the ledger pay off and where the human stays in the loop, see the breakdown of [QuickBooks automation](https://flowrunner.ai/blog/quickbooks-automation), and for the end-to-end picture across the finance stack, the [accounting automation](https://flowrunner.ai/solutions/accounting-automation) overview. To be clear about the boundary, because the category’s marketing blurs it constantly: an orchestration layer is not a bookkeeping platform and it does not replace your ledger. It also does not replace your controller. It changes what your controller spends the day on, from sorting transactions to resolving the exceptions that actually need a person. Anyone selling you software that eliminates the controller is misrepresenting how the work is structured. ### A short checklist before you buy Five checks, in order. Each one maps back to a chore or a capability above. 1. **Map your three biggest recurring manual tasks and confirm the tool handles them end to end, not just the happy path.** A demo that sails through the matched case tells you nothing about the exception, which is the case you are buying for. 2. **Confirm how exceptions surface and who can see them.** The resolution context should travel with the exception. And it should not be a black box owned by a consultant you have to call to understand your own books. 3. **Confirm the audit trail is visible and exportable**, not buried in vendor logs. Finance automation lives or dies on the trail from trigger to decision to approver. ![A finance audit-trail view titled "Approval evidence" showing a clean table with columns for Transaction, Approver, Timestamp, and Decision](https://flowrunner.ai/images/blog/automated-bookkeeping-software-4.webp) 4. **Confirm the tool works with the systems you already run** rather than asking you to migrate to its world. The orchestration layer should not care which ledger you use. 5. **Pilot on one painful workflow before committing to a platform decision.** Prove the exception handling and the audit trail in one real month-end, then expand. For a fuller prioritization framework, see [how to decide what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating). The teams who get automated bookkeeping right are not the ones who bought the most-featured platform. They are the ones who wrote down where their senior finance time actually leaked, then matched a capability to each leak and refused to buy on the happy-path demo. Pull up your last close and find the three places a person still had to touch a transaction by hand. That short list, not the vendor’s feature grid, is the evaluation. ### Quick answers **What does automated bookkeeping software actually automate?** Five capability areas worth evaluating separately: document capture, transaction categorization, reconciliation, approval routing, and exception handling. Most tools automate the matched cases inside their own data model and leave cross-system exceptions for a human, which is where finance time still goes. **Is automated bookkeeping software the same as QuickBooks or Xero?** No. QuickBooks, Xero, NetSuite, and Sage Intacct are the ledger, the system of record. Dedicated platforms like Zeni, Digits, and Pilot, plus connective automation layers, sit around the ledger to capture documents, reconcile across systems, and route exceptions. The ledger holds the books; the rest is the flow into and out of them. **Can automated bookkeeping software replace a bookkeeper or controller?** No, and any vendor claiming it does is misrepresenting how the category works. Automation handles the matched, repetitive cases. The exceptions, the cross-system mismatches, and the approvals that need judgment still require a person. The realistic outcome is a controller who reviews exceptions instead of sorting transactions. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Automated Compliance Reporting: The Report Is the Cheap Part Source: https://flowrunner.ai/blog/automated-compliance-reporting Articles May 28, 2026 Updated May 30, 2026 9 min read Automated compliance reporting is only as defensible as the evidence under it. Capture approver, timestamp, and decision at the moment of work, not at audit. ![A bald cartoon man with three hairs sticking up at the end of a report-printing conveyor belt, turned away from the finished stack and pointing up the line at one unstamped item highlighted in amber while correctly stamped items glow sage green.](https://flowrunner.ai/images/blog/automated-compliance-reporting-hero.webp) Automated compliance reporting is sold as a dashboard, and the dashboard is the cheapest part of the problem. The report that lists who approved what and when is a query. It is only as defensible as the evidence the query runs against, and that evidence is created somewhere the reporting tool cannot reach: inside the approval, the reconciliation, and the exception, at the moment a person made a call. Finance leaders who buy the reporting layer first are buying a better way to reassemble evidence that should never have been scattered in the first place. That is the reframe this article is built on. The report is the output. The control point is the input. Almost everything that goes wrong with compliance evidence goes wrong at the input, and no amount of polish on the output fixes an input that was never captured. ### What CFOs actually mean by automated compliance reporting Strip the category language off and the finance version of this need is narrow and concrete. It is a defensible answer to a small set of questions an auditor, a lender, or an acquirer will ask: who approved this transaction, when did they approve it, what did they see when they approved it, and what happened to the items that did not fit the rule. Across accounts payable, reconciliation, and the monthly close, that is most of the job. The finance leaders we talk to do not describe this as a reporting gap. They describe it in their own words, and the words are telling. One called the automation a consultant had built for them “a little bit of a black box,” because it produced an answer without showing the work. Several describe the recurring failure as things that “fall through the cracks,” noticed after the fact rather than caught in the moment. The evidence exists in pieces. It is just spread across email, chat, the ERP, the bank feed, and someone’s memory, and it has to be reassembled when somebody official asks. ![A finance compliance report view titled "Approval evidence: Q2" showing a clean summary table with columns for Transaction, Approver, Timestamp, and Decision](https://flowrunner.ai/images/blog/automated-compliance-reporting-1.webp) So the honest baseline is not chaos. It is not “finance teams handle this so badly that records are effectively lost.” Records exist. The problem is that they were captured as a byproduct of systems that were never designed to tell an audit narrative, so producing the narrative means stitching the pieces back together under time pressure. That stitching is the cost. It is the controller’s late nights before fieldwork and the back-and-forth on evidence requests, and it recurs every period because the evidence is reassembled rather than retained. Here is the distinction worth holding onto, because the rest follows from it. Reporting is the output layer. Control points are the input layer. A control point is any moment where a human checks, signs, validates, or decides: the bill approval, the duplicate-payment check, the reconciliation of a line that did not match, the review of a flagged exception. The report is downstream of all of them. If the control points captured their own evidence as they ran, the report is a query. If they did not, the report is a reconstruction project wearing a dashboard. ### The control points that need to capture evidence at the moment of action Walk the finance processes auditors care about and the same pattern repeats. The evidence that holds up is the evidence captured at the instant of the decision, written against the record the decision was about. The evidence that costs you is the evidence inferred later from whatever traces survived. Four control points carry most of the weight. - **Bill approvals.** The defensible record is the named approver and the timestamp captured against the bill itself, at the moment of approval, not the final posted journal entry that shows only that a number landed in the ledger. The posted entry tells you the money moved. It does not tell you who decided it should. - **Vendor validation and duplicate checks.** The risk here is the one a finance leader named precisely: that “it would be very easy to double pay on something” if both the company and a distributor paid against the same invoice. The control is the validation that runs before a bill is created. The evidence is the check itself, logged as an event, so you can show not only that the duplicate was caught but that the control was operating. - **Reconciliation steps.** The audit-relevant record is the decision history per line, both the matched lines and, more importantly, the unmatched ones. A reconciliation that retains only its final cleared state has thrown away the part the auditor actually probes: what you did with the items that did not reconcile cleanly, and why. - **Exception handling.** When an item gets flagged, the evidence is who reviewed it, what they decided, and when. This is the control point most likely to live entirely in a chat thread or a hallway conversation, and the one most likely to be the subject of a pointed question later. Notice what these have in common. Each is a moment of human judgment, not a system event a monitoring tool can observe from the outside. A duplicate check can fire automatically, but the decision to override it, or to pay anyway because the controller knows something the rule does not, is a person. That judgment is exactly the thing that needs to be captured, and exactly the thing that is hardest to reconstruct after the fact, because it never lived in a system to begin with. ### Why workflow-level capture beats bolt-on reporting tools There are two ways to produce a compliance report. You can reconstruct the evidence after the fact from system logs that were not designed for an audit narrative, or you can write the audit record as part of the operational step so the report is a query against data that already exists. Bolt-on reporting tools do the first. Workflow-level capture does the second. The reconstruction model is the one most “automated compliance reporting” tools are quietly selling. They connect to your systems, pull what those systems happened to log, and assemble it into a report. This works for evidence that lives cleanly in one system. It struggles with the evidence that matters most, because the most important evidence is the human decision that happened between systems and was never written down anywhere queryable. That gap is not a product flaw. It is structural, and it points at the real problem. A compliance report is a query. The evidence it queries is produced at the moment a person approves a bill, clears a reconciling item, or resolves a flagged exception. That moment lives in the seam between the systems of record: the bill is in the ERP, the approval happens in chat or email, the supporting document is in a drive, the payment clears in the bank feed. No single system owns the moment, so no single system captures it, and the reporting tool downstream can only report what was captured. That seam, the work between the systems and the human judgment inside it, is a category of its own. An orchestration layer is the name for the system that owns it: a layer that sits above the systems of record and captures the decision the instant it happens. It posts the approval into the channel a person already uses, stamps the named approver and timestamp against the bill at the moment they click, and routes the exception to a reviewer with full context attached. FlowRunner is built for that layer. It does not replace the reporting tool or the ERP. It makes the moment of decision a logged event instead of a thing you reconstruct in October. The worked examples are concrete. A Slack approval that records the [named approver and timestamp on every approval](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack), stamped against the QuickBooks bill at the instant the button is clicked, is a control point capturing its own evidence. The same pattern on a different ERP [captures approver identity on every bill approval](https://flowrunner.ai/workflows/automate-with-acumatica) in Acumatica and routes the exceptions for explicit review. A reconciliation that retains [decision history for every reconciliation step](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) keeps the per-item record auditors probe, not just the final cleared state. And the preventive control fires before the damage: [vendor validation and duplicate checks before a bill is created](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack), with the check logged as its own audit event. ![A horizontal workflow diagram with four nodes left to right](https://flowrunner.ai/images/blog/automated-compliance-reporting-2.webp) The human-in-the-loop part is not a workaround that pollutes the clean automation. It is the feature. Every pause, every ask, every override is itself an audit event. When the workflow stops and pulls a person in because the data landed on something the rule could not resolve, the record of who was pulled in, what they saw, and what they decided is precisely the evidence the report needs. The automation handles what is not the work. The exceptions are the work, and capturing how they were handled is what makes the report defensible. ### What to ask vendors and internal teams when evaluating automated compliance reporting The reframe turns into a short list of questions. Ask them of any vendor selling you automated compliance reporting, and ask them of your own internal build before you trust it. The answers separate tools that capture evidence from tools that reconstruct it. - **Where is the named approver captured, and at which step?** The answer you want names a specific moment in a specific workflow, against a specific record. The answer that should worry you is a description of a report that infers the approver from a posted entry. - **Can you produce the full decision history for one transaction, end to end, without rebuilding it from multiple exports?** Pick a real transaction in the demo. If producing its history means pulling three system exports and joining them by hand, that is your audit experience every period, dressed up. - **What happens to exceptions, and is the exception path logged with the same fidelity as the happy path?** Most tools instrument the happy path well and the exception path poorly. The exception path is where audit findings live. - **How are overrides and manual entries recorded?** Finance reality includes the moment one finance leader described to us: “if someone’s out, I might even enter a bill into the system.” Those off-pattern actions are real, they are legitimate, and they need the same rigor as the routine ones, because they are the ones an auditor will ask about. - **Does the platform expose the workflow logic itself, or is the compliance behavior a black box you have to trust?** This is the one the persona names directly. A control you cannot inspect is a control you cannot defend, and “trust us” is not an audit position. None of these questions are about the report. They are all about the capture. That is the point. The quality of automated compliance reporting is decided upstream of the report, at the control points, and a vendor conversation that stays on dashboard features is a conversation avoiding the part that determines whether the evidence holds up. For the deeper version of how these control activities map to a specific framework, the [SOX compliance checklist](https://flowrunner.ai/blog/sox-compliance-checklist) walks the control families auditors test and where each one fails. If your evaluation has reached the stage of comparing a governance platform against a workflow layer, the [honest read on Vanta versus an orchestration layer for SOX](https://flowrunner.ai/blog/flowrunner-vs-vanta-sox-compliance-software) lays out which problem each kind of tool actually solves, because they are not the same problem. ### A practical starting point The mistake is buying a monolithic compliance platform up front to solve a problem you have not yet localized. The better first move costs nothing but attention. Pick the one process where audit pain is highest. For most finance teams that is AP bill approval, month-end reconciliation, or refund and credit approvals. Map the control points that already exist inside it: every place a human currently checks, signs, validates, or chases. They are already there, mostly implicit, mostly uncaptured. Then make each one an explicit, logged step that records the approver, the timestamp, and the decision against the underlying record as it happens. Now apply the only test that matters. Pull one real transaction from that process and ask whether you can answer “who approved this, and when” without further investigation. If the answer is a query, the process is producing audit-ready records. If the answer is a project, you have found the gap, and you have found it on your own schedule instead of an auditor’s. Once the pattern is proven on one process, it extends to the adjacent ones, because the capability is the same: capture at the moment of action, write against the record, log the exceptions with the same fidelity as the happy path. That is a far more defensible path than buying a reporting layer and hoping the evidence underneath it holds. The report was never the hard part. The question that decides whether your automated compliance reporting is worth anything is whether the report is a query or a reconstruction, and that question is answered long before the auditor asks it, at the moment somebody approved the bill. ### Quick answers #### What is automated compliance reporting? It is producing the records auditors and regulators ask for (who approved what, when, and on what basis) without assembling them by hand. The defensible version captures that evidence inside the operational workflow as the work happens, so the report is a query against existing data rather than a reconstruction. #### Is automated compliance reporting the same as compliance automation software? Not quite. Compliance automation software is the reporting and monitoring layer on top. Automated compliance reporting is only as good as whether the underlying control points (approvals, reconciliations, exceptions) captured evidence at the moment of action. The report is the output; the capture is the input that decides whether the output holds up. #### Can you automate compliance reporting without buying a dedicated compliance platform? Yes, when the operational workflows already capture named approver, timestamp, and decision context against the underlying record. Many finance teams get further by making each existing control point a logged step than by buying a reporting layer that reconstructs evidence after the fact. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## The Client Onboarding Checklist Finance Owns, Not Customer Success Source: https://flowrunner.ai/blog/client-onboarding-checklist-for-finance-leaders Articles May 27, 2026 Updated May 28, 2026 9 min read A client onboarding checklist written for the CFO whose first invoice has to land clean. Signed engagement letter to revenue activation, with the gaps named. ![A bald cartoon traffic-cop strawman at a crossroads between a paper engagement letter and an open accounting ledger, waving one signed document through while stopping a second incomplete document mid-air with a raised hand.](https://flowrunner.ai/images/blog/client-onboarding-checklist-for-finance-leaders-hero.webp) The client onboarding checklists that rank first on Google were written for customer success teams, and a CFO reading them is reading the wrong document. Welcome emails, kickoff calls, swag boxes, “stakeholder alignment meetings”: none of it answers the question finance actually owns, which is whether the first invoice goes out clean and on time. The checklist that matters to finance is shorter, narrower, and more consequential. It runs from the moment sales says “we’re closing this one” to the moment the first payment clears, and it is graded by exactly four things: the engagement letter is countersigned, the client record matches the legal entity, the first invoice is correct, and the first payment is reconciled. Everything else is decoration. This article is that second checklist, written for the finance leader who keeps catching items the customer success checklist never mentioned. It is sequenced by the handoffs that actually determine revenue activation, and it is honest about where most onboardings break. ### Why finance owns more of client onboarding than the checklist usually admits Sales closes the deal. Customer success runs the welcome. Finance carries the consequences. The misnamed entity, the missing W-9, the wrong billing contact, the engagement letter that never got countersigned: these are not customer-experience problems. They are revenue activation problems, and they live on the finance leader’s desk regardless of who owns the checklist. The pattern is familiar to anyone in the seat. A senior finance leader who manages multiple client engagements described his ongoing anxiety in exactly this register, talking about a contract that needed a signature: “I can’t let that fall through the cracks. I got to make sure it’s signed.” That is the texture of the work. Not “did we onboard the client well?” but “is the document signed, is the record clean, will the first invoice go out, and will the first payment arrive?” The cost of getting this wrong is not a bad first impression. It is duplicate payments where both sides of a billing relationship pay the same invoice. It is unbilled revenue that sits on a spreadsheet for a month because nobody set up the customer record. It is the engagement letter that exists in someone’s email but not in the system of record, and the first invoice that goes out under the wrong legal name and gets disputed before the relationship has begun. The framing that follows treats the checklist as a sequence of finance handoffs, not a celebration of a new logo. ### Before the engagement letter goes out Most onboarding articles start with intake. Finance starts earlier, before the engagement letter is even drafted. The items below have to be settled before the contract is in motion, because rewriting a signed engagement letter to fix a billing contact or a legal entity name is expensive and slow: - **The legal entity name.** Not the brand name, not the parent company unless the parent is paying, not the DBA. The exact registered legal entity that will appear on the W-9 and on every invoice. Confirm this with the prospect’s controller or finance contact, not with the salesperson. - **The billing address and remit-to.** Where invoices go and how payments are returned. Different from the office address often enough that assuming they match is its own category of mistake. - **The tax ID (EIN or equivalent) and W-9 status.** Required for any vendor payment relationship in the United States. Capture once, store in the client folder, do not chase later. - **The billing contact (a real human with email and phone).** Distinct from the deal contact and from the executive sponsor. The billing contact is who gets the invoice, answers questions about it, and approves payment. - **Pricing structure, billing cadence, and contingent fees.** Written somewhere the billing team reads, not buried in the deal memo. If pricing depends on milestones, headcount thresholds, or success fees, the trigger conditions go in the same document, with examples. - **The named finance owner of the client record.** One person on the finance side owns this engagement from contract to first close. Naming the owner before the engagement letter goes out is the cheapest insurance against the entire downstream sequence. Five minutes of this work before the contract drafts prevents two hours of cleanup after the first invoice goes out wrong. Senior finance time spent fixing onboarding errors after the fact is the most expensive form of onboarding labor any firm has. ### Intake: capture client data once, in a structured way The most common intake pattern is also the worst: an email thread between sales, the new client, and an associate, with attachments arriving in pieces and required fields never explicitly listed. Some never arrive. Some arrive but get missed. The thread closes and the engagement begins with three fields blank. A structured intake form fixes most of this: - One form, one client, one submission. Required fields are required. The form does not submit until every field is populated. - The form writes through to the CRM and the accounting system, or at minimum produces a single record that can be moved into both without retyping. - Duplicate detection runs at submission time. The legal entity name is matched against the CRM and the accounting system. If a similar record already exists, intake pauses for a human to decide whether this is the same legal entity or a related entity that needs its own record. - Incomplete intakes route back to sales with a specific list of missing fields. Not “please complete the intake form.” Specifically: “missing EIN, missing billing contact phone, missing payment method.” That last item is where most intake processes fail. The intake form exists, the form gets submitted with gaps, the gaps get noticed by someone downstream, that person sends a generic “please complete” message, and the cycle stalls. The form-driven pattern in the [Google Forms to HubSpot to Slack intake workflow](https://flowrunner.ai/workflows/orchestrate-google-forms-and-hubspot-and-slack) shows the shape: required fields enforced at submission, duplicate detection against the existing record set, exceptions routed to a named owner with the specific fields missing called out by name. The duplicate detection step is not optional. The anxiety a finance leader carries about duplicate records is grounded in real cost. Allen Taute, a CFO who worked through this kind of exposure with distributor billing, put it this way: “I feel that it would be very easy to double pay on something like that if both we and the distributor made a payment.” Different scenario, same root cause. Two records for the same legal entity in a system that should have one record produces double payments, double invoices, and reconciliation that costs more than the relationship is worth. ### The engagement letter and the countersignature gate Most checklists treat the engagement letter as a milestone. Finance treats it as a gate. Specifically: no countersignature, no client record creation, no first invoice. The countersignature is the moment the engagement legally exists; everything before it is conditional. The failure pattern is consistent across firms. The engagement letter goes out for signature. The client signs. The countersigned copy never makes its way back into the system of record because it sits in a salesperson’s inbox, or in an envelope on someone’s desk, or in a DocuSign envelope nobody is monitoring. Sales has moved on to the next deal. The finance owner did not know the letter went out today. The accounting system was never updated. Three weeks later somebody notices there is no first invoice yet. The behavior on the sales side that produces this pattern is not malicious. It is the predictable mechanics of how a sales team treats paperwork. Allen Taute, again, on rubber-stamped approvals from the sales side: “a lot of the sales guys, it’s just paperwork. They just want to get it off their desk.” When the signature is a step in a process the signer thinks is theater, they will move on the moment the form is clicked. The downstream work that the signature is supposed to gate does not happen automatically just because the signature did. The fix is structural: - The countersignature triggers the next step automatically. The system that holds the signature emits the event; the system of record receives it; the client record is created or unlocked. - The finance owner gets a notification when the countersignature lands. Not a generic mailbox; a specific person whose job includes acting on it. - An unsigned engagement letter older than a defined threshold (often a week, sometimes shorter for time-sensitive engagements) escalates. The threshold lives in the system, not in someone’s memory. - Skipped or stuck countersignatures are visible to finance, not buried in sales. The pattern in the [DocuSign to HubSpot to QuickBooks to Slack engagement-letter workflow](https://flowrunner.ai/workflows/orchestrate-docusign-and-hubspot-and-quickbooks-online-and-slack) is the cleaned-up version: the countersignature event drives client record creation in the CRM and the accounting system, with the finance owner pinged at the moment the signature lands and an audit trail kept of the entire sequence. ### Setting up the client in the accounting system By the time the client record gets to the accounting system, every prior field should already be settled. This step is mechanical. The mistakes are still common: - **Customer name matches the legal entity from the engagement letter exactly.** Not “Acme Inc” when the letter says “Acme, Inc.” and not the DBA when the legal name is different. Trailing punctuation matters. The auditor matches strings. - **Payment terms attached at record creation.** Net 30, Net 45, milestone billing, retainer model. Whatever the engagement letter says, the customer record reflects it from day one. Updating payment terms later is a known source of incorrect invoices. - **Class, project, or department coding configured.** If the firm uses class tracking, project codes, or department dimensions to allocate revenue, the customer record is tagged at creation. Adding the tag later means rerunning revenue reports that the leadership team already saw. - **W-9, ACH authorization, and any tax-exemption documents stored in the client folder.** Not in someone’s email. Not in a shared drive nobody else can find. In the system where the next person who needs it expects it to be. - **The billing contact on the customer record is the billing contact, not the deal contact.** This sounds obvious. It is the most common quiet error in the entire sequence. QuickBooks, NetSuite, Sage Intacct, Xero, and Acumatica all support the configuration above. The systems are not the problem. The problem is the handoff from the contract document to the data that goes in the system; that handoff is where the error lives. ### The first invoice: closing the loop on revenue activation The first invoice is the test. Everything before it was setup; the first invoice is whether the setup worked. Three rules apply: - **The first invoice is scheduled at onboarding, not deferred.** A specific date is tied to a specific event in the engagement letter (signing, kickoff, end of first month, completion of a milestone). “We will bill them next month” is not a date. - **One person reviews the first invoice line by line before it goes out.** The first invoice sets the tone for every dispute that follows. An error caught before the invoice ships is a configuration fix; an error caught after is a credit memo, a relationship moment, and a credibility question. - **Exception handling is set up before the invoice goes out.** If the first payment does not arrive on the expected date, the system flags it. Not the AR clerk noticing during a regular AR review two weeks later. The system, on the date, surfaces the exception with the client name and the amount. The line-by-line review is the step most often skipped. It is also the step most likely to catch the kind of error that gets rubber-stamped through downstream approvals. A reviewer with no structured prompt looks at the invoice, recognizes nothing obviously wrong, and approves. A reviewer with a checklist tied to the engagement letter (does the legal name match, does the payment term match, are the line items priced per the contract, does the billing contact match) catches the mismatch. For ongoing invoice intake once the relationship is established, the [Mailbox to Parseur to QuickBooks to Slack invoice automation workflow](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) is the operational shape: invoices arrive, get parsed, get matched against the customer record set up during onboarding, and route to a human reviewer for anything that does not match cleanly. ### Internal communication and visibility Onboarding milestones happen across at least three teams: sales, customer success, and finance. Without explicit communication, each team sees a partial view. Sales knows the deal closed but does not know whether the first invoice went out. Finance knows the invoice went out but does not know whether the client was kicked off well. Customer success knows the kickoff happened but does not know whether payment cleared. The fix is one shared channel with structured events, not a side conversation: - Milestone events post to a shared Slack channel (or equivalent): engagement letter sent, countersigned, customer record created, first invoice sent, first payment received. - Exceptions post separately, with context. Not “engagement letter is overdue” but “Acme Inc engagement letter sent 14 days ago, no countersignature, owner: M. Thompson, last reminder: 4 days ago.” - Routine events stay quiet. The channel is for state changes that matter, not for every internal log line. A noisy channel is the channel nobody reads. The [QuickBooks to Slack notification pattern](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) is the implementation shape for the accounting-system half of this: routing real events out of the accounting system into a place finance, sales, and operations all see. This is the seam where most onboarding sequences fail in practice. The intake form exists. DocuSign exists. The accounting system exists. The Slack channel exists. None of them natively coordinate. None of them know what the other one did. The handoff between “signature landed” and “customer record created” is a manual step somebody has to remember. The handoff between “customer record created” and “first invoice scheduled” is another manual step somebody else has to remember. Each manual step is where things fall through the cracks. That coordination problem is what an orchestration layer is for. It sits above the systems of record (CRM, e-signature, accounting, communication), listens for what they emit, enforces the handoffs, pauses to ask a human at the moments that need judgment (a duplicate-record question, an ambiguous billing arrangement, an unusual first invoice), and keeps the audit trail intact across the entire sequence. FlowRunner is built for that layer. The same pattern that turns a [signed engagement letter in DocuSign into a clean first invoice in QuickBooks with Slack notifications and human review at the exceptions](https://flowrunner.ai/workflows/orchestrate-docusign-and-hubspot-and-quickbooks-online-and-slack) handles every other handoff on the checklist. The work shape is identical across onboardings. Only the labels change. ### Post-onboarding: the first 30 and 90 days Onboarding does not end at the first invoice. The first invoice is the start of the test, not the end. Two follow-up moments matter: - **At 30 days, confirm the first payment cleared.** If it did not, find out why. The reasons are usually one of three things: the payment method was set up wrong, the invoice was disputed but the dispute lived in someone’s email, or the billing contact never received the invoice in the first place. All three are recoverable in week four; in week eight they have compounded. - **At 90 days, reconcile actual billing against the engagement letter.** Scope creep, retainer overages, contingent fees that triggered without anybody flagging them, milestone work billed differently than written. If the actual revenue does not match the contracted structure, the discrepancy gets resolved now, while both sides remember what the engagement said. The 90-day review is where most firms learn that the engagement letter and the actual delivered work have drifted, and where the conversation about the next engagement letter quietly begins. Treating it as a routine AR review is the most expensive form of optimism. ### What scaling looks like for fractional CFOs and roll-up firms The checklist matters more, not less, for finance leaders who run multiple client engagements at once. The fractional CFO with a dozen clients and the accounting roll-up adding a new firm a quarter both face the same constraint: adding a client adds proportional intake work, and senior finance time is the constraining input. The pain there is not headcount cost. As Ryan Bateman of Platform Accounting Group put it about automation across an M&A roll-up: “it doesn’t feel like it would probably strip out a significant amount of costs, but it would help drive efficiency and productivity and enable us to scale.” That is the shift in framing that this kind of checklist either enables or blocks. A checklist that has to be run manually for every new client caps the practice at whatever number of clients senior finance can babysit. A checklist that runs as a workflow, with named owners, structured intake, enforced countersignature gates, and exception routing, lets the same senior finance team carry many more relationships. The bottleneck moves from “we cannot onboard another client this quarter” to “we are deciding which clients to take on.” For roll-up firms and outsourced finance practices, the onboarding checklist is the operational asset that makes growth possible. It is also the first artifact a new client sees that signals whether this firm runs on processes or on heroics. ### A working blueprint for the finance onboarding checklist Pulling the sequence and the enforcement together into something you can adapt: - **Pre-engagement-letter items**, each with the finance owner named: legal entity confirmed, billing address and remit-to captured, tax ID collected, billing contact identified, pricing structure documented, finance owner of the engagement assigned. - **Intake items**, each with required fields enforced: single structured intake form submitted, duplicate detection run against CRM and accounting, missing-field list returned to sales when incomplete, intake stored in one place finance can find. - **Engagement-letter items**, each with an enforced gate: countersignature tracked in one place, finance owner pinged when it lands, unsigned letters escalate after a defined threshold, no client record created until the countersignature is in. - **Accounting-system setup items**, each with the source of truth named: customer name matches the engagement letter exactly, payment terms set at creation, class and project coding configured, tax documents attached, billing contact set to the billing contact. - **First-invoice items**, each with a date and a reviewer: invoice scheduled at onboarding, line-by-line review against the engagement letter, exception handling configured for the first payment, named owner for the AR follow-up. - **30-day and 90-day items**, each with a calendar trigger: payment confirmation at 30 days, actual-versus-contracted reconciliation at 90 days, onboarding gaps fed back into the checklist for the next client. Note next to each step which is automatable, which requires a human review, and which is pure documentation. Most steps are a mix: the data movement automates, the judgment moments pause and ask. Frame this as a starting template that you adapt to your own accounting system, your own engagement-letter conventions, and your own team structure. The shape of the work is universal even when the labels change. The first clean invoice is not produced by the checklist. It is produced by the discipline of running the checklist, by the named owners enforcing each handoff, and by the layer that catches what slips between them. Get the engagement letter countersigned cleanly, get the customer record matched to the legal entity, send the first invoice line-by-line correct, and the rest of the relationship starts from a position of credibility instead of cleanup. ### Quick answers #### What should a client onboarding checklist for an accounting firm include? Confirm legal entity name and tax ID before the engagement letter is drafted. Track the countersignature as a hard gate. Create the client in the accounting system using the legal name from the letter. Schedule the first invoice with a specific date. Set a 30-day check-in to confirm the first payment cleared. Each step needs a named owner and an exception path. #### Who owns client onboarding, finance or customer success? Customer success owns relationship onboarding. Finance owns billing onboarding. The two checklists run in parallel and the finance one is usually the smaller, more consequential of the two. Misnamed entities, wrong billing contacts, and unsigned engagement letters do not show up in the customer success checklist but they are the items that delay or distort the first invoice. #### What is the most common failure point in client onboarding? The handoff between the signed engagement letter and the client record in the accounting system. The contract gets countersigned and lives in someone’s inbox. The accounting system never gets the update. The first invoice goes out late or never. A named owner on the countersigned-to-customer-record handoff fixes most of it. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner Now Has a Free Plan. Here Is What Changed and Why. Source: https://flowrunner.ai/blog/flowrunner-free-plan-and-starter-pricing Product September 8, 2026 6 min read FlowRunner added a permanent Free plan, a $5 Starter plan, and a 14-day Professional trial. Every plan keeps unlimited users, every integration, AI agents, and human-in-the-loop. Here is what changed, what did not, and why we did it. ![A FlowRunner Pricing Update card on a cream field: the headline A Free Plan. Paid Plans From $5., the subhead Free, Starter at $5, and a 14-day Professional trial, and a six-step price ladder rising from $0 Free through $5 Starter, $45 Growth, $299 Professional, and $999 Business to Custom Enterprise, with the caption Unlimited users on every plan, including Free.](https://flowrunner.ai/images/blog/flowrunner-free-plan-and-starter-pricing-hero.webp) FlowRunner now has a Free plan, a $5 Starter plan, and a 14-day trial of Professional for every new account. The platform itself did not change. Every plan, including Free, still has unlimited users, unlimited workflows, every integration, AI agents on your own keys, and human-in-the-loop. What changed is where the ladder starts. Here is the full picture, what stayed the same, and why we made the change. ### What changed | Plan | Price | Executions a month | Log history | | --- | --- | --- | --- | | Free | $0 | 100 | 24 hours | | Starter | $5 or $15 | 300 or 3,000 | 24 hours | | Growth | $45 to $149 | 12,000 to 60,000 | 7 days | | Professional | $299 | 75,000 | 30 days | | Business | $999 | 250,000 | 90 days | | Enterprise | Custom | Unlimited | Unlimited | Three things are new. - **A permanent Free plan.** $0, no time limit, no credit card. 100 executions a month, one running at a time, 24 hours of visible run history. One Free workspace per account. - **A Starter plan with two rungs.** $5 a month for 300 executions, or $15 for 3,000. Same platform as Free, more volume. - **A 14-day Professional trial.** Every new account starts its first workspace on Professional, with audit trails and role-based access control switched on, and no card required. When the trial ends, the workspace moves to Free and keeps running. One smaller change: you no longer need a company email address to sign up. Any real address works. Disposable domains are still blocked. ### Why we did it Honest version: FlowRunner was already the cheapest platform in its category per run, and it had the highest entry price in that category with no free option. Both of those were true at once because of the billing unit. FlowRunner counts one complete workflow run as one execution, however many steps it has. At $45 for 12,000 executions, Growth worked out to less than half a cent per run. But the first number anyone sees is the sticker, and $45 with no free plan sat above Zapier, Make, and n8n on every price filter, every “cheapest automation tool” list, and every answer a language model gives to “affordable Zapier alternative”. We were being excluded at the sort step, before the per-run math was ever read. That is a shelf-price problem, not a value problem. So we fixed the shelf. The Free plan and the $5 Starter put FlowRunner where the filters can see it. The per-run price still falls at every step up the ladder: $0.0167 on Starter at $5, $0.005 on Starter at $15, $0.00375 on Growth at $45, and $0.00248 on Growth at $149. Every upgrade buys more runs per dollar than the plan you leave. ### What did not change We gated nothing by feature. The Free plan runs the same platform as Enterprise. - **Unlimited users.** No seats, on any plan, including Free. Invite the whole team on day one. - **Unlimited workflows.** Build as many flows as you need. - **Every integration.** The whole catalog of [1,800+ verified integrations](https://flowrunner.ai/integrations) is on every plan. Nothing is behind a higher tier. - **AI agents with your own keys.** Bring your OpenAI, Anthropic, Google, or other provider keys and pay those providers directly, with no markup. - **Human-in-the-loop.** An agent can pause a run and pull in a person over email, Slack, WhatsApp, or phone, then resume with full context. That is the core of the product, and it is on the Free plan. Executions, concurrency, and log history are the only meters. If a plan feels tight, it is because of volume or history, never because a capability was held back. ### One run is one execution The billing unit is where FlowRunner pricing differs from most of the category, and it matters more now that the entry plans are small. [Zapier bills per task](https://flowrunner.ai/compare/flowrunner-vs-zapier), where each successful action step counts. [Make bills per credit](https://flowrunner.ai/compare/flowrunner-vs-make), where each module run consumes credits. Take a 10-step flow that runs 1,000 times a month. That is roughly 10,000 Zapier tasks, roughly 10,000 Make credits, and 1,000 FlowRunner executions. So the 100 executions on the Free plan are 100 whole runs, not 100 steps. A five-step flow gets 100 runs a month on FlowRunner Free and 20 runs on a 100-task free tier. The number on the plan card looks the same. What it buys does not. ### Log history is a window, not a deletion policy The Free and Starter plans show 24 hours of run history. Growth shows 7 days, Professional 30, Business 90, Enterprise everything. Two things to be clear about. First, this is run history and analytics, not audit trails. Audit trails start at Professional and are a separate record built for auditors. Second, runs older than your window are hidden, not deleted. The app tells you how many runs sit outside the window, and moving to a plan with a longer window reveals them. We chose “log history” as the wording on the pricing page for exactly that reason. Retention is a data-handling promise, and the periods after which run data is actually deleted are stated in the [Privacy Policy](https://flowrunner.ai/privacy), not on a plan card. Why 24 hours on the entry plans? Because an overnight failure stays visible and a weekend one does not, and that is the moment a team discovers it needs more history. We would rather that moment be an upgrade prompt than a support ticket. ### The 14-day trial The trial is deliberately simple: one plan, one clock, and a soft landing at the end. Your first workspace starts on Professional for 14 days: 75,000 executions, 30-day audit trails, and role-based access control. No card. If you add a payment method, you land on whichever plan you picked. If you do not, the workspace moves to the Free plan and your flows keep running within Free limits. Nothing is paused and nothing is deleted. One trial per account, and export is not available while a workspace is on trial. The trial runs on Professional rather than Growth so you can evaluate the part of FlowRunner that is not table stakes: the audit trail and the access controls, in your own workflows, before you decide whether they are worth $299. ### If you already have an account Nothing changes for paid plans, and existing workspaces are not moved to any new plan. If you want the lower entry point for a second workspace, the Free plan is available to every account. If you were waiting for a free option before trying FlowRunner, that was the last reason to wait. ### Start free The [pricing page](https://flowrunner.ai/pricing) has the full comparison table, plan by plan. Or skip it and [start on the Free plan](https://app.flowrunner.ai). Build a flow, invite the team, connect the tools you already use, and let an agent ask a person when a decision needs one. No card, no clock. ### Quick answers **Is the FlowRunner Free plan really free?** Yes. $0, no time limit, no credit card. It runs 100 executions a month, one at a time, with 24 hours of log history, and includes every integration, AI agents with your own keys, human-in-the-loop, unlimited users, and unlimited workflows. One Free workspace per account. **Do I need a company email address to sign up?** No. Any real email address works, including personal domains. Disposable and throwaway email domains are not accepted. **What happens when the 14-day trial ends?** If there is no payment method on file, the workspace moves to the Free plan and your flows keep running within Free limits. Nothing is paused or deleted. **Are runs older than my log history window deleted?** No. They are hidden. The app shows how many runs sit outside your window, and moving to a plan with a longer window reveals them. The periods after which run data is actually deleted are stated in the Privacy Policy. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Airbyte for Automating Report Generation Source: https://flowrunner.ai/blog/flowrunner-vs-airbyte-automate-report-generation Comparisons May 28, 2026 Updated September 5, 2026 11 min read Airbyte moves data and gives agents access to it. FlowRunner governs what the agent does with it and pulls a human in on the exception. The honest finance read. ![A bald cartoon man holding a P&L sheet, a robot on a ladder reading hundreds of green source-system binders to his left, and a small desk with an amber variance notification and an unchecked approval box to his right.](https://flowrunner.ai/images/blog/flowrunner-vs-airbyte-automate-report-generation-hero.webp) Ask the question the tool vendors skip: when your weekly P&L lands in front of the board and one number is wrong, who is accountable, and at what moment did they get the chance to catch it? That question decides whether [Airbyte](https://airbyte.com) and FlowRunner are even in the same evaluation. Airbyte is one of the strongest open-source data movement platforms in the market, and it has recently extended into giving AI agents access to your business data. FlowRunner governs what those agents and the people around them are allowed to do with that data, and where a human gets pulled in. Those are different jobs. The confusion is real because both products now say the word “agents,” and a finance buyer typing “automate report generation” into a search bar cannot tell from the homepages which layer they are looking at. So this comparison starts with the honest version. If the problem is consolidating data from hundreds of SaaS and database sources into a warehouse, or giving an AI assistant searchable read access to that data, Airbyte is built for exactly that and FlowRunner is not. If the problem is that the controller is still assembling the Friday P&L by hand, flagging the one variance that does not tie, and chasing a sign-off before it ships, that work sits on a different layer of the stack entirely. Knowing which problem you actually have is the whole evaluation. ### Airbyte and FlowRunner, side by side | | Airbyte | FlowRunner | | --- | --- | --- | | **Category** | [Open-source data replication platform and context layer for AI agents](https://docs.airbyte.com/) | Orchestration layer for coordinating work and AI agents across tools | | **Primary job** | Replicate data from sources into warehouses and lakes; give agents searchable access to that data | Assemble reports, route approvals, escalate exceptions, govern what agents do | | **Built for** | Data engineers feeding a warehouse; developers wiring agents to data | Finance and operations teams who own the report and the sign-off | | **Connectors** | [600+ replication connectors plus 50+ agent connectors](https://airbyte.com/connectors), with a Connector Builder for custom sources | Operational integrations to QuickBooks, NetSuite, Acumatica, Stripe, Slack, email, plus MCP-extensible custom connectors | | **The “agent” story** | [Context Store](https://airbyte.com): a “live, searchable index of your customers, deals, tickets” so agents stop making things up | Agents as first-class workflow nodes that pause and call a human as a tool when uncertain | | **What lands where** | Source rows land in Snowflake, BigQuery, Databricks, lakes; or get indexed for agent retrieval | Reports land in Slack channels, email inboxes, ERP entries, approval queues | | **Human-in-the-loop** | Not the product’s job | Callable action: pause, route to a named approver via Slack, email, WhatsApp, or phone, resume on response | | **Governance** | Sync security, access controls on the data layer | Cross-system audit trail and RBAC on the [Professional plan at $299/mo](https://flowrunner.ai/pricing); SSO on Business | | **Open source** | [Core edition is open source, “Always free”](https://airbyte.com/pricing); 21.3k GitHub stars | Community Edition self-hosted; cloud tiers from a Free plan and $5/mo | | **Pricing model** | Volume or capacity for replication; [Agent Operations](https://airbyte.com/pricing) for agents | Execution-tiered subscription, Free / $5 / $45 / $299 / $999 / Enterprise | The table answers the side-by-side question. The rest of this piece answers the one a checklist cannot: which of these you actually need for report generation, and why both teams keep landing on the same search result. ### What Airbyte does, in Airbyte’s words Airbyte’s documentation is unusually clear about what it is. It describes itself as [“an open source data replication platform and context layer for AI agents”](https://docs.airbyte.com/) that lets you “Replicate data from hundreds of sources into warehouses, lakes, and databases.” The connector breadth is the headline asset: [600+ replication connectors plus 50+ agent connectors](https://airbyte.com/connectors), with a Connector Builder for sources that are not in the catalog. The open-source core has real gravity behind it, with [21.3k stars on the airbytehq/airbyte repository](https://github.com/airbytehq/airbyte) and a community that has been building connectors for years. The newer half of the story is the agent pivot. Airbyte’s homepage now leads with [“Agents that actually know your business”](https://airbyte.com) and introduces the Context Store with a sharp line: “Without context, agents make things up. Airbyte gives every agent a live, searchable index of your customers, deals, tickets, and conversations across your tools.” One MCP connection, they say, “gives Claude, Cursor, or any MCP client access to your full business data.” That is a genuinely useful capability, and it is honestly described. An agent that can search your real data is better than one guessing. ![A simplified diagram showing many labeled source-system tiles (Salesforce, Stripe, Zendesk, NetSuite, HubSpot) on the left, an arrow flowing through a central block labeled "Airbyte: replicate + index" in sage green, splitting into two destinations on the right: a cylinder labeled "Warehouse / Lake" and a magnifying-glass icon labeled "Searchable context for agents"](https://flowrunner.ai/images/blog/flowrunner-vs-airbyte-automate-report-generation-1.webp) What Airbyte is honestly not built for is the moment after the agent has the data. The warehouse is loaded. The Context Store is indexed. The agent can read every deal and ticket you have. And somebody still has to decide what the agent is allowed to do with that, where it must stop, who it asks when a number looks wrong before that number ships to investors, and where the record of that decision lives. None of that is replication, and none of it is access. Airbyte does not claim it is. This is not a gap in Airbyte’s product. It is a different product. ### What “automate report generation” actually means for a CFO Here is what most posts comparing data tools will not say plainly: the fact that both Airbyte and FlowRunner now say “AI agents” is exactly what makes this hard to evaluate, and the distinction that resolves it is access versus accountability. Airbyte gives an agent access to your data. It does not give the agent a boss. For a finance leader, the boss, the place the work stops and a human signs, is the entire point. For a large slice of mid-market finance teams, the report-generation problem is not solved by moving data anywhere or by giving an assistant read access to it. It is solved by getting the right summary, with the right context, into the right person’s hands, at the right time, with a place to escalate when something looks wrong. The pattern is recognizable from conversations with finance leaders running on operational systems like QuickBooks, NetSuite, and Acumatica: - A weekly P&L digest needs to land in the CFO’s Slack by 9am Monday, pulled from QuickBooks, with a one-line variance comment against last week - An AR aging summary needs to reach the controller every Friday, with anything past 60 days flagged for follow-up - A monthly board pack needs operational KPIs assembled from the ERP and the payment processor, with the CFO reviewing variance commentary before it ships to investors - An exception report needs to flag any vendor invoice paid twice, or any distributor billback that does not tie to the bank, and pause for human judgment before it gets posted None of those are “queries against a warehouse,” and none are “an agent searching the Context Store.” They are operational events: a Friday clock tick, a Stripe webhook, an email in a shared inbox, an ERP transaction crossing a threshold. The report is the output of a governed workflow, not a retrieval result. The thread running through all four is the same one that runs through [the difference between AI automation and AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents): coordination and judgment are the work, not the data. A finance team that solves this with a context layer alone ends up one step short. The agent can now see the data, which is real progress. But access is not control. An agent with searchable read access to your books and no governance layer above it will still happily assemble a P&L with a variance it had no instruction to question, and ship it. The duplicate-payment risk does not need better retrieval. It needs a workflow that stops, routes to the CFO, captures the response, and writes the decision to the audit trail. ### Where FlowRunner fits: the layer that governs what happens with the data [FlowRunner](https://flowrunner.ai) is an orchestration layer. It coordinates work and AI agents across the tools a finance team already runs, and report generation is one of the patterns it most directly answers. The shape looks like this: ![A horizontal six-step flow rendered as labeled boxes connected by arrows: "Trigger fires (Friday 9am / Stripe webhook / ERP threshold)" then "Pull context (QuickBooks, NetSuite, Stripe)" then "Assemble report" then "Distribute (Slack / email / ERP)" then a diamond decision node in amber labeled "Variance over threshold?" with one path continuing and one path branching down to "Pause: route to named CFO approver"](https://flowrunner.ai/images/blog/flowrunner-vs-airbyte-automate-report-generation-2.webp) 1. A trigger fires (a scheduled time, an ERP transaction, a Stripe webhook, an inbox message) 2. The workflow pulls the right context from the right systems (QuickBooks, NetSuite, Acumatica, Stripe, parsed documents) 3. The workflow assembles the report (a digest, a variance summary, an exception list, a board-pack section) 4. The workflow distributes the report (a Slack channel, email, an ERP entry, a shared drive) 5. If something looks off, the workflow pauses and calls a human as an action, not a status notification 6. The human gets the report and the flagged item with context already attached, responds with a structured decision, and the workflow resumes with the decision recorded in the audit trail That step 5 is the difference that the access-versus-accountability framing points at. Airbyte’s context layer makes the data _readable_ to an agent. FlowRunner makes the agent’s behavior _governed_: where it pauses, who it asks, what it is allowed to ship without a human, and what gets logged. A context layer answers “can the agent see the number.” An orchestration layer answers “who is accountable when the number is wrong, and where did they get the chance to catch it.” Both matter. They are not the same layer, and one does not substitute for the other. ![A mockup of a Slack message card](https://flowrunner.ai/images/blog/flowrunner-vs-airbyte-automate-report-generation-3.webp) Concretely, this is the pattern behind [an automated Monday P&L digest delivered to a finance team in Slack](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack), where the report assembles from QuickBooks and lands in a channel at a scheduled time. It is the pattern in [reconciling Stripe payments against QuickBooks invoices with a human-in-the-loop escalation](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack), where a mismatch routes to a named reviewer instead of silently posting. And it is the pattern in [everything a finance team can automate with Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica), where ERP-level reporting, exception flagging, and notification happen inside one governed workflow. The category that owns this layer is orchestration as a service. The seam it covers is the one between data being accessible and a decision being accountable. A warehouse stores rows. A context layer makes those rows searchable to an agent. Neither is built to fire on an operational event, assemble a report from the systems of record, deliver it to a named person, pause for human judgment when an exception comes through, and record who decided what. That work is its own category, and FlowRunner is built for it. ### Where Airbyte is the better fit This is the section that, if I could not write it, would mean I had not understood Airbyte. There are clear cases where Airbyte is the right product and FlowRunner is not in the running: - **The actual problem is data centralization** from many SaaS and database sources into a warehouse or lake. Airbyte’s [600+ connector catalog](https://airbyte.com/connectors) is dramatically broader than FlowRunner’s, and replication is what it is engineered to do. - **You need an agent to have searchable read access to your full business data.** Airbyte’s Context Store and [MCP access for Claude, Cursor, or any MCP client](https://airbyte.com) are purpose-built for exactly that, and FlowRunner does not offer an equivalent data-indexing layer. - **You want open source you can run yourself for free.** Airbyte’s [Core edition is “Always free”](https://airbyte.com/pricing) with a mature community behind it. FlowRunner has a Community Edition, but not the same ecosystem depth or [21.3k-star contributor base](https://github.com/airbytehq/airbyte). - **Your data team is large enough** to operate the warehouse and the BI layer, and the report assembly happens inside that stack rather than in operational tools. Two honest limitations of FlowRunner belong here and not buried elsewhere. First, Airbyte’s connector breadth and years of production deployments are a real maturity advantage; FlowRunner is earlier-stage, with fewer publicly documented deployments and a smaller community of contributed integrations. Second, for bulk replication into a warehouse, Airbyte’s replication product is the right tool and FlowRunner is not built for that job at all. If those describe your need, buy Airbyte. It is genuinely good at what it does, and we are not the better product for data movement. We are not in that category. ### Where FlowRunner is the better fit Choose FlowRunner when: - The report comes from operational systems (QuickBooks, NetSuite, Acumatica, Stripe, inboxes) and goes to operational channels (Slack, email, an ERP, an approval queue), without needing to land in a warehouse first - The workflow needs to pause and ask a human when an exception shows up (a duplicate-payment risk, an AR anomaly, a variance outside range), not just make the data available to read - Non-developers in finance ops should configure new reports, schedules, and escalation logic without filing a data-engineering ticket - You need a cross-system audit trail and RBAC on a mid-market budget, [included on the Professional plan at $299 a month](https://flowrunner.ai/pricing) (SSO arrives on the Business plan) rather than gated behind an enterprise procurement cycle - AI agents are appearing in your stack (a forecasting agent, an AP coding agent, a variance-commentary agent), and you need a layer above them that keeps a human in control on the calls that matter For most mid-market finance teams, the two products are complementary, not competing. The same is true of [our comparison against Fivetran](https://flowrunner.ai/blog/flowrunner-vs-fivetran-automate-report-generation), which sits on the pure data-movement side of this same boundary. Airbyte lands and indexes the data. FlowRunner governs the report, the approval, and the exception, regardless of where the source data lives. ### How to decide One question, asked the right way: when the report is wrong, who is accountable, and where did they get the chance to catch it? If your answer points at a data freshness or coverage problem, your analytics team explores the data in a BI tool, and the bottleneck is replication or giving an agent access to read it, then your problem lives on the data layer. Airbyte is the platform to evaluate, and the report is a downstream artifact built on top of what Airbyte delivers. If your answer is “a named person in finance signs off, and right now they catch the bad number by manually eyeballing it on Friday afternoon,” then your problem is orchestration. The category to evaluate is orchestration as a service. The report is the output of a coordinated workflow across operational systems, with a human pulled in at the moment a judgment is required, and an audit trail recording the decision. Access to the data was never the constraint. Accountability for what gets done with it was. Most finance teams past a certain size will run both layers over time, for different work. That is not a hedge. It is the architecture the work actually has, and [knowing what is worth automating first](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) is the useful next read for sequencing the two. If you are weighing extraction and document workflows alongside reporting, [what automating data extraction actually means for finance teams](https://flowrunner.ai/blog/automate-data-extraction) covers the upstream half of the same picture. ### Quick answers #### Is FlowRunner a replacement for Airbyte? No. Airbyte is an open-source data replication platform that moves data from 600+ sources into warehouses, lakes, and databases, and now also serves as a context layer that gives AI agents searchable access to that data. FlowRunner is an orchestration layer that coordinates what agents and people do across the tools a finance team already runs. If the problem is consolidating data into Snowflake or BigQuery, or giving an agent read access to your business data, Airbyte is built for that. FlowRunner is not a replication tool. #### Airbyte now has an agent context layer. How is that different from what FlowRunner does? Airbyte’s Context Store gives an agent access to your data so it stops making things up. That is the read side: the agent can see your customers, deals, and tickets. FlowRunner governs the act side: what the agent is allowed to do with what it sees, where it pauses, who it asks when a number looks wrong, and what gets written to the audit trail. Access is necessary. It is not the same as governance and human-in-the-loop control. #### Can we automate report generation directly from QuickBooks or NetSuite without a warehouse or a context store? Yes, and that is the case FlowRunner is built for. Most mid-market finance teams want a weekly P&L digest, an AR aging summary, or a duplicate-payment flag delivered to Slack or email from the operational systems themselves. FlowRunner connects to QuickBooks, NetSuite, Acumatica, and Stripe directly, assembles the report, distributes it, and pauses for a named approver when a variance crosses a threshold. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Ansarada for M&A Due Diligence: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-ansarada-m-and-a-due-diligence Comparisons May 28, 2026 Updated September 7, 2026 13 min read Ansarada is a 20-year virtual data room built for deals. FlowRunner orchestrates the reconciliation and approval-chasing around the checklist. How to decide. ![A bald cartoon man holding a long unrolled checklist, a green DATA ROOM safe with locked folders on his left and a desk labeled THE WORK AROUND IT with mismatched ledgers, a Slack bubble, and email on his right, looking down at one amber unticked item connected by a loose amber thread.](https://flowrunner.ai/images/blog/flowrunner-vs-ansarada-m-and-a-due-diligence-hero.webp) If you are searching “FlowRunner vs Ansarada” for your M&A due diligence, the honest answer is that you are comparing two tools that do not do the same job, and naming that mismatch is more useful than pretending one wins. [Ansarada](https://www.ansarada.com/virtual-data-rooms) is a virtual data room with twenty years of deal pedigree, built to secure and govern the confidential documents that move through a transaction. FlowRunner is an orchestration layer for the operational work happening around those documents: the reconciliation, the vendor validation, the approval-chasing that has to clear before a diligence item can honestly be marked done. The real decision is not which product is better. It is which of those two jobs is the one actually holding up your close. ### Ansarada and FlowRunner, head to head | | Ansarada | FlowRunner | | --- | --- | --- | | **Category** | Virtual data room for M&A, capital raises, IPOs, and infrastructure procurement | Orchestration layer for work and AI agents across business operations | | **Primary job** | Secure the documents, run structured Q&A, govern access across the deal lifecycle | Coordinate the financial reconciliation, vendor and distributor validation, and approval-chasing around the deal | | **Built for** | CEOs, dealmakers, investment bankers, law firms, M&A advisors, private equity | Operations and finance leaders running cross-system workflows, including the operational layer of diligence | | **Track record** | Founded 2005, twenty years across 170 countries ([about](https://www.ansarada.com/about)) | Newer entrant; no comparable VDR track record | | **Document security** | Granular permissions, AI-Redact, remote self-destruct, dynamic watermarking, real-time activity tracking | Not a VDR feature; FlowRunner does not replicate document-level deal controls | | **Structured Q&A** | Automated Q&A workflows purpose-built for diligence | Not offered as a deal-room feature | | **AI in the product** | AI-Sort, AI-Translate, AI-Redact, AI-Predict, AI-powered risk dashboards inside the room | AI agents that coordinate work across tools, with human-in-the-loop as a callable action when judgment is required | | **Operational diligence work** (revenue-to-cash ties, vendor master cleanup, approval routing) | Out of scope by design; the room governs documents, not the reconciliation feeding them | Native fit: workflows trigger across email, ERP, Slack, parsers, and approvals with a per-item audit trail | | **Governance and compliance tiering** | ISO 27001 certified, comprehensive audit trail for document activity ([about](https://www.ansarada.com/about)) | Core capability (agents, human-in-the-loop, all integrations, BYOK) at Growth $45/mo; audit trails and RBAC at [Professional ($299/mo)](https://flowrunner.ai/pricing); SSO/SAML, 90-day audit retention, and compliance reporting at Business ($999/mo); not a VDR-grade document-security claim | | **Pricing model** | Storage-based, free until the deal goes live, tailored quote ([pricing](https://www.ansarada.com/pricing)) | Execution-based subscription tiers; full operational diligence workflow runs from the Growth tier at $45/mo, with governance features layered in at Professional and compliance posture at Business | The table carries the surface comparison. The rest of this article is the part the table cannot: why these two tools keep showing up in the same search, and how to tell which one your deal actually needs. ### What Ansarada is, in its own words Ansarada describes its data room as [“a secure online workspace used to store and share confidential documents with permission controls, activity tracking, and reporting,”](https://www.ansarada.com/virtual-data-rooms) built for [“high-stakes transactions where security, speed, and constant oversight aren’t just nice to have, they’re mission-critical.”](https://www.ansarada.com/virtual-data-rooms) On its company page it states, [“For 20 years we’ve been helping people get their Deals in order,”](https://www.ansarada.com/about) with a founding year of 2005 and a footprint across 170 countries. That is not framing to argue with. It is an accurate description of a platform built specifically for the dealmaking community over two decades. The feature depth is real and it is deal-specific. AI-Sort for automatic indexing. AI-Translate for cross-border document sets. AI-Redact for sensitive material. AI-Predict, which Ansarada markets as [“AI that predicts deal outcomes with 97% accuracy.”](https://www.ansarada.com/virtual-data-rooms) Remote self-destruct to revoke access after download. Dynamic watermarking. Role-based access controls at the folder and document level. Automated Q&A workflows. Real-time activity tracking with comprehensive audit trails. On security, the company states it has been [“ISO 27001 certified for over 12 years.”](https://www.ansarada.com/about) Its pricing is storage-based rather than per-user, and [“your data room is completely free until your transaction is live,”](https://www.ansarada.com/pricing) which is a genuinely low-friction way to start. ![An Ansarada-style virtual data room interface, described textually rather than screenshotted, showing a left-hand folder tree, document-level permission toggles, a watermark indicator on an open file, and an AI risk-signal panel on the right](https://flowrunner.ai/images/blog/flowrunner-vs-ansarada-m-and-a-due-diligence-1.webp) Said plainly: for a deal team running a live sell-side or buy-side transaction, those are not conveniences. They are the product. A general orchestration platform with no data-room primitives is the wrong tool for securing transaction documents, and recommending FlowRunner for that job would be irresponsible. This article does not make that recommendation. FlowRunner is not a virtual data room, does not host confidential deal documents, and does not replace Ansarada for the secure-room function. The honest reason both names show up in the same search is different, and it is worth saying out loud. ### The diligence checklist is two kinds of work wearing one name Here is what most articles on M&A due diligence will not say plainly. The phrase “due diligence checklist” hides two different kinds of work, and a virtual data room is built for one of them. The first kind is document work. Hundreds or thousands of line items, each mapped to a document that has to be requested, uploaded, permissioned, reviewed, redlined, redacted, and signed off, with Q&A flowing back and forth between counterparties. The bottleneck is access, version control, redaction, and Q&A turnaround. That is exactly what Ansarada has spent twenty years optimizing, and it does it well. The second kind is reconciliation work. Tie the trailing-twelve-month revenue to actual customer payments. Validate distributor billbacks against goods shipped. Confirm the AP queue holds no duplicate vendor records across the two entities. Check that the seller has no unregistered sales-tax exposure in states it forgot about. Run client intake on the acquired book before close. Chase the named approvers for sign-off on the closing schedule. None of that is a document-access problem. The document might already be sitting in the room. The question is whether the numbers underneath it actually reconcile, whether the vendor master is clean, and whether the approver who has to confirm a figure has answered the email. These two checklists share a name and almost nothing else. They live in different systems, get done by different people, and fail in different ways. A data room’s activity log can tell you a document was opened. It cannot tell you whether the reconciliation underneath that document was completed, or whether an exception got routed to the one person who could resolve it. That gap is not a flaw in Ansarada. It is the edge of what a data room is for. ### Where Ansarada is genuinely stronger A comparison that claims FlowRunner wins on every axis would fail your own judgment, so name the axes where Ansarada is plainly the better tool. - **M&A domain depth.** Twenty years of deal-specific workflow, structured Q&A management, deal readiness scoring, and a feature set tuned to how transactions actually run. FlowRunner does not match this and is not pursuing it. - **Document-level security purpose-built for transactions.** AI-Redact, remote self-destruct, dynamic watermarking, and granular file-level permissions are built for sensitive deal documents in a way a general orchestration platform does not replicate. If your risk is “the wrong party sees the wrong page,” that is Ansarada’s home turf, not FlowRunner’s. - **Established trust and third-party social proof.** Ansarada carries a large verified user community on [G2 and Capterra](https://www.g2.com/products/ansarada/reviews) and a recognised brand in the VDR category. FlowRunner is newer and less recognised in deal circles, and for a transaction where counterparty confidence in the data room matters, that recognition has real commercial weight. - **Compliance posture for the document workflow.** ISO 27001 certification held for over twelve years, with comprehensive activity tracking inside the room. For uploading sensitive transaction materials, that substantiated posture is exactly what a deal team needs to see. If those axes decide your version of the checklist, the conclusion is simple. Choose Ansarada, or a peer virtual data room. FlowRunner is the wrong shape for that job, and the rest of this article is context rather than a recommendation. ### Where FlowRunner is the better fit The reconciliation side of the checklist is a different problem, and a virtual data room cannot close it by design. In conversations with finance leaders running acquisition programs, the same three gaps surface over and over, and none of them is a document-access problem. - **Diligence items falling through the cracks near close.** A partner at one accounting firm put it plainly: there are things that should happen during diligence and simply do not, the items that everybody jumps on to jam through at the last minute. The room knows a document was uploaded. It does not know whether the reconciliation underneath it cleared, or whether the approver answered. - **Double-payment risk during the transition.** A CFO described how easy it is to double pay a distributor invoice when both entities are mid-handoff and both AP queues are live. That is a workflow problem between two ERPs that do not yet know each other exist, not a document problem. - **Two finance stacks that have to be made to agree.** Charts of accounts, vendor masters, customer masters, and sales-tax setups must reconcile before the diligence numbers can be trusted at all. Most of that work happens in spreadsheets and inboxes, not in the data room. [FlowRunner](https://flowrunner.ai/) is built for that operational layer. A workflow triggers when a parsed vendor invoice arrives, gathers context across the seller’s ERP and the buyer’s ERP, attempts the mechanical part of the match, and pauses to ask a named reviewer when the result is ambiguous. According to FlowRunner’s [feature documentation](https://flowrunner.ai/features), agents can invoke humans as callable tools in the workflow when they hit uncertainty, which is the design point that separates this from a notification. The human is a step in the logic, not a passive alert at the end of it. ![A Slack-style human-in-the-loop message described textually, showing an agent pausing on an ambiguous distributor invoice match, presenting the two candidate vendor records side by side, and offering Approve, Reassign, and Hold actions to a named finance reviewer](https://flowrunner.ai/images/blog/flowrunner-vs-ansarada-m-and-a-due-diligence-2.webp) A second workflow runs the approval chase. The closing schedule lands in front of the CFO, the named approvers get a message through the channel they actually read with the specific question and full context attached, the response is captured against the source record, and the audit trail spans the entire motion. A third handles client intake for the acquired book, parsing the existing client list into the buyer’s onboarding system with incomplete records routed to a human as exceptions. ![A FlowRunner-style workflow diagram described textually, showing a parsed-invoice trigger branching into an automated ERP match, a human-in-the-loop decision node rendered in amber for ambiguous matches, and a final audit-log write step rendered in sage green](https://flowrunner.ai/images/blog/flowrunner-vs-ansarada-m-and-a-due-diligence-3.webp) That per-item audit trail is the seam worth dwelling on, because it is exactly where a data room’s logging stops. Ansarada’s activity tracking is comprehensive about the document: who opened it, when, from where. It is silent on the reconciliation work that determines whether the document’s contents are even true. The checklist item “TTM revenue ties to cash” is marked done by a human judgment about numbers across two systems, and nothing in the room records whether that judgment was actually made, by whom, or against what evidence. The category that owns that gap is orchestration as a service: a layer above the systems of record that listens for what they emit, gathers the context a checklist row needs to be true, pulls in a named human when the data alone cannot settle the question, and writes one audit trail across the reconciliation work itself. FlowRunner is built for that layer. Ansarada is built for the document and Q&A workflow inside the room. Different jobs, opposite sides of the same checklist. This is the same pattern documented in the [post-M&A integration checklist for the CFO running the combined finance function](https://flowrunner.ai/blog/post-m-and-a-integration-checklist). The work shape does not change at close. Before close it shows up as diligence items the room cannot enforce. After close it shows up as integration items the room is no longer part of. On the cost side, the entire operational diligence workflow described above runs from FlowRunner’s [Growth tier at $45 per month](https://flowrunner.ai/pricing), which includes the AI agents, the multi-channel human-in-the-loop (email, Slack, WhatsApp, phone), all integrations, BYOK for the model layer, and unlimited users. There are no technical limitations at that tier that would prevent the reconciliation work this article describes. Governance features layer in at Professional ($299/mo) when a CFO needs audit trails and RBAC called out as decision-relevant for diligence work. SSO/SAML, 90-day audit retention, and compliance reporting begin at Business ($999/mo). The honest framing matters here: this is mid-market infrastructure a CFO can authorize without an enterprise procurement cycle, not a VDR-grade document-security claim and not a substitute for ISO-certified deal-room controls. ### How to decide One diagnostic, run honestly, settles most of it. **What is your close actually waiting on right now?** Walk the open items on your checklist and name, for each, what the work really is. If the answer is “get the document uploaded, permission it correctly, redact the sensitive pages, turn the Q&A around, sign off,” your bottleneck is document workflow and the question is which data room to run. Ansarada or a peer VDR, and an orchestration layer is the wrong shape for it. If the answer is “tie this number to that system, clean up the vendor master, chase the partner for approval, run intake on the acquired book, get two charts of accounts to agree,” your bottleneck is reconciliation work, and a better data room will not touch it. Most mid-market deals are waiting on both, run by different people. The deal team’s document index belongs in a VDR. The finance leader’s reconciliation ledger belongs in an orchestration layer. That is not a hedge. It is the structural reality on the buy side of most transactions that are not bulge-bracket M&A, and it is why the better question is not “Ansarada or FlowRunner” but “which of these two jobs is the one slipping.” If your diligence keeps slipping on documents, this comparison points you to Ansarada. If it keeps slipping on numbers and approvals, the orchestration layer is the part you are missing. If your deal sits closer to the deal-room end of that line, the companion piece comparing [FlowRunner and Datasite for the diligence checklist](https://flowrunner.ai/blog/flowrunner-vs-datasite-for-m-and-a-due-diligence-checklist) walks the same trade-off against a different VDR. And if you are still deciding whether the reconciliation work is even worth automating, the guide on [how to know what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) frames it as a financial calculation, while the piece on [building automations that hold up under scrutiny](https://flowrunner.ai/blog/the-truth-about-building-automations) covers the engineering discipline a real deal calendar demands. ### Quick answers #### Is FlowRunner a replacement for Ansarada? No. Ansarada is a virtual data room built over 20 years for M&A, capital raises, and IPOs, with document permissions, AI-Redact, remote self-destruct, dynamic watermarking, and structured Q&A. FlowRunner is not a data room and does not host transaction documents. If the job is securing and governing deal documents, that belongs in Ansarada, not FlowRunner. #### Where does FlowRunner fit if Ansarada is already running the data room? FlowRunner orchestrates the operational diligence work that lives outside the data room: tying revenue to actual payments, validating vendor and distributor billing, reconciling two charts of accounts, and chasing named approvers for sign-off, each with human-in-the-loop on the ambiguous items and a per-item audit trail. The data room governs the documents; the orchestration layer governs the reconciliation work feeding them. #### Who should choose FlowRunner instead of Ansarada for diligence? A CFO or VP Finance running an acquisition program where most of the checklist is financial reconciliation, vendor validation, and internal approval-chasing rather than VDR-resident document review. Think accounting roll-ups, family-office acquisitions, and mid-market bolt-ons where items fall through the cracks near close because nothing enforces ownership across systems, not because documents are hard to access. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs AppFolio: An Honest Read on Property Management Automation Source: https://flowrunner.ai/blog/flowrunner-vs-appfolio-property-management-automation Comparisons May 28, 2026 Updated September 7, 2026 13 min read AppFolio runs your property management operation. FlowRunner coordinates the work between AppFolio and the rest of your stack. Here is how to tell which you need. ![A bald cartoon man beside a tidy sage-green filing cabinet labeled AppFolio, holding one amber cord that runs out to a scatter of unconnected objects (a bank, an insurance shield, a turnover-vendor toolbox, a spreadsheet, an owner statement) on the floor around him.](https://flowrunner.ai/images/blog/flowrunner-vs-appfolio-property-management-automation-hero.webp) How many systems does your property management company actually run? Name them. There is the platform that holds your units, leases, and ledger, which for a lot of operators is [AppFolio](https://www.appfolio.com/). Then there is the bank. The insurance verification service. The turnover and maintenance vendors who never made it into your AppFolio vendor list. The owner-reporting tool one of your principals insists on. The portfolio you acquired last year that is still on a different system because the migration keeps slipping. AppFolio runs the first one beautifully. The question that decides whether you need anything beyond it is what happens to the work that has to cross all the others. That question is the whole comparison, and it is why a FlowRunner-versus-AppFolio article is not really a head-to-head at all. ### AppFolio and FlowRunner, side by side | | AppFolio | FlowRunner | | --- | --- | --- | | **What it is** | A unified property management platform: accounting, leasing, maintenance, resident and owner portals, reporting | A coordination layer for AI agents and human reviewers across the systems you already run | | **Layer** | System of record for the property management operation | Sits above and between systems of record, including AppFolio itself | | **Primary job** | Run leasing, maintenance, accounting, and communication for a portfolio | Automate processes that span the platform plus banks, insurers, vendors, and other tools | | **AI model** | Native [Realm-X](https://www.appfolio.com/ai) (Assistant, Flows, Performers); agentic AI inside AppFolio workflows | General-purpose agents configured against any connected system; a human invoked as a callable step on exceptions | | **Property management depth** | Years of real estate domain data, templates, and modules purpose-built for the vertical | None out of the box; horizontal infrastructure with no pre-built real estate workflows | | **Human handoff** | Tasks and approvals inside AppFolio modules | Agent pauses, routes a decision to a named person via Slack, email, WhatsApp, or phone, resumes on the response | | **Cross-system reach** | Deep where AppFolio integrates; bounded by AppFolio’s integration catalog | [MCP](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents) and an open connector model across tools AppFolio may not integrate with | | **Portfolio coverage** | Single-family, multifamily, student, affordable, commercial, HOA, investment management, one platform | Not portfolio-typed; coordinates whatever systems the operator connects | | **Pricing model** | Quote-based, [tiered Core / Plus / Max](https://www.appfolio.com/pricing), 50-unit minimum on Core | Published tiers; core platform (AI agents, human-in-the-loop across email, Slack, WhatsApp, and phone, all integrations, BYOK) from [$45/mo Growth](https://flowrunner.ai/pricing); audit trails and RBAC at $299/mo Professional; SSO at $999/mo Business | | **Best fit** | You need a system to run the property management operation itself | You have the platform and need the work between it and everything else coordinated | The table answers the snippet question. The rest of this article is the part the table flattens: why these two products sit on different layers, and how an operations leader actually tells which problem they have. ### Who each product is for AppFolio is for the company that needs to run a property management operation. That is most property management companies, and AppFolio is one of the strongest answers in the category. Their own [positioning](https://www.appfolio.com/) frames the product as the “AppFolio Performance Platform” and tells buyers to “Move Beyond Property Management Software,” which is marketing for a real claim: accounting, leasing, marketing, maintenance, resident and owner portals, and reporting all live in one system. AppFolio serves single-family, multifamily, student housing, affordable housing, commercial, community associations, and investment management portfolios from that single platform. If you do not have a system that owns your units, leases, and ledger, you do not need a coordination layer yet. You need a platform, and AppFolio is built to be one. FlowRunner is for the operations leader who already has that platform and is still doing manual work around it. The buyer here is the VP or director of operations who can name six processes that touch AppFolio plus something AppFolio does not own, and who is tired of being the person who reconciles the difference by hand. [FlowRunner](https://flowrunner.ai/) does not run a property management operation. It coordinates the work that spans the operation and the rest of the stack, and it pulls a human in when an agent hits something it should not decide alone. Those are different buyers with different problems, even when they work at the same company. ### What most of these comparisons get wrong Here is the thing the category keeps publishing that does not hold up: the side-by-side that lines AppFolio up against an automation tool as if they compete for the same purchase. They do not. AppFolio is a system of record. A coordination layer is a system that sits above records. Comparing them feature-for-feature produces a checklist where AppFolio wins every row that involves running a property, because of course it does, that is what it is for, and the automation tool wins a few rows about connectors, and the reader learns nothing about the decision they actually face. The honest framing is layered, not lateral. The question is not “AppFolio or FlowRunner.” It is “is my automation problem inside the platform, or between the platform and everything else.” Most property management companies have some of both. The useful comparison is figuring out which kind of work is eating your team’s week. ### Axis 1: the AI inside the platform This is the axis where AppFolio is genuinely strong and any honest comparison says so plainly. AppFolio describes its AI as “native agentic AI,” branded Realm-X, and the framing they lead with is “AI Built-In, Not Bolted-On,” with the product “designed from the ground-up with AI as a core building block, not an add-on.” That is not empty. [Realm-X](https://www.appfolio.com/ai) ships in three parts. The Assistant “instantly pulls performance insights and reports, executes bulk actions and communication.” [Realm-X Flows](https://www.appfolio.com/articles/realm-x-flows) is an automation engine for standardizing repeated processes so they run “consistently, automatically, and 24/7.” And Realm-X Performers, in AppFolio’s own words, “use agentic AI to independently observe, interpret and act on signals from your data to keep operations flowing.” \[![a labeled diagram contrasting two layers, the upper band showing in-platform AppFolio Realm-X work (leasing, maintenance, accounting tasks) self-contained inside one boundary, and the lower band showing a separate coordination layer reaching out to four external systems, with the boundary between in-platform and between-systems work drawn as a clear dividing line](https://flowrunner.ai/images/blog/flowrunner-vs-appfolio-property-management-automation-1.webp)\] For work that lives inside AppFolio, that is a serious capability, and it is improving. A maintenance request that comes in through AppFolio, gets triaged, dispatched to a vendor in your AppFolio vendor list, and followed up on is a workflow Realm-X is built to own end to end. FlowRunner does not do that better, because FlowRunner does not have the leasing module, the maintenance module, or the resident portal the workflow runs through. On this axis, if your automation need is in-platform, AppFolio is the answer and you should weigh Realm-X heavily. The boundary worth noticing is the phrase “signals from your data.” Realm-X acts on the data inside AppFolio. The work that involves data AppFolio does not hold is the next axis. ### Axis 2: the work between systems Walk one process that does not fit in a single platform. A resident’s insurance lapses. The proof of coverage lives with a third-party verification service, not in AppFolio. The lease clause that says what happens on a lapse lives in AppFolio. The decision about whether to force-place a policy, send a notice, or call the resident depends on the resident’s payment history, the property’s policy, and a judgment call a human should make. Three systems and a person, for one event. Or owner reporting at month end, where the numbers come out of AppFolio but a principal wants them reshaped in a separate tool with commentary. Or an acquired portfolio still running on a different platform that has to be reconciled against the AppFolio book until the migration finishes. Or a turnover vendor who invoices through a system that is not in your AppFolio vendor list, where the invoice has to be matched against the work order and the unit condition report before anyone pays it. \[![a horizontal flow diagram showing a single event (insurance lapse) entering a coordination layer, the layer pulling context from three labeled systems (AppFolio lease, third-party insurance verification, payment history), an agent attempting a match, then branching to a human review step rendered in amber when the proof of coverage does not match the lease requirement, then writing the resolution back to AppFolio with an audit-trail entry](https://flowrunner.ai/images/blog/flowrunner-vs-appfolio-property-management-automation-2.webp)\] None of these live inside AppFolio, so none of them are jobs for the AI inside AppFolio. They are jobs for a layer that listens to what AppFolio emits, gathers the context AppFolio does not hold, brings a person in when the answer needs judgment, and writes the decision back so the record stays in one place. That layer is a category, not a single product. It sits above the systems of record, coordinates across them, and governs the agents and humans doing the work. FlowRunner is built for that layer. Buying it to automate something that lives entirely inside AppFolio would be a mistake. Buying it because the seams between AppFolio and the other five systems are where your team loses its week is the fit. ### Axis 3: the human-in-the-loop, and where it sits Both products keep humans involved. The difference is architectural. AppFolio’s tasks and approvals are structured handoffs inside AppFolio modules, which is exactly right for in-platform work. FlowRunner’s [human-in-the-loop architecture treats a human reviewer as a callable step](https://flowrunner.ai/blog/the-truth-about-building-automations) inside an agent workflow that can span systems. The agent runs, hits something it should not decide on its own, pauses, routes the decision to a named person with the full context attached, captures the response through Slack, email, WhatsApp, or phone, and resumes from that answer. One operations leader called the pattern a digital andon cord, the pull that stops the line when something is wrong rather than letting the work plow ahead. \[![a Slack approval message preview showing a coordination-layer escalation for a resident insurance lapse, with the lease coverage requirement, the third-party verification result that does not match, and the resident payment history summarized in the message body, plus Approve, Send Notice, and Call Resident action buttons, the human-decision moment rendered in amber](https://flowrunner.ai/images/blog/flowrunner-vs-appfolio-property-management-automation-3.webp)\] The reason this matters for property management specifically is that the exceptions you cannot fully automate are the cross-system ones. An in-platform approval gate handles an in-platform decision. It cannot route a decision that depends on data in your bank feed, your insurance verifier, and a vendor system at the same time, because those are not in the platform. A configurable escalation layer across systems can. That is the capability FlowRunner adds, and it adds it precisely where a single-vendor approval gate runs out of reach. ### Axis 4: pricing models that do not line up AppFolio prices the platform. [Their pricing](https://www.appfolio.com/pricing) is quote-based, packaged as Core, Plus, and Max, “Flexible Plans That Scale With Your Business,” with a stated 50-unit minimum on Core and minimum spend on every tier. You are buying a property management system priced to your portfolio size, and that is the correct way to price a system of record. FlowRunner prices coordination. The core platform, with AI agents, human-in-the-loop across email, Slack, WhatsApp, and phone, all integrations, BYOK, unlimited users, and unlimited workflows, is at the [Growth tier at $45 per month](https://flowrunner.ai/pricing). Audit trails and RBAC come in at the Professional tier at $299 per month. SSO and 90-day audit retention come in at the Business tier at $999 per month. The unit being priced across all tiers is execution volume across whatever processes you orchestrate, not units under management. A per-door comparison is meaningless because the two products meter different things. The honest read is that they are not substitutes on a price sheet any more than they are substitutes on a feature sheet. An operator who needs both pays AppFolio for the platform and FlowRunner for the work around it, and the two line items answer different questions. ### Where AppFolio is stronger, named explicitly A comparison where FlowRunner wins every axis would be dishonest, and on the axes that define property management software, AppFolio wins outright. Three of them, in the order an operator should weigh them: **It is an actual property management platform and FlowRunner is not.** This is the one that matters most and it is not close. AppFolio has accounting, leasing, maintenance, resident and owner portals, and reporting purpose-built for real estate, refined over years against a large customer base. FlowRunner has none of that. It is horizontal infrastructure with no pre-built property management workflows, no resident portal, no general ledger. A property management company that tried to run on FlowRunner alone would have nothing to run on. If you are choosing a system to operate your portfolio, this axis ends the discussion: you need a platform, and FlowRunner is not one. **Vertical depth and data.** Realm-X is trained and tuned against property management data and embedded in the workflows where that data lives. AppFolio’s templates, integrations, and AI reflect years of building specifically for leasing, maintenance, and real estate accounting. FlowRunner’s agents are general-purpose; the operator configures them against their own stack and supplies their own context. For in-platform property management tasks, AppFolio’s domain-specific tuning is a real advantage that FlowRunner does not try to match. **Breadth of portfolio coverage in one system.** Single-family, multifamily, student, affordable, commercial, HOA, and investment management, all from one platform with one packaging model. For an operator with a mixed portfolio who wants a single system of record across all of it, that consolidation is genuine value, and FlowRunner is not a system of record at all. If your evaluation rests on these three surfaces, choose AppFolio, and a coordination layer is a question for later, not now. ### Where FlowRunner is the better fit \[![a coordination-layer overview showing AppFolio at the center as the system of record, with four orchestrated workflows fanning out to a bank feed, an insurance verification service, an off-platform turnover vendor, and an owner-reporting tool, each workflow showing its status and a count of items currently paused for human review](https://flowrunner.ai/images/blog/flowrunner-vs-appfolio-property-management-automation-4.webp)\] Choose FlowRunner, on top of a platform like AppFolio, when: - The work eating your team’s week is between systems, not inside one: insurance lapses, owner-reporting reshaping, acquired-portfolio reconciliation, off-platform vendor invoices - You run more than one system that has to agree, and a person currently reconciles the difference by hand - The exceptions you cannot automate are the cross-system ones, and you want a human pulled in with full context rather than a flat alert - You connect tools AppFolio does not integrate with natively, and you need an open connector model and [MCP](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents) support to reach them - You want operations people, not engineers, to build and change these cross-system workflows - You need audit trails and RBAC on the coordination work at mid-market pricing (Professional at $299/mo), with SSO available at Business ($999/mo), not just inside the platform What FlowRunner does not try to be is a property management system. It will not replace AppFolio’s ledger, portals, or vertical AI, and an operator whose problem is in-platform should not buy it expecting that. The configuration that makes sense is the layered one: AppFolio runs the operation, FlowRunner coordinates the seams. We have not built a packaged AppFolio connector to date, and the honest version of that sentence is that the architectural shape, AppFolio as system of record with a coordination layer listening around it, is the same one we already run against ERPs and line-of-business systems. ### How to decide Three questions, in order. They are designed so you reach the answer, not so I reach it for you. **1\. Do you have a system that runs the operation?** If the honest answer is no, or “we are on spreadsheets and a tool we have outgrown,” stop here. You need a property management platform before you need anything that coordinates one, and AppFolio is one of the category’s strongest answers. Come back to the coordination question once the platform is in. **2\. Where does the work that frustrates your team actually live?** Walk the last five processes that cost an operations person real hours. If they ran entirely inside one platform, AppFolio’s Realm-X is the place to push, and you may not need a second tool at all. If two or three of them spanned the platform plus a bank, an insurer, a vendor system, or a separate portfolio, that work is between systems, and it is the work a coordination layer is built for. **3\. Where do your hardest exceptions sit?** Look at the cases nobody could fully automate. If they resolved cleanly inside an in-platform approval, an in-platform tool is enough. If the hard ones needed someone to weigh data from three systems at once, you need an escalation layer that can reach across all three and bring a person in with the whole picture framed. Most operators find question two splits their answer: some of the week is in-platform and some of it is between systems. That is the normal shape, not a hedge. It is also why the layered configuration is common: the platform for the operation, the coordination layer for the seams. If you want to see how the cross-system pattern looks in practice, our [property management automation overview](https://flowrunner.ai/solutions/property-management-automation-software) walks through it, and the [FlowRunner versus Buildium read](https://flowrunner.ai/blog/flowrunner-vs-buildium-property-management-automation) covers the same layered question against a different platform. If you are earlier than that and still deciding what is even worth automating, [start there instead](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating). The seams are where it shows. The insurance lapse that needed three systems and a judgment call. The owner statement reshaped by hand every month. The acquired portfolio reconciled in a spreadsheet until the migration lands. The off-platform invoice matched against a work order before anyone pays it. The vendor who falls outside the catalog and so falls to a person. None of those are failures of your property management platform. They are the work that lives between it and everything else, and that is the work a coordination layer exists to stop letting plow ahead unwatched. ### Quick answers #### Is FlowRunner a replacement for AppFolio? No. AppFolio is a complete property management platform with accounting, leasing, maintenance, resident and owner portals, and reporting built in. FlowRunner does not provide any of those modules and would need a system like AppFolio to function in a property management context. FlowRunner is a coordination layer that connects AppFolio to the other systems your company runs and pulls a human in on the exceptions. A property management company needs the platform first. #### When does FlowRunner add value on top of AppFolio? When work crosses the boundary of the platform. AppFolio’s Realm-X automates leasing, maintenance, and accounting inside AppFolio. FlowRunner adds value when a process has to span AppFolio plus something AppFolio does not own: a bank feed, an insurance verification service, a turnover vendor outside your AppFolio vendor list, a separate owner-reporting tool, or an acquired portfolio on a different system. It is most useful when a person has to make the call on the exceptions, because the data needed to decide sits in more than one place at once. #### Does AppFolio already do AI and automation? Yes, and the comparison is dishonest if it pretends otherwise. AppFolio ships native agentic AI through Realm-X, with an Assistant, a Flows automation engine, and Performers that, in AppFolio’s words, use agentic AI to independently observe, interpret and act on signals from your data. For a large share of in-platform work, that is all you need on its own. The question FlowRunner answers is what happens to the work that does not live inside AppFolio. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs BILL: An Honest Read on Accounts Payable Automation Software Source: https://flowrunner.ai/blog/flowrunner-vs-bill-automate-accounts-payable-software Comparisons May 27, 2026 Updated May 28, 2026 12 min read BILL is the answer when AP is the whole job. FlowRunner is the answer when AP is one of several finance and ops processes that need orchestrating together. ![A bald cartoon man holding an old-fashioned balance scale, with a neatly stacked green tower labeled BILL on one pan and a loose constellation of amber objects connected by threads on the other, weighing two different shapes of finance automation.](https://flowrunner.ai/images/blog/flowrunner-vs-bill-automate-accounts-payable-software-hero.webp) The fastest way to misread the FlowRunner versus [BILL](https://www.bill.com/product) question is to start with a feature checklist. BILL is the largest pure-play financial operations platform for small and mid-sized businesses in the US market, and a checklist will surface that in fifteen rows before you get to the question that actually decides the purchase: is accounts payable the whole job, or is AP one of many finance and ops processes that need to run together? The first answer points to BILL. The second answer points to an orchestration layer that handles AP as one of several workflows it coordinates, and FlowRunner is built for that layer. ### Side by side, at a glance | | BILL | FlowRunner | | --- | --- | --- | | **Category** | Financial operations platform (AP, AR, spend and expense in one) | Orchestration layer for AI agents and human approvers across any business process | | **Primary job** | Run AP, AR, and spend management as a packaged financial workflow | Coordinate work across AP and adjacent finance, procurement, and ops processes | | **Built for** | SMB and mid-market finance teams whose primary need is financial operations | Finance and ops leaders coordinating work across multiple systems and processes | | **Scope** | Accounts payable, accounts receivable, spend and expense management, vendor payments | Any ops or finance workflow against connected systems (ERP, parsers, Slack, email, WhatsApp) | | **Payment network** | 8.3 million network members; 498,000 businesses; native payment rails | No payment network; orchestrates against existing ERP, bank, and processor connections | | **AI agents** | Domain-specific agents trained on financial data at scale (Invoice Coding Agent, W-9 Agent, Transaction Agent) | General-purpose agents the customer configures against any process; not pre-trained on financial data at packaged accuracy | | **Human approvals** | Transactional approval gates inside the AP and expense product | Callable action: agent pauses, routes a decision to a named approver via Slack, email, WhatsApp, or phone, resumes on response | | **Accountant channel** | 98 of the top 100 US accounting firms partner with BILL; dedicated accountant console | No equivalent accountant console or partner channel for accounting firms | | **Compliance posture** | SOC 1 and SOC 2 Type II, unalterable audit trail on every transaction | Audit trails, RBAC, and SSO at the [Professional tier](https://flowrunner.ai/pricing) ($299/mo); no SOC attestation today | | **Best fit** | AP is the whole job, or AP plus AR and spend management in one product | AP is one of many automated processes the same team needs to orchestrate | The table tells most of the story. The rest of this article is the part the table cannot. ### What BILL is good at, said honestly BILL is the established, scaled answer for SMB and mid-market financial operations. Their own [product positioning](https://www.bill.com/product) frames the platform as integrated AP, AR, and spend and expense management in one place, and the scale claims behind that positioning are real, sourced from BILL’s own [company page](https://www.bill.com/about): approximately 498,000 businesses on the platform, a payment network of 8.3 million members, and 98 of the top 100 US accounting firms in partnership. That is not a feature; that is a moat. A new platform can build payment connectivity, but a payment network of that size is years of accumulated relationships, not a roadmap. The accountant channel is the part most platform comparisons skip. For a wide slice of SMBs, the accountant or fractional CFO is who picks the AP tool. BILL’s accountant console and the deep partnerships with the top US firms mean the buying motion for the platform often goes through a trusted advisor, not a cold evaluation cycle. FlowRunner has no equivalent channel today. That is a real distribution advantage and the honest framing is to name it. Specific things BILL does that FlowRunner does not: - A native, US-scale payment network that lets buyers and vendors transact through pre-built connectivity rather than ERP-to-bank plumbing the customer has to maintain - An integrated AR product alongside AP, so finance teams who need both can manage them in one workflow - A spend and expense management product (the BILL Spend and Expense suite, formerly Divvy) with corporate cards, budgets, and expense reports tied into the same AP and AR core - Deep, in-house-maintained integrations with QuickBooks, NetSuite, Sage Intacct, Xero, and Acumatica built specifically for financial workflows - Domain-specific [AI agents](https://www.bill.com/product/ai-agents) trained on financial data at scale and shipped to nearly half a million customers (Invoice Coding Agent, W-9 Agent, Transaction Agent), embedded directly into the AP and expense workflows where they run daily - SOC 1 and SOC 2 Type II compliance with bank-grade encryption and an [unalterable audit trail](https://www.bill.com/product/security) on every transaction, which matters when the auditor asks for evidence Lead with this honestly because if the rest of this article reads like a competitive attack on BILL, the article is dishonest. It is not an attack. BILL is the right answer for a clearly defined buyer, and that buyer is common. ### What AP looks like when it is not the whole job Here is what most articles framed around AP automation software will not say plainly: for a meaningful slice of the validated finance buyer pool, AP is not a self-contained problem with a self-contained tool answer. It is one node in a graph of finance and ops chores that share the same systems, the same approvers, the same exception logic, and the same evening hours of the same controllers. The pattern that surfaces in conversations across the validated CFO and VP of Finance pool, drawn from roughly a dozen finance-leader discussions, looks like this: - AP invoices land in an inbox, get parsed (or retyped), and need to flow into an ERP that is often Acumatica, NetSuite, QuickBooks Online, or a Sage product - Distributor billbacks arrive separately, sometimes through the same vendor, and need to be validated against contract terms before payment goes out (one CFO described how easy it would be to double pay on something like that when both the distributor and the company can initiate payment) - Freight cost allocation gets done in a pivot table uploaded back to the ERP at month end - Approval routing happens informally over email and Slack, with thresholds written in a policy memo that no system enforces - Customer ops, HR onboarding, and vendor onboarding share approvers, share routing patterns, and share the same audit obligation, but live in disconnected tools - Senior finance staff still enter bills when someone is out, which is what happens when the automation that exists is doing parsing and nothing else For a finance leader running that mix, the right question is not “which AP product wins on a checklist.” It is “what handles the routing, the exceptions, the cross-system handoffs, and the audit trail across AP plus everything that touches it, without buying a separate product for each one?” That seam, between AP and the seven other things finance ops actually does, is where an orchestration layer lives. It is a category, not a product. The category sits above the systems of record, listens for what they emit, gathers context the system did not have at fire time, brings the right human in with the full record framed, and writes the answer back so the audit trail stays in one place. FlowRunner is built for that layer. Buying it for AP alone is overkill; buying it because AP is one of several processes you need to automate against your existing stack is the fit. ### Where FlowRunner fits: AP as one of many orchestrated workflows [FlowRunner](https://flowrunner.ai/) is an orchestration layer for AI agents and humans across the tools a finance and ops team already runs. AP, in that frame, is one of the patterns the platform handles, not the product it sells. The pattern that maps to AP-shaped work looks like this: 1. A trigger fires (an inbox email, a new bill in the ERP, a webhook from a parser) 2. A workflow gathers context from every system it needs (the ERP, the parser, the vendor master file, the PO record, the approval rule) 3. An agent attempts mechanical work (extract fields, suggest a GL code, check for duplicates, match against a PO) and either clears the bill or pauses it 4. Where it pauses, the workflow calls a named human as an action with the full context attached: the bill, the vendor history, the PO reference, what failed validation, what the agent would have done 5. The human responds (approve, reject, adjust, escalate) through Slack, email, WhatsApp, or phone, and the workflow resumes from the response 6. The bill gets written back into the ERP with the approval evidence captured, and the audit trail records who decided what at what time That last point is where the human-in-the-loop architecture diverges from a transactional approval gate. BILL’s approvals work well inside the AP product; they are a structured gate in a fixed financial workflow. FlowRunner’s [callable human-in-the-loop](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) generalizes the same pattern across processes. The same routing, the same escalation, the same audit-trail capture handles a bill approval over threshold, a vendor onboarding document check, a customer ops escalation, and an HR offer-letter sign-off. The shape is the same; the workflow it lives inside is different. Concretely, the [email-to-payment invoice automation we publish for QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) shows the AP-shaped pattern with duplicate detection and named-approver routing. The parallel patterns work for [parsed vendor invoices into Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and for [vendor documents into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack). The [Acumatica workflow catalog](https://flowrunner.ai/workflows/automate-with-acumatica) shows AP as one part of a broader orchestration footprint inside one ERP. Each of these uses the same orchestration layer, with the same audit trail, governed by the same RBAC, and the same layer extends into procurement, customer ops, and HR work that is not AP at all. ### Where BILL is stronger, named explicitly A comparison that crowns FlowRunner the winner on every axis fails the credibility test. Three places BILL is genuinely the better fit, in order of how heavily a finance buyer should weigh them: **Payment network and AR coverage.** BILL’s 8.3 million network members and the AR product alongside AP are not features FlowRunner ships. For a finance team whose primary need is the full financial-operations arc in one platform (AP plus AR plus spend management with payment connectivity built in), BILL wins outright. FlowRunner orchestrates against the bank and the ERP, but does not run a payment network and does not ship an AR product. **Packaged AI accuracy on financial workflows.** BILL’s [AI agents](https://www.bill.com/product/ai-agents) are trained on financial data at scale and shipped to nearly half a million customers. The Invoice Coding Agent, W-9 Agent, and Transaction Agent live inside AP and expense workflows that run daily across the entire customer base. FlowRunner’s agents are general-purpose; the customer configures them against the systems and rules of their own stack. For invoice coding accuracy out of the box, BILL has a packaged advantage. FlowRunner closes the gap with configuration, BYOK models, and the customer’s own historical data; the trade-off is real and the right framing is fit, not parity. **Accountant channel and SOC posture.** 98 of the top 100 US accounting firms in partnership, an accountant console designed for those firms, and SOC 1 and SOC 2 Type II compliance with an unalterable audit trail per [BILL’s security documentation](https://www.bill.com/product/security). A finance team buying through their accountant, or a team whose auditor scope explicitly requires SOC 1 attestation today, should put real weight on this. FlowRunner publishes audit trails, RBAC, and SSO at the [Professional tier](https://flowrunner.ai/pricing) and is designed to meet common audit requirements for mid-market finance operations, but is not making a SOC attestation claim in this article. Naming these honestly is what makes the rest of the comparison readable. A buyer who lands here, sees BILL win three axes that matter to their situation, and chooses BILL is the right outcome for that buyer. ### Where FlowRunner is the better fit Choose FlowRunner when: - AP is one of several finance and ops processes you need to automate against the systems you already have, not the only one - The pain is between systems and around the workflow (distributor billbacks, freight allocation, exception routing, vendor onboarding documentation, customer ops escalations), not inside an AP-and-AR product boundary - You want a callable human-in-the-loop that generalizes across processes, not a transactional approval gate inside one financial product - You want non-developers in finance and ops to configure new workflows against any connected system, without engineering resources, with new connectors building in 30 minutes or less - You need audit trails, RBAC, and SSO at mid-market pricing ($299 per month) for orchestration across any process, not just AP - You see AI agents arriving in your stack from multiple vendors and want a coordination layer above them before that becomes a problem in its own right What FlowRunner does not try to be: a replacement for BILL’s payment network, AR product, accountant console, or packaged financial-domain agent accuracy. A team whose evaluation rests primarily on those surfaces should choose BILL. The honest read on the two products coexisting: a mid-market finance team running BILL for AP, AR, and spend management could absolutely run FlowRunner alongside it for the orchestration work outside that boundary. That is a real configuration; we have not built a BILL connector to date, but the architectural shape (BILL is a system of record for financial transactions; FlowRunner orchestrates around it) is the same as the one we already publish for QuickBooks, NetSuite, and Acumatica. ### How to decide A three-question test, in this order: **1\. What is the scope of the automation problem you are solving?** If it is “we need AP, and we need AR, and we need spend management, and we want them in one product with a payment network already built in,” BILL is the answer. If it is “we need AP automated, and we also need to handle distributor billbacks, vendor onboarding documents, customer ops escalations, and a half-dozen other workflows that touch finance, ops, and procurement,” the answer is an orchestration layer, and FlowRunner is built for that. **2\. Where does your exception logic live?** Walk through the last three AP exceptions your team handled. Did they all resolve cleanly inside an AP-product approval gate, or did one of them need a sales person to confirm a billback amount, another need an account manager to validate a contract term, and a third need an HR review of a new vendor’s tax documentation? The first answer points to an AP product. The second answer points to a layer that calls humans across processes, not just inside one. **3\. Who is recommending the tool?** If it is your accountant or a fractional CFO with multiple clients, the gravity of BILL’s accountant channel is real, and the friction of buying outside that channel matters. If it is your CFO or VP Operations making an internal call about how to absorb the next eighteen months of operational complexity without proportional hiring, the orchestration scope is what is being purchased, and the channel becomes secondary. Most mid-market teams will land on one product or the other based on the answer to question one. A meaningful minority will run both, used for different work, because the scopes do not overlap once you look closely. That is not a hedge. It is what the brief from the validated finance buyer pool actually looks like. ### Quick answers #### Is FlowRunner a replacement for BILL? No. BILL is a financial operations platform with deep, packaged coverage of accounts payable, accounts receivable, and spend and expense management. FlowRunner is an orchestration layer that coordinates AI agents and humans across the tools a finance team already runs. A team whose primary need is end-to-end AP, AR, and spend in one product should buy BILL. #### Where does FlowRunner fit if AP is part of a bigger automation problem? FlowRunner fits when AP is one of several finance and ops processes that need to be automated against the systems you already have. It parses invoices into QuickBooks, NetSuite, or Acumatica, routes approvals through Slack with named approvers, and uses the same orchestration layer for procurement, customer ops, and HR work outside the AP product boundary. #### How does FlowRunner’s pricing compare to BILL? FlowRunner and BILL price different scopes, so a per-seat comparison is misleading. FlowRunner’s Professional tier at $299 a month covers audit trails, RBAC, and SSO for orchestration across any business process. BILL prices per user for AP, AR, and spend management packaged together. The right question is which scope you are buying for. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs BlackLine: An Honest Comparison on Accounts Reconciliation Software Source: https://flowrunner.ai/blog/flowrunner-vs-blackline-accounts-reconciliation-software Comparisons May 27, 2026 Updated May 28, 2026 13 min read BlackLine wins when reconciliation is the only problem and the buyer is enterprise F&A. FlowRunner wins when reconciliation sits alongside cross-functional automation. ![A bald cartoon man at a country path fork choosing between a tall single ledger building on his left and a village crossbeam connecting five differently colored lanterns on his right, considering both shapes of finance work.](https://flowrunner.ai/images/blog/flowrunner-vs-blackline-accounts-reconciliation-software-hero.webp) The buyer evaluating [BlackLine](https://www.blackline.com/) against FlowRunner is not actually comparing two reconciliation products. They are choosing between two different shapes of finance automation, and the right answer depends almost entirely on which shape their finance function actually has. BlackLine is a vertical specialist with two decades of F&A depth and a roster of pre-trained agents named after the workflows they own. FlowRunner is a horizontal orchestration layer that treats reconciliation as one of many patterns coordinated across procurement, support, HR, and finance at once. Picking the wrong shape is more expensive than picking the wrong product. ### Side by side, at a glance | | BlackLine | FlowRunner | | --- | --- | --- | | **Category** | Enterprise financial close and accounting automation platform | Orchestration layer for work and AI agents across business functions | | **Tagline** | [”Trust is in the Balance”](https://www.blackline.com/) (Verity AI) | “Agents That Ask” | | **Scope** | Record-to-Report, Invoice-to-Cash, the structured close | Cross-functional automation: finance, procurement, support, HR, ops | | **Agent strategy** | Pre-built, vendor-trained Verity agents (Accruals, Match, Collect, Flux) | General-purpose Agent Factory; customer builds and owns the agents | | **F&A domain depth** | 20+ years of accounting process knowledge baked into the agents | None at the vendor level; customer brings the domain knowledge | | **ERP integration** | Deep, native into SAP and major ERPs as a core product promise | Broad coverage across mid-market stacks; shallower in F&A-specific ERP context | | **Human-in-the-loop** | Review and approval steps inside the close workflow | Callable action: pause, route to a named approver via Slack or email, capture the response, resume | | **Pricing** | Enterprise, not publicly listed | Published tiers; [Professional at $299/mo](https://flowrunner.ai/pricing) includes audit trails, RBAC, SSO | | **Buyer** | Enterprise CFO with a dedicated F&A automation budget | Mid-market finance leader whose finance pain is one of several cross-functional problems | | **Procurement cycle** | RFP, security review, vendor approval | Self-serve trial, credit card, finance director can authorize | The table covers the structural facts. The rest of this article is the part that decides whether either product is the right purchase for your team. ### What BlackLine is good at, said honestly BlackLine’s positioning is earned. The [Verity AI workforce](https://www.blackline.com/) is “built on 20+ years of deep finance and accounting process knowledge,” and the named agents are not branding decoration. Verity Accruals understands accruals because BlackLine has watched finance teams book them across thousands of close cycles. Verity Match is built for transaction matching. Verity Collect handles invoice-to-cash. Verity Flux is for variance analysis. Each agent ships with the workflow already shaped, the controls already built, the audit trail already wired. A controller can deploy a Verity agent and have it usefully working on the same close cycle. That depth is not something a horizontal platform can fake. A general-purpose orchestration tool can match transactions, but it does not know that a $0.30 fee gap on a Stripe payout is a normal merchant fee deduction, not an exception. A general-purpose tool can route an accrual journal entry, but it does not know that a missing PO accrual in October at a multi-entity SaaS company is a different problem than a missing inventory accrual in October at a CPG company. BlackLine knows. That is not a feature; it is product investment compounded across years. BlackLine’s customer outcomes are real and worth naming. [eBay reported a 70 percent faster close, and Kempinski Hotels reported 50 percent less time on reconciliations](https://www.blackline.com/). These are large, complex finance organizations whose results are not transferable to a mid-market team in a straight line, but they are evidence that the platform delivers what it promises when the buyer fits. The ERP integration story matters too. BlackLine integrates directly and deeply with SAP and other major ERPs as a core product promise, not as a connector built reactively. When the general ledger is the source of truth, the depth of the connection to the GL becomes the difference between a tool that works and a tool that creates a second reconciliation problem on top of the first. FlowRunner’s connector ecosystem is broader across the mid-market stack, but it is shallower in F&A-specific ERP context. That is the honest read, and it is the most important sentence in this article for a buyer whose stack is SAP-anchored. If your finance organization is enterprise, your stack is SAP or another tier-1 ERP, your reconciliation volume is large, your team is dedicated F&A, and your procurement function expects vendor maturity at the BlackLine level, the rest of this article is interesting but probably not actionable. Buy BlackLine. ### Where FlowRunner is honestly weaker Before the case for FlowRunner, the case against. A comparison that hides this section is dishonest, and the buyer can tell. - **No F&A domain depth at the vendor layer.** FlowRunner has no equivalent to 20+ years of accounting process knowledge baked into pre-trained agents. Teams reconciling accounts or managing accruals get more out-of-the-box accuracy from BlackLine’s Verity agents than from anything FlowRunner ships pre-configured. The depth has to come from the customer’s own knowledge of their accounting workflows when they build on FlowRunner. - **No pre-built financial close workflow library.** BlackLine ships named, deployable agents for specific F&A tasks. FlowRunner’s [Agent Directory](https://flowrunner.ai/) is a general-purpose library. A finance team adopting FlowRunner would build and validate reconciliation and close workflows themselves rather than receiving them pre-configured. That is design intent (the customer owns the agents and the logic, not the vendor), but it is also more work upfront. - **Shallower F&A ERP integration story.** FlowRunner connects broadly across mid-market stacks (Stripe, QuickBooks Online, Acumatica, NetSuite, distributor portals, inboxes, parsers, Slack), but it does not integrate to SAP at the depth BlackLine does. If your reconciliation source of truth is a complex SAP general ledger, BlackLine’s connection is materially better. These are not minor caveats. For an enterprise finance org whose only automation problem is the close, BlackLine is the better product, full stop. The case for FlowRunner does not require pretending otherwise. ### Where the architectural fork actually matters Here is what most comparison posts on accounts reconciliation software will not say plainly: the choice between BlackLine and FlowRunner is not really about which one reconciles accounts better. It is about whether reconciliation is the only automation problem your finance function owns, or one of several. In conversations with finance leaders, the pattern that drives a FlowRunner purchase rarely starts with reconciliation alone. It starts with an aggregate of cross-functional friction: - AP coordinators entering vendor bills into the ERP because the parser does not handle the long tail of formats - Senior finance staff building pivot tables to chase down Stripe payouts that should match QuickBooks invoices but don’t - Distributor billbacks arriving on portals that no vendor natively connects to, getting resolved in email chains and spreadsheets - Procurement requests bouncing between Slack, email, and an approval workflow nobody can find later - Support escalating refund decisions to finance with no structured way to capture the audit trail - HR onboarding documents getting handed to finance for vendor setup with three copies in two different systems One CFO described how easy it would be for two people to review the same bill and approve it on the same day, with nobody catching it for weeks. That is not a close-management problem. It is an exception-handling problem that crosses procurement, AP, and finance, and it lives outside the boundary of any single F&A automation platform. The work in these conversations is not the matching. The matching is mostly mechanical. The work is the exception, and the exceptions cross functions. A vertical specialist like BlackLine is built to be excellent inside its function. It is not built to coordinate across them, and there is no reason it should be. That is the architectural fork. ### Where FlowRunner is built to live [FlowRunner](https://flowrunner.ai/) is a horizontal orchestration layer. It coordinates work and AI agents across the tools and functions a mid-market operations stack already runs. The pattern that maps to reconciliation pain looks like this, with the same shape mirroring across procurement, support, and HR pain: 1. A trigger fires (a Stripe webhook, a new invoice in QuickBooks, an email landing in a shared inbox, a Slack form submission) 2. A workflow gathers context from every system it needs (the payment processor, the ERP, the distributor portal, the parsed document, the CRM) 3. The workflow attempts the mechanical work (match, classify, write, route). If everything ties, it writes the entry and moves on 4. If the work has an exception (mismatched amount, missing reference, duplicate risk, an unfamiliar pattern), the workflow pauses and calls a human as a structured action, not as a status change 5. The human receives a Slack message, email, or WhatsApp message with the full context attached and a structured response (approve, reject, adjust, escalate) 6. The workflow resumes from the response, writes the result with the human’s decision captured, and produces an audit-trail record of who decided what at what time The shape of step 4 is the architectural difference. FlowRunner treats human-in-the-loop as a callable action inside the workflow: pause, route, capture, resume. Approval is not a separate gate in a separate product. It is a function call. The audit trail records the named approver, the timestamp, the context the workflow surfaced to them, and the decision they returned. One prospect, a CEO at an automotive services company, called this “a digital andon cord,” AI that stops the line when it hits uncertainty. That metaphor is closer to how the architecture actually behaves than any approval-gate description. Concretely, the same pattern shows up across [payment-to-invoice reconciliation between Stripe, QuickBooks, and Slack](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack), [Stripe-to-QuickBooks reconciliation with pause-on-mismatch](https://flowrunner.ai/workflows/connect-quickbooks-online-with-stripe), [end-to-end invoice processing from inbox to ERP](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack), and [orchestrating accounting workflows in Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica). The same audit trail, the same RBAC, the same human-in-the-loop shape, applied across the long tail of tools where the reconciliation actually happens. The category that owns this layer is orchestration as a service: a system above the systems of record and above the vertical agent platforms, listening for what they emit, gathering context the agents themselves did not have, pulling a human in at the moments that need judgment, and writing a single audit trail across the whole workflow. As enterprises adopt five or six vertical agent platforms (one for the close, one for spend management, one for support, one for sales ops), the coordination problem above them is the one nobody is yet solving inside any single vendor. FlowRunner is built for that layer. BlackLine is built to be excellent inside one of those vertical platforms. ### Governance without an enterprise procurement cycle Mid-market finance teams need audit trails, role-based access control, and SSO without a six-month procurement process. The standard pattern in the enterprise F&A category is to gate those features behind enterprise pricing, an RFP, a security review, and a vendor approval cycle. BlackLine is sold that way because BlackLine’s buyer expects to buy that way. FlowRunner publishes those features at the [Professional tier at $299 a month](https://flowrunner.ai/pricing). Audit trails, RBAC, and SSO are designed to meet common audit requirements for mid-market finance operations. The framing matters: this article does not claim auditor acceptance of any specific compliance framework such as SOC 2 or ISO 27001. It claims that the governance infrastructure exists at a price a finance director can authorize without an enterprise procurement cycle. If your auditor’s scope requires substantive attestation, those conversations are separate and need to happen against Trust-page evidence, not a marketing comparison. The honest read: an enterprise finance org with a BlackLine budget probably already has the governance pedigree they need. Where FlowRunner adds value at this layer is workflow-level governance across the cross-functional orchestration work, available at a price that does not require an enterprise procurement cycle to evaluate. ### Where BlackLine is the better fit Buy BlackLine when: - Your finance function is enterprise-scale and the close is the central automation problem you are solving for - Your stack is SAP-anchored or runs another tier-1 ERP as the source of truth for reconciliation - Your reconciliation volume justifies a vendor-trained agent built specifically for accruals, matching, collections, or variance analysis - Your procurement function expects vendor maturity, security pedigree, and customer reference depth at the enterprise level - Your team is dedicated F&A staff who will operate the close inside a single purpose-built product That set of conditions describes a real and common enterprise finance organization. If it describes yours, BlackLine is almost certainly the right purchase and FlowRunner is the wrong tool for the job you are doing. ### Where FlowRunner is the better fit Choose FlowRunner when: - Reconciliation pain lives between systems and across functions, not inside one accounting product - Your finance function is mid-market and shares automation budget with procurement, ops, or support - Exceptions are the work, and you need a callable human-in-the-loop that pauses the workflow, routes to a named approver, and resumes - You need to orchestrate across Stripe, QuickBooks, Acumatica, NetSuite, distributor portals, inboxes, parsers, and Slack, not just into a tier-1 GL - You want governance infrastructure (audit trails, RBAC, SSO) at mid-market pricing without an enterprise procurement cycle - Your finance team wants to build and own its agents rather than buy pre-configured vertical agents - You see AI agents arriving across your stack (close, spend, support, sales, HR) and want a coordination layer above them before that becomes the next operational problem The two products are not interchangeable, and they are not directly competitive in the same procurement cycle. An enterprise SAP-anchored close team belongs in BlackLine. A mid-market finance leader whose pain crosses functions belongs in FlowRunner. Most evaluations conflate these buyers because the SERP for “accounts reconciliation software” mixes them, but the right product depends on which buyer your function actually is. ### How to decide A three-part pivot test, with the order intentional: **Where does your finance function’s automation problem actually live?** If the answer is “inside the close, which is the central process my F&A team owns,” that is a vertical specialist problem. If the answer is “across the close plus AP exceptions plus distributor billbacks plus procurement plus support refunds plus vendor onboarding,” that is a cross-functional orchestration problem. Vertical depth solves the first. Horizontal coordination solves the second. **Who owns the agents?** Vertical platforms like BlackLine own the agents. The vendor builds them, trains them, configures them, and ships them. Customers operate them but do not author them. A finance team comfortable buying agents pre-built belongs in a vertical specialist. A finance team that wants to build, own, and modify its agents (because the workflows are specific to the company, or because the team wants the orchestration logic transparent and inspectable) belongs in a horizontal platform. **What is the procurement cycle your buyer can run?** A CFO who can authorize a $25K-per-month enterprise platform after a six-month evaluation cycle has options. A finance director who needs to demonstrate value within the quarter, on a published price tier they can put on a corporate card, has different options. Be honest about which one is you. The buyers who try to evaluate BlackLine and FlowRunner against each other on a feature checklist almost always end up frustrated, because the products are not symmetric. The buyers who name which shape their finance function actually has, and pick the matching tool, end up satisfied with whichever they choose. ### Quick answers #### Is FlowRunner a replacement for BlackLine? No. BlackLine is an enterprise-grade financial close and accounting automation platform with deep, pre-trained agents for record-to-report and invoice-to-cash workflows. FlowRunner is a horizontal orchestration layer for coordinating work and AI agents across business functions. A team committed to BlackLine for the close should keep it. FlowRunner is for buyers whose reconciliation pain is one of several cross-functional automation problems, not the only one. #### Where does FlowRunner fit if we already use BlackLine? FlowRunner sits across functions, not inside the close. Where BlackLine owns reconciliation, accruals, and the structured close, FlowRunner coordinates the work between systems and functions that surrounds it: payment-to-invoice matching with human pause on mismatches, document intake from inboxes into ERPs, exception routing to a named approver in Slack with the full context attached, and the same orchestration pattern applied to procurement, support, and HR exceptions alongside finance. #### Is FlowRunner cheaper than BlackLine? FlowRunner publishes pricing; BlackLine sells enterprise. FlowRunner’s Professional tier at $299 a month includes audit trails, RBAC, and SSO, which makes it accessible to a finance director who can authorize the spend without a procurement cycle. BlackLine is an enterprise platform with enterprise pricing and a deeper F&A footprint. The right question is which problem you are buying for, not which line item is smaller. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Buildium for Property Management Automation Source: https://flowrunner.ai/blog/flowrunner-vs-buildium-property-management-automation Comparisons May 28, 2026 Updated September 7, 2026 12 min read Buildium is a purpose-built property management platform. FlowRunner is the orchestration layer above it. They solve different problems. Here is how to tell which you need. ![A bald cartoon man at the door of a tidy sage-green house facing five overflowing slate-blue mailboxes labeled bank, dollar, envelope, wrench, and key leaning at him from the street, with a single amber pause flag planted in the gap between the house and the mailboxes.](https://flowrunner.ai/images/blog/flowrunner-vs-buildium-property-management-automation-hero.webp) If you manage residential or commercial property and you are searching for how to automate the work, the first thing worth saying is that Buildium and FlowRunner are not two answers to the same question. Buildium is a property management application. FlowRunner is an orchestration layer that runs above your applications. Comparing them head to head is useful, but only if you understand that the right answer for most readers is not one or the other. It is Buildium for the property management product, and FlowRunner when the work that has to happen between Buildium and the rest of your systems has outgrown what any single product can hold. Here is the part most comparison posts will not say, because it does not flatter the author: if you do not already run a property management platform, you should buy one, and Buildium is a strong, mature choice. FlowRunner does not replace it and is not trying to. This article is about the seam Buildium does not own, the space between your property management system and your bank, your general ledger, your owners’ inboxes, and the growing pile of other tools your operation runs. ### Buildium and FlowRunner side by side | | Buildium | FlowRunner | | --- | --- | --- | | **What it is** | A [purpose-built property management platform](https://www.buildium.com/), self-described as the “#1 most recommended property management software” | An orchestration layer for work and AI agents across the tools an operation already runs | | **Primary buyer** | Property manager, landlord, or community association manager who needs a system of record | VP or director of operations coordinating work across multiple systems, with humans in the loop on exceptions | | **Core scope** | Accounting, rent collection, leasing, tenant screening, maintenance, [1099 e-filing](https://www.buildium.com/), resident and owner portals, all in one product | No property management features; coordinates whatever systems you run, including a property management platform like Buildium | | **Portfolio coverage** | [Residential, commercial, community association, student housing](https://www.buildium.com/) under one interface | Domain-agnostic; not specific to property at all | | **AI and automation** | [Agentic AI](https://www.buildium.com/blog/examples-of-automated-property-management-systems/) and automation baked into core features; states AI should be “a trusted extension of our platform, not a replacement for human judgement” | Human-in-the-loop as a callable step the agent invokes at decision points, comparable to a [digital andon cord](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents) | | **Governance** | Role-based controls and compliance features within the product | Available as a step-up: audit trails and RBAC from the [$299 a month Professional tier](https://flowrunner.ai/pricing), SSO and longer retention at the $999 Business tier, spanning every system the agent touches | | **Pricing** | Published tiers: [Essential $62, Growth $192, Premium $400 a month](https://www.buildium.com/pricing/); 14-day free trial | Published tiers from a Free plan and $5 Starter; [Growth $45 a month](https://flowrunner.ai/pricing) (AI agents, all integrations, human-in-the-loop, unlimited users and workflows); Professional $299, Business $999 | | **Maturity** | Established base, [Buildium Academy](https://www.buildium.com/), live phone support, years in residential PM | Newer entrant; horizontal platform, carries adoption risk in any single vertical | The table sorts the two products. The rest of this article is the part a table cannot carry: which problem you actually have, and which tool is built for it. ### What Buildium is genuinely good at [Buildium](https://www.buildium.com/) is a real, mature property management platform, and it describes itself as the “#1 most recommended property management software.” That is the vendor’s own framing, and the product depth behind it is the reason a property manager should take it seriously. A buyer evaluating it should know exactly where it is strong, because none of these strengths are in dispute. **It is a complete property management product in one place.** Buildium covers accounting, online rent collection, leasing, tenant screening, maintenance and operations, [1099 e-filing](https://www.buildium.com/), and resident and owner portals inside a single application. It serves single-family, small multifamily, community association, student housing, and commercial portfolios under one interface. FlowRunner provides none of that. There are no lease templates, no CAM charge tracking, no 1099 filing, no violation management in FlowRunner, because FlowRunner is not a property management system. If what you need is the product that runs the property, this is the column you are shopping in, and FlowRunner is not in it. **It has invested in onboarding and support.** Buildium points to [Buildium Academy](https://www.buildium.com/), onboarding specialists, and live phone support, alongside a self-reported 95 percent customer support satisfaction rating. Read the satisfaction number as the vendor’s own marketing claim, because it is. The underlying point still stands: a vertical product with years of customer success investment behind it carries a real advantage when an office manager who does not write software needs the thing to work on Monday. FlowRunner is a newer product, and the maturity gap is a fact to weigh, not a detail to bury. ![A single tidy product dashboard panel on a cream tile, labeled at the top Property Management System](https://flowrunner.ai/images/blog/flowrunner-vs-buildium-property-management-automation-1.webp) **It is building its own automation, and it is honest about the human role.** Buildium describes [agentic AI](https://www.buildium.com/blog/examples-of-automated-property-management-systems/) and automation as “already baked into many of its core features,” covering tasks like listing syndication, autopay and late-fee handling, owner statements, and maintenance work-order routing. It also states plainly that “AI should be a trusted extension of our platform, not a replacement for human judgement.” That is the right instinct, and it is worth pointing out directly rather than pretending Buildium’s automation is marketing veneer. It is not. The distinction FlowRunner draws is not whether Buildium keeps humans involved. It is about where the automation lives and what it can reach. That distinction has a name, and it is the rest of this article. ### The seam Buildium does not own Here is what most property management automation comparisons skip past. The hard part of a growing operation is not the work that happens inside any one product. Buildium runs the work that happens inside Buildium very well. The hard part is the work that happens in the gaps between systems. Think about where the time actually goes in an operation past a certain size: - A payment lands in the operating bank account that does not match any rent record cleanly, because a resident paid through a channel that did not sync, or paid two months at once, or paid short by a fee. - An owner emails asking why a distribution looks off, and answering means reconciling the property management ledger against the actual bank statement and the management agreement, three systems that do not talk. - A maintenance invoice arrives from a vendor by email, needs to be coded to the right property and the right GL account, and needs someone to decide whether it is within the approval threshold before it gets paid. - A new tool gets adopted, an insurance tracker, a separate accounting system the CPA prefers, a banking portal, and now there is one more island that someone has to ferry data to and from by hand. None of those are failures of the property management platform. They are the connective tissue between the property management platform and everything else. That is the seam. And the seam is exactly the kind of work that does not fit inside a vertical product, because a vertical product is built to own its own four walls, not to govern the traffic between yours. ![A horizontal pipeline diagram on a cream tile](https://flowrunner.ai/images/blog/flowrunner-vs-buildium-property-management-automation-2.webp) This is the category FlowRunner is built for. Orchestration as a service is a layer above the systems of record. It listens for what those systems emit, gathers the context that no single system holds on its own, attempts the mechanical part of the work, and calls a human in at the moments that need judgment, writing a structured record of who decided what across the whole chain. Buildium is one of the systems of record this layer sits above. It is not a competitor to the layer, any more than your accounting package competes with the person who reconciles it. For the broader framing of why this is a different category from in-product automation, the companion read is [the difference between AI automation and AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents). ### How the human moment differs in scope Both Buildium and FlowRunner believe a human belongs in the loop. Buildium says so directly, and means it. The difference is not philosophy. It is reach. Buildium’s automation keeps a human involved inside Buildium. A late-fee rule, a maintenance routing rule, an owner-statement schedule: these run inside the product, and when a person needs to weigh in, they weigh in on a Buildium screen about a Buildium object. That is the correct design for work that lives entirely inside the property management system, and it is genuinely useful. FlowRunner’s human-in-the-loop architecture treats a human as a callable step inside an agent that spans systems. The agent reconciling a bank deposit against the property management ledger does not stop at the edge of one product. It pulls the deposit from the bank feed, the expected charges from the property management system, the management terms from wherever they live, attempts the match, and when something does not reconcile, it pauses and routes a structured request to a named person through the channel they prefer. The person sees the deposit, the expected record, and the specific discrepancy in one message, decides, and the agent resumes with that decision captured in the audit trail. ![A single mobile chat message card on a cream tile](https://flowrunner.ai/images/blog/flowrunner-vs-buildium-property-management-automation-3.webp) The difference shows up in three concrete places: - **What the agent can see.** A human moment inside one product carries that product’s context. A human moment inside a cross-system agent carries context the property management system never had, the bank side, the GL side, the agreement. The reconciliation question is unanswerable from inside any one of those alone. - **What the audit trail records.** An in-product action records what happened in that product. An orchestration-step approval records the trigger, the context the agent gathered from every system it touched, the decision it would have made on its own, the person it routed to, the response, and the action taken across systems. That difference matters in an owner dispute or an audit conversation, not in a demo. - **What it can reach next quarter.** When you add the next tool, an in-product automation does not extend to it. An orchestration layer was built to add a system to the chain. That is the entire point of sitting above the systems rather than inside one. For more on keeping multi-step work coordinated as it grows, see [coordinating parallel workflow branches](https://flowrunner.ai/blog/synchronize-block-keeping-automations-in-order). This is not a feature gap with Buildium. Buildium is not trying to be the layer above your bank and your GL. It is trying to be an excellent property management product, and it is one. The two tools are aimed at different altitudes. ### Pricing, and who each one is for A direct dollar comparison is possible here, because both publish pricing. It is also slightly beside the point, because the two are not priced as substitutes. - Buildium publishes [tiered pricing](https://www.buildium.com/pricing/): Essential starting at $62 a month, Growth at $192 a month, and Premium at $400 a month, with a 14-day free trial and no credit card required. The entry tier is built for an individual property manager or a small portfolio that is “taking off.” That low barrier to entry is a genuine strength, and it is the right starting point for a small operator who needs a property management product and nothing more elaborate. - FlowRunner publishes pricing too. There is a Free plan and a $5 Starter, and the [Growth tier at $45 a month](https://flowrunner.ai/pricing) is the full core capability set at production volume: AI agents with BYOK, all integrations, human-in-the-loop across email, Slack, WhatsApp, and phone, unlimited users, and unlimited workflows. The [Professional tier at $299 a month](https://flowrunner.ai/pricing) adds governance (audit trails, RBAC) for operations teams that need formal controls, and the $999 a month Business tier adds SLA tracking, SSO and a longer audit retention window. The right tier is the one whose governance posture matches what the operation actually has to evidence, not the highest tier on the page. The honest read on price is that an individual landlord choosing a property management product is not the buyer FlowRunner is priced for, even at $45 a month, because FlowRunner is not a property management product. An operations leader who already runs a property management system, plus accounting, plus banking, plus a stack of other tools, and who is spending senior time stitching them together by hand, is. For the question of whether a given process is even worth automating in the first place, [this guide on how to decide what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) is the place to start before you price anything. ### Where Buildium is the stronger choice The brief for this article was explicit, and so is this section. There are concrete situations where Buildium is the better choice, and a comparison that hides them is not worth your time. - **You need a property management system and do not have one.** This is the most common case, and the answer is unambiguous. Buildium is a complete, [purpose-built platform](https://www.buildium.com/) with years of domain-specific features. FlowRunner does not provide accounting, leasing, screening, or any of it. If this is you, buy Buildium or a platform like it, and revisit FlowRunner when the work between systems becomes the bottleneck. - **You are a small operator who wants a low entry price and fast onboarding on a vertical product.** Buildium’s [$62 a month Essential tier](https://www.buildium.com/pricing/), 14-day trial, and [onboarding and support investment](https://www.buildium.com/) are built for exactly this buyer. FlowRunner’s orchestration focus is not, even at the $45 a month Growth tier; the issue is not price, it is that FlowRunner does not run your properties. - **You want the established, lower-risk vertical incumbent.** Buildium has brand recognition, a large customer base, and deep property management workflow fit that a newer horizontal platform cannot replicate through agents alone. Switching costs and accumulated trust are real, and for a buyer who weighs maturity heavily, that is a genuine point in Buildium’s column. We will not argue otherwise. These are not throwaway acknowledgments. They are decision-shaping facts. If your situation matches them, Buildium is the right call, and FlowRunner sits alongside it later, if at all. ### Where FlowRunner is the better fit Choose FlowRunner when the problem is not the property management product but the work around it: - You already run a property management platform like Buildium as your system of record, and the friction is in the gaps between it and your bank, your general ledger, your owners, and your other tools. - The work that eats your team’s time is cross-system: reconciling deposits against the ledger, coding and approving vendor invoices, answering owner questions that require pulling from three places, ferrying data to a tool that does not integrate. - Human approval is the moment the work’s value is decided. You want that approval to be a callable step inside an agent that can see every system involved: pause, route with full context, resume with the decision recorded. Not an alert inside one product, about one product. - You want governance, audit trails, RBAC, and SSO, that spans the whole chain of systems an agent touches, at a [price an operations director can authorize](https://flowrunner.ai/pricing) when the governance posture is actually required (Professional $299 a month for audit trails and RBAC, Business $999 a month for SSO and longer retention). - You expect AI agents to start arriving in your stack from multiple vendors, including from inside your property management platform, and you want a coordination layer above them, with humans in the loop, before that becomes the next integration mess. On why that coordination is harder than it looks, [this is what breaks automations in production](https://flowrunner.ai/blog/the-truth-about-building-automations). The pattern is the same one FlowRunner runs in any operation: ingest the trigger, gather context across systems, attempt the mechanical match, call a human on the exception, record the decision. Property is just one place the pattern applies. If you are still deciding which vertical product to standardize on, the [property management automation software overview](https://flowrunner.ai/solutions/property-management-automation-software) frames that choice, and the companion comparison [FlowRunner vs AppFolio](https://flowrunner.ai/blog/flowrunner-vs-appfolio-property-management-automation) runs this same fit question against a different incumbent. ### How to decide Two questions settle this, and you can answer both in a few minutes. **Question one: do you already have a property management system?** If the answer is no, stop here. You need a property management product first, and Buildium is a strong, complete, well-supported choice with a low entry price. FlowRunner is not a substitute and will not run your properties. Come back to it when the systems around your property management platform start to multiply. **Question two: where do your exceptions live?** Look at the last ten things that pulled a senior person off their real work. If they happened inside your property management product, a late fee, a renewal, a maintenance request, then in-product automation, including Buildium’s own, is the right tool, and it is improving. But if they happened in the gaps, a deposit that did not match the ledger, an owner question that needed three systems to answer, an invoice that needed coding and an approval before money moved, that is the seam no single product owns. That is the digital andon cord an operations leader actually wants: the system reaches the deposit it cannot reconcile on its own, pulls up, and asks a named person before anything posts, instead of either guessing or quietly leaving it for a human to find later. Both tools are real answers to real questions. Buildium is the property management product, and a good one. FlowRunner is the orchestration layer above it, built so the human decision is a first-class step that can see across every system involved. The question is not which is better. It is whether the work that is costing you lives inside one product or in the space between them, and whether you have a system of record yet at all. ### Quick answers #### Does FlowRunner replace Buildium? No. FlowRunner is not a property management application and does not provide accounting, leasing, tenant screening, maintenance tracking, 1099 e-filing, or resident and owner portals. Buildium does all of that in one product. If you need a property management system, you need Buildium or a platform like it. FlowRunner is the orchestration layer that sits above your property management system and the rest of your stack, coordinating work across them with a human in the loop on exceptions. #### Is FlowRunner cheaper than Buildium? At entry tiers, FlowRunner is cheaper, but the comparison is not apples to apples. Buildium starts at $62 a month for its Essential plan, built for an individual property manager or small portfolio. FlowRunner has a permanent Free plan and a $5 Starter, and its Growth tier at $45 a month includes AI agents with BYOK, all integrations, human-in-the-loop across email, Slack, WhatsApp, and phone, and unlimited users and workflows. The two are priced for different buyers, not as substitutes for the same job. If you need formal governance (audit trails, RBAC), FlowRunner’s Professional tier is $299 a month. #### Can FlowRunner connect to Buildium? FlowRunner is designed to coordinate work across whatever systems an operation already runs, with a property management platform like Buildium as the system of record. The honest pattern is to keep Buildium as the place property data lives and use FlowRunner to handle cross-system exception routing, reconciliation between Buildium and the bank or general ledger, and human escalation. We do not claim a prebuilt, one-click Buildium connector today; treat integration as a build, not a checkbox. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Celigo for QuickBooks API Automation: An Honest CFO Read Source: https://flowrunner.ai/blog/flowrunner-vs-celigo-quickbooks-api-automation Comparisons May 28, 2026 Updated September 7, 2026 13 min read Celigo is a mature iPaaS with 1,000+ connectors and JavaScript hooks for developers. FlowRunner is for finance teams who own QuickBooks workflows without one. ![A bald cartoon man with ink-smudged hands standing between a slate-blue industrial control panel of hand-set switches spilling punched tape on his left and a compact sage-green dashboard with a single AP Agent tile and an amber pause button on his right.](https://flowrunner.ai/images/blog/flowrunner-vs-celigo-quickbooks-api-automation-hero.webp) If you are a finance leader pricing out how to automate the QuickBooks work that has outgrown your team, the question that actually decides the platform is not which one connects to QuickBooks. They both do. The question is who is going to be on call the day the integration needs to change. Celigo and FlowRunner answer that question differently, and the right answer depends on whether you have a developer in the building. Celigo is a mature, top-ranked integration platform, and for a team with engineering on call it is a serious choice. FlowRunner is built for the finance or operations team that wants to own the workflow itself. Most comparison content in this category picks a winner and hides the trade-off. This one names it. ### Celigo and FlowRunner at a glance | | Celigo | FlowRunner | | --- | --- | --- | | **Category** | Intelligent automation platform / iPaaS, [self-described](https://www.celigo.com/platform/) as “#1 ranked iPaaS for two years” | Orchestration layer for work and AI agents across the tools finance and ops already run | | **Primary buyer** | IT team enabling business users across ecommerce, CRM, and ERP | Finance or operations leader who wants to own a QuickBooks-centered agent | | **Building model** | Prebuilt integration templates plus [JavaScript hooks](https://docs.celigo.com/hc/en-us/articles/360038887192-Hooks-overview) for custom transformation logic | Agents and flows configured visually, with human-in-the-loop as a callable step | | **Connector breadth** | Broad and mature; [1,000-plus prebuilt connectors and templates](https://www.celigo.com/platform/) | Narrower today; concentrated on integrations finance and ops invoke from QuickBooks-centered work | | **Custom logic ownership** | JavaScript hooks, written and maintained by a developer | Visual configuration, owned by the team that owns the workflow | | **Human approval** | AI exception management plus workflow design; Celigo states [AI resolves “95% of errors automatically”](https://www.celigo.com/platform/) | Pause-and-ask as a callable action; routes to a named approver via [Slack, email, WhatsApp, or phone](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack), captures the response, resumes | | **Governance** | Role-based access and compliance features in higher [editions](https://www.celigo.com/pricing/) | Audit trails and RBAC at the [Professional tier ($299 a month); SSO at the Business tier ($999 a month)](https://flowrunner.ai/pricing) | | **Pricing posture** | [Endpoint-and-flow based](https://www.celigo.com/pricing/), not per transaction; dollar figures not public, sold through sales | Published tiers; core platform at Growth ($45 a month); governance overlay at Professional ($299 a month) | | **Maturity** | Established enterprise base and [active community](https://docs.celigo.com/hc/en-us/community/topics) | Newer entrant; [smaller customer base](https://flowrunner.ai/about), carries adoption risk | The table sorts the two platforms. The rest of this article is the part a table cannot carry: when each answer is the right one. ### What “QuickBooks API automation” actually means for a finance buyer Here is what most QuickBooks API comparisons skip past. The buyer searching for this is rarely the developer who wants OAuth scopes and endpoint references. The buyer is the controller or VP of Finance at a company where the QuickBooks-centered work has grown faster than the headcount, and someone said “there’s an API, we can automate this.” There is an API. Intuit publishes the [QuickBooks Online Accounting API](https://developer.intuit.com/app/developer/qbo/docs/develop), and it is the same interface both Celigo and FlowRunner call into to read and write bills, invoices, and journal entries. That is the part worth saying plainly, because it dissolves the most common false premise in this category: the platform choice is not decided by who has better QuickBooks API coverage. Both reach the same API. The choice is decided by who owns the workflow logic that sits on top of it. ![A single horizontal pipeline diagram](https://flowrunner.ai/images/blog/flowrunner-vs-celigo-quickbooks-api-automation-1.webp) The work itself is consistent enough across finance teams to predict. In [conversations with finance leaders](https://midnightflow.ai/portfolio/scale-your-cfo-practice-without-hiring/), the same pattern surfaces: - Bills arrive by email, by vendor portal, and sometimes on paper. AP entry happens manually when staff are out, which means senior finance occasionally enters bills directly. - Distributor billbacks get reconciled across portals and spreadsheets. Duplicate payment is a recurring worry, because the company and the distributor can both pay the same charge and nobody catches it for weeks. - Stripe payouts do not line up cleanly with QuickBooks invoices. The matching is mechanical, but the exceptions (the payment short by a fee, the refund that did not propagate) are the actual work. - Recurring journal entries get drafted in Excel and uploaded monthly. The rule almost never changes once documented. That is the work both platforms are bidding on. Not “automate QuickBooks” in the abstract. Automate AP intake, reconciliation, and journal entry drafting around QuickBooks, with the controller in the loop on the exceptions and out of the loop on the routine. ### What Celigo is genuinely good at [Celigo](https://www.celigo.com/platform/) is a mature integration and automation platform, and it describes itself as the “#1 ranked iPaaS for two years.” That is the vendor’s own framing, and the product depth behind it is real. A finance buyer evaluating it should know exactly where it is strong. **Connector breadth.** Celigo advertises [1,000-plus prebuilt connectors and templates](https://www.celigo.com/platform/) spanning ERP, CRM, ecommerce, and HR systems. For QuickBooks specifically, Celigo ships [prebuilt integration templates](https://www.celigo.com/integrations/quickbooks/) it describes as “pre-packaged flows to jump start your integrations,” covering ecommerce, CRM, and payment syncs with tools like Shopify, Salesforce, and NetSuite. If you need broad out-of-the-box coverage across many SaaS tools, this is a genuine advantage, and FlowRunner does not match the breadth of that catalog today. **Developer-grade customization through JavaScript hooks.** This is the heart of Celigo’s model and a real strength for the right team. Celigo’s integrator.io exposes [JavaScript hooks](https://docs.celigo.com/hc/en-us/articles/360038887192-Hooks-overview) that let developers customize how data is transformed and how a flow behaves, with support for modern JavaScript including object destructuring and spreading following the V8 engine upgrade in the 2026 release line. For a team that needs precise, low-level control over transformation logic, hooks give a developer exactly that. This is not a weakness to be argued away. It is a deliberate design choice that fits teams who have someone to write and maintain the code. **An established base and an active community.** Celigo has a real enterprise customer base and an [active community](https://docs.celigo.com/hc/en-us/community/topics) where product staff engage directly on technical questions. That matters. When a flow breaks at month-end, the difference between a vendor with a responsive community and a vendor without one is the difference between a fixed evening and a lost one. FlowRunner is a newer product, and the maturity gap is a fact a buyer should weigh, not a detail to bury. The honest read most comparison posts in this category will not say out loud is that none of those strengths are in dispute. The comparison is not Celigo versus a strawman. It is a fit question, and the fit question has a name: do you have a developer on call to write and maintain the hooks? ### Where the hook model and the orchestration model diverge The cleanest way to understand FlowRunner against Celigo is to separate the two customization models without arguing about which is smarter. They are different answers to “how does the workflow do something the prebuilt template does not.” **The hook model.** A Celigo flow handles the common case through a template. When the flow needs to do something custom, transform a field, filter a record, branch on a condition the template does not cover, a developer writes a JavaScript hook. The fluency required is “what does this hook receive, what should it return, how do I keep it working when the data shape changes.” That fluency lives in a developer. For a finance team with engineering on call, that is a clean division of labor. For a finance team without one, the hook is the moment the automation stops being theirs. Every future change routes back through a person who has other priorities. **The orchestration model.** A FlowRunner agent owns a workflow end to end. The builder configures what the agent does, the agent gathers context from the systems it needs, and the agent invokes a human as a callable action when the situation warrants it. The fluency required is “what does this workflow do, when does it pause, who decides.” That fluency lives in the finance or operations team that owns the work. The custom logic is configured visually rather than scripted, so the person who owns the workflow can change it without filing a ticket. The shape of the unit itself is also different, and the difference matters for orchestration. A Celigo flow follows the standard iPaaS contract: the flow starts with a trigger (a scheduled run, a webhook, a record event), and the steps after it run in response. A FlowRunner flow can start with any node type, an action, a trigger, a condition, or a group. That means a flow can be defined once and invoked as a callable subflow by another flow (treated as a function in someone else’s workflow), run manually when the controller wants to kick it off, or started by a condition evaluating business state rather than waiting for an event. The same building block becomes a routine, a tool another agent can call, and a step inside a larger orchestration. That compositional flexibility is the practical shape an orchestration layer takes when you sit it above the systems of record rather than between two of them. The distinction matters most for QuickBooks-centered work because the exceptions are not edge cases. They are the work. A workflow that posts the matched bills and pauses for a named approver on the ones that need judgment is exactly what finance leaders describe wanting. The agent does the routine. The human owns the exception. The audit trail records both. The category that owns this pattern is orchestration as a service: a layer above the systems of record that listens for what they emit, gathers the context those systems do not share, calls a human at the moments that need judgment, and writes a structured record of who decided what across the whole workflow. A hook is a developer customization point inside a flow. It is not a place where the judgment call between systems lives, because that judgment is not a transformation, it is a decision. An orchestration layer is built to own that decision, and FlowRunner is built for that layer. For the deeper category framing, the companion read is [FlowRunner vs Workato for QuickBooks API automation](https://flowrunner.ai/blog/flowrunner-vs-workato-for-quickbooks-api-automation), which works the same fit question against a different competitor. ### How human review shows up in each platform The two platforms treat the human moment as different shapes, and the shape is decisive for finance work. FlowRunner’s human-in-the-loop architecture treats human reviewers as callable steps at decision points. Celigo treats the same moments primarily as errors for the platform to resolve. That single architectural difference drives most of what follows. Celigo approaches exceptions primarily as errors to be resolved, increasingly with AI. The platform’s positioning leans hard on automation that keeps running: Celigo states that its [AI exception management resolves “95% of errors automatically.”](https://www.celigo.com/platform/) Read that as the vendor’s own marketing claim, because it is, and read the intent behind it clearly. The goal is to keep the flow running with as little human interruption as possible. For a high-volume ecommerce sync where most exceptions are genuinely mechanical (a malformed SKU, a timeout, a retryable failure), that is the right goal. ![A single mobile chat message card on a cream tile](https://flowrunner.ai/images/blog/flowrunner-vs-celigo-quickbooks-api-automation-2.webp) FlowRunner approaches the human moment as a decision the workflow is designed to route, not an error to suppress. The agent invokes a human as a callable step inside its own orchestration: pause the workflow, route a structured request to a named approver through the channel they prefer ([Slack, email, WhatsApp, or phone](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack)), wait for the response, then resume with the decision captured in the audit trail. The same pattern powers [QuickBooks bill approval in Slack](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack), [Stripe-to-QuickBooks reconciliation with mismatches paused for review](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack), and [inbox-to-QuickBooks invoice processing with duplicate detection](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack). The architectural difference shows up in three concrete places: - **The intent.** Celigo’s exception management is built to resolve and keep running. FlowRunner’s human-in-the-loop is built to pause and ask. In finance, the bill without a matching PO and the suspected duplicate payment are not errors to auto-resolve. They are exactly the decisions you want a human to make before money moves. - **What the audit trail records.** Auto-resolution records that the system fixed it. An orchestration-step approval records the trigger, the context the agent gathered, the decision the system would have made on its own, the human it routed to, the response, and the action taken. That difference matters in an audit conversation, not in a demo. - **Who owns the rule.** Deciding which cases pause for a human, in FlowRunner, is workflow configuration the finance team owns. Tuning exception handling in a hook-based flow is, for the custom cases, developer work. The owner of the pause rule should be the person accountable for the money. That is the controller, not the engineer. This is not a feature gap with Celigo. It is a different starting point. Celigo built a platform to keep automations running across a broad SaaS estate. FlowRunner built a platform to put a human in the loop at the decision points finance cannot delegate to a retry. ### Pricing, and who can authorize the spend A direct dollar comparison is not fully possible, because Celigo does not publish tier pricing. What can be said honestly: - Celigo uses [endpoint-and-flow-based pricing](https://www.celigo.com/pricing/), and it makes a real virtue of this: “Pay for endpoints and flows, not per task or transaction,” with “no overage fees.” That is a genuinely cleaner metering model than per-task pricing, and for a buyer burned by execution-count surprises it is worth weighing. The dollar figures are not public; pricing runs through a sales conversation, and the compliance and access features sit in higher editions. - FlowRunner publishes pricing. The core platform (AI agents, human-in-the-loop across email, Slack, WhatsApp, and phone, all integrations, BYOK, unlimited users and workflows) is at the Growth tier at [$45 a month](https://flowrunner.ai/pricing). The Professional tier at $299 a month adds governance (audit trails and RBAC); SSO is at the Business tier at $999 a month. No developer work required to turn governance on. ![A simple pricing-and-governance dashboard panel on a cream tile](https://flowrunner.ai/images/blog/flowrunner-vs-celigo-quickbooks-api-automation-3.webp) The point is not that one is cheaper. Without a Celigo quote, that comparison cannot be made. The point is who can authorize the spend. A finance director with a software budget can sign for the Growth tier at $45 and have a working QuickBooks-centered agent the same afternoon, or step up to the Professional tier at $299 when audit trails and RBAC are required. An iPaaS contract sold through enterprise procurement, with dollar figures behind a sales call, is a different motion that pulls in IT and frequently legal. For a finance-led purchase, that difference is decisive. For an IT-led integration program, it is irrelevant. Match the platform to who is signing. ### Where Celigo is the stronger choice The brief for this article was explicit, and so is this section. There are concrete situations where Celigo is the better platform, and a comparison that hides them is not worth reading. - **You need broad, out-of-the-box connector coverage today.** Celigo’s [1,000-plus connector library](https://www.celigo.com/platform/) is larger and more mature than FlowRunner’s. If your automation depends on prebuilt connectors across a wide SaaS estate, especially deep ecommerce and ERP coverage, Celigo is built for that and FlowRunner is not yet. - **You have developers and want low-level transformation control.** If your team needs precise scripting control over how data is transformed, and you have engineers to write and maintain [JavaScript hooks](https://docs.celigo.com/hc/en-us/articles/360038887192-Hooks-overview), Celigo gives you that depth directly. FlowRunner’s visual model is a deliberate trade: easier for the finance team to own, less granular than raw code. - **You want the lower-risk, established platform.** Celigo has years in market, an enterprise customer base, and an [active support community](https://docs.celigo.com/hc/en-us/community/topics). FlowRunner is newer and carries the adoption risk any earlier-stage product carries. For a buyer who weighs maturity heavily, that is a real point in Celigo’s column, and we will not argue otherwise. These are not throwaway acknowledgments. They are decision-shaping facts. If your situation matches them, Celigo is the right call. ### Where FlowRunner is the better fit Choose FlowRunner when: - The work that needs automating is QuickBooks-centered AP, reconciliation, invoice intake, or journal entry drafting, with the controller in the loop on exceptions. - The team that will own the build is finance or operations, and they do not have a developer on call to write and maintain hook scripts. They want to own and change the workflow themselves. - Human approval is the moment the workflow’s value is decided, and you want that approval to be a callable step inside the agent (pause, route with context, capture the decision) rather than an error the platform tries to auto-resolve. - You want to start the core platform at a price a finance director can [authorize](https://flowrunner.ai/pricing) (Growth at $45 a month) without an enterprise procurement cycle, and add governance (audit trails and RBAC at Professional, $299 a month) or compliance posture (SSO at Business, $999 a month) when the audit conversation requires it. - You expect AI agents to start arriving in your stack from multiple vendors and you want a coordination layer above them, with humans in the loop, before that becomes the next integration mess. The same orchestration shape extends past QuickBooks. The pattern for [reconciling Stripe against open QuickBooks invoices](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack), [parsing inbox documents into QuickBooks with duplicate detection](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack), or [running equivalent AP work on Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) is identical: ingest the trigger, gather context across systems, attempt the mechanical match, call the human on the exception, record the decision. ### How to decide Two axes settle this, and you can place yourself on both in about five minutes. **Axis one: who is on call when the integration needs to change?** Walk through what happens when a flow needs new transformation logic next quarter. If the answer is “we file it with engineering and they write the hook,” Celigo fits that division of labor cleanly, and you get a mature platform and a deep connector catalog with it. If the answer is “there is no engineering team, and the finance ops person who owns this needs to change it themselves,” that is the constraint FlowRunner was built around. The constraint here is organizational, not technical, and it is the one that actually predicts which platform a finance team can live with. **Axis two: what shape is the human moment?** Look at the last five exceptions in your AP or reconciliation process. Were they mechanical failures you would happily have a system auto-resolve to keep the flow running? Celigo’s exception-management model is built for that. Or were they judgment calls (a new vendor with no history, a suspected duplicate payment, a bill that should have a PO and does not) where the right behavior is to stop the line and ask a named person before anything posts? That second shape is the digital andon cord finance leaders describe wanting, where the system pulls up and says it needs a human rather than plowing on. That is what a callable human-in-the-loop step is for. Both platforms are real answers to real questions. Celigo is the proven, broad, developer-extensible iPaaS. FlowRunner is the orchestration layer a finance team can own without an engineer, built so the human decision is a first-class step rather than an error to suppress. The question is not which is better. It is which buyer is doing the asking, and whether there is a developer standing behind them. ### Quick answers #### Is FlowRunner a replacement for Celigo? Not for an IT team that has standardized on Celigo as its integration backbone across ecommerce, CRM, and ERP. Celigo is purpose-built for that, and it is a mature, top-ranked iPaaS. FlowRunner is for the finance or operations leader who wants to own a QuickBooks-centered AP, reconciliation, and approval agent without writing JavaScript hooks or depending on the team that maintains them. #### Does FlowRunner have as many connectors as Celigo? No, and we will not pretend otherwise. Celigo advertises 1,000-plus prebuilt connectors and templates built over many years. FlowRunner focuses on the integrations finance and operations teams actually invoke from QuickBooks-centered workflows. If your decision turns on the widest possible out-of-the-box catalog, Celigo wins that axis. #### Do I need a developer to automate QuickBooks with Celigo? Not for prebuilt integration templates, which are designed for business users. You do need developer ownership when a flow needs custom data transformation, because Celigo’s customization layer is JavaScript hooks that someone has to write and maintain. FlowRunner’s model keeps that logic in a visual builder, so the finance team that owns the workflow can own the changes too. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Coupa for Vendor Onboarding Software: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-coupa-vendor-onboarding-software Comparisons May 28, 2026 Updated September 7, 2026 13 min read Coupa onboards vendors inside a full spend management suite. FlowRunner orchestrates onboarding across the systems you already run. How to decide between them. ![A bald cartoon man at a reception desk holding a single vendor folder of loose onboarding papers in amber, with a vast green multi-window service hall looming behind him representing Coupa's full spend suite, weighing the small immediate task against the large system.](https://flowrunner.ai/images/blog/flowrunner-vs-coupa-vendor-onboarding-software-hero.webp) When a procurement team shortlists [Coupa](https://www.coupa.com/products/) for vendor onboarding, they are rarely buying vendor onboarding. They are buying a spend management suite, and onboarding is the wedge that gets it in the door. That is not a criticism of Coupa. It is the entire logic of the product, and for a large procurement organization it is often the correct logic. The mistake is treating “what onboarding tool should we buy” and “should we standardize our entire spend stack on one vendor” as the same question. They are not, and the answer to the second one decides the first. ### The two products at a glance | | Coupa | FlowRunner | | --- | --- | --- | | **Category** | [AI Total Spend Management](https://www.coupa.com/platform/ai/) suite, longstanding Business Spend Management (BSM) platform | Orchestration layer for work and AI agents across the systems a team already runs | | **Primary job** | Unify sourcing, procurement, invoicing, expenses, payments, and supplier risk in one platform | Coordinate vendor onboarding, document collection, validation, and exception handling across existing systems | | **Vendor onboarding role** | One capability inside [Supplier and Risk Management](https://www.coupa.com/blog/optimize-supplier-onboarding-coupa-supplier-and-risk-management/), “the original entry point for engaging with a supplier” | The whole job: orchestrate intake, validation, screening, enrichment, and approval routing across the tools that hold vendor data | | **Built for** | Enterprises standardizing their spend stack on a single suite | Mid-market procurement and ops teams orchestrating across the systems they keep | | **Supplier risk and sanctions screening** | Native, inside the suite (OFAC, InfoSec, ABAC, GDPR, ESG monitoring) | Orchestrated by calling the screening source you choose, then routing flagged hits to a human | | **AI framing** | AI-native spend management, touchless processing and fraud detection across source-to-pay | Agents as first-class workflow nodes that invoke humans as callable tools when they hit uncertainty | | **Integration model** | [App Marketplace](https://www.coupa.com/integrations-app-marketplace/) of pre-certified partner apps, connects to up to 160 ERPs | Agent Factory builds new connectors in roughly thirty minutes; broad orchestration substrate | | **Ecosystem maturity** | Deep and curated, years of partner investment | Agent Directory is earlier stage, partner network narrower | | **Enterprise credibility** | [Leader in the 2026 Gartner Magic Quadrant for Source-to-Pay Suites](https://www.coupa.com/newsroom/coupa-named-a-leader-in-the-2026-gartner-magic-quadrant-for-source-to-pay-suites/), network of 10M+ buyers and suppliers | Newer platform, carries the adoption risk that comes with that | | **Pricing surface** | Enterprise, quoted through sales, scoped to modules and spend volume | Core platform (AI agents, all integrations, human-in-the-loop, unlimited users/workflows) at [Growth ($45/mo)](https://flowrunner.ai/pricing); governance tooling (audit trails and RBAC) at Professional ($299/mo); SSO at Business ($999/mo) | | **Time to first workflow** | Weeks to months as part of a suite deployment | Days to weeks, configured visually by non-developers | The table covers the scope difference. The rest of this article is the part the table cannot: which buyer you are, and where the actual onboarding work lives once the demo is over. (For the same comparison against a different incumbent, see the sibling piece [FlowRunner vs SAP for vendor onboarding software](https://flowrunner.ai/blog/flowrunner-vs-sap-vendor-onboarding-software).) ### Who Coupa actually serves, said honestly Coupa describes itself as the [#1 AI Total Spend Management platform](https://www.coupa.com/platform/ai/), an AI-native ecosystem spanning procurement, finance, and supply chain operations. The longer-standing category term for what Coupa built is Business Spend Management: one suite that unifies sourcing, procurement, invoicing, expenses, payments, and supplier management rather than stitching point tools together. Vendor onboarding lives inside that suite, specifically inside [Coupa Supplier and Risk Management](https://www.coupa.com/blog/optimize-supplier-onboarding-coupa-supplier-and-risk-management/), which Coupa frames as “the original entry point for engaging with a supplier.” This is a credible and well-earned position. Coupa was named a [Leader in the 2026 Gartner Magic Quadrant for Source-to-Pay Suites for the third consecutive year](https://www.coupa.com/newsroom/coupa-named-a-leader-in-the-2026-gartner-magic-quadrant-for-source-to-pay-suites/), and it brings a network of more than 10 million buyers and suppliers onto one platform. For a CFO or a head of procurement at a large enterprise, that track record is not marketing noise. It is exactly the kind of signal that de-risks a nine-figure spend decision. Buyers consolidating their spend stack onto a single suite are doing something deliberate, and Coupa is one of the strongest answers to that specific goal. Inside the suite, the onboarding capability is genuinely deep. Coupa’s supplier management centralizes supplier information (tax IDs, certifications, insurance, banking details) and automates compliance checks across domains like OFAC sanctions, InfoSec, anti-bribery and anti-corruption, GDPR, and ESG monitoring, with a self-service portal where suppliers upload documents and maintain their own profiles. The decisive part is that risk and performance monitoring sit in the same platform as purchasing and invoice management. The vendor you onboard is the vendor you transact with, scored against the spend data the suite already holds. That data continuity is the part no horizontal tool replicates for free, and it deserves to be named plainly rather than waved past. ![a clean illustrative depiction of a single unified supplier record inside a spend suite, showing tax ID, certification status, banking detail, and a risk score badge all on one screen, conveying the suite-as-one-platform model](https://flowrunner.ai/images/blog/flowrunner-vs-coupa-vendor-onboarding-software-1.webp) ### The honest case for Coupa, before the contrast A comparison that only lists where FlowRunner wins is marketing, not analysis. Here is where Coupa is the stronger choice, stated without hedging. - **Native supplier risk and sanctions screening.** Coupa ships third-party risk scoring and sanctions screening inside the suite. If your selection criteria require turnkey OFAC, InfoSec, and ABAC screening with monitoring built in, Coupa delivers that natively. FlowRunner orchestrates screening by calling an external source; it does not ship a native risk engine. - **Suite depth FlowRunner does not have.** Coupa is a full Business Spend Management platform. Sourcing, procurement, invoicing, expenses, payments, treasury. FlowRunner is not a spend management suite and does not pretend to be one. It has no native invoicing, sourcing, or expense modules. - **A mature, curated integration ecosystem.** Coupa’s [App Marketplace](https://www.coupa.com/integrations-app-marketplace/) offers pre-certified partner apps and connects to existing IT landscapes including up to 160 ERPs. That is years of ecosystem investment. FlowRunner’s Agent Directory is earlier stage and its partner network is narrower today. - **Enterprise brand trust.** Coupa carries the analyst recognition and the install base that large procurement and finance buyers weight heavily. FlowRunner is a newer platform and [carries the adoption risk](https://flowrunner.ai/about) that comes with being newer. A buyer who ranks vendor maturity as a top criterion should weigh that honestly. If those four points describe your buying criteria, the rest of this article is interesting but not decisive. Coupa is a real answer to a real goal. The comparison gets useful when the goal is different. ### The part most vendor onboarding comparisons skip: the form is not the work The standard way to evaluate vendor onboarding software is to score intake portals and screening checklists against each other and pick the highest total. That framing quietly assumes the hard part is collecting the documents. It is not. The hard part is everything that happens when a submission does not cleanly pass. Walk through a real onboarding rather than a feature grid: - A vendor sends a W-9, a certificate of insurance, and banking details across three emails over a week, none of them in the format the portal expected. - The submitted tax ID looks like a near-match to an existing vendor in the ERP, same legal entity, different DBA, so it is probably a duplicate but not certainly one. - The remit-to bank account on the form differs from the account already on file for what might be the same supplier, which is either a legitimate update or the exact signature of payment fraud. - The category and spend threshold push approval routing down a conditional path the rule was never quite built to handle gracefully. - The COI expires three weeks after go-live, the AP system keeps paying, and nothing notices because the certificate is sitting in a folder. ![a vendor onboarding flow diagram where most steps proceed automatically in sage green (intake parse, tax ID check, COI date check) but one branch diverges to an amber decision node labeled "probable duplicate / banking change" that routes to a person icon, illustrating exceptions as the real work](https://flowrunner.ai/images/blog/flowrunner-vs-coupa-vendor-onboarding-software-2.webp) A clean intake form handles none of that. Screening flags some of it. What none of the checklist tooling resolves is the judgment call: is this a duplicate, is this banking change safe, who decides, and with what context in front of them. Coupa handles a great deal of the mechanical and screening work well, inside its suite, against its data. The judgment calls still land on a procurement coordinator’s desk regardless of which vendor you buy. Here is what most comparison posts on this topic will not say: exceptions are not the failure case of vendor onboarding. Exceptions are the work. The automation handles what is not the work. So the question that actually separates these two products is not whose intake form is nicer. It is what each one does at the moment a submission stops being routine. ### How the suite handles exceptions versus how an orchestration layer handles them Inside a suite, an exception is typically a status. The record stalls in a state, a queue, an approval step, and a human is notified that something needs attention. They open the platform, navigate to the record, reconstruct what tripped, and decide. That model is coherent and it works. It also assumes the human lives inside the suite and that everything the decision needs already lives there too. [FlowRunner](https://flowrunner.ai/) treats the exception differently, because it sits at a different point in the stack. FlowRunner is an orchestration layer above the systems of record, and it treats AI agents as first-class workflow nodes rather than features bolted onto a platform. For vendor onboarding, the shape looks like this: 1. A trigger fires: a supplier form submission, a vendor email landing in a shared mailbox routed through a parser, an event in the contract system. 2. The workflow gathers context across the systems involved: the ERP’s existing vendor list, the contract repository, the parser’s output, the internal approval matrix. 3. The mechanical validation runs: does this tax ID match an existing vendor under a different name, does the COI date check out, is this a banking change on an existing record, does the remit-to address match a known-bad pattern. 4. If everything ties, the workflow writes the vendor into the ERP and posts notifications with no human involved. 5. If anything is ambiguous, the workflow pauses and calls a named procurement reviewer in Slack with the full context attached and a structured response, then resumes from their answer with the decision captured against the audit trail. ![a Slack message preview where a workflow agent pings a named procurement reviewer about a vendor with a mismatched tax ID, showing the attached context (submitted W-9, the near-match existing vendor record, the differing remit-to account) and two structured response buttons, approve and reject](https://flowrunner.ai/images/blog/flowrunner-vs-coupa-vendor-onboarding-software-3.webp) Step five is the whole point. FlowRunner’s agents invoke a human as a callable tool when they hit uncertainty, not as a status that fires on every threshold breach. The human does not go to the work. The work, with its context already assembled, comes to the human in the channel they already use. This is the same shape documented in [vendor validation against Acumatica before bill creation](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and [cross-referencing vendor history in NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack), and it is the difference our prospects describe as a [digital andon cord](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents): the line stops itself when the agent hits something it should not decide alone. The seam this exposes is the one worth naming. A suite owns the front door and asks the rest of your stack to come inside. But vendor data does not live in one place for most teams. It lives in the ERP, the AP system, a parser, a contract tool, and a shared inbox, and the onboarding decision has to reconcile across all of them. That reconciliation, and the human judgment threaded through it, is not a feature any single system of record will ever fully own. It is a layer above them. A system that listens for what those tools emit, assembles the context a static approval rule never had, and pulls a named human in at the moments that need a decision. That layer is the category. FlowRunner is built for it. Coupa, by design, is built to be the system you bring the work into, not the layer that coordinates the systems you keep. ### Connectors in thirty minutes, governance a procurement director can authorize Two practical choices separate FlowRunner from the suite model, and both matter to a team that has been told to ship an onboarding workflow this quarter. The first is connector velocity. FlowRunner’s Agent Factory lets an operations person compose a workflow that combines a parser, an ERP write, a screening call, a Slack approval, and an audit trail without filing an engineering ticket. When the workflow needs to reach a system the catalog does not already cover, a new connector is built in roughly thirty minutes. The bar to add a new exception path is low enough that the process evolves as your vendor mix evolves, instead of waiting on a partner-ecosystem release cycle or an implementation milestone. Coupa’s App Marketplace is broad and pre-certified, which is a real strength for teams that want partner-supported integrations they do not maintain. FlowRunner’s bet is different: build the exact connector you need, fast, and own it. The second is published pricing across the stack. FlowRunner’s [core platform](https://flowrunner.ai/pricing) (AI agents, all integrations, human-in-the-loop across email, Slack, WhatsApp, and phone, unlimited users and workflows) is available from the Free plan up, with the Growth tier at $45 per month as the production anchor. Teams that need governance tooling step up to the Professional tier at $299 per month for audit trails and role-based access control. SSO and SAML sit one tier up at Business ($999 per month). The framing matters and I will keep it honest: $45 buys the platform that does the onboarding job, $299 adds the governance overlay a mid-market procurement operation typically wants when audit trails become a real requirement, and SSO is a Business-tier line item. None of this buys a native supplier risk engine, and none of it is a claim of certification against any named compliance framework. Coupa’s pricing is enterprise, quoted through sales, scoped to the modules and spend volume you license. These are not the same kind of purchase, and a buyer should not pretend they are. What FlowRunner offers is a platform a director can start using directly, and a governance step they can authorize directly when they need it, neither requiring an enterprise procurement cycle. ### Where FlowRunner is the stronger fit Choose FlowRunner when: - Your vendor onboarding work lives across email, a parser, Slack, the ERP (NetSuite, Acumatica, QuickBooks Online, or others), and a contract repository, and the coordination across those systems is the actual bottleneck. - You are not trying to consolidate your entire spend stack onto one suite, and adopting a full BSM platform to fix an onboarding workflow would mean buying a much larger problem than the one you have. - You need a procurement coordinator or ops analyst to build and adjust exception paths without waiting on engineering or a partner release cycle. - You want intelligent human-in-the-loop where agents pause and route to a named reviewer with full context, rather than every threshold breach surfacing as a status in a queue someone has to go check. - You want a published-pricing path you can authorize directly: the core platform at Growth ($45/mo) for the onboarding workflow itself, with the option to step up to Professional ($299/mo) for audit trails and RBAC when governance becomes a real requirement, and Business ($999/mo) for SSO and SAML, rather than an enterprise contract. - You see AI agents arriving in your stack from multiple vendors (extraction agents, supplier-side AI, approval bots) and you want a coordination layer above them before that fragmentation becomes its own problem. The two products can also coexist. A team standardizing the regulated core of its spend on Coupa can still use an orchestration layer for the onboarding edges that touch systems outside the suite: a niche parser, a legacy ERP Coupa is not the system of record for, a Slack-based approval the procurement team actually lives in. The structured spend management flows through Coupa. The cross-system coordination flows through FlowRunner with the context already attached. ### How to decide: walk your last three exceptions Skip the feature grid. Do this instead. Pull the last three vendor onboardings that actually consumed your team’s time. Not the clean ones. The ones that stalled. For each, answer two things. **Where did the work happen?** If the honest answer is “inside one platform, against data that platform already held,” you are describing the suite case, and a unified spend management platform like Coupa is the natural home for it. If the answer is “across a shared inbox, a parser, two Slack threads, a contract envelope, and three exports from the ERP before anyone could approve the record,” you are describing the orchestration case, and the question becomes which orchestration product to buy. **What did the exception actually need?** If it needed turnkey sanctions and third-party risk screening inside one suite, Coupa’s native risk module is a real advantage and you should weight it. If it needed a named human pulled into the decision with the full context assembled, fast, in the channel they already work in, that is the work an orchestration layer is built to absorb. Most mid-market procurement teams will find their painful onboardings lived across systems and their exceptions needed judgment more than they needed another screening checkbox. For those teams the comparison resolves toward FlowRunner, not because it wins on every axis (it does not, and Coupa’s suite depth, native risk screening, and ecosystem maturity are genuine advantages where they apply) but because the decision is a scope decision before it is a feature decision. A separate framework on [how to evaluate which procurement workflows are worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) handles the financial side once you know which case you are in. ### Quick answers #### Is FlowRunner a replacement for Coupa? No, and we do not position it that way. Coupa is a Business Spend Management suite that spans sourcing, procurement, invoicing, expenses, payments, and supplier risk. FlowRunner is an orchestration layer that coordinates work across the systems you already run. They overlap only on the vendor onboarding workflow. If you need a unified spend management platform, Coupa is the right shortlist. If you want to orchestrate onboarding across an ERP, a parser, a contract tool, and Slack without consolidating onto a new suite, that is the FlowRunner case. #### Does FlowRunner do sanctions screening and supplier risk scoring like Coupa? Coupa has native third-party risk and sanctions screening built into its supplier management module, covering domains like OFAC, InfoSec, and ABAC inside one platform. FlowRunner does not ship a native risk-scoring engine. It orchestrates screening by calling the screening service or data source you choose as a step in the workflow, then routes a flagged hit to a named human with the context attached. If turnkey native risk scoring inside one suite is a hard requirement, Coupa has the edge there today. #### How is FlowRunner priced compared to Coupa? Coupa pricing is enterprise and quoted through sales, scoped to the modules and spend volume you license. FlowRunner publishes its tiers: the core platform (AI agents, all integrations, human-in-the-loop, unlimited users and workflows) is available from the Free plan up, with the Growth tier at $45 per month as the production anchor. Teams that need governance tooling step up to the Professional tier at $299 per month for audit trails and RBAC, with SSO and SAML at the Business tier ($999 per month). No enterprise procurement cycle is required to access either the platform or the governance overlay. The two are not the same kind of purchase, which is the point of the comparison. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Datasite for the M&A Due Diligence Checklist: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-datasite-for-m-and-a-due-diligence-checklist Comparisons May 27, 2026 Updated May 28, 2026 12 min read Datasite owns the deal room. FlowRunner orchestrates the financial reconciliation and approval-chasing around it. Which fits depends on the deal. ![A bald cartoon man on the threshold between a green vault labeled DATA ROOM holding stacked files and Q&A, and an open workshop labeled OPERATIONAL DILIGENCE with invoices, vendor statements, and amber threads connecting them across a workbench.](https://flowrunner.ai/images/blog/flowrunner-vs-datasite-for-m-and-a-due-diligence-checklist-hero.webp) The M&A due diligence checklist is two different documents pretending to be one, and which one you are reading determines whether [Datasite](https://www.datasite.com/) or FlowRunner belongs in the answer. If your checklist is the deal team’s diligence index, governing document review across a live sell-side or buy-side transaction, Datasite is the right home for it and FlowRunner is not trying to replace that. If your checklist is the CFO’s operational diligence ledger, governing the financial reconciliation, vendor billing validation, and internal approval-chasing that has to clear before the deal team’s items can be marked done, that work lives outside the data room entirely and a general-purpose orchestration layer fits it better than a VDR ever will. Most articles on the M&A due diligence checklist do not distinguish between those two documents. This one does. ### Side by side, at a glance | | Datasite | FlowRunner | | --- | --- | --- | | **Category** | Virtual data room for M&A dealmaking | Orchestration layer for work and AI agents across business operations | | **Primary job** | Secure the documents, manage Q&A, govern access across the deal lifecycle | Coordinate the financial reconciliation, vendor billing validation, and approval-chasing happening around the deal | | **Built for** | Investment banks, law firms, private equity, corporate dev teams running M&A transactions | Operations and finance leaders running cross-system workflows (including the operational layer of diligence) | | **Document permissioning and redaction** | Granular permissioning, AI-powered redaction, watermarking, all native to the product | Not a VDR feature; FlowRunner does not replicate this | | **Q&A management** | Structured Q&A workflow purpose-built for diligence | Not offered as a deal-room feature | | **AI inside the workflow** | Embedded AI for redaction, summarization, semantic search, citation-backed answers; MCP server with Claude, ChatGPT, Copilot, and Blueflame AI working directly inside the deal | AI agents that coordinate work across tools, with human-in-the-loop as a callable action when judgment is required | | **Deal lifecycle coverage** | Pipeline, Outreach, Prepare, Diligence, Acquire, Archive | Out of scope; FlowRunner does not address the deal lifecycle as a domain | | **Operational diligence work (financial reconciliation, vendor reconciliation, internal approval-chasing)** | Out of scope by design; the data room governs documents, not the reconciliation work feeding them | Native fit: workflows trigger across email, ERP, Slack, parsers, and approvals with full audit trail | | **Track record in regulated financial transactions** | Established over many completed deals with investment banks, law firms, and private equity | None comparable; FlowRunner is not the right venue for a live transaction | | **Mobile and offline** | Full mobile app, offline document access, translation across 17 languages | No mobile-first deal interface; designed as a coordination layer for back-office work | | **Governance pricing** | Enterprise pricing; not publicly listed | Audit trails, RBAC, and SSO at the [Professional tier](https://flowrunner.ai/pricing) ($299/mo) | The table tells most of the story. The rest of this article is the part the table cannot. ### What Datasite is, said in their words Datasite describes itself as [“the digital home of M&A”](https://www.datasite.com/) and positions the platform as [“today’s data room, tomorrow’s deal platform.”](https://www.datasite.com/) The product line maps to the full deal arc: Pipeline for opportunity capture, Outreach for deal marketing, Prepare for transaction readiness, Diligence as the premier sell-side data room, Acquire as the buy-side data room, and Archive for post-close preservation. That is not marketing framing for FlowRunner to argue with. It is an accurate description of what Datasite has built over years of serving the dealmaking community. The depth that comes with that scope is real. Granular document permissioning. AI-powered redaction. Structured Q&A management with role-based visibility. Watermarking and access logging built into the core product. A SOC 2 Type II posture and a policy that deal data is not used to train external AI models. Embedded AI capabilities including document summarization, semantic search, and citation-backed answers inside the room. A continuous-delivery release model that ships new features without downtime. A mobile app that handles deal work in the field, with translation across 17 languages and an App Marketplace for deal-specific extensions. Datasite recently announced an MCP server that, in their own words, [“lets Claude, ChatGPT, Microsoft Copilot, and Blueflame AI work directly inside the deal, setting up rooms, drafting Q&A, and surfacing diligence answers, all without a single document leaving the most trusted environment in M&A.”](https://www.datasite.com/) That is a serious product announcement, and it is not a feature gap on FlowRunner’s side. It is a category Datasite owns and FlowRunner is not competing for. Said honestly: for a deal team running a live sell-side or buy-side transaction, those capabilities are not nice-to-have. They are the product. A general-purpose orchestration layer with no VDR primitives is the wrong tool for that job, and recommending FlowRunner for it would be irresponsible. This article is not making that recommendation. ### The seam: the M&A due diligence checklist is two documents Here is what most posts on the M&A due diligence checklist will not say plainly: the document hiding behind that phrase looks different depending on who is holding it. For the deal team (investment banker, M&A lawyer, PE associate), the checklist is a diligence index. Hundreds or thousands of line items mapped to documents that need to be requested, uploaded, reviewed, redlined, and signed off. The work shape is document-centric. The bottleneck is access, permissions, redaction, Q&A turnaround, and version control. That is a data room problem, and Datasite has spent years optimizing for it. For the operating finance leader (CFO, VP Finance, fractional CFO, head of an accounting roll-up), the checklist is something else. It is a reconciliation ledger. The line items are: tie the trailing-twelve-month revenue figure to actual customer payments, validate distributor billbacks against shipped goods, confirm the AP queue has no duplicate vendor records, verify there is no outstanding tax exposure in states the seller forgot to register in, run client onboarding for the acquired book before close, chase internal approvers for sign-off on the closing schedule. The work shape is data-centric and people-centric, not document-centric. The bottleneck is rarely document access. The bottleneck is that the systems do not agree, the vendor master is dirty, the approver is on a plane, and the email chain has scattered across four inboxes. The deal team’s checklist and the finance leader’s checklist nominally share a name. They live in different work, get done by different people, run on different systems, and fail in different ways. The article that crowns a single product as “the answer for the M&A due diligence checklist” has skipped that distinction. Most do. ### Where Datasite is the right home for the checklist If your version of the checklist is the deal team’s diligence index, choose Datasite (and probably not FlowRunner as a substitute for it) when: - The transaction is a live sell-side or buy-side deal where multiple bidders, advisors, or counterparties need governed access to the same document set - Document permissioning, watermarking, redaction, and structured Q&A are core to how the diligence runs - Investment banks, law firms, or private equity counterparties are involved and the brand recognition of the data room matters to them - The deal team needs AI capabilities purpose-built for diligence (summarization, semantic search, citation-backed answers) operating directly on the document set without data leaving the environment - The transaction will eventually move to archival and the same vendor handling diligence should preserve the project record post-close - Mobile access, offline document review, or multi-language translation across deal participants is part of the workflow - The full deal lifecycle (sourcing, outreach, prepare, diligence, acquire, archive) is in scope and a single integrated platform is the right shape for it That set of conditions describes a real and common M&A engagement. If it describes yours, the right answer is Datasite or a peer VDR, and an orchestration layer is the wrong shape for that work. FlowRunner is not asking deal teams to bet a live transaction on a platform without a track record in regulated financial transactions, and it would be dishonest of this article to suggest otherwise. ### Where FlowRunner is the right home for the checklist For the operating finance leader’s version, the calculus is different. If your version of the checklist is a reconciliation ledger where the line items are financial and operational rather than document-centric, the data room is the wrong center of gravity. Most of the work is happening before any document gets uploaded, or after it has already cleared the room. This is the version of the checklist that finance leaders we have talked to describe most often. The CFO running diligence on an acquired accounting practice, where the work is verifying client-by-client revenue, sorting out duplicate vendor records between the two AP queues, and chasing partners for approvals on closing schedules. The fractional CFO running diligence on a family office acquisition, where the checklist is mostly verifying the books and there is no investment bank involved. The accounting roll-up doing its fifth bolt-on, where the diligence checklist is a known template and the work is execution against it, not negotiation of it. The mid-market operating buyer doing a small-bolt-on where the deal team is a partner at a regional law firm and a CFO running point, not a bulge bracket and a Big Four. In conversations with finance leaders, three operational gaps surface over and over during a live deal, and a virtual data room cannot fix any of them by design: - **Diligence items falling through the cracks near close.** The partner at one accounting firm described it plainly in a recent conversation: there are probably things that should happen during the diligence process that just plain do not, that fall through the cracks. The data room knows whether a document was uploaded. It does not know whether the financial reconciliation underneath that document was actually completed, or whether an approver answered the email asking them to confirm a number. - **Vendor or distributor billing errors creating double-payment risk during the transition.** A CFO in a recent conversation said it would be very easy to double pay on a distributor invoice when both parties were mid-handoff and both AP queues were live. That is not a document problem. It is a workflow problem between two ERPs that do not know each other exist yet. - **Manual reconciliation between systems that should agree but don’t.** The two finance stacks (charts of accounts, vendor masters, customer masters, sales tax setups) have to be made to agree before the diligence numbers can even be trusted. Most of that work is happening in spreadsheets and inboxes, not in the deal room. [FlowRunner](https://flowrunner.ai/) is built for that operational layer. A workflow triggers when a parsed vendor invoice arrives, gathers context across the seller’s ERP and the buyer’s ERP, attempts the mechanical part of the reconciliation, and pauses to ask a human reviewer when the match is ambiguous. A different workflow runs the approval chase: the closing schedule lands in front of the CFO, the named approvers get a Slack message with the specific question and full context attached, the response is captured against the source record, and the audit trail spans the whole motion. A third workflow handles client intake for the acquired book, parsing the existing client list into the buyer’s onboarding system with exceptions routed to the operator when records are incomplete. This is the same pattern documented in the [post-M&A integration checklist for the CFO](https://flowrunner.ai/blog/post-m-and-a-integration-checklist) running the combined finance function, and the [vendor reconciliation work that surfaces in distributor billback management](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating). The work shape is the same on both sides of close. Pre-close it shows up as diligence items the data room cannot enforce. Post-close it shows up as integration items the data room is no longer part of. The category that owns the operational layer of diligence is orchestration as a service: a system above the systems of record that listens for what they emit, gathers the context a checklist row needs to be true, asks a named human when the data alone cannot resolve the question, and writes a single audit trail across the financial reconciliation work that the data room was never designed to own. FlowRunner is built for that layer. Datasite is built for the document and Q&A workflow that sits inside the deal room itself. They are different jobs on opposite sides of the same checklist. ### Where Datasite is stronger and FlowRunner is not trying to compete This section names the axes plainly. The depth here is real and the comparison is only honest if FlowRunner’s limitations get the same plain treatment Datasite’s strengths do. - **M&A domain depth.** Datasite has built years of deal-specific workflow logic, structured Q&A management, granular document permissioning, AI-powered redaction, and watermarking into the core product. FlowRunner does not match this depth for live deal teams and is not pursuing it. Recommending FlowRunner as a substitute for the data room is the wrong recommendation. - **Brand recognition and track record in regulated financial transactions.** Datasite has accumulated trust with investment banks, law firms, and private equity firms over many completed deals. That trust has commercial weight. FlowRunner has no comparable track record in this category and is not asking buyers to take it on faith. - **Deal-specific feature breadth.** A full mobile app with offline capabilities, document translation across 17 languages, the App Marketplace, and the embedded AI feature set built specifically for M&A (Blueflame AI Q&A, citation-backed answers, semantic search across the deal corpus). These are scope choices that reflect Datasite serving the dealmaking community as its primary buyer. FlowRunner does not offer them and does not intend to compete on them. - **Compliance posture for live transactions.** Datasite’s SOC 2 Type II attestation and the explicit policy that deal data is not used to train external AI models are the kind of substantiated compliance claims a deal team needs to see before uploading sensitive materials. FlowRunner has audit-trail, RBAC, and SSO infrastructure at mid-market pricing, but does not currently make the specific compliance-framework claims appropriate for sell-side dealmaking. Use the right tool for the right level of regulated exposure. If those axes are decisive for your version of the checklist, the conclusion is straightforward. Use Datasite. The rest of this article is interesting context but not actionable. ### Where FlowRunner is the better fit and Datasite is not the right shape Symmetrically, on the operational side of the same checklist: - **Coordination across the seller’s and buyer’s systems that touch the diligence work.** Vendor master reconciliation, distributor billback validation, customer-credit inventory, approval routing, manual reconciliation steps that should be running once and ending up in an audit trail. FlowRunner’s orchestration substrate is broad and extensible (Stripe, QuickBooks Online, Acumatica, NetSuite, Slack, email, WhatsApp, parser services, and most of what a mid-market finance stack runs). New connectors take 30 minutes or less when the catalog does not already cover what the team needs. - **Human-in-the-loop as a callable action, not an approval gate at the end of a workflow.** When an agent encounters an ambiguous vendor match or a closing-schedule line that does not tie, it pauses, asks a named reviewer via the channel they actually read (Slack, email, WhatsApp, phone) with full context attached, and resumes when the response is captured. This is [the difference between AI automation and AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents) made operational. The data room’s approval mechanism is a legitimate control for document workflow; it is not built for ambiguous reconciliation decisions on data the room does not own. - **A visual builder for non-developers.** A controller or operations analyst can compose a diligence-period workflow that combines a parser, an ERP read, a Slack approval, and an audit-trail entry without filing an engineering ticket. The article on [building automations that hold up in production](https://flowrunner.ai/blog/the-truth-about-building-automations) covers the engineering discipline this needs to actually survive contact with a real deal calendar. - **Mid-market governance pricing.** Audit trails, RBAC, and SSO at $299 per month on the [Professional tier](https://flowrunner.ai/pricing). The framing matters: this article is not claiming auditor acceptance of any specific compliance framework. It is claiming the governance infrastructure for operational diligence work exists at a price a CFO or fractional CFO can authorize without an enterprise procurement cycle. Specific framework attestations are separate conversations and would be misleading to claim here. - **Audience fit for the operating finance leader.** The fractional CFO running diligence on an accounting bolt-on. The CFO at a roll-up doing its fifth acquisition this year. The mid-market operating buyer whose checklist is mostly financial and operational. The family office CFO doing diligence in-house. None of those buyers are running a Datasite-scale deal room, and most of them never will. The work they have is real, and the orchestration layer is the shape that fits it. ### How to decide Two questions, with the order intentional. **1\. Whose checklist are you actually running?** Walk through the last five line items on your version of the M&A due diligence checklist. For each one, name what the work actually is. If the answer is “request a document, get it uploaded, review it for redlines, send a Q&A back, sign off,” that is data room work and the question is which VDR to choose. Datasite or a peer. If the answer is “tie this number to that system, validate the vendor master, chase the partner for approval, run client intake on the acquired book, reconcile the closing schedule,” that is operational work and the question is which orchestration layer to choose. FlowRunner or a peer. Most live transactions in the mid-market have both. The deal team’s checklist runs in the VDR. The finance leader’s checklist runs around it. **2\. Where is the work falling through today?** If items are slipping because documents are not getting uploaded, getting redacted incorrectly, or getting answered in Q&A out of sequence, the gap is in the data room workflow and a better VDR will close it. If items are slipping because numbers are not getting reconciled, approvers are not responding, vendor records are not getting cleaned up, and the closing schedule is held together by a spreadsheet and a partner’s memory, the gap is in the operational layer and a better data room will not touch it. The honest read is that the second kind of gap is what finance leaders bring up most often when asked what consumed their last deal cycle. Most mid-market M&A engagements will end up using both kinds of tool, in different work, run by different people. Datasite (or a peer VDR) for the document side, including the diligence index, the Q&A workflow, and the deal-team artifacts. An orchestration layer for the financial reconciliation, vendor master cleanup, internal approval chase, and the integration work that begins the moment the deal closes. That is not a hedged answer. It is the honest one, and it is the structural reality on the buyer side of most deals that are not bulge-bracket M&A. ### Quick answers #### Is FlowRunner a replacement for Datasite? No. Datasite is a virtual data room purpose-built for M&A dealmaking, with document permissioning, AI-powered redaction, Q&A management, and a track record across many completed transactions. FlowRunner is an orchestration layer for general business operations. A live sell-side or buy-side transaction belongs in a data room, not in FlowRunner. #### Where does FlowRunner fit if the diligence checklist still has Datasite running the deal room? FlowRunner handles the operational diligence work that lives outside the data room: vendor billing validation, distributor billback reconciliation, internal approval-chasing, and the financial reconciliations that need to clear before the checklist can mark items done. The data room owns the documents; the orchestration layer owns the work around them. #### Who is FlowRunner the right answer for instead of Datasite? Finance leaders running diligence as part of an accounting roll-up, family office acquisition, or mid-market deal where most of the checklist is financial reconciliation, vendor reconciliation, client intake, and chasing internal approvals rather than VDR-resident document review. The deal is real; the document load is not what’s holding up close. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Drata for SOX Compliance Software: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-drata-sox-compliance-software Comparisons May 28, 2026 Updated May 30, 2026 13 min read Drata is a strong continuous-compliance platform, but SOX is not in its framework list. Where a GRC tool fits, where FlowRunner fits, and when you need both. ![A bald cartoon man with three hairs sticking up turning away from a wall of green status lights to rest his hand on a single amber lever tagged approve, as if the lights cannot answer what the lever asks.](https://flowrunner.ai/images/blog/flowrunner-vs-drata-sox-compliance-software-hero.webp) A CFO comparing Drata against FlowRunner for SOX is usually one toggle into a mistake, and the mistake is treating SOX like one more framework you switch on next to SOC 2. SOX is not a framework you enable. It is a set of human decisions inside your close cycle, and a decision is not something software can poll. Drata is very good at proving the state of your systems. SOX 404 turns on proving the judgment of your people: who approved this bill, who signed off this reconciliation, who decided the exception was acceptable and why. Those are different problems, and a finance leader should know which one they are buying for before the demo flatters them into the wrong one. This piece compares the two products on the axes a SOX buyer actually weighs: framework coverage, where the control evidence comes from, integration scope, the auditor experience, pricing, and the buyer each product is shaped for. Drata is a strong product. FlowRunner is a different kind of product. They are not substitutes, and the most useful thing this comparison can do is draw the line cleanly. ### Drata and FlowRunner, side by side on SOX | Axis | Drata | FlowRunner | | --- | --- | --- | | Category | Continuous-compliance and trust management platform | Orchestration layer above the systems of record | | Self-description | [”The Agentic Trust Management Platform”](https://drata.com/), “Continuous Real-Time Trust” | An orchestration layer that runs the control activities and captures the evidence at the moment of work | | Framework coverage | SOC 2, ISO 27001, ISO 42001, GDPR, HIPAA, PCI DSS, DORA, FedRAMP, [“30+ Pre-Mapped Frameworks”](https://drata.com/products/compliance-automation). SOX is not listed. | No pre-mapped frameworks. Designed to support common control-evidence requirements at the moment of the transaction. | | Evidence model | Monitor and pull: [“automated tests across your environment to monitor success, surface failures”](https://drata.com/products/compliance-automation) | Capture at the moment of work: named approver, timestamp, and decision recorded against the record | | What it sees in SOX | The IT general controls slice: access, provisioning, change management on financial systems | The financial-process controls: AP approvals, segregation of duties on transactions, reconciliation sign-offs, exception decisions | | Integrations | [”Hundreds of tools”](https://drata.com/integrations): cloud, identity, HRIS, version control, ticketing | Newer and narrower; centered on ERPs, accounting platforms, payment processors, document parsers, channels | | Auditor experience | Audit Hub: [“centralize auditor collaboration, evidence requests, and approvals”](https://drata.com/integrations) | No auditor portal | | Human-in-the-loop | Review queues and workflow status | Agents pull a named human in as a callable step and resume with that decision as context | | Pricing | [Quote-based, contact sales](https://drata.com/pricing) | Audit trails and RBAC at $299/mo (Professional); SSO at $999/mo (Business) ([source](https://flowrunner.ai/pricing)) | | Honest replacement question | Replaces no part of your financial-process control activities | Replaces no part of your continuous-monitoring or SOC 2 program | Both products can sit in the same stack without stepping on each other. The sections below explain where each one earns its place, and where it does not belong. ### What Drata actually is, in Drata’s own words Drata describes itself as [“The Agentic Trust Management Platform”](https://drata.com/), built to [“leverage autonomous AI agents to automate compliance, manage internal and third-party risk”](https://drata.com/) and deliver [“Continuous Real-Time Trust.”](https://drata.com/) The compliance-automation product promises to [“run automated tests across your environment to monitor success, surface failures and determine remediation plans”](https://drata.com/products/compliance-automation) and to [“automate collection across your tools so your compliance program stays current.”](https://drata.com/products/compliance-automation) It carries strong public proof: Drata advertises 8,000+ global customers and a 4.8 out of 5.0 G2 rating on its homepage. That is a serious continuous-compliance platform, and the description tells you exactly what it is built around. The frameworks Drata names are SOC 2, ISO 27001, ISO 42001, GDPR, HIPAA, PCI DSS, DORA, FedRAMP, and what it calls [“30+ Pre-Mapped Frameworks.”](https://drata.com/products/compliance-automation) SOC 2 is the anchor. ISO 27001 and HIPAA cluster around the same security-and-trust posture. ISO 42001 is the AI governance extension. PCI DSS, DORA, and FedRAMP are the regulated-industry and public-sector additions. Read the list twice and notice what is not on it. SOX is absent. That absence is honest on Drata’s part, and it is the whole reason this comparison exists. SOX is not a security-posture framework. It is an SEC statute about internal controls over financial reporting, and the external auditor tests those controls under [PCAOB AS 2201](https://pcaobus.org/oversight/standards/auditing-standards/details/AS2201). Its control families live in entity-level governance, access and segregation of duties, approvals, reconciliations, the procure-to-pay cycle, and the IT general controls that support the financial systems. Drata is excellent at one slice of that picture. Its connector library reaches [“hundreds of tools”](https://drata.com/integrations) across cloud providers, identity, HRIS, and version control, which gives it real depth on the IT general controls layer of SOX: user provisioning, access reviews, change management, the configuration state of the systems your financial data sits in. If you are already buying Drata for SOC 2 and your SOX program needs the ITGC slice automated, you get genuine adjacency value and you should take it. What Drata is not built to do is sit inside the AP approval, the reconciliation, the journal-entry review, or the exception handling that finance owns. Those activities are what a CFO personally certifies under Sections 302 and 404, and they run in the ERP, the AP system, the close tool, and the channels finance works in. Drata can confirm who has access to the ERP. It cannot tell you who approved which bill, or what they decided when the bill did not match the purchase order. ### What FlowRunner actually is, plainly FlowRunner is an orchestration layer. It sits above the systems of record, the ERP, the AP system, the billing platform, the payment processor, the document repository, and the channels your team works in, and it runs the work that moves between them. In an approval flow, FlowRunner posts the request into the channel approvers already use, captures the response, writes the named approver and timestamp back to the bill record, and routes anything unusual to a designated reviewer with full context attached. FlowRunner is not a GRC platform, and pretending otherwise would not survive a single product evaluation. It has no pre-mapped frameworks. It does not run continuous control monitoring across your cloud infrastructure. It has no Audit Hub, no third-party risk module, no framework-mapped control library. If those are the capabilities you are shopping for, FlowRunner does not have them and a comparison with Drata on those axes is one Drata wins outright. I will say where else Drata wins further down, plainly, because the comparison is worthless if I do not. What FlowRunner does deliver is the moment-of-work evidence a control activity produces, written into the system that owns the underlying record. The named approver captured against the bill. The timestamp of the decision. The exception rationale stored next to the transaction. The reviewer identity logged on the reconciliation step. An audit trail that does not need to be reconstructed during fieldwork because it was never scattered across inboxes in the first place. ![Slack approval message for an accounts payable bill, showing the vendor, amount, the named approver who clicked approve, and a timestamp, with an exception note thread below it](https://flowrunner.ai/images/blog/flowrunner-vs-drata-sox-compliance-software-1.webp) ### Framework coverage Drata is a category leader on framework breadth, and it is not close. The 30-plus pre-mapped frameworks span security, privacy, AI governance, and regulated-industry standards, each with the control mappings and automated tests already built. For a buyer whose primary need is a SOC 2 program manager with framework mapping and continuous monitoring, Drata is the right tool and FlowRunner is not in the conversation. Full stop. FlowRunner offers no framework coverage at all. There is no SOC 2 module, no ISO module, no SOX module, because the product is structurally different. It captures evidence at the point a control executes rather than mapping evidence to a framework after the fact. For a SOX buyer, though, the framework-coverage axis is a trap. SOX is not on Drata’s list, so “Drata covers more frameworks” is true and irrelevant to the specific question on the table. The real question is which product produces the evidence a SOX tester asks for at the line item, and that is a question about where evidence comes from, not how many frameworks a dashboard maps. That is the next axis, and it is the one that matters most. ### Where the evidence is born Here is the thing most comparisons of compliance software will not say, because it cuts against the dominant product narrative in the category: continuous control monitoring, the capability Drata is genuinely best at, is the capability SOX 404 needs least. That sounds like a knock on Drata. It is not. It is a statement about what kind of thing a SOX control is. Continuous monitoring works by polling. It connects to a system, reads the configuration, and tests that state against a control. [“Run automated tests across your environment to monitor success, surface failures.”](https://drata.com/products/compliance-automation) When the control is “access to production is restricted to authorized users,” Drata polls the identity provider and proves the state. That is a real, valuable control, and it is a state. You can read it off a system at any moment and the answer is the same until something changes. Now take the controls SOX 404 actually turns on. “Every bill over the threshold requires named approval from a controller, with the timestamp captured.” “The reconciliation was prepared by one person and reviewed by another.” “The exception was escalated, examined, and accepted with a documented reason.” None of these are states. They are events. They happen once, at a specific moment, when a specific human makes a specific judgment, and then they are over. There is no configuration to poll afterward. A control that is a human saying yes is not a state you can read off a system. It is a decision, and if you did not capture it at the instant it was made, you are reconstructing it from memory and email threads in October. That gap, the difference between a control you can poll and a control that is a decision, is where an entire category lives that neither the GRC dashboard nor the system of record owns. The approval that routes from the AP clerk to the controller to the CFO does not live inside QuickBooks; QuickBooks holds the bill, not the conversation. It does not live inside Drata; Drata monitors the systems, not the judgment. It lives in the seam between them, in the work that moves a decision from one human and one system to the next. An orchestration layer is the category that owns that seam: a layer above the systems of record that listens for what they emit, pulls the right human in at the moment a judgment is required, captures what they decided, and writes that evidence back into the system that owns the record. This is the digital andon cord one operations leader described to us, the cord that stops the line when something needs a person and records exactly what the person did. FlowRunner is built for that layer. It does not replace the tool that monitors your infrastructure state. It captures the decisions that monitoring is structurally blind to. So the evidence models are not competitors. Drata’s model is monitor and pull, and it is strong wherever the SOX evidence lives in infrastructure that holds a state. FlowRunner’s model is capture at the moment of work, and it is strong wherever the SOX evidence lives in a human decision that has to be recorded against a financial record as it happens. A program that covers both slices needs both kinds of tools, or it assembles the moment-of-work evidence by hand at audit time, which is the failure mode the [SOX compliance checklist](https://flowrunner.ai/blog/sox-compliance-checklist) walks through control family by control family. ![A FlowRunner workflow diagram showing a bill ingested from the ERP, an automated threshold check, a branch that routes over-threshold bills to a named human approver as a callable step, and the captured decision written back to the bill record](https://flowrunner.ai/images/blog/flowrunner-vs-drata-sox-compliance-software-2.webp) ### Integration scope Drata wins this axis, and the win is real. A connector library reaching [“hundreds of tools”](https://drata.com/integrations) that pulls compliance evidence automatically from cloud infrastructure, identity providers, HRIS, version control, and security tooling is years of category investment FlowRunner has not matched. A finance team standardizing on FlowRunner for evidence will have fewer pre-built integrations available than it would on Drata. That is the honest read, and I am not going to dress it up. The qualification is what each library is optimized for. Drata’s connectors are built for security-and-trust posture: they cover what a SOC 2 program needs to monitor, which is cloud, identity, and engineering systems. FlowRunner’s connector set is centered on the systems financial work actually runs in: ERPs, accounting platforms, payment processors, document parsers, and the channels approvals route through. A finance team’s SOX evidence problem leans heavily on those finance systems and far less on the cloud-monitoring surface where Drata has its depth. So “hundreds of tools” is the right thing to weigh if your evidence problem is “is our cloud environment configured to standard.” It is the wrong thing to weigh if your evidence problem is “can we prove who approved the bills and signed the reconciliations across the seven systems where the financial work happens.” For that problem, the question is not how many connectors exist but whether the ones that own your transactions are covered. ### The auditor experience This is a genuine Drata strength worth naming on its own. Drata’s [Audit Hub](https://drata.com/integrations) is built to “centralize auditor collaboration, evidence requests, and approvals in one secure hub.” When the auditor can pull evidence directly through a structured workspace instead of chasing the controller for screenshots, the fieldwork cycle compresses. That is a real advantage in the frameworks Drata covers, and it reflects the kind of compounding investment a platform whose entire business is compliance certification can make and a newer entrant cannot match in the short term. FlowRunner has no auditor portal, and it is not building one. What it offers the auditor instead is upstream of the portal question: evidence that is already complete and already attached to the record, so the evidence request is a query rather than a reconstruction project. Those are different contributions to the same audit. Drata makes the handoff to the auditor smoother. FlowRunner makes the evidence exist in defensible form before the handoff. A program can value both. ### Pricing and packaging Drata’s pricing is [quote-based and not publicly listed](https://drata.com/pricing); the buying motion is a sales-led evaluation, which is appropriate for a multi-framework platform sold to security and compliance teams. FlowRunner publishes its tiers. [Audit trails and RBAC start at the Professional tier at $299 per month](https://flowrunner.ai/pricing). SSO and SAML start at the Business tier at $999 per month. I am stating that split precisely on purpose, because it is easy to imply that everything governance-related arrives at $299, and it does not: SSO is a Business-tier capability. The packaging is shaped for mid-market finance teams that need named-approver capture and an audit trail in their day-to-day workflows without standing up a six-figure enterprise program. Comparing the two on price directly is misleading, because they are different categories of product. The signal that matters is what each tier is shaped for. Drata is shaped for security-led, multi-framework compliance programs. FlowRunner at the Professional tier is shaped for a finance team that needs approval and reconciliation evidence captured in the systems it already works in. ### Where Drata is the better fit The comparison is not honest unless it stops and says plainly where Drata wins. It wins in a lot of places. - **SOC 2, ISO 27001, HIPAA, PCI DSS, and the rest of the marketed frameworks.** This is what Drata was built for. There is no version of this comparison where FlowRunner is the better choice for the SOC 2 program-manager role. - **Continuous control monitoring across infrastructure.** Automated tests that watch your cloud and identity configuration and surface failures continuously. FlowRunner does not do this and is not trying to. - **Automated evidence collection from connected systems.** Pulling evidence from hundreds of tools without manual gathering is the core of Drata’s value, and it is real. - **The auditor handoff.** Audit Hub compresses fieldwork in the frameworks Drata covers. FlowRunner has no equivalent. - **Third-party and internal risk management.** Drata has dedicated products for vendor risk and enterprise GRC. FlowRunner has neither. - **A recognized GRC name on the slide.** A CFO whose buying decision has to clear an audit committee that wants an established compliance vendor will find Drata, with its 8,000+ customers, delivers that signal in a way FlowRunner does not. - **A published security certification.** Drata’s business is certification, and it carries the trust attestations a security-conscious buyer expects. FlowRunner does not currently publish a SOC 2 or equivalent certification page, and for a CFO weighing a vendor on exactly that axis, that absence is a fair and material consideration. I am not going to wave it away. If those criteria describe your actual problem, Drata is the right tool and FlowRunner should not be on the shortlist. ### Where FlowRunner is the better fit FlowRunner earns its place in the part of SOX that Drata is not built to reach: the control activities that are decisions, not states. ![A FlowRunner execution log for a reconciliation workflow, showing each step with a timestamp, the named preparer and reviewer, and the per-item decision recorded against the transaction, presented as an auditable trail](https://flowrunner.ai/images/blog/flowrunner-vs-drata-sox-compliance-software-3.webp) - **Approval evidence at the moment of the transaction.** The named approver and timestamp captured against the bill record, not pieced together from email later. The [QuickBooks and Slack approval workflow](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) does exactly this: a named approver and timestamp recorded for every approval, in the system that owns the bill. - **Reconciliation decision history per step.** Bank, merchant-processor, and intercompany reconciliations where each reconciling item carries a logged decision. The [Stripe and QuickBooks reconciliation pattern](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) retains decision history for every reconciliation step, kept against the transaction. - **AP control evidence inside the ERP.** The [Acumatica AP automation pattern](https://flowrunner.ai/workflows/automate-with-acumatica) captures approver identity on every bill approval and routes exceptions for explicit review, producing procure-to-pay control evidence inside the system of record. - **Exception handling as a callable step.** When the automation hits something it should not decide alone, a named human is pulled in with full context, the decision is recorded, and the workflow resumes carrying that decision forward. The exception evidence is the audit trail, not a thread you rebuild in the fall. - **Mid-market governance pricing.** Audit trails and RBAC at $299 per month put evidence infrastructure inside the budget of finance teams that cannot stand up a multi-tool enterprise GRC program on day one. - **Compliance evidence as an automation byproduct.** The same approval and reconciliation workflows that produce SOX evidence are the automation a finance team would justify on efficiency grounds anyway. The [framework for deciding what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) treats compliance evidence as a value multiplier on that work, not a separate line item. The framing FlowRunner supports is “designed to support common SOX control-evidence requirements,” not “auditors accept FlowRunner output as SOX evidence.” SOX compliance is the issuing company’s responsibility, not a vendor’s, and no automation tool satisfies Section 404 on its own. What an orchestration layer delivers is evidence in the form an auditor expects to receive it: named approver, timestamp, context, exception rationale, all retained against the underlying record. This is the same line FlowRunner sits on against [Vanta, the other continuous-compliance platform a SOX buyer will weigh](https://flowrunner.ai/blog/flowrunner-vs-vanta-sox-compliance-software). The pattern repeats across GRC tools because they all share the same architecture: monitor the systems, prove the state. The financial-process controls that are decisions live in the seam those tools do not reach. ### How to decide You can resolve almost every version of this decision with two questions, in order. **First: is the evidence you are missing a state or a decision?** If the gap is “can we continuously prove our cloud and identity systems are configured to standard, across SOC 2 and a stack of other frameworks,” that is a state problem, and Drata is built for it. If the gap is “can we prove who approved the bills, who signed the reconciliations, and what was decided on the exceptions,” that is a decision problem, and a monitoring platform is structurally blind to it no matter how many connectors it has. **Second: are you running a multi-framework compliance program, or tightening financial-process evidence?** A company building or running SOC 2 alongside several other frameworks needs a continuous-compliance platform as the program’s backbone, and Drata is a strong choice for that backbone. A finance leader asked to harden control evidence ahead of an investor diligence cycle, a lender covenant, an audit, or an M&A event usually has a narrower problem: named-approver and reconciliation evidence in workflows that currently live in spreadsheets and inboxes. A full GRC platform is more tool than that problem needs at the buying stage. FlowRunner is shaped for it and priced for it. If both answers point the same way, you have your tool. If the first answer is “both,” and for a company on a public-company path it often is, then both tools belong in the stack and they will not get in each other’s way. Drata watches the systems hold their state. FlowRunner stops the line and records the decision when a human has to make one. Buy the tool that closes the gap you actually have, and if you are honest with yourself about whether that gap is a state or a decision, the rest of the choice makes itself. ### Quick answers #### Is Drata a SOX compliance platform? Drata markets coverage of SOC 2, ISO 27001, ISO 42001, GDPR, HIPAA, PCI DSS, DORA, FedRAMP, and 30+ pre-mapped frameworks. SOX is not among the frameworks Drata lists. Its continuous control monitoring overlaps with the IT general controls slice of SOX, but it is not a SOX 404 program manager and does not run the financial-process controls a CFO certifies under Sections 302 and 404. #### Can FlowRunner replace Drata? No. FlowRunner is an orchestration layer, not a GRC platform. It has no pre-mapped frameworks, no continuous control monitoring across cloud infrastructure, no third-party risk module, and no Audit Hub. If you need a continuous-compliance platform for SOC 2 and adjacent frameworks, you need Drata or a tool like it. FlowRunner solves a different layer. #### Where does FlowRunner fit if a finance team is buying SOX software? FlowRunner runs the control activities SOX 404 turns on: AP approvals, reconciliations, and exception handling. It captures the named approver, the timestamp, and the decision against the record in the system that owns it, at the moment the work happens. It is designed to support common SOX control-evidence requirements at the point of the transaction rather than reconstructing them at fieldwork. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Fivetran for Automating Report Generation Source: https://flowrunner.ai/blog/flowrunner-vs-fivetran-automate-report-generation Comparisons May 27, 2026 Updated September 7, 2026 10 min read Fivetran moves data into your warehouse. FlowRunner generates the report, routes the approval, and chases the exception. The honest finance comparison. ![A bald cartoon man at a counter between a tall green data tank labeled with source system bands flowing into a warehouse box on his left, and a small desk with a single P&L report, phone notification, and clipboard on his right.](https://flowrunner.ai/images/blog/flowrunner-vs-fivetran-automate-report-generation-hero.webp) When a finance buyer types “automate report generation” into a search bar and lands on a [Fivetran](https://www.fivetran.com/) page, something subtle goes wrong on the way to the demo. Fivetran is one of the strongest pure-play data movement platforms on the market, but it does not generate reports. It moves the data that reports are eventually built from. The mistake is not Fivetran’s, and it is not the buyer’s. It is the consequence of “report generation” sitting on top of a stack where every layer claims it, and most of them only own a slice. The honest version of the FlowRunner versus Fivetran comparison starts there. If the actual problem is centralizing data from hundreds of SaaS and database sources into a warehouse, this article will not change your evaluation. Buy Fivetran. If the actual problem is that the controller is still copy-pasting a weekly P&L into an email at 4:30 on Friday, the comparison is not even close. That work lives on an entirely different layer of the stack. ### What you are actually comparing | | Fivetran | FlowRunner | | --- | --- | --- | | **Category** | Automated data movement platform | Orchestration layer for work and AI agents across tools | | **Primary job** | Replicate data from sources into warehouses and lakes | Generate reports, route approvals, escalate exceptions, coordinate handoffs | | **Built for** | Data and analytics engineers feeding a warehouse | Finance, operations, and ops-led report owners | | **Source connectors** | [700+ fully managed connectors](https://www.fivetran.com/connectors) including SaaS, databases, SAP, streaming, files | Operational integrations to QuickBooks, NetSuite, Acumatica, Stripe, Slack, email, plus MCP-extensible custom connectors | | **What lands where** | Source rows land in Snowflake, BigQuery, Databricks, S3, Iceberg, Delta Lake | Reports land in Slack channels, email inboxes, ERP entries, approval queues | | **Replication features** | Log-based CDC, history mode (Type 2 SCD), column hashing for PII, row filtering, schema drift handling | Not offered, not in scope | | **Pricing model** | [Monthly Active Rows (MAR)](https://www.fivetran.com/pricing) per connection, capacity-tiered | Execution-tiered subscription, [Free / $5 / $45 / $299 / $999 / Enterprise](https://flowrunner.ai/pricing) | | **Enterprise reliability** | Regional failover, private networking, customer-managed keys, PCI DSS Level 1 (Business Critical plan only) | Single-region cloud or self-hosted; RBAC, SSO, audit trails on Professional plan | | **Human-in-the-loop** | Not a Fivetran capability | Callable action: pause, route to a named approver via Slack, email, WhatsApp, or phone, resume on response | | **AI agent orchestration** | Not in scope | Native: agents as first-class workflow nodes with human escalation as a callable tool | | **Free trial** | 14 days per connection | Permanent Free plan (100 executions/mo), plus 14 days of Professional on signup, no card | Below the table is what no comparison checklist captures. ### What Fivetran does, in Fivetran’s words Fivetran positions itself as [“the data foundation for AI”](https://www.fivetran.com/), an automated data movement platform built to “securely move, manage, and transform data to power analytics, operations, and AI at scale.” That positioning is accurate, and the product backs it. The depth is real. Fivetran offers [700+ fully managed connectors](https://www.fivetran.com/connectors) spanning SaaS applications, databases, SAP, streaming sources, and file sources, with log-based CDC, history mode for Type 2 slowly changing dimensions, column hashing for PII, row filtering, and automated schema drift handling. The pipelines are idempotent, retried automatically, and managed end-to-end with zero customer-side maintenance. On the [Business Critical plan](https://www.fivetran.com/pricing), regional failover, private networking options (AWS PrivateLink, Azure Private Link, Google Cloud PSC), customer-managed keys for encryption, and PCI DSS Level 1 certification become available. Enterprise customers can purchase through Snowflake, AWS, Azure, or Google Cloud marketplaces with commitment burn-down. This is data engineering infrastructure at a level of maturity that FlowRunner does not attempt to compete with. Finance teams centralizing data from Salesforce, Workday, NetSuite, Marketo, Stripe, and a long tail of SaaS sources into Snowflake or BigQuery should evaluate Fivetran the same way they would evaluate Airbyte or Stitch. FlowRunner does not belong on that shortlist. What Fivetran is honestly not built for is the moment after the data lands. The warehouse is full. The dbt models have run. And now somebody still has to assemble the weekly P&L, post it to a Slack channel, flag the AR aging anomaly, and route a duplicate-payment risk to a named approver before the AP run on Tuesday. None of that is data movement. Fivetran does not claim it is. ### What automate report generation actually means for a CFO Here is what most posts on automated report generation will not say plainly. For a large slice of mid-market finance teams, the report-generation problem is not solved by getting the data into the warehouse. It is solved by getting the right summary, with the right context, into the right person’s hands, at the right time, with a place to escalate when something looks wrong. The pattern is recognizable from conversations with finance leaders running on operational systems like QuickBooks, NetSuite, and Acumatica: - A weekly P&L digest needs to land in the CFO’s Slack DM by 9am Monday, pulled from QuickBooks, with a one-line variance comment against last week - An AR aging summary needs to land in the controller’s inbox every Friday afternoon, with anything aged past 60 days flagged for a follow-up - A monthly board pack needs the operational KPIs assembled from the ERP, the payment processor, and a parsed inbox of vendor invoices, with the CFO reviewing variance commentary before it ships to investors - An exception report needs to flag any vendor invoice paid twice (or any distributor billback that does not tie to the bank) and pause for human judgment before it gets posted None of those workflows are “queries against a warehouse.” They are operational events: a Friday clock tick, a Stripe webhook, an email landing in a shared inbox, an ERP transaction crossing a threshold. The report is the output of the workflow, not a SELECT statement. A finance team building this on a warehouse-and-BI stack ends up with three problems. Dashboards are not reports; they are explored, not received. Scheduled BI emails are brittle, ungoverned, and do not handle exceptions. And nothing in the warehouse layer pauses and asks a human for input when something looks off. That last gap is the one that costs money. A duplicate-payment risk does not need a dashboard. It needs a workflow that stops, routes to the CFO, captures the response, and writes the decision to the audit trail. ### Where FlowRunner fits: the layer that generates the report [FlowRunner](https://flowrunner.ai/) is an orchestration layer. It coordinates work and AI agents across the tools a finance team already runs. Report generation, in that frame, is one pattern among many, and the one that most directly answers the “automate report generation” query for an operational finance team. The pattern looks like this: 1. A trigger fires (a scheduled time, an ERP transaction, a Stripe webhook, an inbox message) 2. A workflow pulls the right context from the right systems (QuickBooks, NetSuite, Acumatica, Stripe, parsed documents) 3. The workflow assembles the report (a digest, a variance summary, an exception list, a board-pack section) 4. The workflow distributes the report (Slack channel, email, an ERP entry, a shared drive) 5. If something looks off, the workflow pauses and calls a human as an action, not a status gate 6. The human receives the report and the flagged item with the context already attached, responds with a structured decision, and the workflow resumes with the decision captured in the audit trail Concretely, this is the pattern behind [automated P&L digests delivered to finance teams](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack), where the report assembles from QuickBooks and lands in a Slack channel at a scheduled time. It is the pattern in [what CFOs can automate with Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica), where ERP-level reporting, exception flagging, and notification all happen in one governed workflow. It is the pattern in [financial data extraction into the ERP with human-in-the-loop](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack), [extracting vendor documents into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack), and [invoice extraction from email to QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack), where extracted figures feed downstream reports and exceptions route to a named reviewer. The category that owns this layer is orchestration as a service. The seam it covers is the gap between data landing somewhere and a decision happening somewhere. A warehouse is built to store rows. A BI tool is built to explore them. Neither is built to fire when an operational event happens, assemble a report from the systems of record, deliver it to a named person, and pause for human judgment when an exception comes through. That work is its own category, and FlowRunner is built for it. ### Where Fivetran is the better fit Buy Fivetran when: - The actual problem is data centralization from many SaaS and database sources into a warehouse or lake - Your analytics team builds reports inside a BI tool (Looker, Power BI, Tableau, Sigma) and the bottleneck is data freshness or replication coverage, not report distribution - You need enterprise data engineering features: log-based CDC, history mode, column hashing for PII, row filtering, regional failover, private networking - You are building a data-foundation-for-AI architecture where downstream agents read from a governed warehouse and Fivetran owns the ingest - Your data team is large enough to operate the warehouse and the BI layer, and the report-generation work happens inside that stack rather than in operational tools That set of conditions describes a real and common architecture for companies past a certain scale. If it describes yours, FlowRunner is the wrong evaluation, and Fivetran is genuinely good at what it does. We are not the better product for data movement; we are not the product for it at all. ### Where FlowRunner is the better fit Choose FlowRunner when: - The report you want generated comes from operational systems (QuickBooks, NetSuite, Acumatica, Stripe, inboxes) and goes to operational channels (Slack, email, an ERP, an approval queue), without needing to land in a warehouse first - The workflow needs to pause and ask a human when an exception shows up (duplicate-payment risk, an AR anomaly, a vendor invoice that does not tie), not just write a row to a table - Non-developers in finance ops are the people who should configure new reports, schedules, and escalation logic without filing a data-engineering ticket - You need audit trails, RBAC, and SSO on a mid-market budget, governance infrastructure without a top-tier procurement cycle - AI agents are starting to appear in your stack (a forecasting agent, an AP coding agent, a variance commentary agent), and you need a coordination layer above them that keeps humans in control on the calls that matter - The reports your team actually needs are received, not explored, and a dashboard is not the right answer For most mid-market finance teams, the two products are complementary, not competing. Fivetran lands the data, if a warehouse is in the picture. FlowRunner generates the reports, routes the approvals, and escalates the exceptions, regardless of where the source data lives. ### How to decide One question, asked the right way: what fires the workflow? If the answer is “a scheduled query against the warehouse, run by an analytics team,” and the warehouse is the system of record for everything downstream, your problem is data movement. Fivetran (or an equivalent ELT platform) is the layer to evaluate. The report is a BI artifact. If the answer is “a Friday clock tick, a Stripe charge, an email landing in shared inbox, an ERP transaction crossing a threshold,” and the report is something a person receives and acts on, your problem is workflow orchestration. The category to evaluate is orchestration as a service. The report is the output of a coordinated workflow across operational systems, with a human pulled in when judgment is required, and an audit trail recording who decided what. Most mid-market finance teams will end up running both layers over time, used for different work. That is not a hedged answer. It is the architecture the work actually has. [How we think about evaluating what is worth automating in the first place](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) is a useful adjacent read for finance leaders sequencing the two investments. ### Quick answers #### Is FlowRunner a replacement for Fivetran? No. Fivetran is a data movement platform that replicates data from hundreds of sources into a warehouse or lake. FlowRunner is an orchestration layer for coordinating work and AI agents across the tools a finance team already runs. If the actual problem is data centralization into Snowflake or BigQuery, Fivetran is the right product. FlowRunner is not. #### Where does FlowRunner fit if we already use Fivetran? FlowRunner sits on the operational side of the data. It pulls report-ready figures from the warehouse Fivetran feeds, formats the Friday P&L, posts it to Slack, escalates an unexpected variance to a named CFO approver, and writes an audit trail of the conversation. The two are complementary. Fivetran lands the data; FlowRunner turns it into the report, the approval, and the exception. #### Can we automate report generation directly from QuickBooks or NetSuite without a warehouse? Yes, that is the use case FlowRunner is built for. Most mid-market finance teams want a weekly P&L digest, an AR aging summary, or an exception flag delivered to their inbox or Slack channel from the operational systems themselves. FlowRunner connects to QuickBooks, NetSuite, Acumatica, Stripe, and ERPs directly, generates the report, distributes it, and pauses for human input when something looks off. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs FloQast: An Honest Read on Account Reconciliation Software Source: https://flowrunner.ai/blog/flowrunner-vs-floqast-account-reconciliation-software Comparisons May 26, 2026 Updated May 28, 2026 11 min read FloQast runs your close. FlowRunner orchestrates the reconciliation exceptions that live between Stripe, QuickBooks, Acumatica, and the inboxes your close software was never built to reach. ![A bald cartoon man at a wide desk with two distinct piles in front of him, a neat green calendar of checkmarks on his left and a scattered pile of faded amber and terracotta receipts and envelopes on his right, considering both as separate kinds of finance work.](https://flowrunner.ai/images/blog/flowrunner-vs-floqast-account-reconciliation-software-hero.webp) Most comparison posts about account reconciliation software pretend two products are competing when they are not. The honest read is that [FloQast](https://floqast.com/products/close/) and FlowRunner solve different parts of the same problem, and finance teams that understand the seam between them stop wasting evaluation cycles on a fight that does not exist. FloQast owns the close. FlowRunner orchestrates the reconciliation exceptions, the cross-system handoffs, and the human judgment calls that pile up before anything hits a close checklist. Most teams running mid-market accounting operations need both, used for different work. ### Side by side, at a glance | | FloQast | FlowRunner | | --- | --- | --- | | **Category** | Financial close management software | Orchestration layer for work and AI agents across tools | | **Primary job** | Run the structured close: reconciliations, checklists, task dependencies, sign-offs | Coordinate exceptions and handoffs between systems before and around the close | | **Built for** | Accounting teams (controllers, accountants, close managers) | Operations and finance leaders coordinating work across multiple systems | | **Native workflow model** | Sequenced close tasks with status tracking and linear enforcement | Branching workflows with callable human-in-the-loop, AI agents as steps | | **Integration shape** | Deep connectors into GLs, ERPs, and reconciliation source systems for the close | Broad orchestration across Stripe, QuickBooks, Acumatica, inboxes, distributor portals, MCP-extensible | | **Exception handling** | Status gates and review steps inside the close workflow | Pause-and-ask called as an action; route to a named approver via Slack, email, WhatsApp, or phone; resume with the response captured | | **Governance** | Close-specific audit trail of reconciliation sign-offs and review | Workflow-level audit trail, RBAC, and SSO starting at the [Professional tier](https://flowrunner.ai/pricing) ($299/mo) | | **Who keeps using what** | Close team owns FloQast | Finance and ops own FlowRunner upstream of the close | The table tells you most of the story. The rest of this article is the part the table cannot. ### What FloQast is good at, said honestly FloQast is a mature, purpose-built [financial close management platform](https://floqast.com/products/close/). It has spent years going deep on the specific workflow of running a structured month-end close: reconciliations matched to GL balances, checklists with task dependencies, status tracking across the close calendar, and linear workflow enforcement that mirrors how an accounting team actually closes the books. The product knows what a tie-out looks like. It knows what an unreconciled difference is. It knows what a controller needs to see on day three of close week. This is not a feature checklist. It is domain depth accumulated by working with accounting teams across years of close cycles. A team that has standardized on FloQast for close management does not migrate that workflow to a horizontal orchestration product. There is no reason to, and any vendor suggesting otherwise is selling against their interest. FlowRunner does not replicate the FloQast close. We do not try to. Specific things FloQast does that FlowRunner does not: - Native, accounting-aware reconciliation matching tied directly to GL accounts - A close calendar with sequenced task dependencies that reflect how accounting teams actually work the close - Linear workflow enforcement built around sign-offs, review, and tie-out - Reusable templates for recurring close tasks across entities and periods - An accounting-team user interface designed for controllers and accountants, not operations generalists Lead with this honestly because if the rest of this article reads like a competitive attack on FloQast, the article is dishonest. It is not an attack. It is a positioning exercise. ### What account reconciliation actually looks like for most CFOs Here is what most posts on account reconciliation software will not say plainly: for a large slice of mid-market finance teams, reconciliation is not a discrete workflow that lives inside a close-management product. It is what happens between systems before anything reaches the close. The pattern is recognizable from conversations with finance leaders across CPG, services, e-commerce, and distribution: - Stripe payouts arrive in batches that do not map cleanly to QuickBooks invoices - Distributor portals report billbacks and chargebacks that do not match what hit the bank - Vendor invoices arrive in inboxes, get parsed (or not), and end up entered in two systems that disagree - The ERP and the GL show the same balance on most days, except the ones that matter - Someone exports a CSV, opens a spreadsheet, and starts an email chain to chase down the difference The work is not the matching. The matching is mostly mechanical. The work is the exception: - The payment that almost matches an invoice but is short by a fee Stripe deducted - The distributor billback that needs an account manager to confirm whether it is valid - The duplicate vendor invoice nobody caught because two AP coordinators opened the same inbox on the same morning In conversations with finance leaders, we hear a recurring fear that two people can review the same item and approve it twice, and nobody would catch it for weeks. One CFO phrased it as how easy it would be to double pay on something like that. A close-management product is not built to live inside that work. Nor should it be. The close-management product is built to certify that the books are right after the exceptions have been resolved. Resolving the exceptions is a different job done with different tools by different people on a different timeline. ### Where FlowRunner fits: the layer that handles the exceptions [FlowRunner](https://flowrunner.ai/) is an orchestration layer. It coordinates work and AI agents across the tools finance and operations teams already run. Reconciliation, in that frame, is one of many orchestration patterns the platform handles, sitting upstream of the close. The pattern that maps to most reconciliation pain looks like this: 1. A trigger fires (a Stripe webhook, a new invoice in QuickBooks, an email landing in a shared inbox) 2. A workflow gathers context from every system it needs (Stripe, QuickBooks, Acumatica, a distributor portal, a parsed inbox document) 3. The workflow attempts to match the records mechanically. If everything ties, it writes a clean entry and moves on 4. If the match has an exception (mismatched amount, missing reference, duplicate risk, unexpected fee deduction), the workflow pauses and calls a human as an action, not a status gate 5. The human receives a Slack message (or email, or WhatsApp message) with the full context attached and a structured response (approve, reject, adjust, escalate) 6. The workflow resumes from the response, writes the entry with the human’s decision captured, and produces an audit-trail record of who decided what at what time The shape of that last point is the thing FlowRunner does that close-management software does not. Human-in-the-loop is a [callable action](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack), not a checklist status. The workflow pauses, routes to a named approver, captures the response, continues. The audit trail records the named approver, not “approved at 3:14 PM by someone with access to the queue.” Concretely, this is the same pattern as [automated payment-to-invoice reconciliation across Stripe and QuickBooks](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack), where matched payments flow through and mismatches pause for a named reviewer. Or [reconciling Stripe payments with QuickBooks and pausing mismatches for review](https://flowrunner.ai/workflows/connect-quickbooks-online-with-stripe). Or [end-to-end invoice processing from inbox to ERP](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack), with duplicate detection and named approver routing for exceptions. Or [orchestrating accounting workflows in Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) where ERP entries and downstream notifications happen in the same workflow, with the same audit trail, governed by the same RBAC. The category that owns this layer is orchestration as a service: a system above the systems of record that listens for what they emit, gathers context the rule did not have, pulls a human in at the moments that need judgment, and writes the audit trail of who decided what across the whole workflow. FlowRunner is built for that layer. FloQast is built for the close that sits downstream of it. ### Where FlowRunner reaches that close software does not The brief from a CFO buyer almost always includes tools no close-management product natively connects to. Distributor portals with no public API. Parsed inbox documents that need to land in the ERP without a human retyping them. Payment processor webhooks that fire seconds after a charge, not on a nightly batch. The list is long and unique to every company. FlowRunner addresses this with two architectural choices the close-management category does not share: - **A broad orchestration substrate.** Stripe, QuickBooks Online, Acumatica, NetSuite, Slack, email, WhatsApp, parser services like Parseur, and most of what mid-market operations stacks actually run. [See the integrations catalog](https://flowrunner.ai/integrations) for the current list. - **Native MCP support and an open Agent Directory.** When the tool you need to reach is not in the catalog, MCP gives you a published protocol for connecting it. The Agent Directory lets finance teams use pre-built agents for common patterns rather than building from scratch every time. This is the differentiator paired with FlowRunner’s honest limitation: our breadth of out-of-the-box integrations is younger and narrower than FloQast’s established accounting ecosystem within the close-management category. We are not the better product for the close itself. We are the better product for orchestrating across the long tail of tools the close-management category does not natively reach. ### Governance that does not require an enterprise procurement cycle Mid-market finance teams need audit trails, role-based access control, and SSO without a six-month procurement process. The standard pattern in the close-management category is to gate those features behind enterprise pricing that demands an RFP, a security review, and a vendor approval cycle. FlowRunner publishes those features at the [Professional tier ($299 a month)](https://flowrunner.ai/pricing). Audit trails, RBAC, and SSO are designed to meet common audit requirements for mid-market finance operations. The framing matters: we are not claiming auditor acceptance of any specific compliance framework. We are claiming that the governance infrastructure exists at a price a finance director can authorize without an enterprise procurement cycle. If your auditor’s scope requires SOC 2 attestation or HIPAA, those conversations are separate. The honest version of this comparison: a team running FloQast for close management probably already has the close-side governance they need. Where FlowRunner adds value is the workflow-level governance across the cross-system orchestration work that lives outside the close. Those audit trails answer different questions than the close’s audit trails do. ### Where FloQast is the better fit Buy FloQast (and probably nothing else for the close itself) when: - Your accounting team is running a structured month-end close with sequenced tasks, sign-offs, and tie-outs - Your reconciliations are GL-account-anchored and the close-management workflow is what you need to manage - Your team is already trained on close-management workflow conventions and the cost of retraining outweighs orchestration benefits elsewhere - The cross-system orchestration work outside the close is small or already absorbed by existing tools That set of conditions describes a real and common finance organization. If it describes yours, the rest of this article is interesting but not actionable. ### Where FlowRunner is the better fit Choose FlowRunner when: - The reconciliation pain is between systems, not inside one accounting product - Exceptions are the work, and you need a callable human-in-the-loop, not a status gate - You need to orchestrate across Stripe, QuickBooks, Acumatica, distributor portals, inboxes, or other tools no close-management product natively connects to - You want governance infrastructure (audit trails, RBAC, SSO) at mid-market pricing without an enterprise procurement cycle - You want non-developers in finance ops to configure new reconciliation logic without waiting on an engineering sprint - You see AI agents arriving in your stack from multiple vendors and want a coordination layer above them before that becomes a problem The two products coexist comfortably in the same finance stack. FlowRunner handles the exceptions and the handoffs upstream and around the close. FloQast handles the close. The clean items hit FloQast already tied out. The exceptions arrive with the context already attached. ### How to decide A two-question test, with the order intentional: **1\. Where is the pain?** Walk through the last three reconciliation problems your team escalated. Did they fail because the close workflow could not certify a tied-out account, or did they fail because two systems disagreed and the difference had to be chased across email, Slack, and a spreadsheet before anything could be certified? The first answer points to close-management software. The second answer points to an orchestration layer. **2\. Where does the exception live?** When a duplicate payment risk surfaces, or a distributor billback dispute arrives, or a Stripe payout does not match an invoice batch, what tool is the resolution happening in today? If the answer is “an email chain and a spreadsheet,” that is the work an orchestration layer is built to absorb. The close-management product receives the resolved entries; it is not built to drive the resolution. Most mid-market finance teams will end up with both products in the stack, used for different work. That is not a hedged answer. It is the honest one. ### Quick answers #### Is FlowRunner a replacement for FloQast? No. FloQast is a financial close management platform. FlowRunner is an orchestration layer that coordinates work and AI agents across the tools around the close. Teams running their close in FloQast should keep it. FlowRunner is for the reconciliation work happening between systems that close software was never built to reach. #### Where does FlowRunner fit if we already use FloQast? FlowRunner sits upstream of the close. It reconciles payments to invoices in real time, parses inbox documents into ERP entries, escalates mismatches to a named approver in Slack, and writes a structured audit trail. The cleanly resolved items hit your books before the close starts. The exceptions arrive with the context already attached. #### Is FlowRunner cheaper than FloQast? FlowRunner and FloQast are different categories of product, so a direct price comparison is misleading. FlowRunner’s Professional tier at $299 a month includes audit trails, RBAC, and SSO. FloQast pricing is not publicly listed and is sold per accounting seat. The right question is not which is cheaper, but which problem you are buying for. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Make.com for Salesforce Data Hygiene: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-make-for-salesforce-data-hygiene Comparisons May 28, 2026 Updated September 7, 2026 13 min read Make wins when Salesforce hygiene is scheduled, rule-based cleanup across many tools. FlowRunner wins when merges and conversions need a human in the loop. ![A bald cartoon man at a desk watching clean records run off a green conveyor belt on his left while two near-identical contact cards joined by a question mark wait on an amber decision tray on his right with an audit clipboard beside them.](https://flowrunner.ai/images/blog/flowrunner-vs-make-for-salesforce-data-hygiene-hero.webp) When a list import drops two records into Salesforce that look like the same company spelled three different ways, which tool should make the call, and who signs off on the merge? That single question separates [Make.com](https://www.make.com/en) and FlowRunner for Salesforce data hygiene more cleanly than any feature table. Make is built to run the cleanup that follows a rule. FlowRunner is built for the part of the cleanup that needs a judgment call and a record of who made it. The honest version of this comparison is not which platform is better at hygiene. It is whether your hygiene workload is mostly rules or mostly exceptions, because the answer points at different tools. ### How the two stack up for CRM hygiene work | | Make.com | FlowRunner | | --- | --- | --- | | **Self-description** | [”The visual AI automation platform”](https://www.make.com/en) | Orchestration as a service for coordinating AI agents and humans across systems | | **Built for** | Scheduled, rule-based scenarios connecting many apps | Coordinating agents and human decisions on cross-system work | | **Salesforce hygiene fit** | Strong for deterministic cleanup: find records, standardize fields, dedupe on clear rules, create and convert | Strong for the judgment calls: ambiguous merges, lead conversion, ownership reassignment with a reviewer in the loop | | **Connector library** | [1,000+ app integrations](https://www.make.com/en/integrations) built over years (catalog now claims 3,000+) | Smaller pre-built catalog today; new connectors built in 30 minutes or less | | **Human review** | Approval and routing built as a gate inside a scenario | Callable step: an agent pauses on uncertainty and calls a named reviewer with full context, then resumes on the response | | **AI agents** | [Real capability layered on the scenario foundation](https://www.make.com/en), with [MCP and connections to Claude, ChatGPT, OpenAI](https://www.make.com/en/help/ai-agents) | Agents are first-class workflow nodes; multi-agent coordination is the base model | | **Entry pricing** | [Free tier (1,000 credits/mo); $12 Core](https://www.make.com/en/pricing) | Free (100 executions/mo); $5 Starter; $45 Growth | | **Audit trails and RBAC** | [Teams tier ($38/mo) and above for audit logs, SSO, roles](https://www.make.com/en/help/organization-and-teams) | [Audit trails and RBAC at $299 Professional; SSO/SAML at $999 Business](https://flowrunner.ai/pricing) | | **Ecosystem** | Mature: established community, Academy, partner network | Newer in the visual workflow space; ecosystem still building | | **Decision audit trail** | Available through audit logs at the Teams tier | Per-record log of every create, update, merge, and human decision | The table covers the shape of the choice. The part that decides it is what happens to the records a rule cannot resolve, and the rest of this article is about that. ### What Make.com is genuinely good at Make is a mature, polished visual scenario builder, and pretending otherwise would not survive contact with the product. Make calls itself [“the visual AI automation platform”](https://www.make.com/en). It frames its job as letting users “connect any app, data source, or AI model” and “build and manage automations and AI agents, visually, in code, or with a prompt.” For Salesforce data hygiene that follows a rule, that is an accurate and capable description. Specific strengths this article is not going to relitigate: - A [large app library](https://www.make.com/en/integrations), with the catalog now claiming 3,000+ integration apps and including the tools a sales ops team already runs alongside Salesforce: Google Sheets, HubSpot, Airtable, Notion, LinkedIn, Slack, and many more. That breadth is a real advantage over FlowRunner’s growing library today, and it matters when hygiene spans tools beyond the CRM. - A visual scenario builder with a long track record and the kind of canvas a non-technical ops user can learn without an engineering background. - A mature ecosystem: an [established community, an Academy, and a partner network](https://www.make.com/en/help) that FlowRunner is still building. Years of market presence count for something real here. - AI agents shipped as a genuine capability, with Make describing [“transparent AI agents that take action and orchestrate complex workflows across 3,000+ apps”](https://www.make.com/en) and supporting [MCP and connections to Claude, ChatGPT, and OpenAI](https://www.make.com/en/help/ai-agents). The agents are real and documented, not vaporware. - Entry pricing that opens at a [Free tier and a $12 per month Core plan](https://www.make.com/en/pricing), accessible for a small team or a solo operator running simple, linear scenarios. FlowRunner’s Free plan and $5 Starter sit alongside it, so the entry price is a match rather than a Make advantage; Make’s edge at this level is its scenario builder and connector directory, not the sticker. For a recurring hygiene job that runs on a schedule, that is a strong toolset. A nightly scenario that finds records matching a clear rule, standardizes country codes and industry picklists, dedupes on an exact email match, and writes the results back to Salesforce is exactly what Make’s scenario model was built to do. If that describes your hygiene problem, Make is a real answer. ![A muted forest green scenario canvas showing a linear chain of connected modules, each a small labeled tile (a Salesforce record icon, a filter, a field-standardize step, an update step), running left to right across a cream background in clean black ink linework, the deterministic cleanup that runs on a schedule](https://flowrunner.ai/images/blog/flowrunner-vs-make-for-salesforce-data-hygiene-1.webp) ### What Salesforce data hygiene actually involves Here is the read most data hygiene content avoids, because it complicates the sales pitch for a dedup engine: the dedup engine is the easy 80 percent. The judgment call is the other 20 percent, and the 20 percent is where the risk and the real work live. Salesforce’s own guidance frames [data quality](https://help.salesforce.com/s/articleView?id=sf.data_quality.htm) around a recurring set of tasks. In practice, ongoing hygiene comes down to a handful of operations: - **Duplicate detection and merging.** Finding records that represent the same entity and combining them without losing field history or related records. - **Lead-to-contact conversion.** Promoting a qualified lead into a contact and account, with the right account match attached. - **Field standardization.** Normalizing country codes, industry picklists, phone formats, and company name casing so reporting holds together. - **Ownership reassignment.** Moving records to the right rep or team when territories shift, an owner leaves, or an account gets reclassified. Two of those four are mostly rules. Field standardization follows a normalization table. Exact-match deduplication follows a key. A scheduled scenario handles both well, and this is squarely Make’s territory. The other two are mostly judgment. A duplicate where the email differs and the company name is spelled three ways is not a rule, it is a decision. A lead that scored hot but may not match a target account is not a rule, it is a decision. An ownership reassignment that affects a rep’s commission is not a rule, it is a decision someone needs to own and stand behind later. These are the cases that produce hygiene debt, because a deterministic scenario has only two ways to handle them: guess, or stop. Guessing creates a bad merge that someone discovers three months later. Stopping leaves the record in limbo until a human notices. That is the seam. Rule-based hygiene runs itself. Judgment-based hygiene needs a person, the right context in front of them, and a record of what they decided. The question is which tool is built for that second shape of work. ### Where FlowRunner fits: the judgment call as a first-class step [FlowRunner](https://flowrunner.ai) treats human review as a first-class workflow primitive, where an agent can pause and call a reviewer on uncertainty rather than only running a deterministic scenario from start to finish. That sentence is the whole architectural difference, so it is worth unpacking against a real hygiene flow. The pattern for ambiguous-merge handling looks like this: 1. An intake event fires (a list import row, an enrichment webhook, a web-to-lead submission). 2. A workflow gathers context: it runs a match query against Salesforce and pulls the candidate plus the closest existing records. 3. An agent attempts the mechanical decision. Clean single match, it updates. No match, it creates. This is the 80 percent, and it runs without a human. 4. On an ambiguous case (two plausible matches, or one match where critical fields disagree), the agent does not guess and does not silently stop. It calls a human as a callable step. 5. The reviewer receives a structured message in Slack, email, or WhatsApp: the candidate record, the closest matches, the fields that overlap, the fields that disagree, and three actions (merge into existing, create new, update existing). 6. The workflow resumes from the reviewer’s choice, writes the decision back to Salesforce, and records who decided what, when, and why in a per-record audit trail. Step four is what a scenario tool does not do natively. In a scenario built for deterministic automation, an approval is a gate you wire in: a router, a notification, a wait state. It works, and it is the right pattern when the rule for escalation is known in advance (“anything over $X needs sign-off”). FlowRunner’s difference is that the agent escalates on its own confidence, not on a fixed threshold. The agent works the merge until it hits the part it should not decide alone, and at that point it hands the reviewer a worked problem rather than a raw alert. The compositional shape of the platform reinforces the pattern. A Make scenario starts with a trigger module by design; the first step is always the event that woke the scenario up. A FlowRunner flow can start with any node type: an action, a trigger, a condition that evaluates the current state of an account, a group of steps invoked together. The same flow that runs on a list-import webhook can be called manually by an ops lead resolving a specific record, scheduled to sweep stale leads on a Sunday night, or invoked as a subflow from inside another agent’s logic when that agent decides it needs the merge resolution as a tool. The reviewer-callable pattern is one expression of the same idea: a flow is a function the platform can call, not only a pipeline a trigger can fire. For hygiene work where exceptions surface from many directions, that compositional flexibility is what lets the same judgment-call logic serve every entry point without rebuilding it per scenario. This is the same shape laid out in the deep dive on treating [Salesforce data hygiene as a control surface rather than a quarterly cleanup project](https://flowrunner.ai/workflows/salesforce-data-hygiene). It is implemented end to end in [Salesforce workflows with human-in-the-loop approvals](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack). The mechanic also shows up one CRM over, where a workflow can [flag duplicate contacts before they ever land in the CRM](https://flowrunner.ai/workflows/orchestrate-google-forms-and-hubspot-and-slack) on the HubSpot side, and in [sales ops orchestration across CRM and calendar](https://flowrunner.ai/workflows/orchestrate-calendly-and-hubspot-and-slack). The records land in a different system; the judgment-call-at-the-seam pattern is identical. There is a category forming around this, and the merge decision is a clean way to see it. When a reviewer resolves an ambiguous merge, the artifact that matters six months later is not the merged record. It is the answer to “why did these two become one, who decided it, and what disagreed at the time.” That answer lives between systems: the intake source, the CRM, the reviewer in Slack. No system of record was built to own it, and no scheduled scenario produces it as a byproduct. An orchestration layer is the category that owns that seam: a layer above the CRM and the intake tools that gathers the cross-system context a rule never had, pulls a human in on the decisions that need judgment, and keeps the decision trail as a first-class artifact. FlowRunner is built for that layer. That is also why the audit trail sits in the platform rather than in a tier: every create, update, merge, and human decision is logged per record, because in hygiene work the decision log is the point. ![A Slack-style review card on a cream background in black ink linework, showing two candidate contact records side by side with a few fields highlighted in faded amber to mark disagreement, and three labeled buttons beneath them reading merge, create new, and update, the structured judgment call an agent hands to a human reviewer](https://flowrunner.ai/images/blog/flowrunner-vs-make-for-salesforce-data-hygiene-2.webp) ### The honest part: where Make is stronger A comparison that gives FlowRunner every axis is not worth reading, so here is where Make is the better tool today, plainly. **Connector breadth.** Make’s library is materially larger. The catalog [claims thousands of integration apps](https://www.make.com/en/integrations) against FlowRunner’s growing set. If your hygiene work reaches across many tools beyond Salesforce, and you need those connectors out of the box today, Make has the immediate advantage. FlowRunner builds new connectors in 30 minutes or less, but a fast build is not the same as a connector that already exists, and honesty requires saying so. **Maturity and ecosystem.** Make has years of market presence, an [established community, an Academy, and a partner network](https://www.make.com/en/help). FlowRunner is newer in the visual workflow space. When something breaks at 11pm, a deep community and a library of existing tutorials is a real asset, and Make has both built up over time. **The deterministic builder itself.** Make’s scenario builder is a polished, proven product for non-AI automation. For scheduled, rule-based hygiene that does not need a judgment step, that maturity is worth a lot, and FlowRunner is the younger product on that specific axis. **Entry price.** [Make’s Free tier and $12 Core plan](https://www.make.com/en/pricing) sit alongside FlowRunner’s Free plan (100 executions per month) and $5 Starter, so neither side wins on sticker. The honest read on the pricing models is that they measure different things and the comparison tilts further toward FlowRunner as flows grow. Make charges per operation, so every module that fires inside a scenario consumes monthly quota: the match query, the candidate fetch, the agent step, the notification, the wait, the write-back, the audit log entry. A single ambiguous-merge flow can spend ten or twenty operations resolving one record. FlowRunner counts executions, where one flow run is one execution whether the flow contains five nodes or five hundred. Tier limits are 12,000 executions on Growth at $45, 75,000 on Professional at $299, and 250,000 on Business at $999. At low volume with simple scenarios, the two entry tiers cost about the same. As flows pick up more steps (more enrichment, more branching, more context for the reviewer), FlowRunner’s per-execution model is the one that holds its cost shape. Neither pricing model is objectively right; they reward different work. None of that is faint praise. For a sales ops team whose hygiene problem is genuinely a scheduled, rule-shaped cleanup across many connected tools, Make is the more sensible buy, and the AI agent layer it has added is a real capability rather than a checkbox. ### Where FlowRunner is the better fit Choose FlowRunner when the hygiene work is shaped by exceptions rather than rules: - **The hard cases are merges, conversions, and reassignments, not standardization.** When the records eating your team’s time are probable duplicates with conflicting fields, leads that need a human to confirm the account match, or ownership changes that affect commission, you need escalation on uncertainty, not a fixed gate. - **Every hygiene decision needs a paper trail.** When “who merged these two accounts and why” has to be answerable months later, a per-record decision log built into the platform beats reconstructing it from a scenario’s run history. - **You want governance at mid-market pricing.** Every FlowRunner plan, from Free through Growth at $45/mo, includes AI agents, BYOK, all integrations, and human-in-the-loop with unlimited users and unlimited workflows, so the core hygiene job (escalating ambiguous merges to a reviewer) runs before you pay anything. [FlowRunner adds audit trails and RBAC starting at the $299 Professional tier, with SSO/SAML available on the $999 Business tier](https://flowrunner.ai/pricing). Make places [audit logs, SSO, and team roles at its Teams tier and above](https://www.make.com/en/help/organization-and-teams). For regulated or audit-heavy teams, where those controls sit in the plan changes the comparison. - **You want non-developers to own the hygiene agent.** Operations teams build and adjust the workflow visually, including the rule for what gets escalated, without filing an engineering ticket. - **The reviewer should get a worked problem, not a raw alert.** When you want the agent to hand a person the candidate, the matches, and the disagreements already laid out, rather than pinging them to go investigate, the callable-human pattern is the difference. These are not mutually exclusive worlds. A team can reasonably run deterministic standardization in Make and route the ambiguous merges through FlowRunner. The architectural difference is real, and the right tool follows the shape of the work rather than a feature checklist. This article focuses on hygiene; the same Make-versus-FlowRunner trade-off shows up across other use cases too, laid out in the [comparison for marketing workflow automation](https://flowrunner.ai/blog/flowrunner-vs-make-for-marketing-workflow-automation). ![A per-record audit trail rendered as a clean vertical timeline on a cream background in black ink linework, each entry a small labeled row (record created, fields standardized, match found, human review requested, merge approved by reviewer), with the human-review row marked in faded amber and the automated rows in muted forest green, the decision log that survives the next audit](https://flowrunner.ai/images/blog/flowrunner-vs-make-for-salesforce-data-hygiene-3.webp) ### How to decide Pull your last twenty hygiene fixes and sort them into two piles. Pile one: the fixes that followed a rule. Standardize this picklist, normalize these phone numbers, merge these exact-email duplicates, create these records that clearly do not exist yet. If your twenty fixes land mostly here, your hygiene problem is deterministic, a scheduled scenario handles it, and Make is the stronger and more cost-effective choice. This article is over and you should go build the scenario. Pile two: the fixes that needed someone to decide. The probable duplicate where the email differed and the company name was spelled three ways. The lead that scored hot but might not match the target account. The ownership reassignment that touched a rep’s number and needed a manager to sign off. Every one of those is a record that sat in limbo, or got a guess that someone unwound later, or pulled a person away mid-task to investigate with half the context. If your twenty fixes land mostly here, your hygiene problem is an exception-handling problem, and the question is which layer lets an agent escalate the call to a human with the candidate, the matches, the disagreement, and a decision trail that survives the next audit. That is the work FlowRunner is built for, and it is the work that does not fit inside a deterministic scenario no matter how many connectors sit behind it. ### Quick answers #### Can Make.com do Salesforce data hygiene? Yes. Make connects to Salesforce and can run scheduled scenarios that find records, update fields, standardize values, and create or convert records. For rule-based cleanup that runs on a schedule across many connected tools, Make is a capable choice. The limit is not whether Make can touch Salesforce. It is what happens to the records that are not a clean rule: the probable duplicate where the email differs, the lead that scored hot but may not match a target account, the ownership reassignment that needs a manager’s call. #### What does FlowRunner do for Salesforce hygiene that Make does not? FlowRunner treats human review as a first-class workflow step. An agent works a hygiene decision (deduplicate, convert a lead, reassign an owner) against its context, and when it cannot resolve the decision confidently it pauses and calls a named reviewer in Slack, email, or WhatsApp with the candidate, the closest matches, and the fields that disagree. The reviewer’s decision writes back to Salesforce and lands in a per-record audit trail. Make can build an approval gate inside a scenario; the difference is that in FlowRunner the agent escalates on its own uncertainty rather than at a fixed rule, and the decision log is part of the platform. #### Is Make.com cheaper than FlowRunner for this? Not on the sticker. Make starts with a Free tier (1,000 credits per month) and a $12 per month Core plan. FlowRunner has a permanent Free plan at 100 executions per month and a Starter plan at $5 for 300 executions or $15 for 3,000, and every plan, including Free, gets AI agents, BYOK, all integrations, and human-in-the-loop (Slack, email, WhatsApp, phone) with unlimited users and unlimited workflows. The units differ: Make’s credits count every module that fires, so a ten-module scenario is roughly ten credits per run and Core’s 10,000 credits are about 1,000 runs, while FlowRunner counts a whole run as one execution. Growth at $45 per month covers 12,000 of them. There are no technical limits on the Free plan that would prevent running a hygiene workflow with agent escalation; the caps are volume, one execution at a time, and 24 hours of log history. The comparison shifts when governance becomes a requirement. FlowRunner includes audit trails and RBAC starting at the $299 Professional tier, with SSO/SAML available on the $999 Business tier. Make places audit logs, SSO, and team roles on its Teams tier and above. For hygiene work where every merge and conversion needs a logged, reviewable trail, where those controls sit in the pricing matters more than the entry price. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Procurify: An Honest Read on Automation in Procurement Source: https://flowrunner.ai/blog/flowrunner-vs-procurify-automation-in-procurement Comparisons May 26, 2026 Updated May 28, 2026 12 min read Procurify is a procure-to-pay application. FlowRunner is an orchestration layer. The honest comparison is about scope, not features, and most procurement teams need a clear answer to which problem they are buying for. ![A bald cartoon man at a workshop table standing between a tall labeled filing cabinet on his left and a tabletop of threads connecting scattered objects on his right, considering both as different shapes of procurement work.](https://flowrunner.ai/images/blog/flowrunner-vs-procurify-automation-in-procurement-hero.webp) Most articles framed as automation in procurement try to crown a winner between products that are not actually competing. The honest version of the [Procurify](https://www.procurify.com/product/) versus FlowRunner question is a scope question, not a feature question. Procurify is a procure-to-pay application with native depth in purchase requests, POs, receiving, AP invoice processing, bill management, and vendor payments. FlowRunner is an orchestration layer that coordinates work and AI agents across the tools a procurement team already runs. A procurement leader who confuses the two ends up either underbuying (an orchestration layer when they needed a P2P system) or overbuying (a P2P application when the real bottleneck was the work happening between systems). ### Side by side, at a glance | | Procurify | FlowRunner | | --- | --- | --- | | **Category** | Procure-to-pay application | Orchestration layer for work and AI agents across tools | | **Primary job** | Run the structured P2P cycle: requests, POs, receiving, AP invoice processing, bill management, vendor payments | Coordinate exceptions and handoffs across procurement and adjacent processes | | **Built for** | Procurement and AP teams in mid-market companies | Procurement, finance, and ops leaders coordinating work across multiple systems | | **Catalog purchasing** | Native PunchOut catalogs with Amazon Business, Grainger, Home Depot, Staples, Uline | Not a procurement feature; orchestration around purchasing happens outside the catalog | | **PO matching** | Two-way and three-way matching against POs, receipts, and invoices | Not a native P2P feature | | **ERP integrations** | Pre-built connectors to QuickBooks, NetSuite, Sage Intacct, Microsoft Dynamics 365 Business Central | Broad orchestration substrate, expandable with new connectors in 30 minutes or less, narrower out-of-the-box than Procurify in the P2P-adjacent finance category | | **Spending Cards** | Physical and virtual cards for decentralized purchasing, tied to the P2P platform | Not offered | | **Mobile app** | Purpose-built for purchasing, with role-specific views for requesters, approvers, receivers | No mobile purchasing interface | | **Human-in-the-loop** | Approval gates inside the P2P workflow | Callable action: agents pause and route to a named human via Slack, email, WhatsApp, or phone with full context, resume on response | | **Workflow scope** | Procurement and AP, end to end | Any ops workflow, with or without procurement involvement | | **Governance pricing** | Tier-gated; pricing not publicly listed | Audit trails, RBAC, and SSO at the [Professional tier](https://flowrunner.ai/pricing) ($299/mo) | The table tells most of the story. The rest of this article is the part the table cannot. ### What Procurify is good at, said honestly Procurify describes itself as [“the leading AI-powered procurement, AP, expense, and payment platform for the mid-market”](https://www.procurify.com/product/) that “standardizes and streamlines the purchasing process from request to payment.” That positioning is accurate. Procurify is a mature, purpose-built procure-to-pay application that covers the full P2P arc in one product, with the kind of domain depth that accumulates from working with procurement teams across years of actual buying cycles. Specific things Procurify does that FlowRunner does not: - **Native PunchOut catalogs with major suppliers.** Procurify [publishes integrations](https://www.procurify.com/integrations/) with Amazon Business, Staples Advantage, Home Depot, Grainger, Uline, and others as embedded purchasing experiences. A buyer punches out to the supplier site, fills a cart, and the cart returns into Procurify as a structured request. FlowRunner has nothing equivalent. - **Two-way and three-way matching against POs, receipts, and invoices.** Procurify positions bill management as [“speed up reconciliation with automated three-way matching and sync approved bills from Procurify to ERP.”](https://www.procurify.com/product/) Three-way match is a core P2P control. FlowRunner does not implement it as a native feature. - **Pre-built, certified ERP and accounting integrations.** Procurify ships connectors to QuickBooks (Online and Desktop), NetSuite, Sage Intacct, and Microsoft Dynamics 365 Business Central with the operational depth that comes from years of mid-market accounting integration work. FlowRunner’s connector library is broad and expandable in 30 minutes or less, but does not match Procurify’s pre-built depth in this specific category. - **Physical and virtual Spending Cards.** Procurify [offers Spending Cards](https://www.procurify.com/spending-cards/) for decentralized purchasing, tied to the same platform that owns the rest of the P2P workflow, with real-time spend visibility and automated reconciliation against the buying record. FlowRunner does not issue cards. - **A purpose-built mobile app with role-specific views.** Procurify’s mobile app handles requests, approvals, receiving, and expenses with views designed for the requester, the approver, and the receiver as distinct roles. FlowRunner has no comparable mobile purchasing interface. - **Domain depth across the full P2P cycle.** Blanket POs, accrual tracking, vendor onboarding flows tied to the buying record, role-specific approval views: Procurify has accumulated these by serving procurement teams as its primary buyer for years. FlowRunner does not. This is not a list of FlowRunner gaps to be filled later. It is the deliberate scope difference between a procure-to-pay application and an orchestration layer. A procurement team that needs a complete P2P system should buy one. If the comparison stopped at the feature checklist, this article would end here with “buy Procurify.” The reason it does not is that the comparison rarely stops there once the buyer walks through what actually happens between the systems. ### What automation in procurement actually looks like for most teams Here is what most posts on automation in procurement will not say plainly: the procurement application owns the structured part of the buying motion. The unstructured work happens around it. For a large slice of mid-market procurement and finance teams, the day-to-day pattern looks like this: - A vendor invoice arrives in a shared inbox in PDF or email body form. Somebody parses it (or asks a teammate to), enters it somewhere, and reconciles it against a PO that may or may not exist - A distributor sends a billback or chargeback statement that does not map cleanly to the goods received, and the dispute has to be chased across email, the ERP, and a spreadsheet before anyone can approve or reject it - A new vendor needs to be onboarded with W-9, insurance certificate, contract reference, banking details: documents that arrive piecemeal over a week and need to land in the right systems before the first PO can clear - A rush order or unusual request lands at a threshold the routing rule does not handle gracefully, and a procurement coordinator manually nudges it through Slack, attaching context that was already gathered but lives in three other places - A request gets approved against a vendor record that has not been reviewed in two years, against a contract reference that may or may not still be valid A P2P application owns the part of this that fits into the request-PO-receive-pay arc. That part is real and Procurify does it well. The parts that do not fit live in inboxes, Slack threads, spreadsheets, and the operator’s head. They are the exception work, and they are typically where procurement teams spend the time they want back. The standard advice on automating procurement is to pick a procurement application. The honest version is that picking one solves the structured part of the problem and surfaces the unstructured part as a separate question. ### Where FlowRunner fits: the layer around the P2P boundary [FlowRunner](https://flowrunner.ai/) is an orchestration layer. It coordinates work and AI agents across the tools a team already runs, regardless of whether those tools are inside or outside the procure-to-pay boundary. Procurement work, in that frame, is one set of orchestration patterns the platform handles. Vendor invoice intake, exception routing, vendor onboarding documentation, and the cross-system handoffs that touch procurement plus adjacent processes are the patterns where FlowRunner does work a P2P application is not built to do. The shape of those patterns: 1. A trigger fires (a vendor email arrives, a parsed document lands, a webhook from a distributor portal posts, an ERP record changes) 2. A workflow gathers context across every system involved (the ERP, the inbox, a parser like Parseur, a distributor portal, an internal vendor record) 3. The workflow attempts the mechanical part of the decision: does this invoice match a PO, is this vendor approved, does this threshold trigger a category review, is this a duplicate of something processed yesterday 4. If the answer is clean, the workflow writes the entry, posts the notifications, and moves on without involving a human 5. If the answer carries an exception (mismatch, missing reference, new vendor, duplicate risk, unusual amount, missing documentation), the workflow pauses and calls a human as a [callable action](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents), not a status gate. The human receives a Slack message, email, WhatsApp message, or phone outreach with the full context attached and a structured response (approve, reject, adjust, escalate) 6. The workflow resumes from the response, writes the entry with the human’s decision captured, and produces an audit-trail record of who decided what at what time Concretely, this is the same shape as [parsing vendor documents into Acumatica with human review on exceptions](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack), where matched documents flow into the ERP and exceptions pause for a named reviewer with the parsed extract attached. The same shape as [processing vendor documents into NetSuite without manual entry](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack), where duplicate and outlier checks happen before a bill is created. The same shape as [invoice processing from email to payment](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack), where the inbox is the trigger and the audit trail spans the parser, the ERP, and the Slack approval. The same shape as [automating accounting workflows in Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica), where ERP entries and downstream notifications happen in one workflow with one audit trail. The category that owns this layer is orchestration as a service: a system above the systems of record that listens for what they emit, gathers context the procurement application did not have, brings a human in only when judgment is required, and writes one audit trail across the whole motion. FlowRunner is built for that layer. Procurify is built for the structured P2P workflow that sits inside it. The honest version of the human-in-the-loop difference, said carefully so it does not become a strawman: Procurify’s approval gates are a legitimate and well-designed control for the procurement workflow they govern. They route requests through structured thresholds, capture approver identity, and write the approval against the PO record. That is a real control and procurement teams rightly value it. FlowRunner’s escalation model is a different shape, suited to a different kind of decision. Where a procurement approval gate fires on a known threshold against a known request type, an orchestration-layer escalation fires when an agent or workflow encounters something it cannot resolve on its own (a vendor not on the approved list, an invoice that almost matches a PO but is short by a freight line, a parsed document with a low-confidence field) and needs a human to decide. Both are forms of human-in-the-loop. They are designed for different work. ### Where FlowRunner reaches that a P2P application does not The procurement team’s actual tool inventory almost always extends past what a P2P application natively connects to. Internal vendor portals with no public API. Parsed inbox documents that need to land in two systems at once. Distributor portals whose statements need to be reconciled before AP can pay. Slack workflows where exception decisions are already happening and the audit trail leaks. Custom approval thresholds that need to combine GL coding plus vendor category plus contract status. FlowRunner addresses this with two architectural choices a procurement application does not share by design: - **A broad, extensible orchestration substrate.** Stripe, QuickBooks Online, Acumatica, NetSuite, Slack, email, WhatsApp, parser services like Parseur, internal APIs, and most of what a mid-market ops stack actually runs. New connectors are built in 30 minutes or less when the catalog does not already cover the system the team needs to reach. - **A visual builder for non-developers.** A procurement operator or ops analyst can compose a workflow that combines a parser, an ERP write, a Slack approval, and an audit-trail entry without filing an engineering ticket. The bar to add a new exception path is low enough that the workflow can evolve with the buying motion instead of becoming a maintenance debt. Said honestly: this is the differentiator paired with a real FlowRunner limitation. The breadth and depth of Procurify’s certified ERP and accounting integrations in the specific finance-systems category exceeds FlowRunner’s pre-built depth there. FlowRunner is not the better choice for the structured P2P workflow itself. It is the better choice for orchestrating across the long tail of tools and exception patterns a P2P application is not designed to reach. ### Governance at a price a procurement director can authorize Mid-market procurement and finance teams typically need audit trails, role-based access control, and SSO without a six-month enterprise procurement cycle. The standard pattern in the procurement-application category is to publish governance features behind tiered pricing that asks for an RFP, a security review, and a vendor approval committee. FlowRunner publishes audit trails, RBAC, and SSO at the [Professional tier](https://flowrunner.ai/pricing) ($299 per month). The framing matters: this article is not claiming auditor acceptance of any specific compliance framework. It is claiming the governance infrastructure exists at a price a procurement director or finance director can authorize without an enterprise procurement cycle. Specific framework attestations (SOC 2, HIPAA) are separate conversations. The honest version: a team running Procurify for P2P already gets procurement-specific audit trails for the buying record inside the Procurify application. Where FlowRunner adds workflow-level governance is across the cross-system orchestration work happening around the P2P boundary, where the exception decisions, the parsed documents, and the Slack approvals would otherwise produce audit gaps that the P2P application is not aware of. ### Where Procurify is the better fit Choose Procurify (and probably not FlowRunner as a substitute for it) when: - You need a structured procure-to-pay system as the system of record for purchase requests, POs, receiving, AP invoice processing, and vendor payments - PunchOut catalogs with Amazon Business, Grainger, Home Depot, Staples, or Uline are core to how your team buys - Two-way and three-way matching, blanket POs, and accrual tracking are required controls in your buying motion - You need physical or virtual Spending Cards integrated with the same platform that owns the procurement record - A purpose-built mobile app with role-specific views is required for field, remote, or distributed buyers - The ERP integration you need is already in Procurify’s certified list and the depth of that integration matters more than orchestrating across the long tail of systems outside it That set of conditions describes a real and common procurement organization. If it describes yours, the rest of this article is interesting but not actionable. The right answer is Procurify or a peer P2P application, and an orchestration layer is a question to revisit later. ### Where FlowRunner is the better fit Choose FlowRunner when: - The procurement work that consumes your team’s time lives between systems rather than inside one P2P application - Vendor invoices, distributor statements, vendor onboarding documents, and chargeback notices arrive across email, parsers, portals, and Slack threads that need to be coordinated into one workflow with one audit trail - You need to route exceptions to humans with full context attached, via the channel the human actually reads (Slack, email, WhatsApp, phone), and capture the decision against the source record automatically - The systems your procurement work touches include tools no P2P application natively connects to, and you need new connectors built in 30 minutes rather than a quarterly product roadmap request - You want non-developers in procurement, finance, or ops to configure new exception routing without waiting on engineering capacity - You see AI agents arriving in your stack from multiple vendors (extraction agents, approval bots, supplier-side AI) and want a coordination layer above them before that becomes its own problem - You want governance infrastructure (audit trails, RBAC, SSO) at mid-market pricing rather than enterprise-tier procurement These two products coexist comfortably in the same procurement and finance stack. Procurify owns the structured P2P workflow. FlowRunner orchestrates the exceptions, the cross-system handoffs, and the work that touches procurement plus everything adjacent to it. The structured part flows through Procurify with the controls procurement teams expect. The exceptions and the off-P2P work flow through FlowRunner with the context already attached. ### How to decide Two questions, with the order intentional. **1\. Where is the system of record for the buying motion?** If you do not have one, or the one you have is a spreadsheet and an ERP module that no one trusts, the question is which P2P application to buy, and an orchestration layer is the wrong starting point. Buy Procurify or a peer, get the structured workflow in place, and revisit orchestration once the structured part is solved. If you already have a system of record (Procurify, a peer P2P application, or a tightly used ERP procurement module), the question is what is happening outside it, and that points to an orchestration layer. **2\. Where does the exception live today?** Walk through the last five things that consumed your team’s time and required judgment. The unusual rush request, the vendor invoice that did not match a PO, the new vendor whose paperwork dragged across three weeks, the distributor billback that took four people to resolve, the request approved against a vendor record that turned out to be stale. For each one, name where the resolution actually happened. If the answer is “inside Procurify’s approval gates,” the P2P application is doing the work it was built to do and the orchestration question is small. If the answer is “across an inbox, a parser, a spreadsheet, two Slack threads, and a phone call before anyone updated the system of record,” that is the work an orchestration layer is built to absorb. The honest read on a procurement automation decision is to use the questions in that order. The decision about whether the structured P2P workflow is worth automating is a different question than the decision about whether the surrounding orchestration is worth automating, and a separate framework on [how to know what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) handles the latter as a financial calculation rather than a gut call. Most mid-market procurement teams will end up with both kinds of tool in the stack, used for different work. That is not a hedged answer. It is the honest one. ### Quick answers #### Is FlowRunner a replacement for Procurify? No. Procurify is a procure-to-pay application with purchase requests, POs, receiving, AP invoice processing, bill management, and vendor payments in one product. FlowRunner is an orchestration layer that coordinates work and AI agents across the tools a procurement team already runs. A team that needs a dedicated P2P system would still buy Procurify or a similar tool. #### Where does FlowRunner fit if we already use Procurify? FlowRunner sits around the P2P boundary. It parses vendor documents into systems Procurify is not the system of record for, routes exceptions that need judgment across Slack or email with full context attached, and coordinates work that touches procurement plus adjacent processes (AP exception handling, distributor billbacks, vendor onboarding documentation) in one workflow with one audit trail. #### Does FlowRunner have PunchOut catalogs or three-way matching? Not as a native procure-to-pay feature. Procurify offers PunchOut catalogs with Amazon Business, Grainger, Home Depot, Staples, Uline, and others, plus automated three-way matching against POs, receipts, and invoices. FlowRunner does not replicate those. If those capabilities are core to your buying motion, Procurify is the right product. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs SAP for Vendor Onboarding Software: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-sap-vendor-onboarding-software Comparisons May 27, 2026 Updated May 28, 2026 13 min read SAP wins for organizations inside the SAP suite. FlowRunner wins when vendor onboarding has to ship this quarter without a multi-year suite rollout. ![A bald cartoon man at a workshop bench with a tall green warehouse silhouette behind him representing SAP's suite depth and a single open vendor folder with loose papers in faded amber in front of him representing the immediate onboarding work, weighing the two different scopes.](https://flowrunner.ai/images/blog/flowrunner-vs-sap-vendor-onboarding-software-hero.webp) A procurement team that shortlists [SAP](https://www.sap.com/products/crm/sales-cloud.html) for vendor onboarding is usually answering a different question than the one they typed into a search bar. The visible question is “what vendor onboarding software should we buy.” The actual question buried underneath is “are we an SAP shop or not, because that decision changes which tool is even on the shortlist.” FlowRunner and SAP are not competing for the same buyer most of the time, and the most useful thing this article can do is tell you which buyer you are before the demos start. ### Side by side, at a glance | | SAP (Sales Cloud + SuccessFactors Learning + S/4HANA vendor master) | FlowRunner | | --- | --- | --- | | **Category** | Enterprise suite spanning CRM, HCM, ERP with vendor-adjacent tooling embedded | Orchestration layer for work and AI agents across the tools a team already runs | | **Primary job** | Run the suite-deep enterprise stack with vendor data unified across sales, HR, and finance | Coordinate vendor onboarding, document collection, and exception handling across existing systems | | **Built for** | Enterprises already running SAP, or committing to a multi-year SAP suite implementation | Mid-market procurement and ops teams that need a workflow shipped this quarter | | **AI framing** | [”Agentic orchestration” with Joule](https://www.sap.com/products/crm/sales-cloud.html) providing contextual guidance to sellers across the sales cycle | Agents invoke humans as callable tools when they hit uncertainty, with the human-in-the-loop step as a first-class workflow component | | **Compliance posture** | Validated SaaS environment for [strict requirements for highly regulated industries with annual updates and validated testing](https://www.sap.com/products/hcm/learning-software.html) | Audit trails, RBAC, SSO designed to meet common audit requirements, not a formally validated environment | | **Integration depth** | Native suite continuity across CRM, HCM, ERP without custom connectors | Broad orchestration substrate, new connectors built in roughly thirty minutes when the catalog does not already cover the system | | **Content and partner ecosystem** | [SAP Content Stream by Skillsoft](https://www.sap.com/products/hcm/learning-software/features.html), open content network providers, preconfigured extensions, years of partner depth | Agent Directory is earlier stage; partner network is narrower | | **Mobile** | AI-enhanced Android and iOS apps with natural language navigation and dynamic visit planning, built for distributed field teams | No purpose-built mobile interface for procurement | | **Pricing surface** | Enterprise procurement cycle, custom pricing | [$299 per month](https://flowrunner.ai/pricing) at the Professional tier for audit trails, RBAC, and SSO | | **Time to first workflow** | Weeks to months as part of suite implementation | Days to weeks, configured visually by non-developers | The table tells most of the scope story. The rest of this article is the part the table cannot. ### Who SAP actually serves, said honestly SAP positions Sales Cloud as enterprise-grade AI-embedded sales software built around [“agentic orchestration” that prepares sellers to execute faster](https://www.sap.com/products/crm/sales-cloud.html), with Joule embedded across the sales cycle as the contextual guidance layer. SuccessFactors Learning is positioned around skills-driven development and automated compliance training that runs inside a validated SaaS environment. The vendor master sits inside SAP ERP or [S/4HANA](https://www.sap.com/products/erp.html), and the data continuity across those products is the part of the SAP value proposition that no horizontal automation tool can replicate without custom integration work. This is real. A buyer already running SAP ERP, S/4HANA, or SuccessFactors HCM does not have a software comparison to make. They have a switching-cost calculation. Vendor records, employee records, contract metadata, and approval matrices already live in the SAP suite; an onboarding workflow built on top of that data lands inside the trust boundary the organization has already drawn. The AI features Joule layers on top arrive against trusted data, which is a different position than an AI feature applied to a clean install at a company that has never had a unified vendor record. Specific things SAP does that FlowRunner does not: - **Native suite continuity.** Vendor master in SAP ERP, employee data in SuccessFactors HCM, customer data in Sales Cloud, contract metadata in SAP Ariba or DocuSign-integrated SAP modules. The data flows through SAP-built connectors, not through reconciliation logic an orchestration layer would have to add. - **Validated SaaS for regulated industries.** SuccessFactors Learning operates inside [“strict requirements for highly regulated industries with annual updates and validated testing”](https://www.sap.com/products/hcm/learning-software.html). For pharma, finance, and healthcare buyers who treat that posture as a hard procurement requirement, that credentialed environment is not a checkbox FlowRunner can replicate with audit trails alone. - **Curated content and partner ecosystem.** [SAP Content Stream by Skillsoft, open content network providers, and preconfigured extensions](https://www.sap.com/products/hcm/learning-software/features.html) represent years of ecosystem investment. The breadth of that catalog is not something a newer platform matches by adding integrations faster. - **Mobile-first field execution.** AI-enhanced Android and iOS apps with natural language navigation and dynamic visit planning are built for distributed field sales teams. FlowRunner has no comparable mobile interface. - **Forecast and pipeline intelligence at enterprise scale.** Sales Cloud combines projected, target, and best-case modeling with real-time pipeline flow, leakage, and conversion data designed for enterprise revenue operations. This list is not a backlog of FlowRunner gaps to close. It is the deliberate scope difference between a multi-product enterprise suite and a horizontal orchestration layer. A procurement team inside an SAP-committed enterprise that types “vendor onboarding software” into a search bar usually already knows the answer is SAP. The honest version of the comparison is what happens when the buyer is not that enterprise. ### What vendor onboarding actually looks like for a mid-market procurement team The standard advice on vendor onboarding software is to compare features against a checklist and pick the highest-scoring tool. The honest version is that most mid-market procurement teams have a more concrete problem that the checklist does not surface. The vendor onboarding work that consumes their time does not live inside any single product today. It lives in fragments: - A vendor sends a W-9, a certificate of insurance, and banking details across three emails over a week - The team manually checks the submitted tax ID against the ERP vendor list (NetSuite, Acumatica, QuickBooks Online, or yes, SAP) and finds something that looks like a duplicate but is not exactly one - The contract goes out for signature through DocuSign; the executed PDF lands in a folder somewhere - Approval routing depends on category, spend threshold, and whether the vendor touches regulated data, and the routing rule does not handle the conditional case gracefully - The COI expires three weeks after onboarding and nothing notices because the COI lives in a folder and the AP system keeps paying A buyer searching for vendor onboarding software is usually looking for the system that absorbs all of that. The conventional answer is to install a portal product that owns the front door. The harder question, the one most comparison posts will not say plainly, is that a portal solves intake and creates a new silo procurement still has to reconcile against the ERP, the AP system, and the contract repository. The intake form is the easy part. The coordination across the systems that already hold vendor data is the work. This is where the buyer decision splits. For an SAP-committed enterprise, the answer is to use the SAP-native pieces (Ariba, SuccessFactors, the S/4HANA vendor master) and accept the implementation timeline because the suite continuity earns it. For everyone else, the answer is a coordination layer over the systems already in place. Adding SAP to a non-SAP stack to solve vendor onboarding is buying the bigger problem. ### Where FlowRunner fits: the layer over the ERP and the channels procurement already uses [FlowRunner](https://flowrunner.ai/) is an orchestration layer that sits above the systems of record. For vendor onboarding specifically, the shape of the pattern looks like this: 1. A trigger fires (a supplier-facing form submission, a vendor email arriving in a shared mailbox routed through a parser, an event in the contract system) 2. The workflow gathers context across the systems involved: the ERP’s existing vendor list, the contract repository, the document parser’s output, internal approval matrix metadata 3. The mechanical part of validation runs: does this tax ID match an existing vendor with a different name, does the remit-to address look like a known fraud pattern, does the COI date check out, is this banking detail change happening on an existing vendor record 4. If everything ties, the workflow writes the vendor record into the ERP and posts the notifications without involving a human 5. If anything is ambiguous (probable duplicate, mismatched tax ID, missing certification, banking change on an existing vendor), the workflow pauses and calls a named procurement reviewer in Slack with the full context attached and a structured response 6. The reviewer decides, the workflow resumes from the response, the ERP record updates with the decision captured against the audit trail The shape of step five is the part that matters. FlowRunner’s agents invoke humans as a callable action when they hit uncertainty, not as a status gate that fires on every threshold breach. Concretely, this is the same shape as the patterns documented in [vendor validation against Acumatica before bill creation](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack), [cross-referencing vendor history in NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack), and [human-in-the-loop on ERP vendor operations](https://flowrunner.ai/workflows/automate-with-acumatica). The full solution shape is documented across the [supplier portal software](https://flowrunner.ai/solutions/supplier-portal-software) and [supplier relationship management software](https://flowrunner.ai/solutions/supplier-relationship-management-software) solution pages, which describe the coordination approach in more detail. A system above the systems of record is the category that owns this work: a layer that listens for what the ERP and the contract system emit, gathers context the approval rule did not have, and pulls a human in at the moments that need judgment. FlowRunner is built for that layer. SAP’s vendor-adjacent tooling sits inside the enterprise suite at a different point in the stack. The honest version of the human-in-the-loop difference, said without strawmanning SAP: SAP’s agentic orchestration is positioned around seller enablement and contextual guidance for users inside the SAP workflow. That framing is real and it is a legitimate design choice for an enterprise sales product. FlowRunner’s human-in-the-loop is a different shape, suited to a different kind of decision. Where seller enablement is about giving a user better context inside a known workflow, an orchestration-layer escalation fires when an agent or workflow encounters something it cannot resolve on its own (a vendor that looks like a duplicate of an existing record, a tax ID that does not validate, a banking change on an established vendor that should not be approved without a callback verification) and needs a named human to decide. Both are forms of intelligent automation. They are designed for different work at different points in the stack. ### Connectors built in thirty minutes, paid for at a price a procurement director can authorize Two architectural choices distinguish FlowRunner from the enterprise-suite model that SAP defines. The first is connector velocity. FlowRunner’s Agent Factory is built so operations teams can compose a workflow that combines a parser, an ERP write, a Slack approval, and an audit trail without filing an engineering ticket. When the team needs to reach a system the catalog does not already cover, new connectors are built in roughly thirty minutes. The bar to add a new exception path is low enough that the workflow can evolve as the buying motion evolves. That is the opposite of the suite implementation timeline that an SAP rollout assumes, and it is the right answer for a procurement team that has been told to ship a vendor onboarding workflow this quarter. The second is pricing transparency at the governance tier. FlowRunner publishes audit trails, RBAC, and SSO at the [Professional tier ($299 per month)](https://flowrunner.ai/pricing). The framing matters: this article is not claiming auditor acceptance of any specific compliance framework, and SAP’s validated SaaS environment for SuccessFactors Learning remains a real advantage for pharma, finance, and healthcare buyers whose regulatory scope demands that posture. FlowRunner’s compliance surface is designed to meet common audit requirements for mid-market procurement operations. It does not constitute a formally validated environment for highly regulated industries. The distinction is honest and it is important. What FlowRunner does offer at $299 per month is the governance infrastructure a procurement director can authorize without an enterprise procurement cycle. That price point is paired with the connector velocity above to produce a different shape of buying decision than the SAP suite assumes. ### Where SAP is the stronger fit Buy SAP (and probably nothing else for vendor onboarding) when: - Your organization already runs SAP ERP, SAP S/4HANA, or SuccessFactors HCM, and the vendor master plus HR data already live inside the suite - Your regulatory scope requires a validated SaaS environment with annual updates and validated testing, particularly for pharma, finance, healthcare, or other industries where that posture is a hard procurement requirement - The breadth of SAP’s content and partner ecosystem (SAP Content Stream by Skillsoft, preconfigured extensions, the partner network) is material to your buying decision - Your sales and procurement teams are distributed and a purpose-built mobile interface with AI-enhanced navigation and dynamic visit planning is a primary requirement - You operate at enterprise revenue scale where Sales Cloud’s forecast and pipeline intelligence (projected, target, and best-case modeling with real-time pipeline flow, leakage, and conversion data) is the level of revenue operations infrastructure your business actually needs This set of conditions describes a real and large class of enterprise buyer. If it describes your organization, the rest of this article is interesting but not actionable. SAP is the right answer and an orchestration layer is a question for a later conversation about extending what SAP already does well. ### Where FlowRunner is the stronger fit Choose FlowRunner when: - Your organization does not already run SAP, and adopting SAP for vendor onboarding would mean buying into a multi-year suite implementation to solve a workflow problem - The vendor onboarding work that consumes your team’s time lives across email, parsers, Slack, the ERP (NetSuite, Acumatica, QuickBooks Online), and a contract repository, with the coordination across them as the actual bottleneck - You need a procurement coordinator or ops analyst to configure new exception paths without waiting on engineering capacity or a partner ecosystem release cycle - You need intelligent human-in-the-loop where agents pause and route to a named reviewer with full context, rather than every threshold breach firing a static approval gate - You want governance infrastructure (audit trails, RBAC, SSO) at mid-market pricing that a procurement or finance director can authorize without an enterprise procurement cycle - You see AI agents arriving in your stack from multiple vendors (extraction agents, supplier-side AI, approval bots) and want a coordination layer above them before that becomes its own problem The two products coexist comfortably in the same procurement stack when the organization is mid-transition. SAP holds the vendor master for the regulated parts of the business; FlowRunner handles the coordination layer for the workflows that touch supplier documents, parsed inbox data, conditional approvals, and exception routing across the channels procurement actually uses. The structured parts of the buying motion flow through SAP. The unstructured coordination flows through FlowRunner with the context already attached. ### How to decide, given where you are on the clock Three questions, in this order. **1\. Are you already on SAP?** If the vendor master lives in SAP ERP or S/4HANA and the HR data lives in SuccessFactors, the answer to “what vendor onboarding software” is the SAP-native pieces and the orchestration question is a separate conversation about extending them. If you are not on SAP, adding SAP to solve vendor onboarding is buying a much larger problem than the one you have. **2\. What is the regulatory scope?** If your organization operates in pharma, finance, healthcare, or another vertical where a formally validated SaaS environment is a hard procurement requirement, SAP SuccessFactors Learning’s validated posture is a real advantage that an orchestration layer with audit trails does not replicate. If your scope is “common audit requirements” rather than “validated environment,” FlowRunner’s $299-per-month governance tier sits at the right price point for the requirement. **3\. Where does the actual onboarding work live today?** Walk through the last three vendor onboardings that consumed your team’s time. For each one, name where the work happened. If the answer is “inside our SAP modules,” extending SAP is the right move. If the answer is “across email, a parser, two Slack threads, a DocuSign envelope, and three exports from the ERP before anyone could approve the record,” that is the work an orchestration layer is built to absorb, and the question is which orchestration product to buy. A separate framework on [how to evaluate which procurement workflows are worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) handles the financial side of that decision. Most mid-market procurement teams will answer question one with “no, we are not on SAP” and question three with “across systems.” For those teams, the comparison this article is staged around resolves to FlowRunner, not because FlowRunner wins on every axis (it does not, and SAP’s validated environment, ecosystem breadth, and mobile depth are real advantages where they apply), but because the buying decision is a scope decision before it is a feature decision. ### Quick answers #### Is FlowRunner a replacement for SAP for vendor onboarding? Not for an organization already running SAP ERP, SAP S/4HANA, or SuccessFactors HCM. Those buyers get native vendor and HR data continuity from the SAP suite that an orchestration layer cannot replicate without custom integration work. FlowRunner is the better fit when the organization does not already run SAP and the goal is to ship a vendor onboarding workflow with intelligent human-in-the-loop checkpoints in weeks rather than after a suite implementation. #### Does FlowRunner offer a validated SaaS environment for regulated industries? No. SAP SuccessFactors Learning carries a validated SaaS posture with annual updates and validated testing for pharma, finance, and healthcare. FlowRunner provides audit trails, RBAC, and SSO starting at $299 per month, which are designed to meet common audit requirements but do not constitute a formally validated environment. If your regulatory scope requires that posture, SAP retains a real edge. #### Can FlowRunner sit alongside SAP rather than replace it? Yes. The most common pattern with SAP customers is to keep SAP as the system of record for vendor master and HR data, and use FlowRunner as the coordination layer that pulls in supplier documents, routes approvals through Slack or email, validates submissions against the SAP vendor record, and escalates exceptions to a named procurement reviewer before the SAP record updates. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Stampli: An Honest Read on Accounts Payable Automation Source: https://flowrunner.ai/blog/flowrunner-vs-stampli-accounts-payable-automation Comparisons May 28, 2026 Updated September 7, 2026 13 min read Stampli wins when AP is the whole job and the invoice is the center of gravity. FlowRunner wins when AP is one of several processes to orchestrate. ![A bald cartoon man comparing two setups on a workbench: a single upright invoice with icons orbiting tightly around it, and a switchboard with cords plugged into several separate small machines, weighing invoice-centered automation against cross-system orchestration.](https://flowrunner.ai/images/blog/flowrunner-vs-stampli-accounts-payable-automation-hero.webp) Ask one question before you compare these two products, and the comparison mostly answers itself: is the invoice the center of gravity, or is it one object among many your team has to move between systems? Stampli is built so the invoice is the center of everything. Coding, approval, fraud checks, the entire conversation about a bill, all of it lives on top of that invoice. If that is your whole problem, Stampli is a serious, mature answer and this article will say so plainly. FlowRunner is built for the other case: when AP is one of several processes that need to run together, and the work keeps crossing out of the invoice into procurement, vendor onboarding, and customer ops. That is an orchestration problem, and FlowRunner is built for that layer. ### Stampli and FlowRunner, head to head | | Stampli | FlowRunner | | --- | --- | --- | | **Category** | Purpose-built AP automation software, AI-powered, invoice-centric | Orchestration layer for AI agents and human approvers across any business process | | **Center of gravity** | The invoice. Collaboration, documentation, and workflows sit on top of each invoice | The process. Workflows span AP plus procurement, onboarding, customer ops, and more | | **Primary job** | Process invoices faster and with more control, end to end | Coordinate work and human judgment across multiple systems and processes | | **Built for** | Finance and AP teams whose primary need is invoice processing | Finance and ops leaders coordinating work across many systems and processes | | **AI assistant** | [Billy the Bot](https://www.stampli.com/ap-automation/), trained on AP tasks (capture, coding, routing, fraud detection), learns existing workflows | General-purpose agents the customer configures against any process; not pre-trained on AP data at packaged accuracy | | **ERP integration** | [Pre-built API and Bridge integrations](https://www.stampli.com/integrations/), “supports all native functionality,” 70+ systems | Broad connector library across ERPs and tools; AP-specific field mapping is configured, not pre-packaged | | **Human approvals** | Structured approval routing inside the invoice workflow | Callable action: agent pauses, routes to a named approver via Slack, email, WhatsApp, or phone, resumes on the response | | **Collaboration model** | The invoice becomes a searchable communications channel and historical record | Decisions and context are captured per workflow step in the orchestration audit trail | | **Pricing** | [Quote-based](https://www.stampli.com/pricing/); figures not published | Public per-execution tiers from a Free plan and $5 Starter; [Growth $45/mo](https://flowrunner.ai/pricing) with AI agents, BYOK, all integrations, and human-in-the-loop; Professional $299/mo adds audit trails and RBAC; Business $999/mo adds SSO/SAML | | **AP track record** | Established public social proof and satisfaction scores in the AP category | Newer in the AP vertical; less public AP-specific proof today | | **Best fit** | AP is the whole job; the invoice is where the work lives | AP is one of many processes the same team needs to orchestrate together | The table draws the line. What it cannot show is why the line sits where it does, and which side of it your team is actually on. That is the rest of this article. ### What Stampli is good at, said plainly Stampli is a purpose-built answer for accounts payable, and the purpose shows. Their own [product positioning](https://www.stampli.com/ap-automation/) describes software that “integrates with ERPs and is powered by artificial intelligence to automate invoice processing, expenses, vendor management, payments, and reconciliation,” and the design idea underneath it is specific and good: bring “all communication, documentation, and workflows into one place” so the invoice itself carries its own history. That last idea is the part most AP tools never get right. Stampli’s framing is that “the invoice itself becomes a communications channel and historical record where all comments, questions, answers, and invoice inputs are documented in a central and searchable location.” Anyone who has tried to reconstruct why a bill got approved by digging through three months of inbox archaeology understands why that matters. The scattered email chain is the default failure mode of AP, and Stampli attacks it directly by making the invoice the place the conversation lives. Billy the Bot is the other half. Stampli positions Billy as “your AI employee, automating capture, coding, routing, fraud detection, and other manual tasks,” an assistant that “assists you across the entire invoice process” and is “always learning.” The framing that matters most for an honest comparison is this one, in Stampli’s own words: “No need to rework your ERP or change your AP processes.” Billy is designed to learn the coding rules and approval patterns a finance team already runs, rather than forcing the team into a new process. That is a real strength, and it is the opposite of a strawman. A finance leader who has been burned by a tool that demanded a process rebuild should weigh it heavily. The ERP integrations back this up. Stampli claims [70-plus pre-built integrations](https://www.stampli.com/integrations/) with “pre-built API (cloud-to-cloud) and Bridge (cloud-to-premises)” connectivity, covering NetSuite, QuickBooks Online and Desktop, Sage Intacct, Oracle, Acumatica, and the Microsoft Dynamics family, among others. Their own description is that the integration “supports all native functionality” with “no development or IT overhead,” “deploying in weeks, not months.” For invoice-specific data flows into a major accounting platform, that is depth FlowRunner does not package the same way. Specific things Stampli does that FlowRunner does not: - A domain-trained AI assistant (Billy) tuned on AP-specific tasks at scale, designed to learn an organization’s coding and approval patterns without a process rebuild - A collaboration hub built on top of each invoice, where every comment, question, and approval lives on the invoice as a searchable record - Pre-built, AP-tuned ERP integrations across 70-plus systems that support native functionality with minimal configuration - An established public track record and customer satisfaction reputation in the accounts payable category specifically, which a newer platform cannot manufacture \[image\[dashboard|An AP automation interface with a single invoice at the center, surrounded by coding fields, an approval status, and a thread of comments attached to that one invoice\]\] That last point deserves its own sentence rather than a competitive dodge. Stampli has years of focused AP-specific reviews behind it across the major software review platforms. FlowRunner is newer and has less public social proof in the AP vertical specifically. A buyer evaluating purely on AP-category reviews and customer count today will find Stampli has more of both. Naming that is the price of being believed on everything else. ### The honest part most platform comparisons skip Here is what a horizontal-platform vendor is not supposed to say in a comparison post: a category-leading, purpose-built tool is the right purchase more often than the platform pitch admits. The standard move in this genre is to argue that a single broad platform always beats a stack of point solutions, because consolidation is tidy and integrations are scary. That argument is mostly marketing. If accounts payable is genuinely the whole job, and the invoice genuinely is where the work lives, then a tool that was designed around the invoice, trained on AP data, and deployed by people who think about AP all day will beat a general-purpose orchestration layer that you have to configure. Buying a horizontal platform to do one vertical job is overkill, and overkill has a cost: configuration time, the absence of a domain-trained model out of the box, and the lack of an opinionated workflow that a specialist tool gives you for free. So the first honest question is not “platform or point solution.” It is “is AP actually the whole job?” For a real slice of finance teams, the answer is yes, and those teams should buy the specialist. The comparison only gets interesting when the answer is no. ### What AP looks like when the work leaves the invoice For a meaningful slice of the validated finance buyer pool, AP is not a self-contained problem that ends at the invoice. The pattern that surfaces across roughly a dozen conversations with CFOs and finance leaders at mid-market companies looks less like a single invoice and more like a graph of chores that share systems, approvers, and exception logic. What that graph actually contains: - Invoices land in an inbox, get parsed, and need to flow into an ERP that is often QuickBooks Online, NetSuite, or Acumatica - Distributor billbacks arrive separately and need validation against contract terms before payment, where the risk is concrete (one finance leader described how easy it would be to “double pay on something like that” when both the company and the distributor can initiate a payment) - Vendor onboarding requires collecting and checking tax documentation before the vendor is ever payable, a step that happens before any invoice exists - Procurement approvals and customer ops escalations share the same approvers and the same routing patterns as AP, but live in disconnected tools - The audit obligation spans all of it, and right now the evidence is spread across an AP tool, an inbox, a Slack channel, and someone’s memory For a finance leader running that mix, the question is not “which AP product processes invoices best.” Stampli might genuinely win that one. The question is “what handles the routing, the exceptions, the cross-system handoffs, and the audit trail across AP plus the half-dozen things that touch it, without buying a separate product for each.” This is where the wedge opens, and it is worth being precise about it. Stampli’s great design choice, putting everything on top of the invoice, is also its boundary. When the work is an invoice, that boundary is a strength. When the work is a vendor that is not yet payable, a billback that is not yet an invoice, or a procurement approval that never becomes one, the work has left the invoice object entirely, and an invoice-centric tool has nowhere to put it. That seam, where the work crosses out of the invoice and into the systems and decisions around it, is where an orchestration layer lives. It is a category, not a product. The category sits above the systems of record, listens for what they emit, gathers the context the source system did not have, brings the right human in with the full record framed, and writes the answer back so the audit trail stays in one place. FlowRunner is built for that layer. Buying it to process invoices alone is overkill. Buying it because invoice processing is one of several things you need coordinated against your existing stack is the fit. ### Where FlowRunner fits: AP as one orchestrated process among many [FlowRunner](https://flowrunner.ai/) is an orchestration layer for AI agents and humans across the tools a finance and ops team already runs. AP, in that frame, is one of the patterns the platform handles, not the product it sells. The AP-shaped pattern looks like this: 1. A trigger fires (an inbox email, a new bill in the ERP, a webhook from a parser) 2. A workflow gathers context from every system it needs (the ERP, the parser, the vendor master file, the PO record, the approval rule) 3. An agent attempts mechanical work (extract fields, suggest a GL code, check for duplicates, match against a PO) and either clears the bill or pauses it 4. Where it pauses, the workflow calls a named human as an action, with the full context attached: the bill, the vendor history, the PO reference, what failed validation, what the agent would have done 5. The human responds (approve, reject, adjust, escalate) through Slack, email, WhatsApp, or phone, and the workflow resumes from that response 6. The bill writes back into the ERP with the approval evidence captured, and the audit trail records who decided what and when The difference from a structured invoice-approval flow is not whether a human is involved. Stampli routes approvals too, and routes them well inside the invoice. The difference is where the human is callable from. FlowRunner’s [callable human-in-the-loop](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) is the same pattern an agent invokes whether it is stuck on a bill over threshold, a vendor onboarding document that fails a check, a procurement request that needs sign-off, or a customer ops escalation. The agent hits uncertainty, pulls the line, and waits for a person. One finance prospect called this a “digital andon cord,” and the term stuck because it is exact: the line stops itself when it does not know, rather than plowing ahead. The routing, the escalation, and the audit capture are the same across all of those processes because they live in the orchestration layer, not inside any one product. \[image\[flow|A workflow diagram with an AI agent node branching to a human approval step, where the agent pauses and routes a decision to a named person before resuming and writing back to the ERP\]\] Concretely, the [email-to-payment invoice automation we publish for QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) shows the AP-shaped pattern with duplicate detection and named-approver routing. The parallel patterns run for [parsed vendor invoices into Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and for [vendor documents into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack). The [Acumatica workflow catalog](https://flowrunner.ai/workflows/automate-with-acumatica) shows AP as one part of a broader orchestration footprint inside a single ERP. Each uses the same orchestration layer, the same audit trail, and the same access controls, and the same layer extends into procurement, onboarding, and customer ops work that is not AP at all. If you want the conceptual version of why a coordinating layer is a different thing from a feature, [AI automation vs. AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents) lays it out. \[image\[message|A Slack approval message showing an invoice flagged for a duplicate-payment risk, with the vendor name, amount, and Approve and Reject buttons, sent to a named approver\]\] ### Where Stampli is the stronger choice, named explicitly A comparison that crowns FlowRunner the winner on every axis fails the credibility test. Three places Stampli is genuinely the better fit, in the order a finance buyer should weigh them: **Invoice processing depth and domain-trained AI.** [Billy the Bot](https://www.stampli.com/ap-automation/) is trained on AP-specific tasks and designed to learn a team’s coding and approval patterns without a process rebuild. For invoice coding, matching, and capture accuracy out of the box, Stampli has a packaged advantage that a general-purpose agent does not match on day one. FlowRunner closes the gap with configuration, the customer’s own historical data, and bring-your-own-key models, but the honest framing is fit, not parity. If the core job is “process invoices faster and more accurately,” Stampli is built for exactly that. **The invoice-centric collaboration hub.** Putting every comment, question, and approval on top of the invoice as a searchable record is a genuinely good design, and FlowRunner does not replicate it. FlowRunner captures decisions and context per workflow step in its audit trail, which is the right model for cross-process work. But if your team’s daily reality is a conversation about a specific invoice, Stampli’s hub is purpose-built for that conversation and FlowRunner’s per-step capture is not the same thing. **AP-specific track record and pre-built ERP depth.** Stampli has an established public reputation in the AP category and [70-plus pre-built integrations](https://www.stampli.com/integrations/) tuned for accounting data flows that “support all native functionality.” FlowRunner is newer in the AP vertical and configures its ERP field mappings rather than shipping them pre-packaged for AP. A team evaluating on AP-category reviews, customer count, and out-of-the-box ERP depth will find Stampli ahead today, and that lead is real. Naming these honestly is what makes the rest of the comparison readable. A buyer who lands here, sees Stampli win three axes that matter to their situation, and chooses Stampli has made the right call for their situation. ### Where FlowRunner is the better fit Choose FlowRunner when: - AP is one of several finance and ops processes you need to automate against the systems you already have, not the only one - The work keeps crossing out of the invoice: vendor onboarding before a bill exists, billback validation before it becomes an invoice, procurement approvals that never do - You want a callable human-in-the-loop that generalizes across processes, where an agent pauses and pulls in a person at any decision point, not a fixed approval gate inside one product - You want non-developers in finance and ops to configure new workflows against any connected system, without engineering resources or a vendor implementation cycle - You want a published entry point: FlowRunner’s [Growth tier is $45 per month](https://flowrunner.ai/pricing) and already includes AI agents, BYOK, all integrations, and callable human-in-the-loop, so the core orchestration job runs there; Professional at $299 adds 30-day audit trails and RBAC when governance becomes a requirement, and Business at $999 adds SSO/SAML and 90-day retention when compliance posture does - You see AI agents arriving in your stack from multiple vendors and want a coordination layer above them before that becomes a problem of its own What FlowRunner does not try to be: a better invoice-processing engine than a tool designed around the invoice. A team whose evaluation rests primarily on invoice coding accuracy, the per-invoice collaboration hub, or out-of-the-box AP-tuned ERP depth should choose Stampli. The two can coexist without conflict, and it is worth saying how. A mid-market finance team could run a specialist AP tool for invoice processing and run FlowRunner for the orchestration work outside that boundary: the vendor onboarding checks, the billback validation, the procurement and customer ops routing that no invoice tool reaches. The architectural shape is the same one we already publish for QuickBooks, NetSuite, and Acumatica: the AP tool owns invoices, and the orchestration layer coordinates everything around them. ### How to decide: draw the boundary, then pick your side Skip the feature checklist. Draw one boundary instead, and see which side your team’s work falls on. The boundary is the invoice. On one side is the work that is an invoice: capture it, code it, match it, approve it, pay it, and keep the conversation about it in one place. On the other side is the work that touches invoices but is not one: onboarding a vendor before they are payable, validating a billback before it becomes an invoice, routing a procurement approval, handling a customer ops escalation that shares your AP approvers. If your work lives almost entirely on the invoice side of that boundary, buy the tool built for the invoice. Stampli is a strong one, its domain-trained AI and invoice-centric hub are real advantages, and a horizontal platform would be overkill. Compare it against other specialists, and if you want the broader AP-software lay of the land, our [honest read on BILL](https://flowrunner.ai/blog/flowrunner-vs-bill-automate-accounts-payable-software) covers the other major purpose-built option, while our [guide to automating AP without losing the judgment that matters](https://flowrunner.ai/blog/how-to-automate-accounts-payable) covers the sequencing before you pick anyone. If your work keeps spilling across the boundary, no invoice-centric tool will own the other side, because the other side is not an invoice. That side belongs to a layer that sits above your systems, calls a human in when an agent hits something it should not decide alone, and keeps one audit trail across all of it. That is the orchestration layer, and FlowRunner is built for it. The andon cord does not care whether the line that stopped is a bill, a vendor document, or a billback. It stops, it asks the right person, and it resumes. That is the work, and it is most of what falls through the cracks today. ### Quick answers #### Is FlowRunner a replacement for Stampli? No. Stampli is a purpose-built AP automation platform where the invoice is the center of gravity, with Billy the Bot trained on AP-specific tasks like coding, matching, and approval routing. FlowRunner is an orchestration layer that coordinates AI agents and humans across the systems a team already runs. A team whose primary need is faster, more controlled invoice processing in one product should look hard at Stampli. #### When does FlowRunner make more sense than Stampli for AP? When AP is one of several finance and operations processes you need to automate against existing systems, not the only one. FlowRunner parses invoices into QuickBooks, NetSuite, or Acumatica and routes approvals through Slack with named approvers, and the same orchestration layer handles vendor onboarding, billback validation, procurement, and customer ops work that lives outside the invoice. #### How does FlowRunner’s pricing compare to Stampli’s? Stampli prices by quote (its pricing page asks you to request a quote and does not publish figures), so a like-for-like number comparison is not possible from public information. FlowRunner publishes per-execution tiers from a Free plan and a $5 Starter up; Growth at $45 a month already includes AI agents, BYOK, all integrations, callable human-in-the-loop across email, Slack, WhatsApp, and phone, and unlimited users and workflows. Professional at $299 a month adds 30-day audit trails and RBAC when governance becomes a requirement. SSO/SAML and 90-day audit retention sit on the Business tier at $999 a month. The honest comparison is model and scope, not a single dollar figure. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Vanta for SOX Compliance Software: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-vanta-sox-compliance-software Comparisons May 27, 2026 Updated May 28, 2026 12 min read Vanta runs your SOC 2 program; SOX 404 is a different problem. When a GRC platform fits, when an orchestration layer fits, when both belong. ![A bald cartoon man with three hairs sticking up standing between two doors in a corridor, one labeled SOC 2 and ISO 27001 in muted teal, the other unlabeled with a warm amber light spilling out, weighing which door to enter.](https://flowrunner.ai/images/blog/flowrunner-vs-vanta-sox-compliance-software-hero.webp) The honest read on the SOX compliance software search results, which most comparison pages will not say out loud, is that Vanta is the highest-profile platform on the page and SOX is not one of the frameworks Vanta markets as covered. That gap is the whole point of this comparison. Vanta is a category leader in security and trust posture. SOX 404 lives in a different category. A CFO buying SOX software needs to know which category they are buying for before they buy. This piece compares the two products on the axes that matter for that decision: framework coverage, where the evidence is produced, integration scope, pricing and packaging, and the buyer the product is shaped for. Both products are legitimate. They are not substitutes. ### How FlowRunner and Vanta compare at a glance | Axis | Vanta | FlowRunner | | --- | --- | --- | | Category | GRC and trust platform | Orchestration layer above systems of record | | Framework coverage | SOC 2, ISO 27001, GDPR, HIPAA, HITRUST, NIST AI RMF, ISO 42001, CMMC, FedRAMP, and more. SOX is not listed. | No pre-built compliance frameworks. Designed to support common control-evidence requirements at the moment of the transaction. | | Primary buyer | Security, GRC, and compliance practitioners | Operations and finance leaders running the actual control activities | | Where evidence is produced | After the fact, by monitoring infrastructure and pulling data from connected systems | At the moment of the work, inside the approval, reconciliation, or exception step | | Integration library | 400+ pre-built connectors across cloud, HRIS, identity, datastores ([source](https://www.vanta.com/integrations)) | Newer and narrower; growing connector set centered on finance and operations tools | | Auditor experience | Dedicated auditor portal, auditor directory, A-LIGN and other partnerships | No auditor portal | | Human-in-the-loop architecture | Workflow status and review queues | Agents invoke human reviewers as callable steps in the workflow | | Self-serve cloud pricing with audit logs, RBAC, SSO | Quote-based | $299 per month at Professional tier ([source](https://flowrunner.ai/pricing)) | | Reasonable replacement question | Replaces no part of your financial control activities | Replaces no part of your SOC 2 program manager | Both products can live in the same stack. The article below explains where each one earns its place. ### What Vanta actually is, in Vanta’s own words Vanta describes itself as an [Agentic Trust Platform](https://www.vanta.com/) that lets companies “Build and prove trust from a single, unified platform.” The product page promise is direct: “Get compliant quickly and painlessly with automation.” The frameworks listed across Vanta’s compliance pages are SOC 2, ISO 27001, GDPR, HIPAA, HITRUST, USDP, NIST AI Risk Management Framework, ISO 42001, CMMC, CJIS, NIS2, DORA, CPS 234, EU AI Act, Essential Eight, Cyber Essentials, FedRAMP, CRI, and custom frameworks. That is a substantial library. It is also a specific category. Notice what is and is not on it. SOC 2 is the anchor. ISO 27001, HIPAA, and HITRUST cluster around the same security-and-trust posture. NIST AI RMF and ISO 42001 are the AI governance extensions. CMMC and FedRAMP are the public-sector frameworks. None of these are SOX 404. SOX is a different beast: an SEC statute about internal controls over financial reporting, governed by [PCAOB AS 2201](https://pcaobus.org/oversight/standards/auditing-standards/details/AS2201), with control families in entity-level governance, access and segregation of duties, approvals, reconciliations, procure-to-pay, and IT general controls supporting financial systems. Vanta is excellent at one slice of that picture. Its connector library [pulls data automatically from 400+ tools](https://www.vanta.com/integrations), which gives it serious depth on the ITGC and access-management layer of SOX: user provisioning, access reviews, cloud-infrastructure controls, system-change management. If your SOX program needs the ITGC slice automated and you are already buying Vanta for SOC 2, you get real adjacency value. What Vanta is not designed to do is sit inside the AP approval, the reconciliation step, the journal entry review, or the exception handling that finance owns. Those are the parts of SOX a CFO actually signs for under Section 302 and 404 certifications, and those activities run in the ERP, the AP system, the close software, and the channels finance works in. Vanta sees the access list to the ERP. It does not see who approved which bill or what they decided when the bill did not match the PO. ### What FlowRunner actually is, plainly FlowRunner is an orchestration layer. It sits above the systems of record (ERP, AP, billing, payment processor, document repository, communication channels) and coordinates the work that runs between them. In an approval flow, FlowRunner posts the approval request into the channel approvers use, captures the response, writes named-approver identity and timestamp back to the bill record, and routes exceptions to a designated reviewer with full context. FlowRunner is not a GRC platform. It does not have pre-built SOX control libraries. It does not have an auditor portal. It does not map controls to frameworks. It does not manage third-party risk. If those are the capabilities a CFO needs to buy, FlowRunner does not deliver them and a comparison with Vanta on those axes is one Vanta wins outright. What FlowRunner does deliver is the moment-of-work evidence the control activity produces, in the system that owns the underlying record. The named approver captured against the bill. The timestamp of the decision. The exception rationale stored next to the transaction. The reviewer identity logged on the reconciliation step. The audit trail that does not require reconstruction during fieldwork because it was never lost in the first place. ### Framework coverage Vanta is a category leader on framework breadth. The published list crosses security, privacy, AI governance, and public-sector frameworks. The auditor portal is a real differentiator: SOC 2 audits run more smoothly when the auditor can pull evidence directly from the platform. Vanta’s partnership with A-LIGN and its broader auditor directory reflect years of compounding category investment that newer entrants cannot match in the short term. FlowRunner offers no framework coverage at all. There is no SOC 2 module, no ISO module, no HIPAA module, no SOX module. The product is structurally different: it captures evidence at the point of control execution rather than mapping evidence to a framework after the fact. For a buyer whose primary need is a SOC 2 program manager with framework mapping and auditor support, Vanta is the right tool and FlowRunner is not in the conversation. For a buyer whose primary need is SOX 404 control evidence inside the financial workflows the CFO certifies, the framework-coverage axis is misleading. SOX is not on Vanta’s list. The actual question is which product produces the audit trail SOX testers ask for at the line item, which is what the next axis covers. ### Where the evidence is produced This is the axis where the two products are most clearly different. Vanta produces compliance evidence by reading from connected systems. The 400+ integration library is the proof point: cloud providers, HRIS, identity providers, datastores, productivity suites. Vanta connects, pulls, and continuously tests configuration and policy state against the framework controls. When the SOC 2 control says “user access is reviewed quarterly with reviewer identity captured,” Vanta knows the review happened because it pulled the access list and the reviewer’s sign-off out of the identity provider and the HRIS. The evidence model is “monitor and pull.” FlowRunner produces evidence by being inside the control activity at the moment it runs. When the SOX control activity is “every bill over the AP threshold requires named approval from a controller with the approval timestamp captured against the bill record,” FlowRunner is the workflow that posts the bill to the controller in Slack, captures the response, and writes the named-approver, timestamp, and rationale back to QuickBooks or Acumatica against the bill record. The evidence model is “capture at the moment of work.” These models are not in competition; they cover different slices of the SOX control universe. Vanta is strong where the evidence lives in infrastructure that can be monitored. FlowRunner is strong where the evidence lives in a human decision that needs to be captured against a financial record. A program covering both slices needs both kinds of tools, or it needs to assemble the moment-of-work evidence manually at audit time, which is the failure mode the [SOX compliance checklist guide](https://flowrunner.ai/blog/sox-compliance-checklist) covers in detail. The structural reason this matters: SOX 404 is full of controls that look like “the controller reviewed the journal entry and approved it” or “the CFO certified the reconciliation.” Those are not infrastructure events. They are human judgment moments. A GRC tool monitoring infrastructure does not see them happen. The work between the systems of record, where approvals route and reconciliations resolve and exceptions get escalated, has no native home in any single system. The orchestration layer is the category that owns that work: a layer above the systems of record that listens for what they emit, routes the human-judgment moments to the people who need to see them, captures the response, and writes the evidence back into the system that owns the record. FlowRunner is built for that layer. It does not replace the GRC tool that handles the framework mapping; it sits where the GRC tool cannot reach. ### Integration scope Vanta wins this axis decisively, and the win is real. A 400+ connector library that pulls compliance evidence automatically from cloud infrastructure, HRIS, identity, and datastores is years of category investment that FlowRunner has not matched. A finance team standardizing on FlowRunner for compliance evidence will have fewer pre-built integrations available than they would on Vanta. That is the honest read. The qualification is what each library is optimized for. Vanta’s library is built for security-and-trust posture: it lists what a SOC 2 program needs to monitor. FlowRunner’s connector set is centered on the systems financial work actually runs in: ERPs, accounting platforms, payment processors, document parsers, communication channels. A finance team’s SOX evidence problem leans heavily on those systems, which is where FlowRunner concentrates depth, and far less heavily on the SOC 2 monitoring surface, where Vanta has its advantage. For a buyer whose audit-trail problem lives in the AP, close, and reconciliation cycle, “we have 400+ connectors” is less load-bearing than “we have the ones that matter for the seven systems where the financial work happens.” For a buyer whose audit-trail problem is “is our cloud environment configured to SOC 2 standards,” the calculus reverses. ### Pricing and packaging Vanta’s pricing is quote-based and not publicly listed. The procurement motion is usually a sales-led evaluation, which is appropriate for a platform sold to security and compliance teams running multi-framework programs. FlowRunner’s [Professional tier is $299 per month](https://flowrunner.ai/pricing) and includes audit trails, RBAC, and SSO at that tier. The packaging is designed for mid-market finance teams that need governance infrastructure without an enterprise procurement cycle. The Business tier at $999 per month adds the depth that larger finance operations need. Direct price comparison between the two is misleading because they are different categories of product. The pricing signal that matters is what each tier of each product is shaped for. Vanta is shaped for security-led compliance programs. FlowRunner at the Professional tier is shaped for finance teams that need named-approver capture and audit trails in their day-to-day workflows without a six-figure annual contract. ### Where Vanta is the better fit The article is not honest if it does not stop and say plainly where Vanta wins: - **SOC 2, ISO 27001, HIPAA, and the rest of the security frameworks.** This is what Vanta was built for. There is no version of this comparison where FlowRunner is a better choice for the SOC 2 program manager role. - **Auditor experience.** Vanta’s auditor portal and A-LIGN partnership are real differentiators that compress audit cycles in the frameworks Vanta covers. FlowRunner does not offer this and is not trying to. - **Trust center, customer commitments, security questionnaires.** These are GTM-facing compliance artifacts. Vanta has dedicated products for each. FlowRunner does not. - **Third-party risk management, vendor security review automation.** Vanta has a full TPRM module. FlowRunner does not. - **Automated control testing against a framework.** Vanta runs continuous tests mapped to the framework controls. FlowRunner produces evidence; it does not test against a framework library because it has no framework library. - **Enterprise GRC scale.** Vanta has the customer base, the brand, and the partner ecosystem in compliance. A CFO whose buying decision needs to clear an audit committee that wants a recognized GRC vendor on the slide will find Vanta delivers that signal in a way FlowRunner does not. If the buying criteria above describe the actual problem, Vanta is the right tool and FlowRunner should not be on the shortlist. ### Where FlowRunner is the better fit Where FlowRunner earns its place is the part of SOX that Vanta is not built to reach. - **Approval evidence at the moment of the transaction.** The named approver and timestamp captured against the bill record, not reconstructed from email threads. The [QuickBooks and Slack approval workflow](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) covers this directly: a named approver and timestamp are captured for every approval, in the system that owns the bill. - **Reconciliation decision history per step.** Bank, merchant processor, and intercompany reconciliations where the per-item decision is logged. The [Stripe and QuickBooks reconciliation pattern](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) captures decision history for every reconciliation step, retained against the transaction. - **AP control evidence in the ERP.** The [Acumatica AP automation pattern](https://flowrunner.ai/workflows/automate-with-acumatica) captures approver identity on every bill approval and routes exceptions for explicit review, producing the procure-to-pay control evidence inside the system of record. - **Exception handling as a callable step.** When the automation hits an exception, a named human is pulled in with full context, the decision is captured, and the workflow resumes. The exception evidence is the audit trail, not an email thread you reconstruct in October. - **Mid-market governance pricing.** Audit trails, RBAC, and SSO starting at $299 per month put the evidence infrastructure inside the budget of finance teams that cannot stand up a multi-tool enterprise GRC program on day one. - **Compliance cost as an automation multiplier.** The same approval workflows that capture SOX evidence are the automation work the finance team would justify anyway. The [framework for deciding what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) treats compliance evidence as a value multiplier, not a separate spend. The framing the FlowRunner platform supports is “designed to support common SOX control evidence requirements” rather than “auditors accept FlowRunner output as SOX evidence.” SOX compliance is the issuing company’s responsibility, not a vendor’s. No automation tool satisfies Section 404 on its own. What an orchestration layer delivers is evidence in the form auditors expect to receive it: named approver, timestamp, context, exception rationale, all retained against the underlying record. ### How to decide Three scenarios cover most of the buying decisions in this comparison. **Scenario one: a CFO at a public or pre-IPO company building a SOX program from scratch.** The right shortlist for the SOX 404 program management function is purpose-built SOX software (AuditBoard, Workiva, Pathlock, Hyperproof, Diligent, AuditBoard’s peers). Vanta belongs on the shortlist if there is significant SOC 2 or ITGC overlap. FlowRunner belongs on the shortlist for the control-execution evidence the SOX program-manager tools do not produce themselves. The three layers are complementary, not substitutes. **Scenario two: a CFO at a company already running Vanta for SOC 2, looking at what else Vanta can do.** Vanta will give you the ITGC slice of SOX cleanly and the access-management evidence you would otherwise build manually. It will not give you the AP, close, or reconciliation control evidence. FlowRunner is the layer that sits in those workflows and produces that evidence. The decision is not Vanta versus FlowRunner. It is whether the workflows where evidence is currently reconstructed at audit are worth the cost of orchestration today. **Scenario three: a finance leader at a mid-market company who is being asked to tighten control evidence ahead of an investor diligence cycle, lender covenant, or M&A event.** A full GRC platform is usually too much tool for the actual problem. The problem is producing named-approver evidence on bill approvals and reconciliations in workflows that currently live in email and spreadsheets. FlowRunner is shaped for that problem and priced for it. Vanta is overkill at the buying stage; it becomes the right tool later if the company moves into SOC 2 or a public-company path. The decision criterion that resolves most of these scenarios is the same one. Where is the evidence currently failing to get captured? If it is failing inside the cloud infrastructure, the identity provider, or the access-management surface, Vanta is the right tool. If it is failing inside the AP approval, the reconciliation step, or the exception handling, FlowRunner is the right tool. If it is failing in both places, both tools belong in the stack and they will not get in each other’s way. ### Quick answers #### Is Vanta a SOX compliance platform? No. Vanta lists SOC 2, ISO 27001, GDPR, HIPAA, HITRUST, NIST AI Risk Management Framework, ISO 42001, CMMC, FedRAMP, and similar frameworks. SOX is not among the frameworks Vanta markets as covered. Vanta is a security and trust platform whose IT general controls overlap with the ITGC slice of SOX, not a SOX 404 program manager. #### Can FlowRunner replace Vanta for compliance? No. FlowRunner is an orchestration layer, not a GRC platform. It does not offer pre-built compliance frameworks, an auditor portal, framework mapping, or third-party risk management. If you need a GRC program manager, you need a GRC tool. FlowRunner sits on a different problem. #### Where does FlowRunner fit if a finance team is buying SOX software? FlowRunner sits where the SOX control activities run: AP approvals, reconciliation steps, exception handling, evidence capture in the system of record. It is designed to support common SOX control evidence requirements at the moment of the transaction, so the audit trail is captured as the work happens rather than reconstructed at fieldwork. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Workato for QuickBooks API Automation: An Honest CFO Read Source: https://flowrunner.ai/blog/flowrunner-vs-workato-for-quickbooks-api-automation Comparisons May 27, 2026 Updated September 7, 2026 12 min read Workato is a mature enterprise iPaaS with a deeper connector library. FlowRunner is an orchestration layer built for finance teams who want to own the agent, not assemble recipes. ![A bald cartoon man between a sprawling slate-blue recipe book of labeled boxes connected by ribbons on his left and a compact sage-green dashboard with a single AP Agent tile and amber pause light on his right, weighing two shapes of automation work.](https://flowrunner.ai/images/blog/flowrunner-vs-workato-for-quickbooks-api-automation-hero.webp) The CFO who searches for “quickbooks api” is almost never asking what Google thinks they are asking. Google reads the query as developer intent and returns Intuit’s [developer documentation](https://developer.intuit.com/app/developer/qbo/docs/develop) at the top of page one. The CFO doing the searching is not building an OAuth flow. The CFO is asking a different question entirely: who is going to own the QuickBooks-centered automation work that has been growing for two years, and which platform makes it possible to own that work without becoming a developer or hiring one? That question has at least two serious answers in 2026. Workato is one. FlowRunner is another. They suit different buyers, and most of the comparison content that ranks for this query never says so plainly. ### Side by side, at a glance | | Workato | FlowRunner | | --- | --- | --- | | **Category** | Enterprise integration and automation platform (iPaaS) | Orchestration layer for work and AI agents across tools | | **Primary buyer** | IT-led integration and automation team in a large enterprise | Finance or ops leader at a mid-market company who wants to own the workflow | | **Building model** | ”Recipes” composed from connectors, triggers, and actions | Agents and flows configured visually, with human-in-the-loop as a callable step | | **Connector breadth** | Significantly larger; deep enterprise coverage including [SAP OData and SAP RFC](https://www.workato.com/integrations) | Narrower today; concentrated on integrations finance and ops actually invoke from QuickBooks-centered workflows | | **AI and agents** | Agentic capabilities through Workbot, Skills, and Agent Studio added to the integration platform | Native AI agent orchestration with an [Agent Directory](https://flowrunner.ai/agents) of pre-built agents and native MCP support | | **Human approval** | Workbot prompt in [Slack or Teams](https://docs.workato.com/workbot.html), with per-user runtime connections | Pause-and-ask as a callable action inside the orchestration; routes to a named approver via Slack, email, WhatsApp, or phone, captures the response, resumes | | **Embedded integration for SaaS vendors** | Mature [embedded product line](https://www.workato.com/embedded) with established partner ecosystem | Not a shipping capability today; stated future direction | | **Pricing posture** | Not publicly disclosed; tiered by feature set, sold through enterprise procurement | [Growth at $45 a month](https://flowrunner.ai/pricing) includes AI agents, BYOK, all integrations, and multi-channel human-in-the-loop; Professional at $299 adds governance (audit trails and RBAC); Business at $999 adds SSO/SAML and 90-day audit retention | | **Who owns the build** | IT-led integration team or a Workato-certified partner | The finance or ops team that owns the workflow | The table tells you most of the story. The rest of this article is the part the table cannot. ### What Workato is actually good at [Workato](https://www.workato.com/integrations) is a mature, enterprise-grade integration and automation platform. It positions itself as the system that connects apps and automates workflows across HR, Finance, Sales, IT, and Operations, and it has the years of product depth behind that positioning to back it up. The connector library is large and the enterprise coverage is real. SAP OData and SAP RFC are not commodity integrations to build, and Workato ships them. NetSuite, Salesforce, ServiceNow, Workday, Slack, and most of what a global enterprise stack actually runs are first-class citizens in the platform. The “recipe” model is the Workato signature. A recipe is a triggered workflow composed from connectors and actions, written in a visual editor that both technical and non-technical users can work in. The model is genuinely flexible, and the IT-led integration teams who have standardized on Workato can describe their entire automation surface area in recipes. That is a real and durable strength. Three more strengths worth naming flat: - **Workbot for Slack and Microsoft Teams** is a sophisticated [conversational automation product](https://docs.workato.com/workbot.html). It supports per-user connection management, including runtime user connections that authenticate per-user credentials inside a recipe. For an enterprise that has standardized on Slack or Teams as the chat layer, Workbot is a real cross-channel approval and invocation surface, not a token feature. - **The [embedded integration product line](https://www.workato.com/embedded) for SaaS vendors** is mature. SaaS vendors who need to ship integrations to their own customers can build on Workato’s embedded platform with an established partner ecosystem behind them. FlowRunner does not have a comparable shipping product. If embedded integration for SaaS vendors is what you are buying for, you should be evaluating Workato, not FlowRunner. - **Enterprise brand recognition and existing IT relationships.** Workato is a known quantity in enterprise IT procurement. The procurement cycle for adding a Workato workload to an existing Workato account is materially shorter than the cycle for bringing in any newer entrant, regardless of feature quality. That is an honest advantage and worth naming. The brief is explicit: do not minimize it. Most comparison articles in this category will not say any of that out loud. This one does, because the rest of the comparison is dishonest if we do not. ### What CFOs running QuickBooks are actually trying to solve Here is what most QuickBooks API and automation comparisons skip past. The buyer searching for “quickbooks api” alongside terms like “automation” and “accounts payable” is almost never the developer. The buyer is the operating finance leader at a mid-market company where the QuickBooks-centered work has grown faster than the team. The pattern shows up in [conversations across finance leaders](https://midnightflow.ai/portfolio/scale-your-cfo-practice-without-hiring/) and is consistent enough to predict: - Bills come in by email, by vendor portal, and sometimes still on paper. AP entry happens manually when staff are out, which means the CFO is occasionally entering bills. - Distributor billbacks and chargebacks are reconciled across portals, spreadsheets, and email chases. Duplicate payment risk is a recurring worry because the company and the distributor can both pay the same charge and nobody catches it for weeks. - Stripe payouts do not line up cleanly with QuickBooks invoices. The matching is mostly mechanical, but the exceptions (the payment that is short by a fee, the refund that did not propagate) are the work. - Recurring journal entries (freight allocation, prepaid amortization, intercompany transfers) get drafted in Excel and uploaded into QuickBooks once a month. The rule almost never changes once it is documented. This is the work agent orchestration is actually for. Not “automate QuickBooks.” Automate the AP intake and the reconciliation and the journal entry drafts around QuickBooks, with the controller in the loop on the exceptions and out of the loop on the routine. The honest read on this work, which most comparison articles will not say, is that the platform fit is decided less by connector breadth and more by who owns the build. A finance team that has to route every change request through an IT-owned integration platform is a finance team that ships less automation than it could. The constraint is organizational, not technical. ### Where the recipe model and the orchestration model diverge The cleanest way to understand FlowRunner against Workato is to separate the two product models without arguing about which is better. They are different shapes of the same job. **The recipe model.** A recipe is an integration primitive. The builder defines a trigger, picks connectors, configures actions, and ships the workflow. The fluency required is “what trigger fires when, which connector reads what, how do the fields map.” That fluency lives most naturally in an IT-led integration team or a partner who has it. The output is a connection between two systems with a behavior in the middle. Approvals fit into recipes as Workbot prompts in Slack or Teams. That is a real capability and not a primitive failure. It is a different architectural choice. **The orchestration model.** An orchestration is an agent that owns a workflow end to end, with human-in-the-loop as a callable step. The builder configures the agent’s job, the agent gathers context from the systems it needs, and the agent invokes a human as an action when the situation warrants it. The fluency required is “what does the workflow do, when does it pause, who decides what.” That fluency lives more naturally in the finance or ops team that owns the workflow. The output is an agent the team owns, not a connection between two systems. One compositional consequence is worth naming because it shapes how the work gets built. A Workato recipe must begin with a trigger. That is the model: trigger fires, connectors run, actions execute. A FlowRunner flow can begin with any node type. A trigger, yes, but also a manual run, a condition that evaluates business state, an action that pulls data on a schedule, or another flow invoking it as a callable subflow. The same flow can be a webhook-triggered automation in one context and a function the controller calls from inside another agent’s logic in another. That flexibility is not a feature checklist win; it is what makes flows compose. Subflows treated as functions, called by other agents, run manually for a one-off reconciliation, or kicked off by a state condition: that compositional shape is hard to assemble when the only legal first step is a trigger. The distinction matters most for QuickBooks-centered work because the exceptions are not edge cases. They are the work. A workflow that posts the matched 70 percent of bills and pauses for a named approver on the 30 percent that need judgment is the workflow finance leaders describe wanting. The agent does the routine. The human owns the exception. The audit trail records both. The category that owns this pattern is orchestration as a service: a layer above the systems of record that listens for what they emit, gathers context the systems do not share, calls a human at the moments that need judgment, and writes a structured record of who decided what across the entire workflow. The recipe-and-connector model can be configured to do parts of this. It was not built for it. The orchestration model was. FlowRunner is built for that layer, and the AI agents in the [Agent Directory](https://flowrunner.ai/agents) are the deployable instances of it. For the deeper category framing, see [AI automation vs. AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents). ### How human approval actually shows up in each platform The brief is precise about this and the article will be too. Workato handles human approval through Workbot, which is a conversational interface in Slack or Microsoft Teams where users invoke recipes and approve actions through chat. It is not a broken capability. It is well-engineered and works as designed. Per-user authentication via [runtime user connections](https://docs.workato.com/workbot.html) is a real piece of identity infrastructure built into the workflow layer. FlowRunner handles human approval differently. The agent is the unit of work. The agent invokes a human as a callable step inside its own orchestration: pause the workflow, route a structured request to a named approver across the channel they prefer (Slack, email, WhatsApp, phone), wait for the response, resume with the response captured in the audit trail. The same pattern that powers [QuickBooks bill approval in Slack](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack), [Stripe-to-QuickBooks reconciliation with mismatches paused for review](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack), and [inbox-to-QuickBooks invoice processing with duplicate detection](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack). The architectural difference matters in three specific places: - **Where the workflow’s intelligence lives.** In the recipe model, the recipe is the workflow and Workbot is the human interface to it. In the orchestration model, the agent owns the workflow and the human channel is one of the agent’s callable tools. The agent can decide, based on the data and the rules, whether the situation needs a human at all and which human and with what context. - **What the audit trail records.** A chat-prompt approval records the chat. An orchestration-step approval records the trigger, the context the agent gathered, the decision the system would have made on its own, the human it routed to, the response, and the action taken. That difference matters in audit conversations, not in marketing copy. - **Who has to be on Slack.** Workbot requires the approver to be in the chat tool. FlowRunner can call across channels in the same workflow, so the controller who refuses to use Slack can be reached by email and the operations lead can be reached on WhatsApp without two different workflows. This is not a feature gap with Workato. It is a different architectural starting point. Workato chose to make the chat layer the conversation. FlowRunner chose to make the agent the conversation. ### Pricing posture and who can authorize the spend A direct dollar comparison between Workato and FlowRunner is not possible because Workato does not publicly disclose tier pricing. What can be said honestly: - Workato’s enterprise feature set, including capabilities like runtime user connections, sits inside pricing tiers that require a sales conversation. The procurement cycle reflects the platform’s enterprise positioning. That is consistent with how iPaaS is bought. - FlowRunner publishes pricing. The Growth tier is [$45 a month](https://flowrunner.ai/pricing), a Free plan and a $5 Starter sit below it with the same platform, and Growth already includes the core capabilities needed for QuickBooks-centered work: AI agents with BYOK, all integrations, multi-channel human-in-the-loop (Slack, email, WhatsApp, phone), unlimited users, and unlimited workflows. There is no technical limitation at Growth that prevents a finance team from running the AP, reconciliation, and journal-entry agents described above. The Professional tier at $299 a month adds the governance layer a finance team typically wants once the workflows are in production: 30-day audit trails and RBAC. The Business tier at $999 a month adds the compliance posture for organizations whose IT or audit function requires it: SSO/SAML and 90-day audit retention. Enterprise is custom, with self-hosted, unlimited audit retention, and Midnight Flow consulting bundled in. There is a second dimension to the pricing question that matters more as the work gets complex: what the meter actually counts. Workato charges per task. Each connector action a recipe executes consumes monthly quota. A recipe that touches five systems to process one bill consumes five times the budget of a recipe that touches one system. FlowRunner charges per execution. A flow run is a flow run. A flow with five nodes and a flow with five hundred nodes count the same against the tier limit: 12,000 executions on Growth, 75,000 on Professional, 250,000 on Business. The honest version of this trade-off cuts both ways. For a small set of simple recipes at low monthly volume, Workato’s entry pricing can come out cheaper. As QuickBooks-centered AP and reconciliation work grows in complexity, where a single bill might touch the inbox, a parser, a duplicate-detection check, an approval routing step, and a posting action, the per-execution model stops penalizing the agent for doing more work inside a single workflow. That is a CFO conversation about cost-of-scale, not cost-per-seat. The other point is not that one is cheaper. The point is who can authorize the spend. A finance director with a $5,000 monthly software budget can sign for Growth at $45 without a procurement conversation, and can sign for Professional at $299 once governance becomes a requirement. The same finance director cannot sign for an iPaaS contract that requires IT, legal, and sometimes a CIO sign-off. That difference is decisive for finance-led purchases and irrelevant for IT-led purchases. Match the platform to who is buying. ### Connector breadth, embedded integration, and the parts where Workato wins The brief is explicit and the article will be too. There are concrete axes where Workato is the stronger platform today, and a comparison that does not name them is dishonest. **Connector breadth.** Workato has built its [connector library](https://www.workato.com/integrations) over many years. The library is significantly larger than FlowRunner’s, and the difference is sharpest at the deep enterprise end: SAP OData, SAP RFC, niche enterprise applications that mid-market companies do not run but global enterprises do. FlowRunner’s roadmap covers the integrations finance and operations teams invoke most often from QuickBooks-centered workflows. We have not matched Workato’s enterprise tail and likely will not for some time. If the buying decision turns on a connector to a system in that tail, Workato is the answer. **Embedded integration for SaaS vendors.** Workato’s [embedded product](https://www.workato.com/embedded) is mature, with a partner ecosystem and a long track record of SaaS vendors shipping integrations to their own customers on top of it. FlowRunner’s partner and community publishing model for the Agent Directory is a stated future direction. It is not shipping today. A SaaS vendor evaluating an embedded integration platform should not include FlowRunner in the consideration set for that use case. **Enterprise brand recognition and procurement gravity.** Workato has earned a position in [enterprise customer logos](https://www.workato.com/customers) and IT organization procurement lists. That position reduces procurement friction in a way no newer entrant matches regardless of feature quality. For an enterprise IT buyer adding a new workload to an existing Workato account, the procurement cycle is materially shorter than the cycle for evaluating a new vendor. That advantage is real and ongoing, and the right buyer for it is the enterprise IT team, not the mid-market CFO. These are not throwaway acknowledgments. They are decision-shaping facts. If your situation matches them, Workato is the right platform. ### Where FlowRunner is the better fit Choose FlowRunner when: - The work that needs orchestration is QuickBooks-centered AP, reconciliation, invoice intake, or journal entry drafting, with the controller in the loop on exceptions. - The owner of the build is a finance or operations leader who wants to ship the agent without becoming fluent in recipes-and-connections or waiting on the IT team that owns the iPaaS. - Human approval is the moment the workflow’s value is decided, and you want that approval to be a callable step inside the agent, not a chat-prompt prompt added around an automation. - You want the core capability (AI agents, BYOK, all integrations, multi-channel human-in-the-loop) available at a published entry price the finance team can authorize on its own, and a clear governance step (audit trails, RBAC) you can move up to when the work warrants it, without an enterprise procurement cycle. - You expect AI agents to start arriving in your stack from multiple vendors (a Salesforce agent, an ERP-vendor agent, a billing-platform agent) and you want a coordination layer above them before that becomes the next integration mess. - You want pre-built deployable agents in an Agent Directory to start from, and the option to build new ones visually rather than composing recipes from primitives. The same orchestration shape extends past QuickBooks. The pattern for [reconciling Stripe payments against open QuickBooks invoices](https://flowrunner.ai/workflows/connect-quickbooks-online-with-stripe), [parsing inbox documents into QuickBooks with duplicate detection](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack), or [orchestrating equivalent AP work on Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) is the same shape: ingest the trigger, gather the context across systems, attempt the mechanical match, call the human on the exception, record the decision. The platform is built around that shape. ### How to decide Three questions, in order: **1\. Who is going to own the build?** If the answer is the IT-led integration team that already runs your iPaaS, and the new workload is one more recipe alongside many others, Workato is the natural extension. If the answer is the finance or ops team that owns the workflow and wants to ship without waiting on IT, FlowRunner is the more direct path. **2\. What is the shape of the human moment?** Walk through the last five exceptions in your AP or reconciliation workflow. Did the resolution happen in a chat prompt, with the approver picking from buttons in Slack? Or did the resolution require context the chat prompt could not carry (a vendor history, a duplicate-payment heuristic, a freight allocation rule), a named approver across multiple possible channels, and a record of why the decision was made? The first answer fits Workbot. The second fits a callable-action human-in-the-loop. **3\. Where do you expect to be in 18 months?** If the answer is “Workato is our integration backbone and we will keep adding recipes,” that is a defensible posture and FlowRunner does not displace it. If the answer is “we are going to be running half a dozen vendor-shipped AI agents on top of our finance stack and we need a layer above them that the finance team can govern without depending on IT for every change,” that is the bet FlowRunner is built for. The orchestration layer is where that work lives. The recipe model can compose pieces of it. The orchestration model is what owns it. Both products are real. Both are honest answers to real buyer questions. The question is which buyer is doing the asking. ### Quick answers #### Is FlowRunner a replacement for Workato? Not for an enterprise IT team running Workato as their integration backbone across HR, sales, finance, and IT. Workato is purpose-built for that. FlowRunner is built for the finance leader who wants to own the QuickBooks-centered agent without routing every change through the IT team that owns the integration platform. #### Does FlowRunner have as many connectors as Workato? No, and the brief is honest about this. Workato has a significantly larger connector library, including deep enterprise connectors like SAP OData and SAP RFC. FlowRunner focuses on the integrations finance and operations teams actually invoke from QuickBooks-centered workflows. If your priority is breadth across a global enterprise stack, Workato wins on that axis. #### Is FlowRunner cheaper than Workato? A direct dollar comparison is not possible because Workato does not publicly disclose tier pricing. FlowRunner’s Growth tier at $45 a month already includes the core capabilities a finance team needs for QuickBooks-centered work: AI agents with BYOK, all integrations, multi-channel human-in-the-loop, unlimited users and workflows. The Professional tier at $299 a month adds governance infrastructure (30-day audit trails and RBAC) at a price a finance director can authorize without an enterprise procurement cycle. SSO/SAML and 90-day audit retention come in at the Business tier of $999 a month. Whether any of those is cheaper than your Workato quote depends on the quote. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Zapier for Salesforce Data Hygiene: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-zapier-for-salesforce-data-hygiene Comparisons May 28, 2026 Updated September 7, 2026 13 min read Zapier wins on app breadth and easy setup for linear Salesforce cleanup. FlowRunner wins when merge and dedup decisions need human judgment plus an audit trail. ![Two nearly identical paper contact cards overlap with an amber question mark between them and an amber pull-cord beside them, while a separate sage green tray holds a clean aligned stack, showing the ambiguous duplicate that stops the line for a human.](https://flowrunner.ai/images/blog/flowrunner-vs-zapier-for-salesforce-data-hygiene-hero.webp) The real question when you put FlowRunner and Zapier side by side for Salesforce data hygiene is not which tool is better. It is which half of the problem you are actually trying to solve. Salesforce hygiene splits cleanly into two parts: the automatic part, where a record arrives, matches a known record, and gets updated without anyone looking, and the judgment part, where a record probably matches something but the email is different and the company name is spelled three ways. Zapier is built for the first half and the long tail of apps around it. FlowRunner is built for the second half, where a human has to make the call and someone has to be able to explain it later. That split is the whole comparison. Hold it in mind and the rest follows. ### FlowRunner vs Zapier for Salesforce data hygiene at a glance | Axis | Zapier | FlowRunner | | --- | --- | --- | | Primary buyer | Anyone automating apps, from solo operators to RevOps teams | COO, VP Ops, RevOps, and sales ops who own data quality and need governance | | App breadth | ”9,000+ apps” advertised on its apps directory | Focused connector catalog, smaller today, deeper on orchestration | | Salesforce actions | New Record, Updated Record, Find Record, Find or Create Record, Create and Update | Find Record by Query, Update Contact, Update Lead, Convert Lead to Contact, Create Note, and more | | The automatic match-and-update | Strong. Linear Zaps do this cleanly | Strong. Same path runs as a flow | | The ambiguous merge decision | No native human-in-the-loop; assembled from Delay, Filter, and an approval app | Native human-in-the-loop: agent pauses, routes candidate and matches to a named reviewer, resumes on the decision | | Workflow shape | Linear trigger-action Zaps with Paths for branching | Multi-branch flows with branch synchronization, subflows callable as tools | | Audit trail per record | Task history per Zap run | Per-record execution log of every create, update, merge, and human decision | | Governance | SAML SSO on Team, audit logs on Enterprise | Audit trails and RBAC from $299/mo (Professional), SSO/SAML from $999/mo (Business) | | Entry price | Free tier (100 tasks), then Professional from $19.99/mo | Free plan (100 executions), then Starter from $5/mo; Growth $45/mo (all include AI agents, human-in-the-loop, all integrations, BYOK, unlimited users) | | Community and learning | Large, mature, peer-to-peer | Smaller, newer | The table satisfies the side-by-side most people came for. The rest of this piece is the opinion the table cannot carry: where each tool is genuinely the right choice, and the one architectural reason the ambiguous merge decision is the line that separates them. ### Who each product is for Zapier describes itself as “the world’s most connected AI platform” and frames its core product as “Do-it-yourself automation for workflows,” running “No-code automation across 9,000+ apps” (per its [apps directory](https://zapier.com/apps)). That is an honest description of a genuinely excellent product. Zapier is for anyone who needs to connect two or more SaaS tools and move data between them without writing code, and it serves an enormous range of buyers, from a solo founder wiring a form to a spreadsheet up to a RevOps team running hundreds of Zaps. FlowRunner is narrower on purpose. It is built for the operations leader, the VP of Ops, the RevOps or sales ops owner who is accountable for data quality and increasingly for who-changed-what. The buyer is not looking for the most connectors. They are looking for automation that knows when to stop and ask a person, and that can show an auditor or a manager exactly why a given Salesforce record looks the way it does. \[image\[dashboard\]: Side-by-side conceptual contrast of a linear Zapier-style trigger-action chain versus a FlowRunner multi-branch flow that splits clean matches from ambiguous matches and routes the ambiguous branch to a human reviewer.\] If your Salesforce hygiene work is mostly the automatic part, Zapier is very likely the right answer and the rest of this article will tell you so plainly. If the judgment part is where your hygiene debt actually accumulates, keep reading. ### App breadth and connector depth Start with the axis where Zapier wins outright, because pretending otherwise would waste your time. Zapier advertises “9,000+ apps” on its apps directory. FlowRunner’s connector catalog is smaller today, and that gap is real. If your hygiene workflow has to reach a long tail of niche or legacy SaaS tools, an industry-specific data vendor, a regional enrichment provider, an older marketing system, Zapier is far more likely to already have a pre-built connector waiting. For breadth of reach, this is not close. What FlowRunner trades breadth for is depth on the Salesforce object model and on what happens between systems. The platform exposes Salesforce actions like Find Record by Query, Update Contact, Update Lead, Convert Lead to Contact, and Create Note, and composes them inside a single multi-branch flow rather than a chain of separate Zaps. Both tools can read and write Salesforce records competently. Zapier’s Salesforce integration covers New Record, Updated Record, Find Record, and a useful Find or Create Record action, positioned by Zapier as “enterprise-grade security, compliance, and performance” (per its [Salesforce integration page](https://zapier.com/apps/salesforce/integrations)). The honest read on this axis: if the count of apps you must touch is the binding constraint, Zapier wins. If the binding constraint is what you do with the Salesforce records once you have them, the connector count stops being the thing that matters. ### The automatic match-and-update Here the two tools are closer than most comparison posts will admit. The bread-and-butter hygiene task is straightforward. A record arrives, you look for an existing match, and if you find a clean one you update it; if you do not, you create one. Zapier does this well. A trigger fires on a new or updated record, a Find or Create Record action does the lookup, and an update action writes the fields. For a high-volume, low-ambiguity stream where almost every record is either a clean match or a clean miss, this is exactly the right tool and FlowRunner offers no meaningful advantage on the happy path. FlowRunner runs the same logic as a flow. Find Record by Query returns zero, one, or several candidates; clean single matches route to Update Lead or Update Contact; zero matches route to Create. On the automatic path, this is the same work in a different editor. So if your stream is overwhelmingly clean, stop here. You do not need what comes next, and Zapier’s simplicity is a feature, not a limitation. The difference between the two platforms only starts to matter at the exact moment the match stops being obvious. ### The ambiguous merge decision This is the axis the whole comparison turns on. The hard cases in CRM hygiene are not “definitely the same record” or “definitely different.” They are “probably the same, but the email is different, the company name is spelled three ways, and one of them has an opportunity attached.” Resolving that case is a judgment call. It needs a person who can see the candidate and the closest existing records together and decide: merge, create new, or update. Zapier has no native human-in-the-loop step. That is not a knock; it is a design choice consistent with its linear, do-it-yourself model. To build a review gate in Zapier, you assemble it from parts: a Delay step to hold the record, a Filter to gate what proceeds, Paths to branch, and typically a separate approval app or a Slack message a person reacts to out of band. The record waits, someone eyes it somewhere, and a human nudges the Zap forward. It works for low volumes. What it does not do is make the human decision a first-class, recorded part of the workflow. The approval lives in a delay-and-filter workaround, and the reasoning lives in someone’s memory or a Slack thread. There is a structural reason this is hard in Zapier and it is worth naming. Every Zap begins with a trigger. That is the shape of the platform: an event fires, then actions run. It is a clean model for the linear case, and it is part of why Zapier is fast to learn. The constraint shows up when the work needs to be composed rather than triggered. A FlowRunner flow can start with any node type: an action, a trigger, a condition, or a group. A flow can be invoked as a callable subflow from another flow, treated like a function. It can be started manually by a reviewer. It can be started by a condition evaluating business state on a schedule rather than waiting for an external event. That flexibility is what lets the reviewer-routing pattern below be composed: the ambiguous branch can call a subflow that handles the human-decision step, and that subflow is reusable across other hygiene flows because it does not require its own trigger to exist. FlowRunner treats the ambiguous case as a real workflow step. AI agents are first-class nodes in the flow, not add-ons bolted onto a trigger-action chain, and one of the things an agent can do is invoke a human as a callable tool when it is uncertain. The pattern, concretely: - The agent runs the match query and classifies the candidate as clean, no-match, or ambiguous. - On ambiguous, it pauses that branch and posts a structured card to the reviewer’s Slack channel: the incoming candidate, the closest Salesforce matches, and the specific fields that agree and disagree. - The reviewer, a named person with the right permissions, picks merge, create, or update. - FlowRunner executes the chosen Salesforce action and records the decision and who made it on the record. - The clean and no-match cases never waited. Only the ambiguous branch paused. \[image\[flow\]: A FlowRunner flow diagram showing an intake record entering a match-query step, then branching three ways into clean match to update, no match to create, and ambiguous to a paused human-review node, with the ambiguous branch resuming after the reviewer’s decision and writing back to Salesforce.\] This is the andon cord that operations leaders keep describing when they talk about automation they would actually trust: the line stops itself when it hits something it should not decide alone, and a person with context makes the call. The full implementation shape of this pattern is documented in [Salesforce workflows with human-in-the-loop approval](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack), and the same shape one CRM over is [routing CRM duplicates to a human reviewer](https://flowrunner.ai/workflows/connect-hubspot-with-slack). Here is what most comparison posts on this topic will not say plainly: a duplicate-detection rule is not the hard part of data hygiene, and neither is the automatic update. Both tools clear those. The hard part is the case that needs a human, and the question that actually separates the platforms is whether the human decision is a native step with a record attached, or a workaround stitched out of delays and filters. That is the line. ### The audit trail Closely related, and the second place the platforms diverge for the operations buyer. Zapier keeps a task history for each Zap run, which tells you the Zap executed and what it did. That is useful for debugging an automation. It is organized around the Zap, not around the Salesforce record. FlowRunner records every create, update, merge, and human decision in a per-record execution log. When the question comes back six months later, why is this Account linked to this Contact, who approved the merge, what fields disagreed at the time, the answer is in the log rather than in someone’s memory. For a sales ops team that has to explain its CRM to a manager, a new RevOps hire, or an auditor, the unit of accountability is the record, and that is the unit FlowRunner logs against. FlowRunner adds governance features in two layers above Growth. Professional at $299/mo adds 30-day audit trails and RBAC. Business at $999/mo adds SSO/SAML and 90-day audit retention. On Zapier, SAML SSO sits on the Team plan and the audit-log depth lands on Enterprise, per its [pricing page](https://zapier.com/pricing). FlowRunner bundles audit trails and RBAC lower on the price ladder than Zapier does. If you do not need that posture, this axis does not matter to you. If audit trails and RBAC are the binding requirement, compare against Professional. If SSO is the binding requirement, compare against Business. \[image\[message\]: A Slack review card showing an incoming Salesforce contact candidate next to two existing records, with matching fields and conflicting fields visually distinguished, and three action buttons reading merge, create new, and update.\] ### Pricing and the cost of starting Now back to an axis where Zapier wins for a large set of teams, stated without softening because the brief is right that softening it would be dishonest. Zapier has a free tier (100 tasks per month) and a Professional plan starting from $19.99/mo billed annually, per its [pricing page](https://zapier.com/pricing). Zapier defines a task as something “counted every time a Zap successfully moves data or completes an action for you automatically,” and notes that polling for new data does not consume tasks. FlowRunner’s Free plan is 100 executions a month, Starter is $5/mo for 300 or $15 for 3,000, Growth is $45/mo for 12,000, Professional is $299/mo for 75,000, and Business is $999/mo for 250,000. The two pricing meters are not measuring the same thing, and the difference matters most as flows grow. Zapier counts tasks: each action inside a Zap that moves data or completes a step consumes quota. FlowRunner counts executions: one flow run is one execution, whether the flow contains five nodes or fifty. A hygiene flow that enriches a candidate, queries Salesforce for the closest matches, scores them, branches three ways, routes the ambiguous case to a reviewer, writes the decision back, and logs the result, is one execution on FlowRunner and ten or more tasks on Zapier. At low volume with linear Zaps, Zapier’s entry price wins on absolute cost and you should start there. As flows acquire nodes (enrichment, matching, branching, write-back), FlowRunner’s per-execution meter holds steadier while the per-task meter scales with the shape of the work. For an individual or a small team running simple, linear Salesforce hygiene, both platforms start at $0. Zapier’s free tier counts 100 action steps; FlowRunner’s Free plan counts 100 whole runs, so a five-step hygiene flow gets five times as many runs on FlowRunner before either plan is exhausted. The fair way to compare price is not entry cost in isolation. It is entry cost against the job. For the simple, linear job, start free on either and let the volume decide. For the job where the ambiguous merge needs a named reviewer in the loop at production volume, Growth at $45/mo is the right anchor on the FlowRunner side, because human-in-the-loop, AI agents, all integrations, BYOK, and unlimited users are included on every plan and Growth adds the volume and concurrency to run it every day. The $299 Professional tier comes into the comparison only when 30-day audit trails and RBAC become binding requirements. The $999 Business tier comes into the comparison only when SSO/SAML becomes a binding requirement. Pick the FlowRunner tier that matches the job you are actually buying for, not the highest tier on the page. ### Community, learning, and time-to-first-workflow One more axis where Zapier is genuinely stronger, and it would be a strawman to skip it. Zapier has a mature learning ecosystem and a large peer community. Its [community](https://community.zapier.com/) is organized into categories including “How Do I…?”, “Troubleshooting”, “Code & Webhooks”, and a “Developer Zone”, with tens of thousands of topics and replies, plus Zapier Learn courses, a blog, and webinars. When a non-technical user hits a wall on a filter, a path, or a formatter, the odds are high that someone has already asked and answered the question. That depth of self-serve troubleshooting is years in the making, and FlowRunner does not match it today. This is a real reason a non-technical team might reach for Zapier first, and it deserves weight in the decision. FlowRunner’s counterweight is not a bigger forum. It is the Midnight Flow consulting arm, which builds the first workflow on your data with you, so the time-to-first-working-flow does not depend on you finding the right forum thread. Those are different answers to the same need. Which one fits depends on whether you would rather self-serve or have it built with you. ### Where Zapier is stronger To put the honest axes in one place, because a comparison that hides them is not worth reading: - **App breadth.** “9,000+ apps” on Zapier’s directory versus FlowRunner’s smaller catalog. If your hygiene work touches a long tail of niche tools, Zapier is more likely to already connect. - **A familiar free tier.** Free at 100 tasks per month, Professional from $19.99/mo. FlowRunner’s Free plan matches it at 100 executions, counted per run rather than per step, so this is a tie on price and a Zapier edge only on name recognition. - **Community and learning depth.** A large, mature, peer-to-peer community with tens of thousands of answered topics, plus Zapier Learn, a blog, and webinars. - **Time-to-first-Zap for simple cases.** For a clean match-and-update with no human review, Zapier’s linear builder is fast and there is nothing to govern. If those four are the axes that map to your situation, Zapier is the right tool and you do not need to talk yourself into a heavier one. ### Where FlowRunner is the better fit And the symmetric list, scoped just as tightly: - **The ambiguous merge decision needs a human in the loop.** FlowRunner makes the reviewer a native, recorded step instead of a delay-and-filter workaround. - **You need a per-record audit trail.** Every create, update, merge, and approval is logged against the Salesforce record, not just against the automation. - **You need governance without enterprise pricing.** Audit trails and RBAC from $299/mo on Professional; SSO/SAML from $999/mo on Business. - **The work is multi-branch, not linear.** Branch synchronization, subflows callable as tools, and agents that escalate to humans suit hygiene that is more than a straight line. The deciding logic is simple. If hygiene for you is mostly the automatic match-and-update across many apps, Zapier. If the part that costs you is the judgment call on the ambiguous record, and being able to prove who made it, FlowRunner. That deciding line is not really about features. It sits at the seam between “a trigger fired” and “a record changed,” the gap where a person has to decide whether two records are the same thing. Native CRM rules force a binary at save time. A linear automation tool runs straight past that gap by design, because handling it would mean stopping the line and waiting for a human, which is not what a trigger-action chain is shaped to do. That gap, the coordination across intake, enrichment, and a human decision, is the work an orchestration layer exists to hold. Orchestration as a service is the category that owns that seam: a system that sits above the CRM and the apps feeding it, listens for what they emit, gathers the context the native rules never had, and pulls a person in at the exact moment the merge needs judgment. FlowRunner is built for that layer. Zapier is built to move data between the systems on either side of it, and it is very good at that. ### How to decide You do not need a feature matrix to make this call. You need to answer three questions about your own Salesforce hygiene. 1. **What fraction of your incoming records are genuinely ambiguous?** If it is small and most records are clean matches or clean misses, the automatic path is your whole problem, and Zapier solves it cheaply. If the ambiguous cases are where your team actually loses time, you are buying for the judgment part, and that is FlowRunner’s axis. 2. **Does anyone ever have to explain why a record looks the way it does?** If a manager, an auditor, or the next RevOps hire will ask who merged what and why, you need a per-record trail, and that points to FlowRunner. If nobody will ever ask, Zapier’s task history is enough. 3. **How many apps beyond Salesforce does the workflow have to touch?** If the answer is a long tail of niche tools, weigh Zapier’s breadth heavily. If it is Salesforce plus a couple of mainstream systems, breadth is not your constraint and the decision comes back to questions one and two. If you want the broader framing behind why agent orchestration is a different category than trigger-action automation, see [why agent orchestration is a different category than trigger-action automation](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents). And if you are weighing a third option, the same analysis applied to a different builder lives in our [FlowRunner vs Make for Salesforce data hygiene](https://flowrunner.ai/blog/flowrunner-vs-make-for-salesforce-data-hygiene) comparison. Zapier is a genuinely strong product and the correct choice for a large share of hygiene work. This comparison is not an argument that it is not. It is an argument that the ambiguous merge decision, and the audit trail behind it, is a different shape of problem, and that the tool you pick should match the half of the problem that actually costs you. ### Quick answers **Can Zapier do Salesforce data hygiene?** Yes, for the linear part. Zapier can trigger on a new or updated Salesforce record, run a Find or Create Record action, and update fields automatically. Where it gets thin is the ambiguous case: a probable duplicate with a different email and three spellings of the company name. Zapier has no native human-in-the-loop step, so a reviewer gate has to be assembled from delays, filters, and a separate approval app, and the decision is not recorded as a first-class part of the run. **Is FlowRunner a Zapier alternative for CRM data cleanup?** For CRM data cleanup specifically, yes, when the cleanup needs judgment and a paper trail. FlowRunner runs the same automatic match-and-update path Zapier does, then treats the ambiguous match as a real workflow step: the agent pauses, posts the candidate and the closest Salesforce matches to a named reviewer in Slack, waits for the merge-or-create decision, executes it, and logs it. Zapier remains the better pick when the job is simple and needs to reach a long tail of apps FlowRunner does not connect to yet. **Which is cheaper, FlowRunner or Zapier, for Salesforce work?** Neither, at the entry point. Zapier has a free tier at 100 tasks per month and a Professional plan starting from 19.99 dollars per month per its pricing page. FlowRunner has a Free plan at 100 executions per month, where an execution is a whole run rather than one action step, and a Starter plan at 5 dollars for 300 executions or 15 dollars for 3,000; Growth is 45 dollars per month for 12,000 executions. Every FlowRunner plan, including Free, includes AI agents, native human-in-the-loop on multiple channels, all integrations, BYOK, unlimited users, and unlimited workflows. Professional at 299 dollars per month adds 30-day audit trails and RBAC. Business at 999 dollars per month adds SSO/SAML and 90-day audit retention. If your hygiene need is simple and linear, both start free, and FlowRunner’s Free plan counts whole runs. If the ambiguous merge needs human review at production volume, compare Growth at 45. If audit trails and RBAC are the binding requirement, compare Professional at 299. If SSO is the binding requirement, compare Business at 999. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## FlowRunner vs Zapier for Marketing Workflow Automation: An Honest Comparison Source: https://flowrunner.ai/blog/flowrunner-vs-zapier-marketing-workflow-automation Comparisons May 28, 2026 Updated September 7, 2026 13 min read Zapier owns connector breadth and fast setup. FlowRunner wins when a marketing workflow needs a human to approve a send or spend before it fires. How to decide. ![A bald cartoon man at a fork in a marketing conveyor belt: one track runs straight off the panel while he holds a lever controlling a raised gate on the second track, deciding whether to release a waiting parcel.](https://flowrunner.ai/images/blog/flowrunner-vs-zapier-marketing-workflow-automation-hero.webp) Most marketing automation runs forward and never has to stop, and that is exactly why the failures hurt: the workflows that send an email to a list, move a budget, or hand a hot account to a rep are the ones you cannot take back once they fire. [Zapier](https://zapier.com) is very good at running marketing work forward. The honest question for a marketing ops team is not whether Zapier can connect your tools, because it connects almost everything. It is whether your hardest workflows need a place to stop, ask a person, and only then continue. This is a comparison of [Zapier](https://zapier.com) and FlowRunner for marketing workflow automation, written for the marketing ops or demand gen lead deciding which platform fits the work they actually run. If you also want the FlowRunner read against a different connector-first builder, the [companion comparison with Make.com](https://flowrunner.ai/blog/flowrunner-vs-make-for-marketing-workflow-automation) covers that pairing. ### The decision in one table | | Zapier | FlowRunner | | --- | --- | --- | | **Self-description** | [”AI automation, governed”](https://zapier.com) / “Your tools. Your rules. Any AI.” | Orchestration as a service for coordinating AI agents and humans across tools | | **Core model** | Forward-running Zaps (trigger, then a chain of tasks), with conditional Paths and AI Agents layered on | AI agents as first-class workflow nodes that coordinate, escalate, and resume | | **Connector library** | [9,000+ apps claimed](https://zapier.com/apps), the broadest in the category | Smaller pre-built catalog today; new connectors built fast; MCP support for unlisted systems | | **Human approval** | Notify-and-wait steps and [Paths conditional logic on Professional and up](https://zapier.com/pricing) | Agent invokes a human as a callable tool on uncertainty, hands over context, resumes from the decision | | **Entry pricing** | [Free $0/mo; Professional from $19.99/mo billed annually](https://zapier.com/pricing) | Free $0 (100 executions); $5/mo Starter; $45/mo Growth; $299/mo Professional | | **SSO/SAML** | [Team plan, from $69/mo](https://zapier.com/pricing) | $999 Business tier | | **Audit logs** | [Enterprise only](https://zapier.com/pricing) | $299 Professional tier (30-day); $999 Business tier (90-day) | | **RBAC and SLA tracking** | Higher tiers | $299 Professional tier | | **Expert ecosystem** | [Mature Solution Partners directory](https://zapier.com/experts) and large community | Developing; partner network and community still early | | **Best fit** | Broad app connectivity and fast forward-running automations | Workflows that must stop for human judgment before a send, a spend, or a handoff | The table settles the structural questions. What it cannot settle is which shape of marketing work is yours, and that is the rest of this article. ### What Zapier is genuinely good at Zapier is not the tool the older comparison posts describe. It has [repositioned around the line “AI automation, governed”](https://zapier.com) and the promise “Your tools. Your rules. Any AI.” It now markets AI agents and guardrails, claims hundreds of thousands of agents built on the platform, and is “built for everyone, from the Fortune 500 to first-time founders.” Pretending Zapier is a simple no-AI connector tool would be a strawman, and it would be wrong. Here is what Zapier does well that this article is not going to argue against: - The broadest connector library in the category. Zapier [claims 9,000+ apps](https://zapier.com/apps), and that breadth genuinely beats FlowRunner today, including the niche and legacy marketing tools most platforms never get around to building. The marketing stack tools you already run are there: Mailchimp, HubSpot, Facebook Lead Ads, Salesforce, Google Analytics. - A setup experience that a non-technical marketer can navigate alone. For a single forward-running automation, Zapier is fast to first result and the learning curve is gentle. - A familiar entry point. Zapier opens with a [Free plan at $0/mo and a Professional plan from $19.99/mo billed annually](https://zapier.com/pricing) that unlocks multi-step Zaps and Paths conditional logic. FlowRunner’s Free plan and $5 Starter match that entry and include multi-step flows from $0, so the sticker is a tie; Zapier’s advantage here is recognition, not price. - A mature ecosystem around the product. Zapier runs a [Solution Partners directory](https://zapier.com/experts) of “consultants, freelancers, and agencies who specialize in streamlining business processes with automation,” with partner tiers and a deep well of community content. FlowRunner is newer, and its peer support and third-party expert network are still developing. If you lean on community answers and hireable experts, that gap is real and it favors Zapier. That is a capable, well-supported product. A marketing ops decision that waves it away is not honest. The reason to keep reading is that connector breadth and fast setup are not the constraint on the workflows that actually keep marketing ops up at night. ### The workflows that should not run forward Here is what most Zapier-versus-anything posts will not say. The real axis is not simple versus complex, and it is not no-AI versus AI, because Zapier now does AI. The axis is whether a workflow runs forward to completion or has to stop, hand a decision to a person, and continue from what that person decided. Most marketing automation can and should run forward. A form fills a HubSpot record, the record adds a Mailchimp subscriber, a closed-won deal posts to a Slack channel. None of that needs a person in the middle, and Zapier handles it cleanly. The workflows that hurt when they go wrong are the ones that touch a send or a spend: - An email or SMS blast goes to a segment, and the segment query was built off a lead-score field that three systems disagree about. Once it sends, it is sent. - A paid-campaign budget shifts automatically based on a performance signal, and the signal was an attribution artifact, not a real lift. The spend already moved. - A hot inbound account routes straight to a rep, except the enrichment was low-confidence and the account is actually an existing customer’s new domain. The handoff already happened, and now two reps think they own it. - A nurture sequence fires off negative-sentiment language in a reply because no step paused to ask whether this contact should have been escalated to a human instead. The pattern under all four is the same. The mechanical part of the work is easy to automate. The part that needs judgment is a single decision point, it shows up unpredictably, and the cost of getting it wrong is a thing you cannot un-send. This is the shape of work that [marketing ops carries across lead scoring, attribution reconciliation, and sales handoff](https://flowrunner.ai/solutions/marketing-workflows), and it is the shape that a forward-running task chain handles least gracefully. The pattern is visible in published FlowRunner workflow guides like [Salesforce lead conversion with human approval](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack), where the conversion decision is the exception and the routing around it is the automated part. ### How FlowRunner treats the agent and the human [FlowRunner](https://flowrunner.ai) is built around that decision point rather than around the connector. In FlowRunner, AI agents are first-class workflow nodes, not add-ons grafted onto a trigger-action model built for app connectivity. An agent has a job, it works that job against the context it can gather, and the platform records what it did. The pattern that maps to the send-and-spend problem looks like this: 1. A trigger fires (a new lead in HubSpot, a campaign performance signal, a reply landing in a shared inbox, a budget-change condition met). 2. The workflow gathers context from every system involved, not just the one that fired the trigger. 3. An agent attempts the mechanical decision: score the lead, deduplicate the account, classify the reply, validate the attribution behind the budget move. 4. If the agent is confident, it completes the action and continues without involving anyone. 5. If the agent hits uncertainty (a low-confidence enrichment, an account that might already be a customer, an attribution signal that does not hold up, sentiment that reads as a churn risk), it calls a human as a callable tool. The named owner gets a Slack message, email, WhatsApp, or phone contact with the full working context attached: what the agent tried, what it found, the options, and a recommendation. 6. The workflow resumes from the human’s structured response, writes the decision back to the source system, and produces an audit-trail record of who decided what and when. Step five is the architectural difference, and it is worth being precise about it because Zapier supports human steps too. FlowRunner’s human-in-the-loop pattern lets an agent invoke a person the way it would invoke any other tool, mid-reasoning, and then continue from a richer context that now includes the human’s decision. That is not the same as a notification followed by a wait. The same pattern carries the [HubSpot and Slack workflow with human-in-the-loop duplicate handling](https://flowrunner.ai/workflows/connect-hubspot-with-slack), where the agent catches a possible duplicate before it multiplies rather than after. One more compositional difference matters for marketing ops teams that build a lot of related workflows. In Zapier, every Zap begins with a trigger. The first step is always something happening in a source system. FlowRunner removes that constraint: a flow can start with any node type. A trigger, yes, but also an action that runs on a schedule, a condition that evaluates business state, or a group of steps invoked as a callable subflow from another flow. That last one is what compounds. The lead-scoring subflow you build once becomes a tool the deal-routing flow calls, the win-loss-attribution flow calls, the churn-risk flow calls. The “kick off a manual run for one record” pattern stops requiring a fake trigger. Neither model is more or less correct, but for orchestration that has to coordinate multiple workflows over the same lead lifecycle, the freedom to start a flow from somewhere other than an inbound event is the difference between writing the same logic three times and writing it once. The category that owns this is orchestration as a service: a coordination layer that sits above the marketing tools and the agents inside them, holds the workflow open at the moment a send or a spend needs a human yes, and resumes the instant that yes arrives. Every marketing tool is now shipping its own AI and its own automation, Zapier included, which makes the coordinating layer above them the thing worth choosing carefully. FlowRunner is built for that layer. The control point before money goes out or a message reaches a list is precisely where a forward-running chain has no natural place to stand, and it is exactly where this layer earns its keep. ### The human-in-the-loop difference, said fairly This is the place comparison content cheats, so let me be careful. Zapier has human approval. It supports notify-and-wait steps, and it supports [conditional branching through Paths on its Professional plan and above](https://zapier.com/pricing). A marketing ops team can build a Zap that routes a lead into a Slack channel, waits for a yes or no, and continues based on the answer. Saying Zapier cannot pause for a person would be false, and the honest contrast is not capability-presence. The contrast is where the human sits in the model. In Zapier, a human step is a node inside a forward-running task chain: the Zap reaches the step, sends the notification, holds, and resumes when the response lands. That works, and for a lot of marketing approvals it works well, especially the rule-based ones (“if the discount exceeds X percent, ask for a yes”). In FlowRunner, the human is something an agent reaches for when its own judgment runs out. The agent is not just executing a step that happens to require a click. It is working a problem, recognizing that the next move is one it should not make alone, and escalating with the reasoning already structured. The two models converge on simple rule-based approvals. They diverge as the decision gets less rule-shaped and more judgment-shaped, which is exactly the direction marketing work has been moving as more of it runs through AI. Both patterns are legitimate. Many marketing workflows want both. The question is which one is the bottleneck in the workflows that scare you. ### Where Zapier is the better fit Choose Zapier when: - Your marketing automation is mostly forward-running connectivity: a trigger, a chain of app-to-app steps, a defined outcome, and no decision in the middle that a person has to own. - The tools you need to connect are the constraint, and you want the [broadest app library in the category](https://zapier.com/apps) so the niche and legacy connectors are already built. - You are a solo marketer or a small team where the [Free plan or the $19.99/mo Professional plan](https://zapier.com/pricing) matters more than governance that ships at a higher tier. - You rely on community answers and hireable help, and Zapier’s [Solution Partners directory](https://zapier.com/experts) and large user base are worth real money to you. - Your approval flows are rule-based (“over this threshold, ask”), which a notify-and-wait Path handles cleanly. That describes a large and legitimate share of marketing ops work. If it describes yours, Zapier is the right call and an orchestration layer is a question for later. ### Where FlowRunner is the better fit Choose FlowRunner when: - Your hardest workflows touch a send or a spend, and the cost of firing them on a wrong assumption is something you cannot take back. - The judgment in those workflows is not rule-shaped. The escalation is triggered by an agent’s uncertainty (low-confidence enrichment, conflicting attribution, a churn-risk sentiment flag, a hot account that might already be a customer), not by a fixed threshold. - You want AI agents as first-class workflow nodes that coordinate across systems and invoke a human as a callable tool, rather than approval steps wired into a forward chain after the fact. - You want governance infrastructure without an enterprise procurement cycle. [FlowRunner’s $299/mo Professional tier](https://flowrunner.ai/pricing) includes 30-day audit trails, RBAC, and SLA tracking by design; SSO/SAML and 90-day audit retention arrive at the $999/mo Business tier. On Zapier, SSO is the Team tier ($69/mo) and advanced audit logs are Enterprise-only, so Zapier reaches SSO cheaper while FlowRunner reaches built-in audit trails and roles sooner. - You want a per-run record of who approved which send or spend and why, because someone downstream is going to ask. - Your flows are getting complex enough that the cost of adding more steps inside one workflow is starting to matter. [FlowRunner counts executions](https://flowrunner.ai/pricing), not actions inside a flow. Growth is $45/mo for 12,000 executions, Professional is $299/mo for 75,000, Business is $999/mo for 250,000, and a five-node flow and a five-hundred-node flow consume the same one execution. [Zapier prices by tasks per Zap run](https://zapier.com/pricing), so each enrichment lookup, conditional branch, or notification step inside a single workflow draws down the monthly quota. The honest read on both: at low volume with simple Zaps, Zapier’s entry pricing wins on raw dollars. The model flips as the workflows get richer, because adding a step in FlowRunner does not change the bill while adding a step in Zapier does. These are not exclusive worlds. A realistic outcome is Zapier for the broad forward-running connectivity and FlowRunner for the workflows that have to stop and ask before they commit. The architectural difference is real, and the right tool depends on which of those two jobs is your actual problem. ### How to decide Walk one list before you choose. Pull up your last five marketing automations that either fired when they should have paused or sat stalled when they should have moved, and put each one in a bucket: - It broke because a connector was missing or a step was misconfigured. That is a connectivity problem, and the breadth and polish of Zapier point at it. - It broke because a send went out on a bad assumption, a budget moved on a soft signal, or a handoff happened that a person should have caught first. That is a stopping-point problem, and a forward-running chain is the wrong shape for it. - It broke because nobody could say afterward who approved it. That is a governance problem, and where audit trails and roles sit in the price ladder decides it. Count the buckets. If most of your pain is the first bucket, Zapier is the better tool and this article is over. If your real pain is the send that cannot be unsent, the spend that already moved, or the approval nobody can trace, the question stops being which connector library is bigger. It becomes which platform was built to hold a marketing workflow open at the exact moment a human has to say yes, and to remember that they did. ### Quick answers #### Is FlowRunner a replacement for Zapier? Not for most of what Zapier does well. If your marketing automation is mostly connecting apps in forward-running steps (a form fills HubSpot, HubSpot adds a Mailchimp subscriber, a deal change posts to Slack), Zapier’s 9,000+ app library and fast setup make it the better tool. FlowRunner is the better fit when a workflow needs a human to approve a send, a spend, or a high-stakes handoff before it fires, with the agent gathering the context and the platform keeping the audit trail. Some teams run both. #### Does Zapier have human approval steps? Yes. Zapier supports conditional branching with Paths on its Professional plan and up, and you can build a notify-and-wait approval into a Zap. Saying Zapier cannot pause for a human would be wrong. The difference is architectural: in Zapier a human step is a node added into a forward-running task chain, while in FlowRunner an AI agent invokes a human as a callable tool when it hits uncertainty, hands over its working context, and resumes from the decision the human returns. #### Is Zapier cheaper than FlowRunner for marketing automation? Not on the sticker. Zapier has a Free plan ($0/mo, two-step Zaps, 100 tasks) and a Professional plan from $19.99/mo billed annually that unlocks multi-step Zaps and Paths. FlowRunner has a Free plan at $0 for 100 executions that ships the full core platform (multi-step flows, AI agents, human-in-the-loop, all integrations, BYOK), a Starter plan at $5 for 300 executions or $15 for 3,000, and Growth at $45/mo for 12,000. The Professional tier at $299/mo adds the governance overlay (30-day audit trails and RBAC); SSO/SAML is at Business $999/mo. For a solo marketer wiring simple Zaps, the two free plans cost the same, and FlowRunner’s counts whole runs rather than steps. The comparison shifts when you need governance built in, since on Zapier SSO is the Team tier ($69/mo) and audit logs are Enterprise-only. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## HIPAA Compliance Software Checklist for Finance Leaders Source: https://flowrunner.ai/blog/hipaa-compliance-software-checklist Articles May 26, 2026 Updated May 28, 2026 9 min read Most HIPAA software checklists are written for developers. This one is for the CFO signing the BAA, with six questions to ask before contract. ![A bald cartoon man with three hairs sticking up checking a noticeboard of vendor name cards with a clipboard, his pen pointed at one unstamped card circled in faded red.](https://flowrunner.ai/images/blog/hipaa-compliance-software-checklist-hero.webp) Most HIPAA software checklists are written for the team building a patient-facing app. This one is written for the CFO signing the Business Associate Agreement. The two readers have different jobs, and conflating them is why finance teams at healthcare-adjacent companies keep ending up with vendor contracts that look complete on paper and fall apart the first time an auditor asks who touched what. If you run finance at a healthcare practice, a healthtech vendor, a healthcare-adjacent SaaS company, or any business where Protected Health Information moves through your AP, billing, payroll, or approval workflows, the question is not whether to architect an encryption scheme. The question is whether you can answer six concrete things about every system in the stack before you put your name on the contract. This is that checklist. ### Why finance leaders need a different HIPAA checklist Walk through the top of the search results for HIPAA compliance checklists and you find the same article published under fifteen logos: a developer’s guide to building HIPAA-compliant software. Encrypt PHI at rest. Hash passwords. Use TLS. Implement audit logging at the application layer. Those checklists are useful for the team writing code. They are not the checklist the CFO needs. The finance leader’s exposure is structural and contractual, not architectural. PHI flows through systems you did not build and rarely chose for their PHI-handling characteristics. AP automation. ERP. Billing platforms. Document storage. The automation platform someone in operations spun up two quarters ago to chase invoices. Each of those is a potential business associate under HIPAA, and each one you authorize to touch PHI without the right paperwork in place is exposure that lives on your audit trail, not the developer’s. Here is what most HIPAA software checklists will not say plainly: encryption is the easy part. The hard part is keeping track of which systems are touching PHI, whether a Business Associate Agreement covers them, and whether you could produce the audit evidence to prove it twelve months after the fact. That is a finance and operations problem, not a security-team problem, and it does not get solved by a TLS cipher recommendation. ### The six-item HIPAA software checklist for finance Run this against every system in the stack that might touch PHI. Run it before you sign the contract, not after. #### 1\. A signed BAA on file for every vendor that touches PHI A Business Associate Agreement is the contract that obligates a vendor to handle PHI under HIPAA’s rules. The U.S. Department of Health and Human Services [explicitly requires](https://www.hhs.gov/hipaa/for-professionals/covered-entities/sample-business-associate-agreement-provisions/index.html) one between a covered entity and any vendor that creates, receives, maintains, or transmits PHI on its behalf. The work is the inventory, not the signature. Walk every system that touches the finance side of the operation and ask which ones are likely to see PHI even incidentally: - ERP (patient statement printing, EOB processing, payment posting) - AP automation (medical-supply invoices with patient identifiers, EOBs that arrive as invoices) - Billing platform (the obvious one) - Document storage and shared drives (anywhere a screenshot, fax, or PDF lands) - Approval workflows that route patient-linked transactions - Email systems that handle PHI in attachments - Any automation or AI platform that reads, writes, or routes data from the above For each one, the BAA is either signed, available on request, or missing. Missing is the gap you act on. Available-on-request is the second gap. #### 2\. Access controls and role-based permissions Finance staff should not see clinical data. Clinical staff should not see banking detail. The system either supports that distinction or it does not. Ask the vendor to walk a sample permission matrix that lets a billing clerk see the line items needed to reconcile a payment without exposing the diagnosis codes, treatment notes, or any element of PHI not required for the financial task. If the answer is “everyone with login access can see everything,” the system is not in scope for a healthcare-adjacent finance stack regardless of what else it does well. #### 3\. An audit trail covering who accessed what and when HIPAA’s documentation retention rules, anchored in the [HHS regulatory text](https://www.hhs.gov/hipaa/for-professionals/privacy/laws-regulations/index.html), require covered entities and business associates to retain documentation for a minimum of six years. That includes audit logs, policies, and the records of access and modification that prove the system is operating as the contract says. The question to ask the vendor is concrete: produce a sample audit log for one transaction. Walk it from the moment a record entered the system to the moment it left or was last modified. If the vendor cannot show that log inside the demo, six years from now you cannot produce it for an auditor either. #### 4\. Encryption at rest and in transit This is the part the developer checklists handle well. It is also table stakes, not a differentiator. Ask the vendor for documentation rather than taking marketing claims at face value: which encryption protocols, which key management, who holds the keys, how key rotation works. Then move on. Encryption is the floor; the BAA, the access controls, and the audit trail are the building. #### 5\. A breach notification process HIPAA’s Breach Notification Rule, [published by HHS](https://www.hhs.gov/hipaa/for-professionals/breach-notification/index.html), requires business associates to notify covered entities of a discovered breach without unreasonable delay and within 60 days. That is the regulatory floor. It is not a maximum acceptable internal target. The questions to ask: - What is the vendor’s contractual notification window? It must be at or below 60 days. - What is their internal detection process? A 60-day notification window is meaningless if it takes 55 days to notice the breach. - Who at the vendor is responsible, and how is the notification delivered? If the vendor cannot answer in concrete terms, the rule is on paper only. #### 6\. Exception handling and human review for automated workflows This is the item most checklists do not have, and it is the one that most affects finance teams running AI agents and automation platforms over PHI-adjacent data. Any automation touching PHI eventually hits an edge case the rule did not anticipate: a duplicate patient identifier, an unusual claim format, an invoice with mixed PHI and non-PHI line items. The system either silently processes the edge case (and your audit trail loses the moment of judgment) or it pauses and pulls a human in with the full context attached. The second pattern is what holds up under audit. The first pattern is what creates the breaches nobody noticed until the quarterly review. Ask the vendor what happens when their automation hits an exception involving PHI: who gets pulled in, what context they see, and how the decision is captured against the underlying record. ### Where finance software stacks fail the checklist The six-item list is the easy 70% of the work. The gaps that recur across the healthcare-adjacent finance stacks I look at: - **Legacy ERPs predate HIPAA-aware access controls.** Older Dynamics NAV and Great Plains installations rely on perimeter security rather than row-level permissions. The system “is HIPAA compliant” in the sense that nobody outside the network can see anything, but every finance user inside the network sees every patient line item the ERP holds. - **AP automation tools without a signed BAA.** Finance assumed the vendor was not handling PHI because AP is “just invoices.” Then a stack of EOBs and patient statements started flowing through and the BAA question never got re-asked. This is the single most common BAA gap on the finance side. - **Approval workflows that live in email and shared inboxes.** A clerk emails the controller for approval on a patient refund. The controller replies “approved.” Twelve months later the audit asks who approved that refund and on what basis, and the answer requires inbox archaeology. Approval evidence stored in email is harder to reconstruct than approval evidence captured against the underlying record. The same architectural problem covered for purchase orders in [PO Approval Workflow: What It Is and Where Most of Them Break](https://flowrunner.ai/blog/po-approval-workflow) applies double when the transaction is patient-linked. - **Piecemeal AI agents built by consultants.** A consultant set up an agent to do invoice matching three quarters ago. It works. Nobody knows what data it touches, where the prompt is stored, or how to produce its execution log. That fails the audit trail test on its own merits and fails the BAA test on top. Each gap is patchable in isolation. The pattern that compounds is what one of the finance leaders I talked to called the piecemeal problem: there is no broad strategy, only point fixes accumulating across the stack. ### How to evaluate automation platforms specifically The automation layer deserves its own treatment because it is the seam between every other system on the list. An automation platform reads from the ERP, writes to the billing system, pulls vendor data from AP, and routes approvals through email or chat. If the platform is in scope for PHI, every connection it owns is in scope. The questions to ask a candidate automation vendor: - **Do you sign a BAA, and at what plan tier?** A vendor whose BAA only comes at the top tier is fine if that tier is your tier; if it is not, the conversation ends here. - **Can I see a sample audit log for a workflow execution?** It should show which integrations were called, what data was passed, who or what triggered the execution, and what the outcome was. Not a system uptime log. The execution log for one run of one workflow. - **How does a human reviewer get pulled in when an automation hits an exception involving PHI?** This is the human-in-the-loop question. The vendor’s answer is either a coherent description of how the platform pauses, routes the exception to a named person with full context, and captures the decision back against the record, or it is a marketing slide. - **What happens to the data after the workflow runs?** Is it retained inside the platform, and for how long? Six-year retention is the HIPAA floor on documentation; the data itself has its own retention question that needs a direct answer. FlowRunner is designed against this list deliberately. Audit logs and human-in-the-loop exception handling are designed to meet common audit requirements rather than treated as bolt-on features above a tier line. That is a category position, not a HIPAA-certification claim; HIPAA has no certification program, and any vendor offering one is selling something other than HIPAA compliance. The broader point is structural. A finance stack at a healthcare-adjacent company has half a dozen systems of record: ERP, billing, AP, document storage, payroll. None of them see each other. Native ERP modules do not see the AP automation. The AP automation does not see the billing platform. The billing platform does not see the document storage. Each one holds part of the story. The audit trail asks for the whole story, tied to one patient or one transaction. That is the category problem. The work between the systems of record, and the human judgment between those systems, has no native home in any of them. An orchestration layer is the category name for the home it needs. It sits above the systems of record. It subscribes to what they emit, joins the context the individual rules never had, and pauses the workflow when the data lands on an exception. Then it writes the answer back against the record so the trail is whole when an auditor asks. FlowRunner is built for that layer. It does not replace the ERP or the billing platform. It makes their outputs reconstructible end to end and the exceptions handlable without inbox archaeology, which is the part the six-item checklist exists to surface. For the AP-side mirror of the same exception pattern in non-PHI finance work, the same architecture handles [vendor invoices flowing into the ERP with exception routing](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and the [approval-routing pattern](https://flowrunner.ai/use-cases/approval-routing) that captures approver identity against the underlying record. The PHI-adjacent versions of those workflows live on the same plumbing; FlowRunner’s [healthcare industry page](https://flowrunner.ai/industries/healthcare) covers the broader pattern. ### A 30-day action plan Run the checklist as a project, not a meeting. - **Week 1.** Inventory every software system that touches PHI or PHI-adjacent finance data. Walk every integration, every shared drive, every approval channel, every automation. The list is almost always longer than memory suggests. - **Week 2.** Pull the BAA status for each vendor on the list. Mark each one signed, available, or missing. Flag every “missing” or “available on request” as an open item. - **Week 3.** Run the six-item checklist against each system. Document gaps in plain language tied to which item failed. For automation platforms specifically, ask the four questions in the section above. - **Week 4.** Prioritize the gaps by exposure. A vendor processing PHI without a BAA is higher priority than a system with weak audit logs. A system with no exception-handling for AI-driven workflows on patient data is higher than either. Work the list top-down. The gaps that get closed in the first week of work are the same gaps that would have shown up in the first hour of an audit. The point of the checklist is not to produce a tidy spreadsheet for the next compliance meeting. The point is to make the moment of inspection survivable. A finance stack that can answer the six questions about each system, with the BAA on file and the audit trail reconstructible, is a stack the CFO can sign for. A stack that cannot is one quarterly review away from a much harder conversation. ### Quick answers #### Is HIPAA compliance certification a real thing for software vendors? No. HIPAA has no certification program. A vendor can be designed to support HIPAA workflows and willing to sign a Business Associate Agreement, but no government body certifies software as HIPAA compliant. Treat any vendor claiming certification with caution. #### Does every finance vendor need a BAA? Only the ones that create, receive, maintain, or transmit Protected Health Information on your behalf. Many AP, billing, and document-storage tools touch PHI without finance realizing it, which is why the inventory step matters more than the contract step. #### How long do HIPAA audit logs need to be retained? HIPAA requires covered entities and business associates to retain documentation, including audit logs and policies, for a minimum of six years from creation or last effective date. Verify your vendor can produce a log on request and that retention matches that floor. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## How to Automate Accounts Payable Without Losing the Judgment That Matters Source: https://flowrunner.ai/blog/how-to-automate-accounts-payable Articles May 26, 2026 Updated June 19, 2026 9 min read AP automation is a sequencing problem, not a tool problem. Get capture, coding, approval, and payment in the right order, and exceptions become the work. ![A bald cartoon man with three hairs sticking up at a small workshop bench, sorting invoices into four labeled bins (capture, coding, approval, payment) and holding back one invoice marked with an amber tag for closer examination.](https://flowrunner.ai/images/blog/how-to-automate-accounts-payable-hero.webp) The honest reason most accounts payable automation projects produce a “black box” feeling six months in is that the project picked a tool before deciding which of four sequential decisions it was actually automating. AP looks like one process from the outside. From the inside, it is four: capture, coding, approval, and payment. Sequence them in the wrong order, and each one quietly inherits the mess of the one before it. This article is for the finance leader weighing whether to commit, not the AP clerk picking software. The argument is that AP automation is a sequencing problem and an exception-handling problem, in that order, and that the tool conversation is the last one to have, not the first. ### Why AP automation stalls for most finance teams The pattern is recognizable across roughly a dozen conversations with CFOs and finance leaders at mid-market companies. Invoices get captured by an OCR product. Structured data lands somewhere. Everything downstream is still manual coding, chasing approvals, and Friday payment runs in the ERP. One finance leader described his own AP day this way: “if someone’s out, I might even enter a bill into the system.” When a senior finance person is the backup AP clerk, the automation that exists is doing capture and capture only. The honest baseline this article assumes is real, not strawmanned: - Invoices arrive by email and PDF, sometimes by EDI for the bigger vendors. - Approval routing happens informally, over email and Slack, with thresholds that exist in policy memos but not in any system. - Reconciliation happens at month-end, in spreadsheets, against the bank feed and the GL. That is not a broken process. It is the process most ten-to-two-hundred-person companies actually run, and it works for a while. It stops working when transaction volume rises faster than headcount, or when a duplicate payment slips through against a distributor who also paid, or when the audit asks who approved a specific bill on what basis and the answer requires three months of inbox archaeology. The “black box” framing one CFO used to describe a prior AI engagement is exactly what happens when piecemeal point tools get stitched together without a sequencing decision underneath. A scanner here, a payment portal there, an approval app over there, no unified audit trail across the three. The finance team can no longer reconstruct what happened to any given bill without opening four systems. ### The four decisions that make up AP, and which ones to automate first Treat each of these as a discrete decision with its own automation question. The sequencing matters because the output of each step is the input to the next. **1\. Capture.** Getting invoice data out of email, PDF, and EDI into structured fields: vendor, invoice number, date, line items, totals, GL hints. This is the highest-volume, lowest-judgment step in the chain. Automate it first. Document parsing into a structured ERP record is the workable starting point for AP automation; the [email-to-payment workflow we publish for QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) shows the shape, with parallel patterns published for [parsed invoices into Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and [vendor documents into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack). The mechanics are similar; the ERP-specific quirks live in the field mapping. **2\. Coding.** Assigning GL account, cost center, and project to each line item. This is partial-automation territory. Most invoices from recurring vendors will code identically every month; rules and history can suggest the right code with high confidence. Some invoices will not. Automate the suggestion with a confidence threshold, route low-confidence items to a human, and let the human’s correction feed back into the rule. Coding is where AI assistance pays for itself, not where it should run unattended. **3\. Approval.** Routing by amount, vendor, department, or category, with named approvers and a written delegation rule. Automate the routing. Keep the judgment human. The approver is committing company funds; that signature is the point of the workflow, not a step to be removed. A [bill approval routed through Slack with named approvers](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) is the right shape: the system handles the routing and the audit capture, the human handles the actual approval decision, and the result writes back against the bill record in the ERP. **4\. Payment.** Scheduling and executing payment through the ERP or the bank. Automate the mechanics, including duplicate-payment checks at the moment of payment release. Keep cutoff and release as explicit human actions, especially for any payment above a tier threshold or to a vendor flagged as new or stale. This is the step where automation that runs unattended generates real money loss, fast. The cost of a single duplicate payment to a major distributor will outweigh the labor cost of every AP clerk hour you would have saved in a quarter. The sequencing rule that follows from this: do not automate step N until step N-1 produces output you trust. Automating approval routing before coding is reliable means routing miscoded bills to approvers. Automating payment before approval is reliable means paying bills with broken approval evidence. ### Where exceptions belong, and why this is the actual ROI Here is what most articles on AP automation will not say plainly: [automation exceptions](https://flowrunner.ai/concepts/automation-exceptions) are the work. The automation handles what is not the work. The five percent of bills that need a human eye is where the controller earns the title; the ninety-five percent that flow through is overhead the automation should make invisible. What that means in practice is that the design conversation should center on the exception queue, not on the throughput rate. The four exception types every AP automation worth committing to should handle: - **Duplicate invoice detection.** Match on vendor identity, amount, invoice number, and date before the bill ever reaches an approver. One CFO put the underlying concern this way: “I feel that it would be very easy to double pay on something like that if both we and the distributor made a payment.” The risk is real, the math is straightforward, and the check belongs at the moment of capture, not at month-end reconciliation when the money is already out. - **Three-way match against PO and receipt.** When those records exist, the bill matches against them automatically and clears. When they do not, the bill pauses and waits for a human to either find the missing record or override the match with a written reason that the audit trail captures. (Three-way match against invoice, PO, and receipt is standard AP control practice in the IOFM and AICPA guidance most controllers know. The automation question is whether your tooling will enforce it before approval, or pretend the bill cleared.) - **Distributor and vendor billback validation.** Billbacks are notoriously a back-and-forth between what the vendor charged and what the contract says they could charge. Compare claimed amounts against contract terms before the payment goes out, not after the AR team raises a credit memo three months later. - **Mismatch flags everywhere else.** Vendor name doesn’t match the master file. Amount differs from the PO by more than a tolerance. GL code suggested by the rule sits below the confidence threshold. The flag fires, the bill pauses, the human gets the full context attached: original invoice, vendor history, PO reference, what failed validation, what the system would have done. The ROI framing follows from the exception design, not from the throughput rate. One CFO described the actual value driver this way: “it doesn’t feel like it would probably strip out a significant amount of costs, but it would help drive efficiency and productivity and enable us to scale.” The honest version of AP automation ROI for a mid-market company is not labor reduction; it is the ability to absorb the next twelve months of transaction volume without adding an AP clerk, while catching the duplicate payments and billing errors that would have hit the bank before anyone noticed. The financial framing for that decision is laid out in [how to know what’s worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating). ### What a working AP automation stack actually looks like A working stack has four layers that map cleanly onto the four decisions. Naming them helps because vendor pitches conflate them constantly. - **System of record.** The ERP or accounting platform that owns the bill: QuickBooks, NetSuite, Acumatica, Sage Intacct, or a higher-end equivalent. This is where the bill, the approval evidence, and the payment posting all need to land, in one place, reconstructible. - **Document parsing layer.** Whatever extracts invoice data from email and PDF into structured fields. Parseur is one option; the ERP-native capture modules are another. The choice matters less than getting the structured data into the ERP cleanly. - **Approval routing layer.** Where the rule lives that says this bill goes to that approver, with that threshold, in that channel, with that delegation rule. Slack and email are the channels; the rule itself sits above them. ERP-native approval modules handle the simple cases. Anything that branches on category, vendor risk, or cross-system context needs something on top of the module. - **Exception queue.** The pile of bills that did not clear automatically, with the context attached for the human who has to resolve them. The exception queue is where the AP team’s day actually happens once the rest is automated. If the queue is not designed deliberately, it becomes another inbox, with all the same problems an inbox has. What you see in the [Acumatica capability catalog](https://flowrunner.ai/workflows/automate-with-acumatica) is the same architecture applied to one ERP: the ERP holds the system of record, the parsing layer feeds it, and the rules above it handle the routing and the exception cases. That seam, between what the ERP can do natively and what the AP team actually does between the ERP, the inbox, the approver, and the bank, is where an orchestration layer lives. We call this category [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service): a platform category that coordinates, governs, and supervises multi-agent environments while keeping humans in control of the decisions that require judgment. It sits above the systems of record and listens for what they emit: a new bill arrived, a PO closed, a vendor was flagged. It gathers the context the ERP’s native rule did not have at fire time. It brings the right human in with the full bill framed and the validation result attached, in the [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) pattern that pauses, assembles context, routes to a person via the right channel, and resumes on response. It writes the answer back against the ERP record so the audit trail stays in one place. FlowRunner is built for that layer. It does not replace the ERP’s bill table or the parsing tool’s extraction; it makes their outputs reconstructible and their exceptions handlable without inbox archaeology. ### How a CFO should evaluate AP automation before committing The evaluation conversation should start from your bill volume and your exception rate, not from the vendor’s feature list. The vendor’s feature list always looks impressive; the question is whether the features map onto the four decisions you actually need automated, in the order you can absorb. Things to ask, in roughly this order: - **What is my current monthly bill volume, and what is my realistic projection over the next eighteen months?** If the answer is roughly flat, the ROI from automation comes mostly from exception detection. If the answer is growing meaningfully, the ROI comes from absorbing the growth without adding headcount. - **Where do my current exceptions go?** If the answer is “they get figured out informally,” that is what is at risk of falling through the cracks as volume rises. The exception queue is what you are buying. - **What does my approval routing look like today, honestly?** If thresholds live in a policy memo and approvers are picked by who answers Slack first, the workflow rule does not exist yet, regardless of what the tool can support. - **What is the relationship between my AP process and my ERP?** Bills that get captured outside the ERP and reconciled back later are fundamentally different from bills that live in the ERP from the moment they arrive. The first arrangement creates two sources of truth that need to be kept in sync; the second creates one. Aim for one. Red flags during a vendor evaluation, in declining order of severity: - Opaque AI with no audit trail. If you cannot reconstruct why a specific bill cleared or got flagged, the tool is a control gap, not a control. - No human-in-the-loop on payment release. Anything that releases payments unattended above a low threshold will eventually pay a duplicate or a fraudulent bill, and the cost of that one event is most of the value the tool would have delivered. - No exception queue, or one that lives outside the ERP. Exceptions handled in a separate app create exactly the piecemeal feeling that the project was supposed to fix. - No way for the finance team to own the routing logic without a consultant. If every rule change requires a vendor ticket, the workflow is rented, not owned. ### A practical sequencing plan for the first 90 days The pattern that actually works is layered, not parallel. Build trust at each layer before adding the next one. **Weeks 1-3. Capture only.** Get invoices flowing from email into structured data in the ERP. Keep coding and approval manual to establish a baseline of how the AP team actually moves bills today. The point of these three weeks is not to save labor; it is to make sure the structured data is right, and to give the AP team time to notice what the capture step gets wrong on the edges. **Weeks 4-7. Add coding suggestions with a confidence threshold.** Suggested GL coding shows up alongside the bill; the AP coder either accepts or corrects. Measure the acceptance rate per vendor; tune the rules where the acceptance rate is low. Do not bypass the human on coding yet, even for high-confidence vendors. The acceptance-rate data is what tells you when bypassing is safe. **Weeks 8-10. Add approval routing in Slack or email with named approvers.** The routing follows the rule set you wrote during weeks 1-3, with the delegation rule built in from the start. Keep payment release manual. The approval routing is the step most likely to surface organizational issues (approvers who do not actually want the responsibility, thresholds that no one believes in); surface them before you wire payment release to anything. **Weeks 11-12. Add duplicate detection and exception flagging.** Run the queue weekly with the AP team to tune the flag rules. Expect the first week’s queue to be noisy; expect the third week’s queue to be the actual signal. What not to automate in the first 90 days: payment release at any meaningful threshold, new vendor onboarding, anything that touches tax filings, anything tied to a vendor where the contract terms are still being negotiated. Each of these is a thing where the cost of a wrong automated action is high and the labor cost of keeping a human in the seat is low. The math works against automating them in the first quarter. The standard advice on AP automation is to pick a vendor and migrate the team. The honest version is the opposite: define the sequencing, define the exceptions you care about, design the queue, and the vendor is whatever survives those three decisions intact. The work is in the design, not in the demo. That work is the part that decides whether a year from now your AP looks like one connected system with a small, well-handled exception queue, or like a black box that everyone has stopped trusting and no one has the time to rebuild. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Human in the Loop Review: What a Good One Actually Looks Like Source: https://flowrunner.ai/blog/human-in-the-loop-review Articles May 28, 2026 Updated June 19, 2026 9 min read A human in the loop review is not an approval queue. It is a pause that hands one decision to one named person with the context already attached. ![A bald cartoon man with three hairs sticking up at a service window receiving one amber-flagged folder labeled review, while faded routine paper slips stream past on their own in the background.](https://flowrunner.ai/images/blog/human-in-the-loop-review-hero.webp) A human in the loop review is worth nothing if the person doing the reviewing cannot see, in the moment they are asked, the one thing that made the case worth asking about. That is the whole game, and most writing on this topic skips it to talk about the approve button. The approve button is the cheap part. The expensive part, the part that decides whether the review is real or theater, is the context that arrives with the request. This is the read for the CFO, VP of Finance, or controller who has been handed black-box automation before and wants the opposite: judgment kept exactly where it belongs, without putting senior finance time back on every transaction. It defines what a review step actually is, names the three pieces a good one needs, and walks the finance cases (AP, duplicate payments, invoice exceptions) where the call gets made every week. For the related question of whether a person should review before the action or monitor after it, the [in-the-loop versus on-the-loop distinction](https://flowrunner.ai/blog/human-in-the-loop-vs-human-on-the-loop) covers that dial. This article is about the review moment itself. ### What a human in the loop review actually is Strip the jargon and a [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) review step is three things in sequence. A workflow reaches a point it was told not to pass alone. It hands one specific decision to one named person, with the case assembled for them. It captures their answer and resumes. That is the pattern in compact form: pause autonomously, assemble the relevant context and the decision choices available, route to a human via their preferred channel, resume the moment the human responds. That is the definition. Note what is not in it. - It is not a generic approval queue that every transaction files into. A queue that holds everything is a queue nobody reads. - It is not a rubber-stamp signoff, where the reviewer’s job is to acknowledge rather than judge. - It is not a manual fallback for when the system breaks. The review is part of the design, not the failure case. A review step has a precise shape: a trigger condition that defines when to pause, the context delivered to the reviewer at that pause, and a captured decision tied back to the record. Miss any one of the three and you have something that looks like oversight and functions like ceremony. ![a three-stage flow diagram of a human in the loop review](https://flowrunner.ai/images/blog/human-in-the-loop-review-1.webp) This is the difference between automation that holds up beyond the demo and automation that only handled the [happy path](https://flowrunner.ai/blog/the-truth-about-building-automations). A review step is one of the things that makes the difference. It also sits in a specific place between two extremes. Full automation with no human is one end. An AI agent acting on its own judgment is the other. The review step is the deliberate seam where the system says it has reached the edge of what it should decide alone and asks. The routing logic that decides whether to pause is a different kind of decision from the action that runs afterward, which is the same line that separates [AI automation from AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents). Routing decides whether to ask. The action either runs or waits on the answer. ### Why finance leaders care about this specific pattern Finance work is full of decisions that resist full automation by a hair. A rule engine cannot make the call. An AI cannot make it with enough confidence to bet money on. A person can, in seconds, but only if a person is not asked to make every call. Across conversations with finance leaders, the recurring pattern is the same: senior finance time gets spent on processing that needs just enough judgment to prevent full automation, and that is the exact gap a review step is designed to fill. A CFO running AP at a growing company put the floor of that zone in one sentence: “if someone’s out, I might even enter a bill into the system.” The judgment to enter the bill is too thin to fully delegate to a tool, but the keystrokes are real work, and they land on the most expensive person in the room. That is precisely the gap a review step closes. The system does the data movement and the pre-checks. The person makes the small judgment the data movement cannot make on its own. The bill posts without the CFO doing the typing. The concern here is not “replace people.” It is “stop spending senior finance time on processing that still needs a sanity check.” Without a review step, finance teams get pushed to one of two bad outcomes. They approve everything blindly because reviewing it all is impossible, or they refuse to automate at all because they do not trust an unattended system with money. Both are failures. The review step is the design that lets the routine majority run while the cases that need a human get one. There is a third reason finance leaders care, and it shows up later than the others, at close and in diligence. A captured human decision, with the reviewer’s name and the reason attached to the record, is defensible. “The system did it” is not an answer when a controller or an auditor asks who approved a specific payment six months after the fact. A review step that captures identity produces an answer that is a name attached to a record. That is the same property that makes a [purchase order approval workflow](https://flowrunner.ai/blog/po-approval-workflow) survive an audit, applied to AP and reconciliation. ### The part most posts on this topic will not say Here is the claim most human-in-the-loop writing tiptoes around: the review is not the safety feature. The context attached to the review is the safety feature. The pause itself is trivial. Any tool can stop and wait for a click. What separates a real control from a liability is whether the person who clicks can actually see the case. A review with no context is a rubber stamp with a timestamp. It is worse than no review at all, because it manufactures audit evidence that does not reflect any actual judgment. The same CFO above named the failure mode from the other side of the desk: “a lot of the sales guys, it’s just paperwork. They just want to get it off their desk.” A loop that always asks but never gets a real answer is theater. It produces a log entry that says “approved by Jane” when Jane never had a chance to judge anything. So the design question is not “where do we add an approval step.” It is “what does the reviewer need to see to make this a real decision in twenty seconds, and how does it get in front of them without making them go hunt for it.” Everything good about a review step follows from answering that question well. Everything bad about approval queues follows from never asking it. ### What a well-designed review pause looks like Four properties separate a review step that adds value from one that adds latency. **The trigger is specific, not vague.** “Pause for review when something looks off” is not a trigger; it is a wish. A real trigger is a condition the system can evaluate: the bill exceeds a dollar threshold, the vendor name does not match prior history, a duplicate invoice number appears within thirty days, the extraction confidence falls below a bar. The narrower the condition, the more the pause means when it fires. **The context travels with the request.** This is the property that decides everything. The reviewer should not receive a notification that says “a bill needs approval” and then go open the ERP, find the vendor, pull the payment history, and locate the original document. All of that should arrive attached: the specific anomaly that was flagged, the vendor’s prior payments, the original invoice, and the exact action that will run on approval. The duplicate payment case makes this concrete. The same CFO said it plainly: “I feel that it would be very easy to double pay on something like that if both we and the distributor made a payment.” A review step that catches that has to show the reviewer the suspected prior payment next to the new one. The pause without the comparison is useless. **The reviewer is reached where they already work.** A separate approval portal that nobody logs into is a backlog with a login screen. The review request should land in the channel the person already lives in, Slack or email, with the decision available right there. The point of meeting people where they are is not convenience. It is that a review which requires a context switch gets deferred, and a deferred review is the backlog you were trying to avoid. **The decision is captured back to the record.** The answer and the reason cannot live in a chat thread that scrolls away. They have to be written back to the system of record, attached to the bill, the invoice, or the transaction, so that the next person who looks at that record sees who decided and why. A decision that is not captured did not happen, as far as the audit trail is concerned. ![a Slack human-in-the-loop review request for a flagged AP bill](https://flowrunner.ai/images/blog/human-in-the-loop-review-2.webp) ### Worked examples in finance workflows The pattern is easier to see in the specific. Each of these is a defined trigger, context delivered at the pause, and a captured decision. - **Near-duplicate AP detection.** An invoice arrives that resembles one already paid. The flow flags it, surfaces the candidate match beside the new bill to a reviewer, and posts only after explicit confirmation. The [human-in-the-loop step on near-duplicate detection in Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) is this pattern inside an ERP. - **Bills above a threshold.** A bill crosses the amount that warrants a person. The flow [pauses the bill for a named approver in Slack](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) before payment runs, with the bill context attached so the approver has something to approve against rather than a number in a vacuum. - **Parsing exceptions on extracted invoices.** Document extraction is reliable until it is not. When a parse fails or an amount looks anomalous, the flow [routes the exception to a human reviewer with the original document attached](https://flowrunner.ai/workflows/connect-parseur-with-slack) and captures the reviewer’s identity with the decision. - **Email-to-payment that escalates only on a mismatch.** An invoice comes in by email, gets parsed, and runs against the purchase order or prior invoices. When the totals line up it flows through. When they do not, the [exception surfaces with the vendor history attached](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) and waits for a person. In every one of these, the reviewer’s experience is the same: a specific case, the reason it stopped, the evidence to judge it, and a decision that gets recorded. That consistency is the product. The captured reasoning across these flows is also what builds the [audit trail that reviewers reach exceptions with full ERP context](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and that holds up at close. ### How this differs from the approval workflow you already have Most finance teams already have approval routing. So the fair question is what a review step adds that a traditional approval chain does not. Three things. | | Traditional approval routing | Human in the loop review | | --- | --- | --- | | What triggers it | Every transaction of a type | Only the conditions the system cannot safely decide alone | | What the reviewer sees | The request, often without context | The flagged case, the anomaly, the history, the source document | | Where the answer lives | Email or chat, loosely tied to the record | Captured back to the system of record with identity | | What it optimizes | Coverage (everything gets a signoff) | Judgment (the few cases that need it get a real one) | The deeper difference is what you are measuring. A traditional approval workflow is measured by coverage: did everything get signed off. A review step is measured by exception resolution: how fast did the cases that genuinely needed a human get resolved, and did the routine majority flow without a human touching it. Counting approval-button clicks tells you how much ceremony you generated. Counting exception-resolution time tells you whether the control is working. Those are different numbers, and the second one is the one that matters. ![a finance operations view contrasting two metrics for the same AP process](https://flowrunner.ai/images/blog/human-in-the-loop-review-3.webp) A review step does not replace your ERP’s approval module. The cases that belong inside the ERP can stay there. It adds the layer for the cases the ERP’s native rules were never built to catch: the [automation exception](https://flowrunner.ai/concepts/automation-exceptions) that arrives from outside the system, the anomaly that needs context from a different system to evaluate, the decision that needs a person pulled in mid-flow without leaving the channel they work in. ### Where the orchestration layer comes in Look closely at the property that decides whether a review is real: the context has to be assembled at the moment of the pause, from wherever it lives. The flagged bill is in the AP system. The vendor’s payment history is in the ERP. The original document came in by email or a parser. The suspected duplicate is a different record entirely. The reviewer needs all of it in one place, in Slack, in the few seconds they have. No single system of record owns that assembly. QuickBooks knows its own bills. Acumatica’s approval module knows the bills inside Acumatica. The parser knows the document. None of them reaches across the others to gather the case and put it in front of a person. That gather, between the systems, at the moment a decision is needed, is a job that lives above all of them. That is the work an orchestration layer exists to do. This is the category we are defining: [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service), a platform category that coordinates, governs, and supervises multi-agent environments while keeping humans in control of the decisions that require judgment. An orchestration layer is a system that sits above the systems of record, listens for what they emit, assembles the context that no single system had on its own, pulls a named person in when the case earns a pause, runs unattended when it does not, and writes the decision back so the trail stays in one place. The review step is the human moment inside that layer. FlowRunner is built for that layer. The pause-and-ask is not a feature bolted onto an automation tool. It is the point where the orchestration layer hands a complete case to a human and waits for the answer, because that is the moment the system was designed to recognize and respect. ### How to decide where a review step belongs You do not add review steps everywhere. You add them where the cost of a wrong unattended action is higher than the cost of a person’s twenty seconds. A short way to find those places: - **Start with the exceptions you already catch by hand.** Duplicate payments, billback errors, bills that fall through the cracks, vendor mismatches. Those are the cases where you already know a human adds value, because you are already the human. Those become your first triggers. - **Name the condition that separates routine from judgment.** For each process, what is the line? The dollar threshold, the new-vendor flag, the confidence score from the extraction step, the duplicate-window rule. If you cannot name the condition, you cannot build the trigger, and you will end up reviewing everything. - **Choose the reviewer by role, not by name.** Route to “the AP manager,” not to a specific person, so the workflow survives turnover and vacations. The judgment lives in the role. - **Decide what happens when no one answers.** A review step that stalls forever is a new way for things to fall through the cracks. Set the rule up front: does the case wait, escalate to a second reviewer after an interval, or hold in a defined state. The default behavior on silence is part of the design, not an afterthought. Then revisit it. The right set of triggers moves as the log accumulates. A condition you reviewed nervously for three months because you were not sure the system was catching the right things can relax once the record shows it is. A condition you let run unattended can earn a review step if the after-the-fact log keeps surfacing expensive mistakes. The triggers are not set once. The finance leaders who get the most from a review step are not the ones who add the most approvals. They are the ones who add the fewest, each one earning its pause, each one arriving with the case already assembled, so the person being asked is making a real decision in the few seconds they have rather than acknowledging a request they had no way to judge. Design the review backward from those seconds. The reviewer’s attention is the scarcest thing in the workflow. Spend it only on the cases that deserve it, and when you spend it, give the reviewer everything they need to make it count. ### Quick answers #### What makes a human in the loop review good instead of a rubber stamp? The context arrives with the request. A good review surfaces the specific anomaly, the prior history, and the original document at the moment of the decision, so the reviewer is judging the case rather than acknowledging it. A review with no context is a rubber stamp with a timestamp. #### What context should a reviewer see at the pause? The specific reason the case was flagged, the source document, the relevant history the system already has, and the action that will run on approval. The reviewer should be able to decide without leaving the message to go hunt in three other systems. #### Should every transaction get a human in the loop review? No. Reviewing everything is how you train people to stop reading. A good design pauses only on the conditions the system cannot safely decide alone and lets the routine majority run, with the full record written to the log for after-the-fact review. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Human in the Loop vs Human on the Loop: A Finance Leader's Guide Source: https://flowrunner.ai/blog/human-in-the-loop-vs-human-on-the-loop Articles May 27, 2026 Updated June 19, 2026 8 min read Human-in-the-loop and human-on-the-loop are not rival philosophies. They are settings on the same dial. Finance leaders set them per workflow. ![A bald cartoon man with three hairs sticking up adjusting an amber pointer on a large workshop dial labeled in the loop on one end and on the loop on the other, with four small framed workflow labels (AP, refunds, reconciliation, close) on the wall behind him.](https://flowrunner.ai/images/blog/human-in-the-loop-vs-human-on-the-loop-hero.webp) Treating human-in-the-loop and human-on-the-loop as rival philosophies is the single most common reason a finance team’s AI oversight policy looks coherent on a slide and falls apart the first time a controller asks where a specific bill got paid. They are not rival philosophies. They are two settings on the same dial, and the choice is per workflow, not per company. This is the read for the CFO, VP of Finance, or controller deciding how much oversight to wire into each piece of finance automation. It defines both models, names the three variables that actually decide where each workflow sits between them, and walks the AP, refund, and reconciliation cases where finance leaders already make this call every day, whether they call it that or not. ### Two oversight models, one spectrum Strip the AI safety language away and you are left with two operating modes that finance has actually run for decades, just with different tools. **[Human in the loop](https://flowrunner.ai/concepts/human-in-the-loop).** The system pauses and waits for a person to approve, edit, or reject before it continues. The action does not happen until the human says yes. The approval gate on a bill above a threshold is human-in-the-loop. The credit memo a billing manager has to sign off on before it issues is human-in-the-loop. The pause-and-ask step in an [AP near-duplicate detection flow](https://flowrunner.ai/workflows/automate-with-acumatica) where a flagged potential duplicate is surfaced to a reviewer before it posts is human-in-the-loop. **Human on the loop.** The system acts on its own while a person monitors and can intervene after the fact. The action happens. The human reviews the log, the exceptions report, or the dashboard on a cadence and steps in when something looks wrong. The auto-categorization rule that posts $40 transactions to a GL code without asking is human-on-the-loop, even if nobody has ever called it that. So is the [routine fulfillment exception that resolves itself while only high-value cases escalate](https://flowrunner.ai/workflows/orchestrate-shipbob-and-quickbooks-online-and-slack). The standard framing on this topic positions these as opposing camps in an AI governance debate, with one camp prioritizing safety and the other prioritizing throughput. That framing is a category error. Here is what most posts on human-in-the-loop vs human-on-the-loop will not say plainly: the actual variable is not how comfortable you are with AI autonomy. It is reversibility. An unreviewed wire to a new vendor is hard to claw back. An unreviewed Slack ping is not. The same finance leader who insists on a named approver for the wire will happily let the ping go out automatically. Both decisions are correct, and they are correct for the same underlying reason. The work is in naming the reason and applying it consistently. | | Human in the loop | Human on the loop | | --- | --- | --- | | When the human acts | Before the system acts | After the system acts | | Default state if no human responds | Workflow waits | Workflow proceeds | | Best fit | High-dollar, irreversible, or audit-exposed | High-volume, reversible, low-confidence-tolerant | | Failure mode if misapplied | Bottleneck; rubber-stamping | Material loss before review catches it | | Finance examples | Wire approvals, refunds above threshold, new-vendor onboarding | Routine categorization, low-dollar reconciliation breaks, expected-variance handling | ### Why this matters for finance leaders specifically Finance work is full of just-enough-judgment tasks that block full automation. A pure rule engine cannot make the call. A pure AI cannot make the call confidently. A human can, but if a human has to make every call the team never gets to the work they were hired for. A CFO running AP at a growing company described the floor of the in-the-loop zone in one sentence: “if someone’s out, I might even enter a bill into the system.” The work has not been delegated to a tool. The judgment to enter is too thin to delegate, but the entering is real. That is exactly the gap human-in-the-loop oversight is designed to close: the system handles the data movement and pre-checks, a person makes the small decision the data movement cannot make on its own, and the bill gets entered without the CFO doing keystrokes. The same finance leader described why the in-the-loop side of the dial is not optional for AP near-duplicates: “I feel that it would be very easy to double pay on something like that if both we and the distributor made a payment.” That is not an AI safety concern. That is a finance reality. The pause is justified by the cost of reversal, not by ideology. The cost of getting oversight wrong is asymmetric in both directions. An unreviewed bill payment is harder to claw back than an unreviewed Slack notification. A bottleneck that pauses every routine exception for review is harder to recover from than a missed low-dollar variance. Audit trail and reversibility constraints push different finance workflows toward different points on the spectrum, and the right answer for AP is not the right answer for reconciliation. There is also a failure mode on the in-the-loop side that finance leaders see and dev teams routinely miss. Approval queues where reviewers rubber-stamp without actually validating are a real risk in finance. The same CFO above put it directly: “a lot of the sales guys, it’s just paperwork. They just want to get it off their desk.” A loop that always asks but never gets a real answer is worse than no loop at all, because it manufactures audit evidence that does not reflect actual review. In-the-loop oversight has to be designed for the cases where review genuinely adds value, not draped over every action as ceremony. The framework that makes all of this manageable is the same one that separates [AI automation from AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents): the routing decision is not the same kind of decision as the action itself. Routing decides whether and how to pause. The action either runs or waits. Treating both as one undifferentiated “AI doing finance work” is how the policy debate goes off the rails. ### When human-in-the-loop is the right call The pattern that earns an in-the-loop pause: - **High-dollar or irreversible actions.** Bills above a dollar threshold, wires, payments to new vendors, refunds above a customer-impact level. The FlowRunner pattern for this in QuickBooks is to [pause large bills for a named approver in Slack](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) before payment runs, with the bill context attached so the approver actually has something to approve against. - **Customer-facing financial decisions.** Refund and credit decisions where the customer relationship and revenue impact justify a routed review. The companion pattern for Stripe is [refund approvals routed to billing managers](https://flowrunner.ai/workflows/connect-slack-with-stripe) above a threshold before they execute, with the original charge and customer history surfaced inline. - **Low-confidence document extraction.** AP near-duplicate detection, vendor invoice parsing where the model surfaces a likely match but the cost of a wrong post is high. The pattern is to flag, surface to a reviewer with the candidate match alongside the underlying bill, and post only after explicit confirmation. - **Audit-exposed events that the team will need to reconstruct later.** Anything that an external auditor, a board member, or a future controller will need to trace back to a named approver. If the question “who approved this” will be asked, the answer needs to be a name attached to the record, not a system action. - **Anything genuinely new.** A new vendor, a new payment category, a new bill type the workflow has not seen before. The cost of an in-the-loop pause on an unfamiliar case is low. The cost of an on-the-loop default on an unfamiliar case is whatever the worst plausible outcome of that case is. ### When human-on-the-loop is the right call The pattern that earns autonomous resolution with after-the-fact review: - **High-volume routine [automation exceptions](https://flowrunner.ai/concepts/automation-exceptions) where pausing every case re-creates the bottleneck the automation was supposed to fix.** Reconciliation breaks under a dollar threshold, expected category mismatches, known-pattern variances. The ShipBob fulfillment workflow above is one example; the [reconciliation flow that runs Stripe payments against QuickBooks invoices with selective human review](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) is another. - **Fully reversible actions with complete logging.** If the action can be undone and the system writes a clean trail of what it did, a reviewer the next morning can catch and unwind a bad decision without material loss. The on-the-loop bar is reversibility plus reviewability, not absence of risk. - **Escalation-by-threshold patterns.** The honest hybrid: the workflow handles the routine ninety percent on its own and routes the outlier ten percent to a human. This is what most well-designed finance automation actually looks like in production, even when the team would describe it as “fully automated.” - **Cases where the realistic baseline is informal approval, not careful review.** If the alternative to on-the-loop is a queue where exceptions sit until someone has time, and that someone rubber-stamps because they are behind, on-the-loop with a real log review on a cadence is the more honest control. The control nobody enforces is worse than the control everyone reviews on Friday. That last point is the honest comparison most articles on this topic skip. The baseline for finance teams today is rarely careful human review of every exception. It is a backlog, an informal approval that bypasses the queue, and a controller checking the bigger items by hand. Measured against that baseline, autonomous resolution with a clean log and a scheduled review is the more defensible control, not the less. Reviewing what the system actually did beats not reviewing what people manually said they did. ### Configuring the spectrum across a finance stack The single most useful mental shift here: this is not a stance you take as a CFO. It is a configuration you set per workflow. AP can be in-the-loop above a threshold and on-the-loop below it. Refunds can be in-the-loop in one product line and on-the-loop in another. Reconciliation can sit between the two with selective review for breaks that exceed a dollar or confidence threshold. The practical inputs to the per-workflow decision: - **Dollar threshold.** Above what amount is the case worth a pause? Set this against actual spend distribution, not a template. If most of your AP is under $2,000 and the threshold for in-the-loop review is $25,000, the workflow treats almost everything as routine and you have effectively delegated the AP control to the rule engine. - **Reversibility.** Is this action easy to undo, hard to undo, or impossible to undo? Payments to vendors are hard. GL entries are easy. Wires are impossible without negotiation. The harder the reversal, the further toward in-the-loop the workflow sits. - **Audit exposure.** Will an auditor or a controller ask who approved this specific record six months from now? If yes, in-the-loop with named-approver capture is the safer place on the dial. If the action is one of ten thousand routine reconciliations nobody will ever ask about individually, the audit value of pausing each one is near zero. - **Confidence score from the upstream step.** Document extraction, near-duplicate detection, and category classification produce a confidence score. A workflow can route high-confidence cases on-the-loop and low-confidence cases in-the-loop on the same flow, using the score as the gate. - **Frequency of true exception.** How frequently does the case actually need human judgment versus how frequently does the human just say yes? If true exceptions are rare, pausing every case is mostly ceremony. If they are common, pausing every case is the work. What both models require to be defensible is the same regardless of where on the dial the workflow sits: - A complete execution log that captures what the system did, when, and against what data - Clear ownership of who reviews what, on what cadence - Approver identity captured against the record for in-the-loop steps - The ability to change the configuration of any one workflow without rebuilding the others Native ERP and accounting platforms expose pieces of this. QuickBooks rules handle the routine categorization on-the-loop end. Acumatica’s approval module handles the named-approver in-the-loop end on bills inside the ERP. Neither one owns the seam where exception data comes in from outside, the seam where the routing decision uses context from a different system, or the seam where a human in Slack gets pulled in mid-flow without leaving the channel they already work in. That seam, between what the systems of record route on their own and what a finance team actually does between them, is where an orchestration layer lives. This is the category we are defining: [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service), a platform category that coordinates, governs, and supervises multi-agent environments while keeping humans in control of the decisions that require judgment. An orchestration layer is a system above the systems of record that listens for what they emit, gathers context the native rules did not have at fire time, pulls a named approver in when the case earns a pause, runs autonomously when the case does not, and captures everything against the record so the audit trail stays in one place. FlowRunner is built for that layer. The pause-and-ask step is the in-the-loop instance. The autonomous run with notifications and a clean log is the on-the-loop instance. The configuration is per workflow, not per platform. ### A short decision guide When you sit down to set the dial for a specific workflow, the question to start with is not “do we want AI to do this.” The question is: if this action is wrong, how hard is it to reverse, and who notices? - **If reversal is hard or the action is high-dollar:** route to a named approver in the channel they already use, with the case context attached so the approval is real and not paperwork. This is the in-the-loop end. - **If reversal is easy and volume is high:** run it autonomously, write a clean log, and review the log on a cadence the team will actually keep. This is the on-the-loop end. - **If volume is high and a subset is irreversible:** use the threshold and the confidence score to split the flow. Most of the population runs on-the-loop. The minority that exceeds the threshold pauses in-the-loop. The split is the point. - **Revisit the configuration as confidence and history accumulate.** The right point on the dial moves over time. A workflow that needed in-the-loop oversight for the first three months because the team was uncertain about the routing rules can move toward on-the-loop once the log shows the rules are catching what they need to catch. A workflow that started on-the-loop can move toward in-the-loop if review consistently finds expensive mistakes. The dial is not set once. The CFOs and VPs of Finance who get the most value from AI oversight are not the ones who pick a side. They are the ones who treat the dial as part of the workflow definition, set it deliberately per case, and revisit it on the same cadence they revisit the rules themselves. The work is not in choosing between human in the loop and human on the loop. The work is in knowing which workflows belong where on the dial today, and being honest about why. ### Quick answers #### What is the difference between human in the loop and human on the loop? Human in the loop means the system pauses and waits for a person to approve, edit, or reject before it acts. Human on the loop means the system acts on its own while a person monitors and can intervene after the fact. The first is a gate. The second is a guardrail. #### Which one should finance use for accounts payable? Neither, as a single answer. AP has both. Bills above a dollar threshold or from a new vendor route in the loop to a named approver. Routine matched bills under threshold can run on the loop with the full posting written to the audit log for review. The split is per bill, not per company. #### Is human on the loop just AI replacing humans? No. The person is still accountable for the outcomes and reviews the log on a cadence. The difference is whether the review happens before the action or after it. Reversibility is what makes one or the other defensible. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Marketing Automation Workflow Examples That Actually Hold Up in Production Source: https://flowrunner.ai/blog/marketing-automation-workflow-examples Articles May 28, 2026 Updated August 11, 2026 9 min read Most marketing automation example lists stop at the email nurture. Seven cross-tool workflows marketing ops owns, and where the human checkpoint belongs. ![Cartoon traffic cop at a four-way intersection waving most arriving leads down three automated sage-green lanes while personally directing one lead down an amber-barriered lane by hand.](https://flowrunner.ai/images/blog/marketing-automation-workflow-examples-hero.webp) Most marketing automation example roundups grade the wrong thing. They list a welcome series, an abandoned-cart sequence, a re-engagement campaign, and they measure each one by the lift it produced. That makes for a tidy article and a useless template, because every one of those examples lives inside a single email tool and never has to hand a lead to another system. The workflows that actually consume a marketing ops lead’s week are the cross-tool ones, where a form submission, a score change, or a booked meeting kicks off a chain of routing, enrichment, and handoff that spans HubSpot, Salesforce, Slack, and the rest of the stack. Those are the examples worth studying, and a useful one names three things: the trigger, the steps, and the human checkpoint. That last item is the one almost every list omits. Here is what most marketing automation guides will not say: an example without a checkpoint is not a finished workflow, it is a demo. Real lead flow has ambiguous ownership, failed enrichment lookups, duplicate records, and target accounts that should never be auto-routed. A workflow that pretends those cases do not exist breaks in week two. So each example below names where a human belongs, because that is the part that determines whether the thing survives contact with production. A note on scope before the examples. This article describes patterns observable in the public documentation for HubSpot, Salesforce, Calendly, and similar platforms. It is not a report of FlowRunner customer outcomes, and there are no measured results claimed here. The point is the shape of the workflow, not a number. ### What counts as a marketing automation workflow (and what most lists miss) There are two things people call a marketing automation workflow, and conflating them is why the example lists feel thin. The first is a single-tool campaign sequence. A welcome series inside your email platform. A drip that fires on a schedule. These run entirely within one tool’s boundary, and that tool automates them well. They are real, but they are not the operational work. The second is a cross-tool workflow. A form submission lands in one system, gets enriched from a second, scored against signals from a third, routed by a rule in the CRM, and announced in Slack. This is the layer marketing ops actually owns: routing, enrichment, handoff, attribution, and exception handling. HubSpot’s own workflow documentation frames automation as triggers, conditions, and actions chained together ([HubSpot: create workflows](https://knowledge.hubspot.com/workflows/create-workflows)), and the moment those actions need to touch a system outside the tool that fired the trigger, you have left the campaign layer and entered operations. Every example that follows is the second kind. Each one names the trigger, the steps, and the checkpoint. If you only remember one thing from this article, remember that a workflow example without a named checkpoint is incomplete. ![a horizontal flow diagram with four labeled stages reading SIGNAL, CRM, SCORE/ROUTE, NOTIFY, with a small amber diamond between SCORE/ROUTE and NOTIFY labeled "human checkpoint"](https://flowrunner.ai/images/blog/marketing-automation-workflow-examples-1.webp) ### Example 1: Inbound form to qualified lead in the CRM **Trigger:** a form submission, whether from a HubSpot form, a Google Form, or an embedded landing-page form. **Steps:** enrich the contact from a data provider, check for an existing duplicate against the CRM, assign a lead score, and set an owner by territory or product line. **Human checkpoint:** low-confidence enrichment and fuzzy duplicate matches do not auto-merge. They route to a marketing ops queue where a person decides, because a wrong merge corrupts the record permanently and a wrong owner assignment sends the lead to the wrong rep. This is the most common marketing automation workflow on earth, and it is the one most often shipped without the checkpoint. The enrichment lookup fails silently, the duplicate logic guesses wrong, and three weeks later someone is untangling two contact records that should have been one. The reference build is [qualifying inbound leads from a form into HubSpot](https://flowrunner.ai/workflows/orchestrate-google-forms-and-hubspot-and-slack), which shows the enrichment-and-route chain with the Slack notification carrying the context the rep needs. ### Example 2: New CRM lead to sales rep in Slack **Trigger:** a new contact created in HubSpot or Salesforce that meets MQL criteria. **Steps:** route to the correct rep’s Slack channel based on territory or account ownership, attach context (source, score, recent activity), and check for an existing open opportunity so you do not ping a rep about an account they already work. **Honest baseline:** the realistic before-state is not chaos. It is a lead-alert email or a CRM mobile notification that the rep has trained themselves to ignore. A Slack message with the actual context, posted to the channel where the rep already lives, closes the gap that the ignored email leaves open. Routing rules in major CRMs assign by territory, product line, or round-robin ([Salesforce: lead assignment rules](https://help.salesforce.com/s/articleView?id=sf.customize_leadrules.htm&type=5)). The rule decides who. The workflow decides that the who actually finds out in time to act. The reference build is [routing new HubSpot leads to Slack with duplicate checks](https://flowrunner.ai/workflows/connect-hubspot-with-slack). ![a Slack message preview posted to a rep channel reading "New MQL: contact name, source Webinar Q2, score 87, last activity pricing page" with a small "Claim" button, not a generic "new lead" alert](https://flowrunner.ai/images/blog/marketing-automation-workflow-examples-2.webp) ### Example 3: Meeting booking to pre-call prep **Trigger:** a Calendly booking confirmed. **Steps:** pull the contact and account record from the CRM, summarize recent touches and product interest, and post the brief to the rep’s Slack DM ahead of the call. **Why it belongs in a marketing automation list:** this is the seam where marketing-sourced bookings meet the sales conversation, and it is exactly where attribution arguments start. If marketing booked the meeting but the source never reaches the rep, marketing loses the credit in the room where credit is assigned. Calendly bookings can trigger downstream automation through documented webhooks ([Calendly: event type integrations](https://developer.calendly.com/api-docs/EventTypeIntegrations)). The reference build is [automated meeting prep from Calendly bookings](https://flowrunner.ai/workflows/orchestrate-calendly-and-hubspot-and-slack). The checkpoint here is light: if the booked contact cannot be matched to a CRM record at all, flag it rather than posting a brief with empty fields. ### Example 4: Lead scoring threshold to sales handoff with attribution intact **Trigger:** a lead score crosses the MQL threshold in HubSpot or Salesforce. **Steps:** stamp the campaign and source onto the opportunity, notify the assigned rep in Slack, and create a follow-up task with an SLA so the handoff has a clock on it. **The failure this prevents:** attribution data lives in marketing tools and dies at the lead record. When the opportunity gets created, the campaign and source fields do not come along, and revenue can no longer be traced back to the campaign that produced it. Salesforce documents campaign influence as the mechanism for connecting campaigns to opportunity revenue ([Salesforce: campaign influence](https://help.salesforce.com/s/articleView?id=sf.campaigns_influence_overview.htm&type=5)), and the whole model collapses if the IDs never propagate across the lead-to-opportunity boundary. This is the example that separates a marketing ops lead who gets budget from one who fights for it every quarter. The reference build is [lead scoring and campaign attribution in Salesforce](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack), which carries the source field through the handoff instead of dropping it. ![a flow showing a Lead record with fields "campaign: Spring Webinar / source: paid social" arrowing into a new Opportunity record with the same two fields preserved, a green check on the carried-over fields, contrasted with a faded path where the fields arrive blank](https://flowrunner.ai/images/blog/marketing-automation-workflow-examples-3.webp) ### Example 5: Event or webinar registration to multi-channel follow-up **Trigger:** a registration captured in the event tool. **Steps:** sync the registrant to the CRM with the event as the source, add them to a post-event nurture, and notify the account owner if the registrant belongs to a target account. **Human checkpoint:** target-account registrants do not get dropped into an automated nurture and forgotten. They generate a personal-outreach task for the owner. The whole point of a target-account motion is that a human touches it; automating that touch away defeats the strategy you spent a quarter building. The pattern is the same as the others: machine handles the volume, human takes the cases where a human is the strategy. ### Example 6: Customer milestone to lifecycle marketing trigger **Trigger:** a product-usage milestone or an approaching renewal date, surfaced from the product database or the CS tool. **Steps:** update the lifecycle stage in the CRM, suppress prospecting campaigns so you stop marketing to an existing customer, and enroll the account in an expansion or renewal sequence. **Why it belongs here:** lifecycle marketing only works when the trigger comes from a real signal rather than a calendar guess. “Send the renewal email 60 days out” is a guess. “Send it when usage drops below the threshold that predicts churn” is a signal. The difference is whether the trigger lives in a system that actually sees the customer’s behavior, which is rarely the email tool. ### Example 7: Content distribution off a publish event **Trigger:** new content published, via an RSS feed or a CMS webhook. **Steps:** post to X and other social channels, notify the content channel in Slack, and log the publish in the content calendar. Posting to X from an automation uses the documented Create Post action and supports text-only or image variants ([X: manage Tweets](https://developer.x.com/en/docs/x-api/tweets/manage-tweets/introduction)). Keep the scope honest here: this is a distribution helper, not a demand-gen workflow on its own. It saves the manual copy-paste across channels. It does not generate pipeline, and an example list that dresses it up as a growth engine is selling you something. ### What separates the workflows that hold up from the ones that break Look back across the seven examples and the same four structural properties decide which ones survive production: - **Duplicate handling at the CRM write step, not after the fact.** Catching a duplicate before it lands is a workflow rule. Catching it afterward is a cleanup project. - **A human checkpoint for the cases that need judgment.** Low-confidence routing, failed enrichment, and target-account exceptions go to a named owner with full context. This is the digital andon cord: the workflow stops the line when it hits a case it should not decide alone, rather than plowing ahead and creating a mess someone finds later. - **Stable IDs carried through every step.** Campaign, source, and owner have to survive every handoff, especially the lead-to-opportunity boundary, or attribution dies quietly. - **An audit trail per run.** When someone asks “what happened to that lead,” you want an answer, not a forensics project across five tools. Notice what those four properties have in common: not one of them lives inside any single marketing tool. HubSpot governs what happens in HubSpot. Salesforce governs what happens in Salesforce. The duplicate check that has to read both, the checkpoint that pulls a human in between them, the IDs that have to cross the boundary intact, the trail that has to span the whole run: that work lives in the space between the systems, and no native feature inside any one of them will ever own it. That space is a category of its own. An orchestration layer is the thing that sits above the marketing stack, listens for what each tool emits, combines the signals, and pulls a named human in at the moments that need judgment. FlowRunner is built for that layer. It triggers on real events from the tools you already run, keeps the human in the loop on the calls that require one, and leaves an audit trail per run so the routing and attribution decisions are reviewable later instead of assumed. If you want the operational version of this argument rather than the example library, the [marketing workflows solution page](https://flowrunner.ai/solutions/marketing-workflows) walks through where the stack breaks and how the layer above it holds. And if you are weighing tools to run these patterns, the honest comparisons against [Make.com](https://flowrunner.ai/blog/flowrunner-vs-make-for-marketing-workflow-automation) and [Zapier](https://flowrunner.ai/blog/flowrunner-vs-zapier-marketing-workflow-automation) for marketing workflow automation lay out where each fits. The examples are not the hard part. Any of these seven can be sketched on a whiteboard in five minutes. The hard part is the checkpoint, the IDs, and the trail, because those are the parts that decide whether the workflow you ship is still running, and still trusted, a quarter from now. ### Quick answers **What is a marketing automation workflow?** A sequence that triggers on an event, runs a series of steps, and ends in an action or notification. The ones worth studying span more than one tool: a form or signal source, a CRM, a scoring or routing layer, and a channel like Slack. **What is the difference between a marketing automation workflow and an email nurture?** An email nurture lives inside one email tool and sends messages on a schedule. A cross-tool workflow coordinates routing, enrichment, scoring, and handoff across systems. The nurture is one step inside the larger workflow, not the whole thing. **Where should a human review step go in a marketing automation workflow?** At the low-confidence moments: ambiguous lead ownership, a failed enrichment lookup, a duplicate match, or a target-account handoff. Automate the clear cases and route only the judgment calls to a named owner with full context. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## PO Approval Workflow: What It Is and Where Most of Them Break Source: https://flowrunner.ai/blog/po-approval-workflow Articles May 26, 2026 Updated May 28, 2026 8 min read A PO approval workflow is the routing, thresholds, and audit trail behind every purchase. The ones that hold up are designed around exceptions. ![A bald cartoon man with three hairs sticking up at his desk, holding a purchase order and pointing at one circled amber line item, with a stack of approved purchase orders sitting untouched beside him.](https://flowrunner.ai/images/blog/po-approval-workflow-hero.webp) A PO approval workflow that only handles the happy path is the single most common reason a procurement team’s approval process looks clean in a diagram and falls apart in a quarterly review. The diagram shows the rule. The audit asks where the approver identity lives, what happened when the approver was out, and which POs got split to stay under a threshold. Those questions decide whether the workflow holds. This is an explainer for procurement operators evaluating whether the current process is worth formalizing or automating. It defines the workflow, walks the standard steps, names where the seams open, and gets to the design questions that actually matter. ### What a PO approval workflow actually is A PO approval workflow is the routing rules, approval thresholds, and system-of-record updates that move a purchase request from submitted to issued PO. It records who approved what, against the PO record, in a form an auditor can reconstruct later. It is not the full procure-to-pay process. P2P starts with sourcing and ends with vendor payment and posting. The approval workflow is one segment of that arc, sitting between the requester and the vendor. Receiving, three-way match, and AP payment live downstream and have their own controls. The typical actors: - **Requester.** The person who needs the goods or service and submits the request with line items, vendor, GL coding, and supporting documentation. - **Budget owner.** The cost-center or department head who owns the spending authority for the GL code being charged. - **Department head.** Often the same as the budget owner at smaller companies; separate at larger ones with delegated approval matrices. - **Finance.** Validates GL coding, budget availability, and policy compliance before or alongside the approval routing. - **Procurement.** Owns the vendor relationship, the approved vendor list, and the contract reference if one applies. In smaller organizations these roles collapse into one or two people. The workflow still needs to name them, because the audit trail asks who did what, not who could have done what. ### The standard steps in a PO approval workflow 1. **Request submission.** The requester enters line items, the vendor, GL coding, and any supporting documentation (quotes, statements of work, contract references). At this stage the workflow either accepts the request or rejects it back to the requester with a reason. 2. **Validation checks.** The system or a finance reviewer confirms budget is available against the chosen GL code, the vendor is on the approved list (or routes through new-vendor onboarding if not), and any required contract reference exists. 3. **Routing.** The request is routed to the right approvers. Routing dimensions commonly exposed in ERP approval modules include amount threshold, cost center, category (IT, marketing, professional services), and vendor risk tier. Microsoft’s [purchase approval workflow documentation](https://learn.microsoft.com/en-us/dynamics365/business-central/walkthrough-setting-up-and-using-a-purchase-approval-workflow) shows how one ERP models these dimensions; the patterns are recognizable across NetSuite, Acumatica, SAP, and others. 4. **Approval capture.** Each approver in the chain reviews and approves or rejects. The system records approver identity against the PO record, with a timestamp and any comments attached. NetSuite’s [PO Approval Workflow SuiteApp states](https://docs.oracle.com/en/cloud/saas/netsuite/ns-online-help/section_4158616141.html) document the same pattern in their terminology. 5. **PO issuance.** Once all approvals are captured, the system generates the PO number, marks the record as Approved, and dispatches the PO to the vendor through the chosen channel (email, EDI, vendor portal). 6. **Closing the loop.** The requester gets notified the PO is live. Downstream systems (receiving, AP) inherit the PO record with its approval history attached so three-way match has the data it needs later. Every guide on this topic publishes some version of those six steps. The article ends here for most of them. That is exactly where the work starts. ### Where most PO approval workflows break Here is what most posts on PO approval workflows will not say plainly: the standard six-step diagram is the easy 70% of the design. The remaining 30% is where the workflow either holds under real-world conditions or quietly stops being a control. The recurring failure modes: - **Approvers out of office with no defined delegation rule.** The request sits. The requester chases over Slack. Eventually someone with a corner of the right authority approves it informally, and the audit trail loses a step. Delegation rules need to exist in writing before they are needed, not be invented during a stalled approval. - **Requests that cross thresholds mid-flight.** A line item gets added, the vendor changes, the quantity goes up. The approval that already cleared was for a different request. Workflows either re-trigger the routing or quietly carry forward an approval that no longer applies. - **Approvals captured in email or chat.** The PO record shows Approved with no link back to the actual conversation. Months later, an auditor asks who approved a specific PO and on what basis, and the answer requires inbox archaeology. Approval evidence stored in email or chat is harder to reconstruct than approvals captured against the PO record inside the ERP. - **Splitting POs to stay under approval thresholds.** Two POs for $9,000 each go through department-head approval; one PO for $18,000 would have required the CFO. Splitting POs to stay under approval thresholds is a recognized control weakness in procurement, and it is almost always invisible until someone runs a vendor-spend report. - **Vendor or contract data that is stale at the moment of approval.** The approved vendor list has not been reviewed in two years. The contract reference points to an expired MSA. The approval clears against data that no longer reflects the relationship. Each of these can be patched in isolation. The workflows that survive an audit are the ones designed around them upfront, not the ones that grow patches over twelve months until the rule set lives in a half-documented mesh of policy memos and inbox conventions. ### Designing approval rules that match how your team actually buys The most common design mistake is borrowing thresholds from a template. The thresholds that work are the ones set against the real shape of your spend. Things to decide before you write the rule: - **Set thresholds against actual spend distribution.** If 80% of your POs are under $5,000, putting the first threshold at $10,000 sends almost everything to the requester’s direct manager and treats finance review as ceremonial. Pull the last twelve months of POs, look at the spend distribution, and place thresholds where they actually segment the population. - **Decide which categories need category-owner review.** Pure dollar-threshold routing misses risk that lives outside dollars. A $2,000 IT purchase introducing a new SaaS vendor with data access carries more risk than a $20,000 office supplies order on a known contract. Category-owner review (IT for software, marketing for agencies, legal for new MSAs) catches what the dollar threshold misses. - **Write the delegation and escalation rules before they are needed.** Every approver has planned absences and unplanned ones. The rule needs to name the delegate, the trigger (calendar OOO, manual flag, time-since-routed), and the escalation path if the delegate also does not act. - **Keep approval levels proportional to risk.** Three signatures on a $1,500 PO does not catch risk; it consumes time. Most of what the third signer would catch was already visible to the first two. Extra layers of sign-off feel like control and operate like friction. The exercise that surfaces all of these in one pass: walk a sample of last year’s POs through the proposed rule set on paper, and ask which ones would have routed sensibly and which would have stalled, escalated badly, or skipped a review the workflow needed. If the rule set fails on five out of twenty real POs, ship the rule set anyway and you ship the failure with it. ### What to automate and what to keep human This is where the design decision actually pays off. Automation should remove the work nobody wants to do and preserve the work that needs judgment. Automate: - The routing decision itself, based on the rules you have written down. - The threshold and budget checks against the live GL and vendor data. - The reminders to approvers, with full context attached, through the channel they actually read. - The system-of-record updates that capture approver identity, timestamp, and any comments against the PO record. - The closing-the-loop notifications back to the requester and forward to receiving and AP. Keep human: - The actual approval decision. The approver is signing off on a commitment of company funds; that signature is the point of the workflow, not a step to be automated away. - Exception handling that has not been seen before. New vendor, new category, unusual urgency: the workflow should route these to a human with the full context, not auto-approve them on a soft rule. - Any judgment call about whether to split, defer, or escalate a request. The decision about whether your current approval process is even worth formalizing is its own question; the framework in [how to know if an approval process is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) handles it as a financial calculation rather than a gut call. For ERP-specific patterns, two examples that show the shape: - **Acumatica users** can [capture PO approvals with approver identity inside your ERP](https://flowrunner.ai/workflows/automate-with-acumatica) so the audit evidence lives where the PO lives, not in a parallel approval app that has to be reconciled later. - **Vendor-invoice intake** sits adjacent to PO approvals and shares the same exception-routing problem. The pattern for [vendor invoices flowing into the ERP with exception routing](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) is the AP-side mirror of what a PO approval workflow does on the procurement side. - **NetSuite users** facing duplicate-bill risk on the AP side can see how [duplicate and outlier checks before a bill is created in NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) sit in the same exception layer as PO approvals. These are all variants of the same architectural question: the ERP holds the system of record, and the workflow that decides what enters it, who approves it, and what happens when something does not fit a rule has to live somewhere. That somewhere is rarely the ERP’s native approval module alone. The native module owns the happy path. The exceptions are where the work is. That seam, between what the ERP can route on its own and what the procurement team actually does between systems, is where an orchestration layer lives. A system above the systems of record that listens for what they emit, gathers context the rule did not have at fire time, brings the right human in with the request fully framed, and captures the answer back against the PO record so the audit trail stays in one place. FlowRunner is built for that layer. It does not replace the ERP’s approval module; it makes the module’s outputs reconstructible and the exceptions handlable without inbox archaeology. ### A checklist for evaluating your current PO approval workflow Run this against the workflow you have today. Each question is a small audit on its own. - Can you reconstruct who approved any PO from the last 12 months without leaving the ERP? If the answer requires opening an inbox, the audit trail is not where it needs to be. - Are thresholds reviewed against actual spend at least annually? Spend shifts. Thresholds that segment the population today miss the segmentation in eighteen months. - Is there a written delegation rule that covers planned and unplanned absences? Written, not informally understood. - Do exceptions (rush orders, new vendors, contract overruns) have a defined path, or do they get worked around? Worked-around is the failure state. Defined is the goal. - Is the approval evidence linked to the PO record, or stored in inboxes? This is the question that decides whether the workflow is a control or a habit. A workflow that answers all five with the right answer is one that holds up. A workflow that answers any of them with the wrong one is one quarterly review away from being rebuilt under pressure. The standard advice on PO approvals is to map the steps and pick a tool. The honest version is the opposite: name the exceptions first, name where the approver identity lives, name what happens when the rule does not fit, and the tool is whatever survives those three questions. ### Quick answers #### What is a PO approval workflow? It is the routing rules, approval thresholds, and system-of-record updates that move a purchase request from submitted to issued PO, with approver identity captured against the record for the audit trail. #### What steps belong in a PO approval workflow? Request submission with line items and GL coding, budget and vendor validation, routing by threshold or category, approval capture with approver identity, PO issuance to the vendor, and closing the loop back to the requester and downstream receiving and AP. #### Where do PO approval workflows usually break? Approvers out of office with no defined delegation, requests that cross thresholds mid-flight, approvals captured in email or chat with no link back to the ERP, and split POs that stay under approval limits by design. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Post-M&A Integration Checklist for the CFO Who Has to Run the Combined Finance Function Source: https://flowrunner.ai/blog/post-m-and-a-integration-checklist Articles May 26, 2026 Updated May 27, 2026 10 min read A post-M&A integration checklist for the CFO running the combined finance function, not the deal team: day one, 30 days, 90 days, and what slips. ![A bald cartoon carpenter in a forest green sweater and apron standing between two mismatched cabinets in olive and rust, nailing a checklist board across the seam to join them, with one amber item slipping through a gap below a crooked nail.](https://flowrunner.ai/images/blog/post-m-and-a-integration-checklist-hero.webp) The post-M&A integration checklists that show up first on Google were written for corporate development, and a CFO inheriting the combined finance function the day after close is reading the wrong document. Corp dev’s checklist closes the deal. The finance leader’s checklist starts when corp dev’s ends, runs on a different clock, and is graded by entirely different work: cash that does not bounce, payroll that does not miss, vendors that do not get paid twice, and a first combined month-end close that ties out cleanly. This article is that second checklist. It is sequenced by the three horizons a finance leader actually has to manage, and it is honest about the items that consistently fall through the cracks even when there is a checklist nominally in place. ### What a post-M&A integration checklist actually has to cover Three different documents get called a “post-M&A integration checklist,” and they are not the same: - **The corp dev integration plan.** Workstreams across HR, IT, legal, finance, ops, and sales. Owned by an integration management office or a deal lead. Ends roughly when the combined org is operational. - **The IT systems migration plan.** Identity, network, endpoint, data, applications. Owned by IT. - **The CFO’s finance integration checklist.** Cash, AP, AR, payroll, chart of accounts, approvals, system consolidation, combined close. Owned by the head of finance. Narrower than the corp dev plan, deeper in finance specifics, sequenced by finance milestones rather than deal milestones. The rest of this article is the third one. The bar for every item on it is the same: a named owner, a due date, and a definition of done. Without all three, items slip. The finance practitioners we talk to describe this consistently. Ryan Bateman of Platform Accounting Group, in a conversation about diligence and integration, put it plainly: there are things that should happen during the diligence process that just plain do not, they fall through the cracks. The checklist is the easy part. The enforcement around it is where every CFO has been burned at least once. Three time horizons map to the work: - **Day one** (the day the deal closes): cash, payroll, signatories, vendor payment authority. - **First 30 days**: chart of accounts, AP, AR, vendor and customer master consolidation, exception process for mismatches. - **First 90 days**: ERP path decision, approval workflows for the combined org, tool stack consolidation, planning the first combined month-end close. Anything that drifts past its horizon gets harder to fix, not easier, because by then the next horizon’s work has already started. ### Day one finance checklist The point of day one is continuity. Cash moves. Payroll runs. Nobody gets locked out of an account they need, and nobody pays the same vendor twice during the handoff. - **Bank account access and signatory updates.** Confirm every operating account, payroll account, and merchant account for both entities. Authorized signers, online banking permissions, wire approval lists, and ACH origination rights. Update where ownership has legally transferred. Document where it has not, with the date it will. - **Approver lists across both entities.** Who can authorize a wire, sign a check, release a payment file. The acquired company’s approvers stay live until the combined approval matrix is published; assume nothing about delegated authority transferring automatically. - **Payroll continuity for the acquired company’s next pay cycle.** Pay date, payroll provider, tax registrations in every jurisdiction the acquired entity operates in. New jurisdictions trigger new state registrations that take real time to obtain; finance owns confirming they are in flight, not assuming HR has it handled. - **Vendor payment authority during the handoff.** This is the duplicate-pay window. Both entities have AP queues. Both may have unpaid invoices for vendors that serve both companies. The risk is that the acquired entity pays an invoice the parent has already paid, or vice versa, because the vendor masters have not been reconciled yet. Practitioners describe this risk in exactly those terms: it would be very easy to double pay if both we and the other party made a payment on the same vendor. Freeze ambiguous payments to shared vendors until the master is consolidated, or route them through a single approver who can see both queues. - **Open POs, unpaid invoices, and customer credits in both systems.** Inventory every one before any data migration starts. Open POs are commitments; missing them affects accruals at month-end. Customer credits left on the acquired company’s books are real liabilities and easy to forget. - **A single source of truth for cash position.** Day one cash visibility across both entities. A shared spreadsheet that reconciles to bank balances is acceptable for week one. What is not acceptable is two cash views and no one whose explicit job is to reconcile them. If even one of those items is undefined at close, the day-one checklist is not done. There is no version of “we will get to that next week” that ends well. ### First 30 days: AP, AR, and the reconciliation work nobody scoped The first 30 days are where the work corp dev did not scope shows up. Most of it is reconciliation. The two finance stacks, including charts of accounts, vendor masters, customer masters, sales tax setups, and bill payment systems, have to be made to agree before consolidation is even possible. - **Side-by-side chart of accounts mapping.** Pull both COAs into a single sheet. Decide which structure the consolidated business will adopt. Map every account on the deprecated COA to its new home. This is the work that drives most of the next 30 days; pretending it is straightforward is the most common 30-day mistake. - **Vendor master consolidation.** Pull vendor lists from both AP systems. The same legal entity will appear under different display names, different addresses, and different remit-to instructions. Build a consolidated master with one record per legal vendor. Flag the duplicates and decide which record survives. - **Customer and distributor billing arrangements.** Both entities may have separately invoiced or been paid by the same counterparty. Reconcile open AR balances against the counterparty’s own AP records where you can get them. Resolve before the first combined statement goes out. - **An exception process for invoices and payments that do not match.** During the transition window, invoices will arrive that do not match a PO in the system you expect, or that match an old vendor master record. Set up an explicit exception path. A human reviewer looks at the mismatch, decides whether to clear it manually, push it back to the vendor, or route it for approval. Auto-clearing is the wrong default during a transition; the cost of clearing a bad match is higher than the cost of pausing. - **Document the manual reconciliation steps you are doing.** Every workaround your team is running during the transition is a workflow that will need to be either eliminated or automated in the 90-day horizon. If you do not write it down while you are running it, you will rebuild it from memory in three months and the documentation will be wrong. The first month-end close as separate entities, with the parent consolidating the acquired numbers through journal entries, is where the 30-day work gets tested. If JE volume is exploding and reconciliations are not tying, the cause is the COA mapping or the vendor master before it is the systems. ### First 90 days: system consolidation and approval workflow decisions By day 90, the finance leader has enough operating data from the combined entity to make the structural decisions that day one was too early for. - **The ERP path.** Three realistic options: keep both ERPs and consolidate at the GL through reporting, migrate the smaller entity onto the parent’s ERP, or run parallel through year-end close and migrate after. Year-end parallel is a common choice because it avoids forcing a mid-year migration on top of a brand-new combined close. None of the three is the right answer in the abstract; the right answer depends on transaction volume, complexity overlap, ERP versions, and how much in-flight upgrade work either entity already has. - **Approval workflows for the combined org chart.** Approval limits, segregation of duties, and spend thresholds were defined for two different companies with two different risk appetites. The combined version is not the union of both. It is rebuilt for the new structure, with explicit ownership of who can approve what, against which budget, in which entity, with which delegate when they are out. Get it written, signed, and configured in the systems within 90 days. - **Tool stack consolidation across both finance functions.** AP automation, expense, close software, BI, payroll, treasury, tax. Inventory both stacks. Decide which contracts to consolidate, which to terminate, and what the renewal timing looks like. Vendor seat audits where both entities have overlapping licenses are routine post-M&A money left on the table; do them in the 90-day window, not after first renewal. - **Cross-system reconciliations where consolidation is not immediate.** If both ERPs are staying for a year, the gap is filled by a reconciliation cadence. Revenue, AP, and cash positions need to agree between systems on a defined frequency. This is the [pattern shown in the Stripe to QuickBooks to Slack workflow](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack): match what can be matched, flag exceptions, route mismatches to a human reviewer with context attached. - **The first combined month-end close planned as a project.** The first close where both entities are on the consolidated COA, with consolidated vendor masters and consolidated approvals, deserves its own checklist, its own dry run a week ahead, and its own post-mortem after. Treating it as a routine close is the most expensive optimism in the entire 90-day plan. A finance team that hits 90 days with a clear ERP path, a published approval matrix, a consolidated tool stack plan, and a successful first combined close has done the work. The rest of the year is normal operating, with a few cleanup tasks. ### What consistently falls through the cracks These are the items the corp dev checklist tends not to surface, that the IT plan does not own, and that finance ends up holding by default. Worth a dedicated section in your own checklist: - **Sales tax nexus changes.** The acquired entity’s physical presence, customer footprint, and remote workforce can trigger nexus in states the parent never registered in. The penalty for missing this is real, and the lookback windows are long. - **Software seat audits.** Both entities licensing the same SaaS vendors at separate contracts. Consolidate before renewal; vendors will not proactively offer the discount. - **Intercompany transactions starting at close.** From day one, there are now intercompany payables, receivables, and transfers between the two legal entities. They need a documented process, a defined elimination treatment, and someone owning the reconciliation before the first month-end. - **Insurance, benefits, and 401(k) plan consolidation.** Open enrollment timing, plan year alignment, and Form 5500 implications for retirement plans. Easily pushed to HR, but the deadlines and the dollar consequences fall on finance. - **Customer contracts with change-of-control clauses.** Some require notification within a window, some require formal consent. Missing one can void the agreement or create a renegotiation opening the customer will not waste. - **Bank covenants and lender notifications.** Many credit agreements require notification of material changes in ownership or structure. Check the documents the week before close, not the week after. - **Audit firm transitions.** The acquired entity may have a different audit firm, a different audit year-end, or unfinished prior-year work. Decide who audits the combined entity, when, and against which opening balance sheet, before audit planning season makes the decision for you. Most of these belong to no one on the corp dev plan because they are post-close finance work. They belong to no one on the IT plan because they are not systems. They belong to finance, and unless they are on the finance checklist with an owner and a date, they slip. ### Making the checklist actually run Here is what most post-M&A integration articles will not say: the checklist itself is not the safeguard. A spreadsheet with 137 rows and a header that says “PMI integration tracker” has never prevented a missed sales tax registration. The safeguard is enforcement: due dates that fire reminders, ownership that does not silently transfer, exceptions that route to a human, and an audit trail when something gets skipped on purpose. This is the gap every CFO has felt during diligence and again during integration. The tracker exists. The tracker is updated when someone remembers. Items get marked done that were not actually done. Emails about blocked items get buried in inboxes. By the time the gap shows up, it is in the form of a duplicate payment, a missed registration deadline, or a covenant breach, and the audit trail for how it happened is gone. Static checklists fail in four specific ways: - No enforcement of due dates. Nothing chases the owner when an item is overdue. - No visibility into what is blocked. Items sit at “in progress” indefinitely. - No audit trail when an item gets skipped or marked done prematurely. - No connection to the systems doing the work. An item marked complete in the tracker does not mean the state change actually happened in the ERP, the AP system, or the bank portal. The fix is to treat checklist items as workflows, not rows. Each item has an explicit handoff, an explicit approver, a due date that escalates if it is missed, and an audit log of every state change. The items that need human judgment (chart of accounts mapping decisions, ambiguous vendor reconciliation, exception clearing) pause and ask a finance reviewer rather than auto-resolving. The items that connect to systems (vendor master updates, approval matrix changes, bank signatory updates) write through to those systems so the tracker reflects reality instead of intent. This is exactly the seam where checklist software ends and operational reality begins. The checklist tool tracks state; the underlying systems hold the actual record; nothing natively coordinates between them, escalates a missed handoff, or routes an exception with the right context attached. That coordination problem is what an orchestration layer is for. It sits above the systems of record, listens for what they emit, enforces due dates, pings the right human when judgment is needed, attaches full context to the ask, and keeps the audit trail intact across every handoff. FlowRunner is built for that layer. The same pattern that handles a [Stripe to QuickBooks reconciliation with exceptions routed to a human](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) handles a post-M&A integration item with a deadline, an owner, an approval gate, and an audit log. The work shape is identical; only the labels change. ### A working blueprint for the finance integration checklist Pulling the horizons and the enforcement together into something you can adapt: - **Day one items**, each with a named owner and a confirm-by-close timestamp: bank access, signatory updates, payroll continuity, vendor payment authority, open PO and AR inventory, day-one cash visibility. - **30-day items**, each with a due date inside the first month-end window: COA mapping, vendor master consolidation, customer billing reconciliation, exception process stood up, manual workaround documentation started. - **90-day items**, each with a defined decision date: ERP path decision, approval matrix published and configured, tool stack rationalization plan, cross-system reconciliation cadence running, first combined close planned with a dry run. - **Cross-the-cracks items**, owned explicitly by finance and not by HR or IT: sales tax nexus changes, seat audits, intercompany process, benefits and 401(k) timing, change-of-control contract review, bank covenants, audit firm transition. The framing in [how to know what is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) applies to each of these in the 30 and 90-day windows. The manual reconciliations you are running during transition are exactly the workflows worth evaluating against the math, before normal operations resume and the workarounds calcify into permanent process. If the consolidated entity is moving onto a single ERP, the [Acumatica AP automation pattern](https://flowrunner.ai/workflows/automate-with-acumatica) and the [QuickBooks to Stripe revenue reconciliation](https://flowrunner.ai/workflows/connect-quickbooks-online-with-stripe) are reference shapes for what the cleaned-up version of those reconciliations looks like. The combined finance function is not built by the checklist. It is built by the discipline of running it, by the people enforcing the handoffs, and by the layer that catches what slips between them. Get the first day right, get the first month-end clean, and the 90-day decisions are made from a position of clarity instead of recovery. ### Quick answers #### What is post-M&A integration in finance? The work the finance function owns after legal close to combine two companies into one operating entity. It runs across day one (cash, payroll, signatories), the first 30 days (AP, AR, chart of accounts), and the first 90 days (system consolidation and combined close). It is distinct from corporate development’s integration plan, which is broader and ends earlier. #### How long does post-merger financial integration take? Day one items must clear by close. The first month-end as a combined entity is the real test, 30 to 45 days after close. ERP consolidation lives on the 90-day horizon at the earliest, and running parallel through year-end close is a common path that avoids forcing a mid-year migration on top of a brand-new combined close. #### What is the biggest risk in the first 100 days after close? Items that lack a named owner and a due date. Duplicate payments and missed sales-tax-nexus changes happen because both entities are mid-handoff and no one has explicit ownership of the seam between them. The checklist itself is not the safeguard; the enforcement around it is. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Predictive Lead Scoring in HubSpot: What It Is, What You Cannot See Inside It Source: https://flowrunner.ai/blog/predictive-lead-scoring-hubspot Articles May 26, 2026 Updated May 28, 2026 8 min read HubSpot's predictive lead scoring is a closed model that ranks contacts by likelihood to close. What it does, what it hides, and where to use it carefully. ![A bald cartoon technician at a workshop bench peers with a magnifying glass at a closed deep-ochre black box with one wire of contact records going in and one wire going out toward a sales record, the model's score visible on a small amber dial but the inside of the box sealed.](https://flowrunner.ai/images/blog/predictive-lead-scoring-hubspot-hero.webp) Turning on HubSpot’s predictive lead scoring is a trade: you accept a model whose internals you cannot inspect in exchange for pattern recognition your manual rubric will never reach. Most articles on this topic treat that trade as a feature decision. It is an operational one, and the trade has consequences for how SDRs work the score, how managers defend dispositions, and where the score belongs inside the larger qualification flow. This article exists to argue that predictive lead scoring is a useful signal, a poor decision, and never a complete routing layer. Knowing the difference is the work. ### What HubSpot predictive lead scoring actually is HubSpot ships two distinct lead scoring mechanisms, and the names are close enough that they get conflated in onboarding all the time. - **HubSpot Score** is the manual property. You write the rules (page visits add 5 points, a specific job title adds 10, an unsubscribe subtracts 20). Every rule is visible, every weight is editable, every contact’s score is traceable back to the rules that touched it. - **HubSpot Predictive Lead Scoring** is a machine-learned probability. It is a separate contact property called _Likelihood to close_. The model is trained on your own historical contacts (closed-won and closed-lost) and produces a numerical likelihood that any given contact will eventually become a customer. The setup, the property, and the gating are documented in [HubSpot’s knowledge base article on determining likelihood to close](https://knowledge.hubspot.com/properties/determine-likelihood-to-close-with-predictive-lead-scoring). The two coexist on a contact. Most teams that move to predictive scoring keep the manual score running alongside it, because the two answer different questions. The manual score reflects your team’s stated theory of the ICP. The predictive score reflects the patterns your historical conversions actually exhibited, which is rarely the same thing. Predictive scoring is gated. Per HubSpot’s documentation, the _Likelihood to close_ property is available on **Marketing Hub Enterprise** and **Sales Hub Enterprise**. Free, Starter, and Professional tiers do not include it. If a brief on your desk says “use HubSpot’s predictive score,” step one is confirming you are on an Enterprise Hub that exposes the property. ### How it differs from rule-based scoring The differences are not subtle. They reshape who the score is for and how it can be used. | | **HubSpot Score (manual)** | **Predictive Lead Scoring** | | --- | --- | --- | | **Logic** | Rules you write | Model trained on your historical conversions | | **Visibility of weights** | Every rule is editable in the UI | Per-contact weights and feature importance are not exposed | | **What it picks up** | What you explicitly told it to | Patterns in your data, including ones you did not encode | | **Maintenance** | Rule list grows; you prune it | Model retrains on new conversion data | | **Data dependency** | Works on day one with no history | Requires sufficient closed-won and closed-lost volume to produce a usable score | | **Auditability** | A rep can ask why a contact scored high and get a real answer | A rep can ask the same question and the platform does not have one to give | The first row is the obvious one. The last row is the one that changes how the score sits in your workflow. A rule-based score is a number with a transparent provenance. A predictive score is a number with a sealed provenance. Both are useful. They support different conversations with the team. ### What you need to turn it on A short list, not because the setup is hard, but because the prerequisites are the part teams skip and then troubleshoot for a week. 1. **The right Hub edition.** Marketing Hub Enterprise or Sales Hub Enterprise. Confirm the _Likelihood to close_ property is visible in your contact properties list. 2. **Enough historical conversion data.** HubSpot’s model needs a minimum volume of both closed-won and closed-lost contacts to produce a usable score. The platform documents the current threshold; do not memorize a specific number from a blog post (this one or any other), because the threshold has moved over time. Read the [current HubSpot documentation](https://knowledge.hubspot.com/properties/determine-likelihood-to-close-with-predictive-lead-scoring) at the moment you plan to turn it on. 3. **A clean enough conversion history that the model is not learning your noise.** If your closed-lost contacts are mostly junk leads that should never have been logged as closed-lost, the model learns to optimize against junk. Garbage in, sealed-box garbage out. 4. **A working theory of where the score goes once it exists.** A score sitting on a contact property nobody routes from is shelfware. Decide which list filters, which workflow triggers, and which rep-facing surfaces will use the score _before_ you turn it on. The score lives as a contact property, which means it shows up in list filters, workflow triggers, dashboards, and reports the same way any other contact property does. The mechanics of consuming the score are familiar. The mechanics of trusting it are the new part. ### The real limitations sales ops should know Here is what most posts on HubSpot’s predictive scoring will not say plainly: the score is a useful input and a closed black box, and pretending otherwise produces an SDR team that does not trust the field and a sales ops team that cannot defend it. Four limitations worth internalizing before you make the score load-bearing. **Model opacity.** HubSpot does not expose per-contact feature weights or model explanations. When a rep asks why Contact A scored 87 and Contact B scored 23, the answer your team can give is “the model said so.” For top-of-funnel sorting that is acceptable. For disqualification decisions, deal-stage reviews, or rep performance conversations, it is not. **Data dependency and historical bias.** The model is trained on what your team has historically closed. If your historical conversions overrepresent a customer segment, the model will keep recommending that segment, even after you have decided as a company to expand into a new one. The model does not know your ICP is moving. It knows what closed. **Latency between event and score update.** A contact’s score does not update at the instant a new conversion event lands. The model retrains on a cadence, and a contact’s recalculation happens on its own cadence inside that. For day-to-day routing this is a non-issue. For “we just decided this lead is hot, do something now” decisions, the score lags behind the action. **The score is a probability, not a routing decision.** This is the limitation that matters most for how the score is operationalized. A _Likelihood to close_ of 87 does not assign an owner, set an SLA timer, notify a rep, attach context, or move the contact through a stage gate. It is a number on a record. Every workflow that consumes the score has to make its own decisions about what to do with it. The score is upstream of the routing layer, not the routing layer. That last limitation is where the predictive score connects to the rest of a sales operation, and where it stops being a self-contained feature. ### Where the workflow layer enters A predictive score is the answer to one question: _how likely is this contact to close, given our history?_ It is not the answer to any of the operational questions that immediately follow. Who owns this contact? What channel does the rep see it in? What context arrives with it? What happens if the score is high and the assigned rep is at capacity? What is the SLA, and who hears about it when the SLA breaks? When the score moves from 30 to 85 overnight on an existing contact, what triggers and who responds? None of those are scoring questions. They are routing, escalation, and orchestration questions. The CRM owns the field. The work that the field is supposed to set in motion lives between systems. Between the CRM and the channel the SDR works in. Between the CRM and the enrichment service. Between the CRM and the manager who has to disposition the borderline cases. That gap, between a number on a record and the action the number is supposed to provoke, is the seam this article is really about. Closed-box scoring models live in CRMs. The orchestration of what happens because of the score does not. An orchestration layer is the category that owns that gap: a system above the systems of record that listens for what they emit, gathers the context the field alone does not carry, and pulls a human in for the cases the model alone cannot defend. FlowRunner is built for that layer. So is the broader category, which is still forming and still missing from most RevOps stack diagrams. The practical shape of operationalizing a predictive score, regardless of which tool plays the orchestration role: - **Use the score as one input alongside firmographic filters and explicit qualification signals.** Not as the sole gate. A predictive 90 on a non-ICP contact is still a non-ICP contact. - **Pair the score with structured routing thresholds.** High-score plus ICP-match goes to AE. Mid-score plus ICP-match goes to SDR for live qualification. Low-score plus weak fit goes to nurture, not to disqualification. - **Surface the score in the channels reps already work in,** with the context attached. A number on a contact record is a number. The same number arriving in Slack with the source, the recent activity, and an open-record link is something a rep can act on. The pattern is the same one we walk through in [route scored HubSpot leads into Slack](https://flowrunner.ai/workflows/connect-hubspot-with-slack). - **Keep a human in the loop for borderline scores.** The opacity that makes the model uncomfortable for disqualification reviews also makes it valuable for _pre-qualification_. Treat scores in the middle band as a flag for human attention, not as a decision. The same shape applies upstream of scoring, too. If you are still capturing inbound through forms and routing manually, the score will sit on top of a noisy data layer. The mechanics of cleaning that up before the score gets near it are covered in [qualify inbound HubSpot leads without manual research](https://flowrunner.ai/workflows/orchestrate-google-forms-and-hubspot-and-slack), and the cross-CRM shape of the same pattern is in [how lead scoring works in Salesforce-driven workflows](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack). ### When manual or hybrid scoring is the better choice Predictive scoring is not the right answer for every team that _could_ turn it on. Three situations where manual or hybrid scoring earns its keep: - **You are below the data threshold.** If the model cannot produce a usable score yet, manual scoring is not the inferior option, it is the only option. Start with explicit rules, accumulate conversion history, then revisit predictive once the model has enough to learn from. - **Your ICP is narrow and well understood.** Teams with a tight, well-documented ideal customer get more value from explicit rules than from a learned model. The rules encode what you already know; the model would discover patterns you already encoded. - **Your ICP is actively moving.** If sales just opened a new vertical or moved upmarket, your historical conversions reflect the old motion. Manual rules let you steer toward the new motion immediately. Predictive scoring will catch up, but on its retraining schedule, not yours. The hybrid pattern is the underused middle ground: manual rules for must-have firmographics (region, employee count, vertical), with the predictive score layered on top to catch behavioral patterns the rules miss. The two answer different questions and the answers compose. The framework for whether the underlying routing work is worth automating at all is in [decide whether lead routing is worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating). The upstream routing-mechanics version of the same conversation, in a different CRM, sits in [Salesforce lead assignment rules: what they do, where they stop](https://flowrunner.ai/blog/salesforce-lead-assignment-rules). Predictive scoring is one component of a qualified pipeline, not the pipeline itself. Treat it that way, and the closed-box nature of the model becomes a manageable trade. Treat it as the whole answer, and the day a borderline disposition lands on a manager’s desk with no explanation behind the number is the day the field starts losing the team’s trust. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Property Management Automation: What to Automate, What Needs a Human Source: https://flowrunner.ai/blog/property-management-automation-guide Articles May 28, 2026 Updated May 30, 2026 10 min read Property management automation is two purchases pretending to be one: the platform that runs your portfolio, and the layer for work between systems. ![A bald cartoon man at a sorting bench beside a conveyor belt of property-task tickets; most tickets in sage green glide past on their own while he holds three amber tickets pulled off the belt for his judgment.](https://flowrunner.ai/images/blog/property-management-automation-guide-hero.webp) Property management automation is two purchases pretending to be one. There is the platform that runs your portfolio, and there is the layer that handles the work between your platform and everything else. Most operations leaders buy the first, assume it covers the second, and then spend the next year doing the second by hand. The platform was never the problem. The work it does not reach is. This article is for the VP or director of operations evaluating how to automate a property management operation, not the platform itself. If you are choosing the system of record, that is a different and well-covered decision. The decision nobody walks you through is what to automate inside it, what genuinely needs a person, and where the work lives that no single platform was built to close. ### What property management automation software actually covers The category, as it shows up when you search for it, is the lifecycle of running rental property: tenant onboarding, rent collection, maintenance ticketing, lease renewals, accounting, and owner reporting. The vertical platforms that own this category are the names you already know. [AppFolio](https://www.appfolio.com/), [Buildium](https://www.buildium.com/), and [DoorLoop](https://www.doorloop.com/) anchor the mid-market and up, with TenantCloud and RentRedi serving smaller portfolios and independent landlords. Each one is a complete system of record for properties, leases, residents, and the money moving through them. These platforms are good at what they do, and that should be stated plainly before anything else. For the core lifecycle, they are the right starting point for nearly every portfolio, and most operators do not need to look further to automate the bulk of their week. What gets lost is that this is one layer of two. The platform is the system of record. Sitting on top of it is the connective work: the data moving between the platform and the systems it does not own, and the exceptions that route to people the platform cannot reach. The category markets itself as if those two layers are the same product. They are not, and conflating them is how teams end up with a strong platform and a stubborn backlog of manual coordination they cannot explain. ### The two layers operations leaders should evaluate separately Layer one is the system of record. It holds your units, leases, residents, ledger, and the standard workflows that run against them. This is what you are buying when you buy AppFolio or Buildium, and it is the larger, more visible purchase. Layer two is the connective layer. It is the work that routes exceptions to people, reconciles data between the platform and adjacent systems, and pulls a human in when a decision needs judgment. It is quieter, it rarely shows up in a platform demo, and it is almost never on the RFP. Which is exactly why it goes unsolved. Here is what most articles on this topic will not say. The standard advice treats automation as a feature checklist you turn on inside your platform: switch on autopay, switch on maintenance routing, switch on owner statements, count the hours saved. That advice is fine as far as it goes, and it covers the first layer well. But it quietly assumes the last stretch of manual work is a feature you have not enabled yet. It is not. The work that resists automation is structural. It lives between systems and between people, in the space no single-vendor platform was designed to own, and no amount of toggling features inside the platform reaches it. That is the seam worth naming, because it is where an entire category sits. The layer that listens to what your platform of record emits, gathers the context the platform does not hold, and brings a person in at the moments that need a decision is not a feature of the platform. It is a category of its own, an orchestration layer that sits above the systems of record and coordinates the work and the people moving across them. [FlowRunner](https://flowrunner.ai/) is built for that layer. The reason to evaluate the two layers separately is that one of them, the connective one, will not get evaluated at all if you fold it into the platform purchase. \[![a two-band diagram](https://flowrunner.ai/images/blog/property-management-automation-guide-1.webp)\] ### What the vertical platforms automate well Be specific about the first layer, because the platforms genuinely earn it. Inside a system like AppFolio or Buildium, the core lifecycle automates cleanly: - **Rent collection and reconciliation.** Autopay, late-fee assessment, late notices, and matching incoming payments against the ledger, all inside the platform. - **Maintenance.** Request intake from residents, vendor dispatch from your vendor list, and status updates back to the resident. - **Leasing.** Lease document generation, e-signature, and renewal workflows tied to lease dates. - **Owner reporting.** Standard owner statements and accounting exports on a schedule. For these, the platform is the answer and a connective layer would be overkill. A maintenance request that comes in through the platform, gets triaged, dispatched to a vendor already in the system, and followed up on is work the platform owns end to end. If the automation you need is in-platform, enable it there and do not buy a second tool to do what the first one already does. The detail that matters is the boundary condition. All of this works because the data and the people involved live inside the platform. The resident is in the platform. The vendor is in the vendor list. The ledger is the platform’s ledger. The moment a workflow needs data the platform does not hold, or a person the platform cannot reach, you have crossed into the second layer. ### Where the vertical platforms leave gaps The gaps are not platform weaknesses. They are the predictable edge of any single system of record, and they cluster in a few recognizable places. **Routing to people outside the system.** The decision that needs a regional manager, an owner, outside counsel, or a third-party vendor who never made it into your platform is the decision the platform struggles to route. One pattern operations leaders describe repeatedly is the inability to tag or route work to people who sit off the core system. The platform can assign a task to a platform user. It cannot easily pull in someone who does not have a seat. **Cross-system reconciliation.** When the platform’s numbers and your general ledger or bank feed disagree, that disagreement surfaces as reconciliation work, not as a clean sync error you can ignore. Someone has to look at both, understand why they differ, and decide what is correct. As one operations leader put it about a different stack, it is not really a sync error, it is catching inconsistencies that were entered somewhere after the fact, where the data simply does not make sense. Property management has the same shape: the platform, the bank, and the books each hold a version, and a person reconciles the difference. **Disputes and edge cases that need context-rich review.** A resident dispute, a complaint, a damage claim contested by a tenant. These need a human reading the full context, and the context is rarely in one place. **Approvals that must pause and ask.** An approval where the right answer depends on the lease, the payment history, and a property-level policy at the same time cannot be rubber-stamped from inside one module. It needs the workflow to stop, gather the pieces, ask a person, and resume with the answer attached. **Custom fee and commission logic.** Management fees, leasing commissions, and owner splits that vary by property, owner, or contract often live in a spreadsheet beside the platform because they do not fit the platform’s standard model. \[![a Slack approval message preview showing a connective-layer escalation for a contested resident charge, with the lease clause, the payment history, and the property policy summarized in the message body, and Approve, Waive, and Call Resident action buttons, the human-decision moment rendered in amber](https://flowrunner.ai/images/blog/property-management-automation-guide-2.webp)\] Every one of these has the same signature. Data in more than one place, a judgment call a person should make, and no native home inside a single platform. This is the work that one operations leader, describing automation they had already built, summed up by saying it got them to about 85% of where they wanted to go, and the missing piece was the intelligent human in the middle. Treat that number as one operator’s description of their own state, not a benchmark for the category. The shape of it, though, is what recurs: the platform handles the bulk, and the last stretch is exceptions and cross-system routing. ### How to evaluate the connective layer You evaluate the second layer by looking at where the first one already failed you, which means looking somewhere other than the platform demo. Start with the work that currently lives outside any system. The spreadsheets, the recurring email threads, the Slack channel where exceptions get sorted out by hand. That residue is the connective layer made visible. It is the work the platform did not reach, and it is the most honest specification you will get for what a second layer needs to do. Then map which of those exceptions actually need a person and which only feel like they do. Some can run end to end once something coordinates the systems involved. Others need judgment and should route to a human every time. The goal is not to automate the judgment away. It is to automate everything around the judgment so the person only touches the decision. If you have not yet sorted your processes into worth-automating and not, [how to decide which workflows are worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) is the place to start before you evaluate any tool. When you do evaluate candidate tools, hold them to two specific tests: - **Can it call your platform’s API and act on what it returns?** A connective layer that cannot read and write to your system of record is not connective. - **Can it route work to people who are not users of that platform?** This is the test most automation tools quietly fail. Routing to a platform user is easy. Routing a decision, with full context, to an owner or a regional manager who lives in email is the capability that closes the gap. One more test, learned the hard way by anyone who has shipped automation into production: the human-routing step is where automations break when it is bolted on as an afterthought rather than designed in. [Why automations fail in production](https://flowrunner.ai/blog/the-truth-about-building-automations) is worth reading before you assume the escalation path will take care of itself. It will not. It is the part that needs the most design, because it is the part that runs when something has already gone sideways. ### A practical buying sequence The order of operations matters more than the tool selection, because buying in the wrong order is how the second layer never gets solved. 1. **Pick the vertical platform first.** Choose the system of record that fits your portfolio size and asset mix. AppFolio, Buildium, DoorLoop, and the rest each anchor on a slightly different segment. This is the larger decision and it should come first, because everything connective sits on top of it. 2. **Run it for a full cycle.** A month, a quarter, a full turnover season. Do not buy anything else yet. Let the platform do what it does and watch where work still falls through the cracks. 3. **Document the residue.** Write down every process that still lives in a spreadsheet, an email thread, or a person’s head. That list is your connective-layer requirements, written by reality instead of by a vendor. 4. **Layer connective automation against the specific gaps.** Buy the second layer to close the residue you documented, not to chase capabilities you might theoretically use. Speculative automation is how teams end up with tools nobody adopts. 5. **Pilot one exception-routing workflow end to end.** Pick the single most painful cross-system exception, build it, route it to a real person, and run it for real. Prove the pattern on one workflow before you expand. Proof beats a roadmap. This sequence is deliberately slow at the start, because the most expensive mistake is buying the connective layer speculatively, before the platform has shown you exactly where it leaks. ### Where a connective layer like FlowRunner fits By the time you reach the second layer, the job is well defined. You need something that sits above your property management system of record, reaches into the systems it does not own, and pulls a person in when a decision needs judgment. That is the category. FlowRunner is one platform built for it, not a replacement for your platform of record, and the distinction is the whole point. \[![a connective-layer overview with the property management platform shown at the center as the system of record, and four coordinated workflows fanning out to a bank feed, a general ledger, an insurance verification service, and an off-platform turnover vendor](https://flowrunner.ai/images/blog/property-management-automation-guide-3.webp)\] Concretely, the connective layer is what handles the cases the earlier sections named: - Routing a maintenance escalation to the right regional manager with the full ticket context attached, even though that manager is not a seat in the platform. - Reconciling rent roll data against the general ledger and surfacing the mismatches a person should review, instead of letting them accumulate until month end. - Pausing an approval to ask an owner before issuing a credit, then resuming the workflow with the owner’s answer recorded and attached. In each case an agent does the gathering and the routine work, and a human is invoked as a callable step at the exact moment judgment is required. The response comes back through Slack, email, WhatsApp, or phone, and the workflow resumes from the answer with an audit trail behind it. One operations leader called this pattern a digital andon cord: the pull that stops the line when something is wrong, rather than letting the work plow ahead unwatched. If you are still sorting out where AI handles the routine and where a person stays in control, [the difference between AI automation and AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents) draws the line that matters here. If your operation already runs on AppFolio or Buildium, the layered question gets specific fast, and the platform-by-platform reads cover it directly: [FlowRunner versus AppFolio](https://flowrunner.ai/blog/flowrunner-vs-appfolio-property-management-automation) and [FlowRunner versus Buildium](https://flowrunner.ai/blog/flowrunner-vs-buildium-property-management-automation) both walk through what stays in the platform and what crosses into the connective layer. For the full cross-system pattern in one place, the [property management automation overview](https://flowrunner.ai/solutions/property-management-automation-software) lays it out end to end. The reason to keep the two layers separate in your head, and in your buying, is the one this article opened with. The platform is not where your team loses its week. The work between the platform and everything else is, and that work will keep falling to a person by hand until something is built to catch it. Decide on the platform first. Then go find the seams, because the seams are the part that was never going to fix itself. ### Quick answers #### What can you actually automate in property management? Inside a platform like AppFolio, Buildium, or DoorLoop: rent collection and autopay reconciliation, late notices, maintenance request intake and vendor dispatch, lease document generation and e-signature, renewals, and standard owner statements. These are mature capabilities and the right starting point for most portfolios. #### What still needs a human in property management automation? The work that crosses systems or needs judgment: routing an exception to a regional manager or owner who sits outside the platform, reconciling the platform against your general ledger or bank feed, deciding a resident dispute, and approvals where the answer depends on context from more than one place at once. A common gap operations leaders describe is the inability to tag or route work to people off the system. #### How do you start automating property management? Pick the vertical platform that fits your portfolio first and run it for a full cycle. Then document where work still falls through the cracks, in spreadsheets, email threads, and Slack. Layer connective automation against those specific gaps rather than buying it speculatively, and pilot one exception-routing workflow end to end before expanding. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Quick Guide: How to Know What's Worth Automating Source: https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating Articles October 4, 2025 Updated May 26, 2026 6 min read Automation isn't magic. It's math. Use this framework to identify workflows with the highest return on investment. ![A bald cartoon man at a chalkboard with chalk in hand, an ROI equation written out and one result line circled in amber, weighing whether the math justifies automating the work.](https://flowrunner.ai/images/blog/quick-guide-how-to-know-whats-worth-automating-hero.webp) Not everything is worth automating. Rather than automating randomly, teams should use financial calculations to identify workflows with the highest return on investment. ### The Core Principle Automation isn’t magic. It’s math. Quantify pain points before building automation solutions. ### The Basic Formula **Annual Hours = Frequency x Duration x 52 x People** With typical labor costs of $60-$90 per hour, even modest time savings compound quickly. ### Three Cost Multipliers Labor savings alone understate the value. Add these multipliers: - **Revenue uplift** from faster processing - **Error-related expenses** (refunds, penalties, rework) - **Compliance and regulatory risks** ### Illustrative Example: Lead Routing Imagine a lead routing workflow processing 40 leads per week, 6 minutes each, across 2 people. That generates roughly 416 annual hours, which translates to a five-figure annual labor cost at typical loaded rates. Now layer the multipliers. Faster lead response can lift conversion revenue. Avoided compliance penalties protect against fines. The total annual value of automating that single workflow can climb well past the labor savings alone. ### Making the Case Don’t tell executives a process “saves time.” Tell them what it’s worth in dollars. Financial framing changes the conversation from cost center to investment. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## QuickBooks Automation: Where It Pays Off, Where It Quietly Breaks Source: https://flowrunner.ai/blog/quickbooks-automation Articles May 26, 2026 Updated June 19, 2026 8 min read An honest CFO read on QuickBooks automation: which workflows around AP, invoicing, reconciliation, and journal entries scale, and where human judgment stays. ![A bald cartoon man watching a Rube-Goldberg contraption of teal pipes and gears feeding paper documents into a rust-colored ledger book, the orchestrated flow from intake to entry.](https://flowrunner.ai/images/blog/quickbooks-automation-hero.webp) QuickBooks automation pays off only where it eliminates the repetitive processing that still needs just enough judgment to keep the finance leader in the loop, and almost none of that work happens inside QuickBooks itself. The leverage is in the orchestration layer around QuickBooks: between the bookkeeping system, the inboxes that feed it bills, the processors that fund it, and the people who have to sign off before money moves. ### What CFOs actually mean by QuickBooks automation The phrase “QuickBooks automation” gets used two completely different ways and the confusion costs finance leaders months of misallocated effort. The first way is bookkeeping shortcuts inside Intuit: bank feed rules, recurring transactions, memorized journal entries, and the new AI features Intuit ships into QuickBooks Online. These work well for what they are. Categorization, recurring entries, and surfacing anomalies are real wins inside the four walls of the app. The second way, and the one CFOs of growing companies actually need, is the orchestration around QuickBooks. The work between systems. Bills come in by email and from vendor portals. Payments land in Stripe, the bank, sometimes both. Approvals live in Slack or someone’s head. Source systems (ERPs, fulfillment platforms, distributor portals) hold the source of truth for transactions that have to end up reconciled in the general ledger. QuickBooks is one node in a five-node graph. The native QuickBooks rules engine cannot see the other four. Finance leaders consistently point at four workflows when asked what’s worth automating in this layer: - Accounts payable intake and bill approval - Customer invoice intake and AR-side reconciliation - Bank and payment processor reconciliation against the GL - Recurring journal entries (freight, prepaid, intercompany) These are the four. The honest read that most QuickBooks automation guides won’t say is that [automation exceptions](https://flowrunner.ai/concepts/automation-exceptions) in these workflows are not the failure case. Exceptions _are_ the work. The repetitive core moves itself once the rules are set. What consumes senior finance time, week after week, is the bill that doesn’t match a PO, the Stripe payout that doesn’t reconcile cleanly, the new vendor who needs approval, the freight allocation that hits the wrong cost center. Automation that only handles the easy core, and silently fails on exceptions, is what makes finance leaders distrust automation in the first place. ### Accounts payable automation in QuickBooks without the rubber stamp problem The honest baseline for AP in most mid-market finance teams running QuickBooks: a mix of Bill.com or similar, PDFs forwarded to a shared inbox, and manual entry when someone is out. Approvals get routed by email or pulled into a tool, and the approver signs off without much context because the context is sitting in three other places. Finance leaders describe still entering bills manually when staff are out, which is a signal that the AP process isn’t fully automated even at companies that already pay for an AP tool. The deeper failure mode is the rubber-stamp problem. When the approver gets a notification that says “Approve this $4,200 bill from Vendor X?” without seeing the PO, the prior history with that vendor, or the budget the line item should hit, the approval is procedural. Sales leaders and operators across finance call out the same pattern in their own world: it’s just paperwork, and people want to get it off their desk. Approvals that look like governance and feel like a chore stop catching the exceptions that matter. What automation should actually do in AP: - Ingest the bill from email, vendor portal, or upload - Extract line items and match against the PO or prior history with the vendor - Route to the right approver with the full context attached (PO, vendor history, budget code, line-item detail) - Post to QuickBooks on approval, with a record of who saw what and when Where the human stays in the loop is the part that matters more than the automation itself: - A new vendor with no payment history - A duplicate that looks like the same bill paid twice (duplicate payment risk is a recurring concern in vendor and distributor billing relationships, where both parties can issue payments against the same charge) - An amount above a defined threshold for that vendor - A bill without a matching PO when one is expected - Anything touching a compliance-sensitive account In those cases the workflow pauses and asks. It does not auto-post. The point isn’t that the automation can’t handle them. The point is that finance owns the exception list, the system enforces the pause, and the audit trail records the decision. For a concrete pattern, see how FlowRunner is configured to [automate invoice processing from email into QuickBooks](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) with approval gates before the bill is posted. ### Invoice automation and the AR side of QuickBooks The phrase “invoice automation in QuickBooks” gets read as vendor-bill intake, which is half the picture. The other half is the AR side: customer invoices going out, and payments coming back in that have to be matched against open invoices and reconciled in the GL. On the AR side, the high-leverage workflow at most mid-market companies is matching Stripe (or another processor) payouts against open QuickBooks invoices, flagging mismatches, and writing the reconciliation entry. The structure of the work is identical to AP at the level that matters: a stable rule for the matched cases, a defined escalation for the unmatched. FlowRunner can [reconcile Stripe payments against QuickBooks invoices](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) and escalate mismatches to a human reviewer with the full context attached. The [QuickBooks and Stripe reconciliation with human-in-the-loop](https://flowrunner.ai/workflows/connect-quickbooks-online-with-stripe) pattern is one of the most common requests finance leaders make once they see the AP side working. The technical point worth naming here, because it shapes every buying decision in this category: anything outside the QuickBooks native UI requires API-level access. Intuit publishes a [developer API](https://developer.intuit.com/app/developer/qbo/docs/get-started) that lets third-party orchestration read and write transactions, bills, and journal entries programmatically. AI bolt-ons that only read screens or scrape exports hit a ceiling fast. An orchestration layer working against the QuickBooks API doesn’t. ### Bank reconciliation and journal entry automation Manual reconciliation between source systems and the GL is the canonical CSV-dump-into-Excel chore. The pattern most finance leaders describe: export from a source system (a distributor portal, a fulfillment platform, an ERP running alongside QuickBooks at mid-size companies), dump it into a CSV or Excel table, then upload or paste into QuickBooks for matching. Multiply that by month-end, by line of business, by entity. The aggregate of small recurring chores is what consumes senior finance time, not any single big task. The automation pattern is structural and stable: - Pull source transactions from the system of record on its own schedule - Normalize the data into a shape that maps to QuickBooks accounts - Match against QuickBooks entries on stable keys (invoice number, transaction ID, date plus amount) - Flag the unmatched, with the context attached - Draft the journal entry for human review before posting Recurring journal entries (freight allocation, prepaid amortization, intercompany transfers) are particularly good automation candidates because the rule is stable and the exceptions are rare but material. Freight cost allocation done manually in a pivot table and uploaded to the GL is a workflow finance leaders explicitly call out as ripe for automation, and one where the rule almost never changes once it’s documented. Here is where the line has to be drawn cleanly: drafting and proposing journal entries is acceptable. Posting to the GL without human review is not, for entries that hit compliance-sensitive accounts. The pattern is a draft + approve + post sequence, not a fire-and-forget. AI agents that post to QuickBooks without visibility are exactly the kind of black-box system finance leaders flag as a deal-breaker. The same orchestration pattern shows up on adjacent ERPs; see [how AP automation works on Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) for a parallel structure. ### Where compliance and human-in-the-loop actually matter Finance automation lives or dies on the audit trail. Every automated action against QuickBooks needs to be traceable to a trigger, a decision, and an approver. Orchestration around QuickBooks can be designed to produce a per-action audit trail covering exactly that: what fired the workflow, what the system decided, who approved at each checkpoint, what was written to QuickBooks. This is a design property of the orchestration layer, not a bolt-on report. Human-in-the-loop is the design principle, not a feature. Defining the checkpoints up front is the work that actually de-risks the automation: - Dollar thresholds above which any post requires approval - New vendors (first-time-seen) - Duplicate-payment heuristics (same vendor, similar amount, close date) - Anything touching compliance-sensitive accounts (payroll, tax, intercompany) - Anything where the source-system data and the QuickBooks data disagree The category-level point this article is making is the one most QuickBooks automation guides skip: native rules and standalone AI features handle low-judgment categorization well, and they fall over on cross-system exception handling because that’s not the problem they were designed to solve. The work between QuickBooks, the inboxes that feed it, the processors that fund it, and the approvers who have to sign off lives in a layer above any single system. We call this category [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service): a platform category that coordinates, governs, and supervises multi-agent environments while keeping humans in control of the decisions that require judgment. An orchestration layer is what owns that seam: a system above the systems of record that listens for what they emit, gathers the context the native rules don’t have, and pulls a human in at the moments that need judgment, in the [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) pattern that pauses, assembles context, routes to the right person, and resumes with the answer recorded. FlowRunner is built for that layer. For approval routing specifically, the most common pattern at mid-market companies is to [route QuickBooks approvals through Slack](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack), because that’s where the approver already lives and the context can be attached to the message instead of asking the approver to chase it across tabs. ### How to decide what to automate first in QuickBooks The prioritization filter that holds up across every CFO conversation: high volume, repetitive, judgment-light on the core path, judgment-heavy on the exceptions. AP and reconciliation win on this filter at almost every company that has grown past a single bookkeeper. Customer invoicing wins for companies with a clean processor (Stripe to QuickBooks is the canonical case). Journal entries win only after the rule is documented and the controller is comfortable with a draft-and-approve flow. Anti-patterns to avoid in the first 90 days: - Trying to automate month-end close end-to-end on day one. Pick one workflow, prove the audit trail, then expand - Replacing the controller’s review step entirely. The review is what makes the automation safe to grow - Layering AI on top of a process nobody has documented. Document the rules and the exception list first, then automate the rules and route the exceptions - Treating compliance posture as a checkbox. The audit trail is the deliverable; certifications belong on a Trust page, not in a workflow description The pragmatic sequence is one workflow at a time. Start with invoice intake or Stripe-to-QuickBooks reconciliation. Prove the audit trail in a real month-end. Expand to adjacent surfaces (the bills the first workflow doesn’t cover, the journal entries that draft from the same source data). For a deeper framework on prioritization, see the guide on [deciding what’s worth automating](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating). The CFOs who succeed with QuickBooks automation aren’t the ones who automated the most. They’re the ones who defined the exception list before they shipped anything, and they own that list as it evolves. The automation is the easy part. The exception list is the asset. ### Quick answers **What does QuickBooks automation actually cover for a finance team?** Four high-leverage workflows: accounts payable intake and approval, customer invoice and payment reconciliation, bank and payment processor reconciliation, and recurring journal entries. Anything outside categorization lives between QuickBooks and another system. **Can AI post journal entries to QuickBooks automatically?** AI can draft and propose journal entries based on stable rules like freight allocation or prepaid amortization. Posting to the GL without human review is not advisable for entries that affect compliance-sensitive accounts. Drafting and proposing is the safe pattern. **Why are QuickBooks native rules not enough for AP and reconciliation?** Native rules handle categorization well inside QuickBooks. They do not coordinate intake from email and bank feeds, route approvals to the right person with context, or compare QuickBooks data against an outside source system. That orchestration work is what mid-market finance teams actually need automated. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Reconciliation Tools: What the Category Actually Sells, and What Most CFOs Are Really Buying Source: https://flowrunner.ai/blog/reconciliation-tools Articles May 27, 2026 Updated May 28, 2026 9 min read Most reconciliation-tool roundups rank by match rate. For mid-market finance teams, the work lives in the 5 percent of items that miss. This is how to evaluate the category honestly. ![A bald cartoon cartoon strawman in a traffic-cop hat at a four-way intersection of paper documents, waving the matching items through in dusty blue and pulling the mismatched items aside in warm rust to inspect one in his hand.](https://flowrunner.ai/images/blog/reconciliation-tools-hero.webp) The phrase “reconciliation tools” sells two completely different products to two completely different buyers, and the search results treat them as one category. That confusion costs CFOs evaluation cycles they cannot afford. Open the top results today and you find [Trintech](https://www.trintech.com/), [BlackLine](https://www.blackline.com/products/financial-close/account-reconciliations/), [Numeric](https://www.numeric.io/product/reconcile), HighRadius, Sage, and a Gartner Magic Quadrant. All of them are real products, all of them describe themselves as reconciliation software, and almost all of them are selling the same thing: financial close management for controller teams reconciling GL accounts at month end. That is one definition of the category. A second definition lives entirely outside that SERP. It is what a CFO or VP of Finance at a 10 to 500 person company means when they say their team “needs better reconciliation tools.” They are not talking about closing the books faster. They are talking about Stripe payouts that do not match QuickBooks invoices, distributor billbacks against shipments that do not tie out, AP bills that arrive in inboxes and post twice, freight cost allocations that someone is still building in a pivot table. That work happens before anything reaches a close calendar, and a close-management product is not built to live inside it. This article is for the second buyer. If you are buying for the first, the article still tells you where the line is. ### What finance leaders actually mean by “reconciliation tools” Both definitions are legitimate. They just describe different work done by different people on different timelines. **Close-management reconciliation.** Standardized account reconciliations, certification workflows, sign-offs, and tie-outs for the monthly close. The buyer is a controller. The user is an accountant. The output is a certified GL account inside a structured close calendar. [BlackLine](https://www.blackline.com/products/financial-close/account-reconciliations/) and Trintech have built mature products against this definition for years, and a team that has standardized its close on one of them does not migrate. **Operational reconciliation across systems.** Catching mismatches between Stripe and QuickBooks before they become cash problems. Validating distributor billbacks. Detecting duplicate AP entries before a bill posts twice. Allocating freight costs across SKUs. The buyer is a CFO or VP of Finance. The users are AP coordinators, finance analysts, ops leads, and sometimes the CFO directly. The output is not a certified account. It is a clean handoff between systems that should agree but routinely do not. The buyers and the products in the second category exist, but they are scattered across other categories. AP automation tools. Payment-platform integrations. Custom n8n flows. Spreadsheets. Email chains. None of them shows up in a “reconciliation tools” search result, because the SERP is dominated by the first definition. The result is that the CFO doing the search reads ten close-management vendor pages and concludes either that the category does not solve their actual problem, or worse, that they need to buy a close platform they do not need. The cleanest framing is to name both definitions up front and let the reader pick which one applies. The rest of this article is mostly about the second. ### Where traditional reconciliation tools fit, and where they leave gaps The close-management category has earned its position. Account-level reconciliation matching, certification, audit-ready sign-offs, and sequenced close tasks are real domain depth, built around how accounting teams actually run a structured close. For a controller team running month-end across multiple entities, a purpose-built close product is the right tool. Nothing in this article is going to convince you otherwise, and any vendor suggesting otherwise is selling against their interest. The gap most close-management tools have not closed, and arguably are not trying to close, is the work that happens before items reach the close at all. Specifically: - Real-time exceptions across third-party systems the close platform does not natively integrate with: Stripe webhooks firing seconds after a charge, distributor portals with no public API, payment processors with their own dispute timelines, parser services like Parseur that turn inbox PDFs into structured data - Routing those exceptions to the person who can actually resolve them, on the channel they live in (Slack, email, sometimes WhatsApp for the account manager who never opens the close calendar) - Recording what happened to each exception, by whom, with the context attached, in a way that ties back to the workflow that surfaced it Honest baseline of where most mid-market companies sit today: reconciliation across systems is a mix of spreadsheet exports, ledger pulls, and informal email follow-up, not a fully manual catastrophe and not a fully automated workflow. Finance leaders describe this routine the same way across companies. A CFO running operations at a CPG distributor described the daily reality as exporting from one source system into a CSV, opening a spreadsheet, and reloading the result into the ERP. That is not a process anyone designed. It is a process that emerged because no tool in the stack was built to live in the seam between the systems. ### A different evaluation frame: exceptions, routing, audit Most reconciliation-tool roundups rank by match rate. That is the wrong headline metric for a finance leader whose team already knows that 95 percent of transactions reconcile automatically. The work lives in the remaining 5 percent. Match rate flatters every vendor in the category. It does not separate them. A 96 percent matcher and a 99 percent matcher look like meaningfully different products in a sales deck and produce identical workdays for the AP coordinator clearing the queue. The only number that matters in the exception queue is how fast a real human resolved an item, and that depends on the things match rate does not measure. Four questions separate a category-confused reconciliation tool from a useful one: 1. **Which systems does it connect to natively?** A tool that reconciles two of your six systems is not solving your problem. It is solving a fraction of your problem while shifting the rest to email. 2. **How does it decide an item is an exception?** Threshold rules? Configurable logic? Outlier detection on its own? An AI agent that flags items that look like prior duplicates? Each of these produces a very different exception queue. 3. **Who does it route the exception to, and through what channel?** A queue dashboard nobody opens is functionally the same as no routing. The exception has to find the named person who can resolve it, on the channel they already check. 4. **What does it record for the audit trail?** Specifically: who decided what, when, with what context attached, and is that record queryable in the order it actually happened? That fourth question is where most operational reconciliation breaks down. A spreadsheet does not have an audit trail. A Slack thread is not an audit trail. The exception got resolved, the entry got posted, and three weeks later nobody can tell the auditor or the next CFO who approved which adjustment and why. The seam where the matcher stops and the human resolution begins is its own category of work. It is not the close. It is not the matching engine. It is the orchestration of context, routing, decision, and record. The category name for the system that lives at that seam is an orchestration layer: a platform above the systems of record that listens for what they emit, gathers context the matcher did not have, pulls a human in at the moments that need judgment, and writes the audit trail of who decided what across the whole workflow. [FlowRunner](https://flowrunner.ai/) is built for that layer. That is also why a pause-and-ask pattern beats a batched exception queue for the items that risk actual cash leakage. A batch queue says: here are 47 things to look at tomorrow morning. A pause-and-ask pattern says: this specific Stripe payout does not match this specific QuickBooks invoice by $12.50, here is the customer, here is the fee deduction Stripe took, approve the adjustment or escalate. The reviewer makes one decision with full context, the workflow resumes from the decision, and the audit trail records who decided what at what time. ### Common reconciliation patterns CFOs run into The frame above is useful in the abstract. It is more useful tied to the patterns finance leaders actually run. - **Stripe payments to QuickBooks invoices.** Most match cleanly. Mismatches are usually fee deductions, currency conversion, partial refunds, or customer-side payment splits. The pattern is [automated reconciliation across Stripe and QuickBooks with Slack escalation when amounts or customers do not tie](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) and [Stripe to QuickBooks reconciliation with human-in-the-loop on the mismatches](https://flowrunner.ai/workflows/connect-quickbooks-online-with-stripe). - **AP and invoice ingestion with duplicate-reference detection.** Two AP coordinators open the same inbox on the same morning. Or the same vendor invoice comes in by email and through a portal upload. The pattern is [catching duplicate invoices before they post](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack) by checking parsed invoice metadata against recent history and pausing on outliers. - **Vendor history lookups against an ERP of record to catch double payments.** This is the pattern one CFO described as the place where it would be very easy to double pay if both the company and a distributor made a payment. The orchestration pattern is [NetSuite vendor matching with SuiteQL](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) for a cross-reference before any new bill posts. - **AP automation with near-duplicate detection inside an ERP workflow.** [Near-duplicate detection in AP automation built on Acumatica](https://flowrunner.ai/workflows/automate-with-acumatica) is the same pattern with the ERP doing the heavy lifting on the matching side. - **Distributor billback validation and freight cost allocation.** Almost never fits a close-management tool because the data lives in distributor portals and an account manager’s inbox, not in the GL. The workflow is: pull the billback, check it against the underlying shipment, route to the account manager for validation if anything looks off, post to the ledger only after confirmation. Each of these is an orchestration pattern, not a close pattern. The work happens before the close, and the close inherits clean entries with the resolution context attached. ### How FlowRunner approaches the same problem, honestly FlowRunner is not a close-management replacement. We do not certify GL accounts and we do not run a structured month-end close. A team running [BlackLine](https://www.blackline.com/products/financial-close/account-reconciliations/) or FloQast for close management should keep doing so. We have written that comparison elsewhere ([the honest read on FloQast vs FlowRunner](https://flowrunner.ai/blog/flowrunner-vs-floqast-account-reconciliation-software)), and the conclusion is the same in both directions: the two product categories coexist comfortably for finance teams that need both. What FlowRunner does is the orchestration layer for the work that lives outside the close. Reconciliation patterns are one application. Specifically: - A trigger fires (a Stripe webhook, a new bill in QuickBooks, a parsed invoice landing from a mailbox watcher, a webhook from a distributor portal). - A workflow gathers context from every system it needs. If the match is clean, the workflow writes the entry and moves on. - If the match has an exception (mismatched amount, missing reference, duplicate risk, unusual fee deduction, billback that does not tie to a shipment), the workflow pauses and calls a human as an action. Not a status gate. An actual pause on a specific transaction, routed to a specific named approver on the channel they live in. - The reviewer receives a Slack message (or email, WhatsApp, phone for some patterns) with the full context attached and a structured response. Approve, reject, adjust, escalate. - The workflow resumes from the response, posts the entry with the decision captured, and writes a per-execution record of who decided what at what time. The audit-trail framing in that last bullet is meant to describe what the workflow records, not to make a compliance claim. FlowRunner is designed to give finance leaders a queryable record of what each workflow saw, what it decided automatically, and which human reviewed which exception. That is not the same thing as audit-ready certification for SOX or SOC 2. If your auditor’s scope requires framework attestation, that conversation is separate, and you should look at our [pricing page](https://flowrunner.ai/pricing) and the trust pages we publish before assuming any specific bar. The honest version of FlowRunner’s coverage in this category: we are the orchestration layer between the systems where reconciliation breaks. We are not the close, and we are not pretending to be. For finance teams whose pain is operational reconciliation across multiple systems, the orchestration layer is the right tool. For finance teams whose pain is the close itself, a dedicated close platform like the ones on the SERP today is the right tool. Most growing mid-market finance organizations end up with both, used for different work. ### How to choose, given what you actually reconcile The decision collapses to one diagnostic question, asked honestly: **What are you actually reconciling, and where does the resolution happen today?** If the answer is “GL accounts, certified by my accounting team, inside a structured month-end close,” you are buying close-management software. [Close-management software](https://flowrunner.ai/solutions/close-management-software) is a real category, with mature vendors, and an orchestration layer is not going to do that job better. If the answer is “Stripe payouts, distributor billbacks, AP bills, parsed inbox invoices, freight allocations, or anything else that crosses systems before it reaches the GL,” you are buying an orchestration and exception-routing tool. Evaluate it by integration coverage and exception routing, not by match rate. The questions are the four above: which systems, how exceptions are decided, who routing reaches, and what is recorded. If the answer is both, the realistic stack runs both, used for different work. The clean items hit your close software already tied out. The exceptions arrive with the context already attached, the named reviewer already named, and a record of how each one got resolved before the close team ever opens it. The reconciliation-tools SERP is not going to make that category split obvious. The category is going to keep selling close-management products under a category name that means two different things to two different buyers. The CFO doing the search has to make the split themselves. What separates the buyers who get the evaluation right from the ones who waste a quarter is not which product they buy. It is whether they named the work first. ### Quick answers #### What counts as a reconciliation tool? The term covers two distinct product categories that share a SERP. Close-management platforms like BlackLine, Trintech, and Numeric reconcile GL accounts during the monthly close. Orchestration and exception-routing tools handle the mismatches between systems before anything reaches the close. Most CFOs at companies between 10 and 500 employees mean the second when they search the term. #### Is match rate a useful metric when comparing reconciliation tools? Not on its own. For finance teams running real volume, 95 percent of items reconcile mechanically and the work lives in the 5 percent that miss. The better questions are which systems the tool connects to natively, how it decides an item is an exception, who it routes that exception to, and what it records for the audit trail. #### Do I need a close-management platform and an orchestration layer at the same time? Often yes. Close-management software certifies the books after exceptions are resolved. Orchestration handles the resolution itself across Stripe, QuickBooks, distributor portals, inboxes, and the other systems the close product was never built to reach. The realistic stack runs both, used for different work. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Return Result Block: The Key to Clear & Structured Output Source: https://flowrunner.ai/blog/return-result-block-the-key-to-clear-and-structured-output Guides September 29, 2025 Updated June 19, 2026 5 min read Transform disorganized workflow output into clear, labeled fields that teams can understand and reuse. ![A small pedestal at center holding an ochre nameplate labeled RESULT, with three rows of green structured tags fanning out from the base, the named output and the labeled fields that follow from it.](https://flowrunner.ai/images/blog/return-result-block-the-key-to-clear-and-structured-output-hero.webp) FlowRunner’s Return Result feature addresses a fundamental challenge in workflow automation: transforming disorganized output data into clear, labeled fields that teams can understand and reuse effectively. ### The “Mystery Output” Problem When automation workflows produce unstructured data, it creates downstream problems: confusion about which fields contain actual values, errors in connected systems, and debugging headaches that consume hours. ### Core Functionality Return Result operates as an organizational layer that lets you: - Assign meaningful names to data fields instead of cryptic references - Group related information into logical structures - Display cleaned-up results in FlowRunner’s TestMonitor debugging tool - Simplify integration with downstream automations ### A Practical Example Instead of presenting raw API responses with unclear hierarchy, Return Result lets you define outputs like “LeadEmail,” “CompanySize,” and “LeadScore” as clear, accessible fields. The contrast between messy nested output and structured results is immediate. ### Strategic Value The feature reflects FlowRunner’s broader philosophy: making automation development feel more like professional software engineering, which is what [Orchestration as a Service](https://flowrunner.ai/concepts/orchestration-as-a-service) requires at scale. Structured outputs reduce friction when sharing workflows with teammates or delivering client solutions. Explore Return Result in your flows through the TestMonitor interface. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Choosing a Salesforce Data Cleansing Tool: What the Term Hides Source: https://flowrunner.ai/blog/salesforce-data-cleansing-tool Articles May 28, 2026 Updated May 30, 2026 11 min read A Salesforce data cleansing tool means three different jobs. Which one is yours decides whether you need a batch matcher, native rules, or an intake layer. ![A forest-green hand-cranked sorting machine feeding mixed paper records, with an amber inspection window holding two near-identical pages under a magnifier while clean sheets stack below in teal and doubtful pages go to a rust side tray.](https://flowrunner.ai/images/blog/salesforce-data-cleansing-tool-hero.webp) The phrase “Salesforce data cleansing tool” is doing more work than it looks, and picking the wrong tool starts with not noticing that. The term bundles three different jobs, and the right product for one of them is the wrong product for the other two. Before you compare vendors, the question worth answering is which job is actually yours. This article is about that question. Not a ranked list of cleansing apps, but a way to tell what you are really buying, and where a tool that checks records at intake fits next to one that scrubs them in bulk. ### What sales ops actually means by a Salesforce data cleansing tool Three separate jobs hide under the one phrase: - **Deduplication.** Finding the records that are the same real-world entity under different spellings, emails, or domains, and merging them into one. This is the job most people picture when they say cleansing. - **Field normalization.** Making the values consistent. Country names, industry picklists, job-title variants, phone formats. The data is not duplicated, it is just inconsistent enough to break reporting and routing. - **Stale-record detection.** Finding the records that have gone cold or were never finished. Missing required fields, no activity past a threshold, owners who left the company. Most of the dedicated tools on the AppExchange lead with the first job, and they run it as a scheduled batch: pull a large set of records, score them against matching heuristics, surface or auto-merge the duplicates, then run again next week or next quarter ([AppExchange data cleansing category](https://appexchange.salesforce.com/explore/business-needs?category=dataCleansing)). For a one-time cleanup of a historical mess, that batch shape is exactly right. A specialized matching engine running fuzzy logic across hundreds of thousands of rows is purpose-built work, and nothing in this article suggests doing that by hand. Here is the part most cleansing-tool roundups skip. A batch sweep fixes records after they have already broken. Between sweeps, new leads keep arriving: web forms drop them in, enrichment vendors push them back, reps create accounts by hand, list imports load a few thousand in an afternoon. Every one of those is a fresh chance to create the next duplicate, the next malformed phone number, the next half-filled record. The sweep cleans the cohort, and the intake quietly refills it. That gap, the time between when a bad record lands and when the next sweep catches it, is where the recurring cleanup project actually comes from. So the first evaluation question is not “which tool has the best match rate.” It is: **is my primary problem the historical mess that already exists, or the intake quality that keeps producing more of it?** Those are different problems, and they point at different tools. ### Where FlowRunner fits relative to dedicated cleansing tools Let me be precise about what FlowRunner is and is not here, because the honest framing is the whole point. FlowRunner is an orchestration layer, not a record-matching engine. It does not replicate the fuzzy-matching scale of a specialized tool built to dedupe a million-row historical database. If that is your job, use the specialized tool. What FlowRunner does is sit one step upstream of the Salesforce write, intercept the intake event, check the candidate record against Salesforce before it lands, and pull a human in when the match is ambiguous. It enforces hygiene at the point of creation rather than sweeping for it later. The two approaches answer different questions, so a side-by-side is the clearest way to see the split. | | Batch cleansing tool | FlowRunner (intake orchestration) | | --- | --- | --- | | **Primary job** | Dedupe and standardize records that already exist | Catch duplicate, malformed, low-confidence records before they land | | **When it runs** | On a schedule (weekly, monthly, quarterly) | At the moment of intake or write, on every record | | **Best at** | Large historical cleanup, fuzzy matching at scale | Preventing the next bad record from being created | | **Ambiguous matches** | Surfaced in a report or auto-merged by heuristic | Routed to a person in Slack with both records in context | | **Audit trail** | The tool’s own merge log | Per-record execution log of every check and decision | | **What it does not do** | Stop bad records arriving between runs | Replace fuzzy matching across a giant existing database | Read across the bottom row. Neither approach does the other one’s job. A batch tool cannot police the intake; an orchestration layer is not built for million-row historical fuzzy matching. The teams that get this right do not pick one. They run native rules underneath, a batch engine for the historical cleanup if they have one, and an intake layer for the ongoing quality. The three do not compete. For the deep treatment of the intake-as-control-surface idea, the companion piece on [treating Salesforce data hygiene as a control surface rather than a cleanup project](https://flowrunner.ai/workflows/salesforce-data-hygiene) walks the same architecture from the hygiene angle. This article stays on the evaluation question: which tool for which job. ### What the intercept-at-intake pattern looks like The mechanic is straightforward, and it leans on Salesforce actions that already exist in the integration. FlowRunner’s Salesforce connector exposes Find Record by Query, Find Records by Query, Create Lead, Update Lead, and Update Record, among others, and those can be sequenced with conditional logic and a human review step ([FlowRunner Salesforce integration](https://flowrunner.ai/integrations/salesforce-pro)). The shape: 1. **Trigger on the intake source, not on the Salesforce save.** A form submission, an enrichment webhook, a list-import row. The candidate record is in flight, not yet committed to Salesforce. 2. **Normalize the candidate in the workflow.** Standardize the country code, map the industry picklist, format the phone, lower-case the company name, extract the domain from the email. This happens on the way in. 3. **Query Salesforce for matches.** Run Find Record by Query on email, normalized company name, and domain. The query returns zero, one, or several candidates. 4. **Branch on match confidence.** A clean single match updates the existing record. A no-match creates a new one. An ambiguous result, several plausible matches or a single match where critical fields disagree, goes to a person. 5. **Resolve the ambiguous case in Slack.** Post both records side by side to the channel sales ops watches, with the overlapping and conflicting fields highlighted. The reviewer picks merge, create, or update. FlowRunner executes the chosen Salesforce action and logs the decision. ![FlowRunner workflow canvas for the intercept-at-intake pattern](https://flowrunner.ai/images/blog/salesforce-data-cleansing-tool-1.webp) The branch in step four is where the design lives. The logic is a confidence threshold, and the shape of that condition matters more than any other choice in the build. Illustratively, the branching config reads like this: ``` { "branch_on": "match_result", "routes": [ { "when": "match_count == 0", "do": "Create Lead" }, { "when": "match_count == 1 && conflicting_fields == 0", "do": "Update Lead" }, { "when": "match_count >= 2 || conflicting_fields >= 2", "do": "human_review_slack" } ] } ``` That snippet is an illustration of the decision logic, not a literal export of FlowRunner’s internal workflow format. The exact node-and-edge JSON shape is a platform detail best read from the [FlowRunner documentation](https://docs.flowrunner.ai) rather than reproduced here, and the build-order specifics live in the [Salesforce plus Slack human-in-the-loop workflow guide](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack). What is real and verifiable is the set of Salesforce actions being sequenced and the fact that a human review step sits between the query and the write. ### Why the ambiguous cases need a human, not a better heuristic The hard cases in CRM cleansing are not “definitely the same” or “definitely different.” Those are easy, and any matching engine handles them. The hard cases are “probably the same, but the email is different, the company name is spelled three ways, and one record has a phone the other does not.” Resolving that well usually means looking at context a matcher does not weigh: where the lead came from, what the enrichment vendor returned, whether there is prior activity on one of the records. A sales ops reviewer with that context in front of them makes a better call than a confidence score does, because the score is guessing at exactly the thing the human can see. This is not a knock on matching engines. It is the boundary of what a heuristic can know. ![Slack review card posted by FlowRunner for an ambiguous match](https://flowrunner.ai/images/blog/salesforce-data-cleansing-tool-2.webp) This is the part most posts on Salesforce data cleansing will not say plainly: a cleansing tool cannot fix the intake that produces the mess, and no matching heuristic, however good, will resolve the cases that actually need judgment. Cleansing is partly an automation problem and partly an exception-handling problem. The exception-handling part is where the recurring debt accumulates, and it is the part a batch sweep is structurally unable to own, because by the time the sweep runs the ambiguous record is already months old and already worked by a rep. That gap, the ongoing intake quality and the judgment calls inside it, is a seam no native rule and no batch tool was built to hold. It is cross-system coordination: an intake event from one place, a query against the system of record, a person in a messaging channel, a write back. That coordination is the work between systems, and it is exactly the kind of work that grows as a sales stack matures. An orchestration layer is the category that owns that seam, a system that sits above the systems of record, listens for what they emit, gathers the context the native rules never had, and pulls a person in at the moments that need a decision. FlowRunner is built for that layer. Salesforce keeps its scope as the system of record. The batch tool keeps its scope as the historical scrubber. The orchestration layer takes the intake. ### What this approach trades off No architecture is free, and the honest read includes the costs. - **It is slower than a batch tool for one-time historical cleanups.** If your job is deduping six years of accumulated records across hundreds of thousands of rows, a specialized matching engine is faster and more thorough. Use it. Run the intake layer afterward so the cleaned database stays clean. - **You define the match and normalization rules up front.** A dedicated tool ships prebuilt matching heuristics tuned by the vendor. With an orchestration layer, you decide which fields, in which combinations, count as a match, and what turns “possible duplicate” into “send to a human.” That is more design work at the start. It is also why the result fits your data instead of a vendor’s averaged assumptions. - **A human review gate adds latency.** Ambiguous records wait for sales ops. For most intake that is the right trade, since a record that needs judgment gets it before it lands. For high-velocity inbound where a five-minute delay costs revenue, tune the ambiguity rule to escalate fewer borderline cases. - **It only polices intake paths that route through it.** If a rep hand-creates records directly in Salesforce, only the native rules apply. The workflow governs the paths wired to it, not every keystroke in the org. - **It pays back differently than a sweep.** A batch tool shows its value in a big one-time number of merged duplicates. The intake layer shows its value as a shrinking cleanup surface over months, plus a reviewable per-record audit trail. The payoff is real but slower to read on a dashboard. The match query is the load-bearing piece. Weak criteria flag too much as ambiguous and bury the reviewer. Strict criteria let near-duplicates slip through as no-match and become next quarter’s dedupe problem. That query deserves real design time, not a default. ![FlowRunner execution log filtered to the intake-hygiene workflow](https://flowrunner.ai/images/blog/salesforce-data-cleansing-tool-3.webp) ### Keep the native rules; they are the baseline None of this displaces what Salesforce already gives you. The platform ships native [duplicate rules and matching rules](https://help.salesforce.com/s/articleView?id=sf.duplicate_rules_overview.htm&type=5) that match incoming records and either alert the user or block the save, and they run across the standard Sales Cloud editions (Starter, Pro Suite, Enterprise, Unlimited, Unlimited Plus, and Developer). Every org should keep them on. They are the last-mile check on whatever any tool writes. What they do not do is resolve ambiguous matches or run before a record reaches Salesforce. A duplicate rule at save time either alerts the rep, who often lacks the context to decide, or blocks the save, which loses the record with no path to resolution. That is not a flaw. It is the documented scope of a save-time check. The intake layer adds an upstream check; it does not replace the native one. The same pattern works one CRM over, where you can [flag duplicate contacts before they reach HubSpot](https://flowrunner.ai/workflows/orchestrate-google-forms-and-hubspot-and-slack) from a Google Form intake. The system of record changes; the shape of the work does not. ### How to evaluate this against a dedicated cleansing tool Four questions sort the decision faster than any feature matrix. 1. **Is the primary problem historical mess or ongoing intake quality?** Historical mess points at a batch matching engine. Ongoing intake quality points at an orchestration layer. Most established orgs have both, in which case the answer is both, used for different work. 2. **Do reviewers need context to resolve duplicates?** If your ambiguous cases get resolved well only when someone can see lead source, enrichment data, and prior activity, a tool that surfaces matches in a report is weaker than one that routes the case to a person with the records side by side. If your duplicates are clean enough to auto-merge by rule, the report is fine. 3. **Does the team already live in Slack for decisions?** A human review step is only useful if the human actually sees it. If sales ops works out of Slack, routing ambiguous matches there fits the existing habit. If approvals happen elsewhere, route them elsewhere, but make sure the channel is one people watch. 4. **Where does the audit trail need to sit?** Inside Salesforce field history, in the cleansing tool’s merge log, or in an orchestration log that records every check and human decision per record. If the question “why did this account end up linked to this contact, and who decided” needs an answer six months later, decide now where that answer will live. ### The takeaway “Salesforce data cleansing tool” is not one product category, it is three jobs wearing one name, and the batch-sweep tools that dominate the search results are built for only one of them. Deduplicating a historical database is real, specialized work, and a dedicated matching engine does it well. Keeping the database clean afterward is a different job, and a batch sweep cannot do it because it runs after the damage, not before. So the decision is not which cleansing tool wins. It is which jobs you have. If yours is a one-time historical scrub, buy the matching engine. If yours is the steady drip of malformed and duplicate records at intake, the better shape is a layer that checks every write, resolves the easy cases automatically, and routes the judgment calls to a person before anything lands. Keep the Salesforce native rules underneath either way. The records belong in the CRM. The work of keeping them clean as they arrive belongs to the layer above it. ### Quick answers **What is the best Salesforce data cleansing tool?** There is no single best one, because the term covers three different jobs. For a one-time historical cleanup across hundreds of thousands of records, a dedicated matching engine like the ones on the AppExchange is the right fit. For preventing bad records at intake, an orchestration layer that checks every write is the better shape. Most teams need both, plus Salesforce native duplicate rules underneath. **Can FlowRunner replace a dedicated Salesforce data cleansing tool?** No, and it is not meant to. A specialized matching engine is built for large historical dedupe at scale. FlowRunner is an orchestration layer that intercepts intake and writes, checks records against Salesforce before they land, and routes ambiguous matches to a person. It complements a matching engine and the native rules rather than replacing either. **Does Salesforce have built-in data cleansing?** Salesforce ships native duplicate rules and validation rules that match and reject records at save time across the standard Sales Cloud editions. They are a real baseline every org should use. They do not resolve ambiguous matches or run before a record reaches Salesforce, which is where most cleansing debt accumulates. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Salesforce Intelligent Document Processing: Who Owns the Workflow When the Extraction Is Wrong Source: https://flowrunner.ai/blog/salesforce-intelligent-document-processing Articles May 28, 2026 Updated May 30, 2026 12 min read Salesforce can read a contract or order PDF. The harder question for finance is who reviews a low-confidence extraction before it becomes a CRM record. ![An editorial cartoon of forest-green weighing scales dropping check-marked olive document pages into a CRM in-tray while two doubtful amber pages are tipped onto a rust side shelf with a hand-bell to summon a reviewer.](https://flowrunner.ai/images/blog/salesforce-intelligent-document-processing-hero.webp) Search “salesforce intelligent document processing” and the first screen is Salesforce selling you Salesforce: Intelligent Document Automation, Intelligent Document Reader, Document AI inside Agentforce. All of it real, all of it answering one question. Can AI read the PDF? Yes. That question was settled a while ago. It is not the question a finance leader is actually asking when an order form, a contract, or a distributor billback needs to become a record in the CRM. The question that decides whether this works is the unglamorous one: who owns the workflow when the extraction is wrong? A document is read, fields come out, and now a record is about to be created or updated in Salesforce on the strength of what a model believed. If the model was sure and correct, nobody needs to do anything. If it was sure and wrong, or unsure and right, somebody in finance needs to see it before it lands. This article is about that somebody, and about the workflow that knows when to call them. ### What CFOs actually mean by intelligent document processing for Salesforce Strip the vendor language and the scope is narrow and concrete. There is a set of recurring documents that need to land as Salesforce records: order forms that map to opportunities or custom order objects, signed contracts that attach to accounts, distributor billback statements with line items, vendor forms that update a record finance already has. The documents arrive in the messy real world (email, a shared drive, a vendor portal, an upload) and need to end up as structured data in the system of record. The finance-side pain is specific, and it is not “we cannot read PDFs.” In conversations with finance leaders, the same picture keeps surfacing. The existing AI effort is piecemeal, with no broad strategy behind it, and the opaque vendor tools they have looked at read like a black box they are asked to trust. Underneath that, the day-to-day is more ordinary. Distributor billing has real errors in it, and if both sides post a payment against the same statement, that is a duplicate payment, not a rounding issue. Near month-end, items fall through the cracks. And when the person who normally handles a vendor’s bills is out, a senior finance staffer enters them by hand, because the judgment in that work has never actually been removed, only assigned to whoever is around. That last detail is the whole story. The judgment did not go away. Intelligent document processing for Salesforce is not the project of making the judgment disappear. It is the project of letting the routine documents flow into Salesforce on their own and getting the judgment cases in front of a person, on purpose, before they become records. ### Native Salesforce IDP versus an orchestrated workflow Because the SERP is dominated by Salesforce’s own features, the distinction is worth drawing flat before anything else. Salesforce native document AI and an orchestrated IDP workflow are not the same product solving the same problem at different price points. They sit at different layers, and one is built to call the other. | | Salesforce native document AI | Orchestrated IDP workflow (FlowRunner) | | --- | --- | --- | | **Core job** | Read a document, extract and classify fields | Coordinate intake, extraction, validation, the write, and human review around it | | **Where the document comes from** | Documents already inside Salesforce or pushed to it | Any source: email, shared drive, vendor portal, upload | | **Validation against existing records** | Limited to what the configured feature performs | Explicit query against Salesforce before the write to avoid duplicates and match parents | | **Low-confidence result** | Handled inside the feature’s own logic | Pauses the workflow and routes the document to a named reviewer with context | | **Who owns the threshold** | Configured within the Salesforce feature | Finance owns it as a workflow rule, editable without a developer | | **The extractor itself** | Salesforce’s model | Pluggable: can be Salesforce’s own document AI, or another extractor | Read the bottom row. The orchestration layer does not replace the extractor. It can call Salesforce’s native [Intelligent Document Automation and Intelligent Document Reader](https://help.salesforce.com/s/articleView?id=ind.document_automation.htm&type=5) as the reading step and then own everything around it. If Salesforce’s document AI is the best reader for your contracts, use it as the reader. The workflow question (where the document came from, whether it matches an existing record, what happens when the extraction is shaky) lives one layer up regardless of whose model does the reading. ### The four moving parts of a defensible IDP workflow Every workflow that lands document data in Salesforce without a reviewer regretting it later is built from four parts. Naming them separately is the point, because the value lives in the seams between them, not in any one box. - **Intake.** The triggering event. An email arrives from a known distributor, a file lands in a watched drive folder, a form posts from a vendor portal, someone uploads a contract. The document is in flight, not yet anything in Salesforce. - **Extraction.** A document model converts the file into structured fields. This is the part the search results are obsessed with, and it is genuinely the easiest of the four to source. It can be Salesforce’s own document AI, a dedicated parser, or an LLM-backed extraction prompt. What matters downstream is that it returns confidence per field, not one score for the whole document. - **The Salesforce write.** The extracted data becomes a record. Depending on the document, that is a new record, an update to an existing one, or a parent record with child line items. This is where the connector earns its keep, and where matching against what already exists prevents the duplicates. - **Exception handling.** When the extraction is low-confidence or a business rule is violated, the document does not write. It pauses and routes to a person, with the extracted fields and the source document attached, and resumes once they decide. The brief most teams write for this kind of project spends ninety percent of its words on extraction and a sentence on the other three. Reverse that ratio and you have a workflow finance will actually trust. ### Why a black-box agent is the wrong shape for finance Here is the part most write-ups on Salesforce IDP will not say plainly. The industry frames the goal as accuracy, as straight-through processing, as the percentage of documents that never touch a human. Push the accuracy number high enough and the exceptions vanish, the story goes. But the exceptions are not the failure case. The exceptions are the work. A higher accuracy number does not retire the judgment in a duplicate billback or a contract whose terms do not match the account on file. It just moves that judgment somewhere quieter, and quieter is exactly where finance does not want it. This is why a single self-directed agent that ingests a document and decides on its own what to write is the wrong shape here, however capable it is. The finance leaders describing their AI efforts as a black box are not being technophobic. They are naming a real requirement: they need to see each step, and they need a defined place where the work that requires a person stops and waits for one. The unit of trust is not the model’s average accuracy across a thousand documents. It is the trail behind this document, the one about to become a payable record. That requirement is what separates the extraction step from the layer above it. Reading the document is a contained task, and Salesforce, a parser, or a model can each do it well. Deciding what the read result means against everything else is a different job. The account it should attach to, the billing period that may already have posted, the contract version on file, the reviewer who signs off when the numbers are soft: that is coordination across systems and people no single document-AI feature was built to own. That coordination is a category of its own. An orchestration layer is the system that owns it: it picks the document up wherever it arrives, runs it through whichever extractor fits, checks the result against the system of record, and stops to bring a person in at the one moment that needs a person. FlowRunner is built for that layer. Salesforce stays the system of record and, if you choose, the document reader. The orchestration layer takes the workflow between the inbox and the record, which is the part that was never going to fit inside the CRM. For the broader framing of why coordinated automation beats a do-everything agent, the piece on [AI automation versus AI agents](https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents) draws the same line at the category level. ### What FlowRunner does at each step The mechanics are concrete, and they lean on Salesforce connector actions that already exist. FlowRunner’s Salesforce integration exposes Find Record by Query, Find Record, Find Child Records, Create Record, Create Child Records with line item support, Update Record, and Create Attachment, among others ([FlowRunner Salesforce integration](https://flowrunner.ai/integrations/salesforce-pro)). Mapped to the four parts: - **Intake.** A trigger on the document source. For recurring vendor documents that arrive by email, that is a mailbox trigger filtered by sender and subject. For portal or drive sources, the corresponding trigger fires the run. - **Extraction.** The reading step returns structured fields with a confidence score per field. The extractor is your choice; the workflow expects per-field confidence so it can route on the field that matters rather than on an average. - **Match before you write.** Before creating anything, **Find Record by Query** checks Salesforce for the account or contract the document refers to, so a known distributor updates the existing account instead of spawning a second one. **Find Child Records** pulls the related agreements or prior line items the new document should reconcile against. - **Write the record and its lines together.** A clean extraction with a matched parent uses **Create Record** for the header and **Create Child Records (with line item support)** for the lines, or **Update Record** when the document corresponds to an existing opportunity or custom object. **Create Attachment** keeps the source document on the record. Salesforce supports writing a parent and its related children in a single request through its composite request patterns ([Salesforce composite REST API](https://developer.salesforce.com/docs/atlas.en-us.api_rest.meta/api_rest/resources_composite.htm)), which is what the Create Child Records action is built on. - **Pause for a person when it is not clean.** When a field’s confidence is below the threshold, or a finance rule is violated (a billing period that already posted, a total that does not match the line items), the workflow does not write. It pings a reviewer in Slack or by email with the extracted fields and the document attached, and resumes once they confirm, correct, or reject. The decision rule in the fourth step is the part finance owns. Illustratively, the routing logic reads like this: ``` { "branch_on": "extraction_result", "routes": [ { "when": "min_field_confidence >= 0.90 && period_already_posted == false", "do": "write_salesforce_record" }, { "when": "min_field_confidence < 0.90 || header_vs_lines_mismatch == true", "do": "route_to_reviewer" }, { "when": "period_already_posted == true", "do": "route_to_reviewer_duplicate_risk" } ] } ``` That snippet is an illustration of the decision logic, not a literal export of FlowRunner’s internal workflow format. The exact node-and-edge shape is a platform detail best read from the [FlowRunner documentation](https://docs.flowrunner.ai) rather than reproduced here. What is real and verifiable is the set of Salesforce actions being sequenced and the fact that a human review step sits between the extraction and the write. ![FlowRunner workflow canvas for document-into-Salesforce processing](https://flowrunner.ai/images/blog/salesforce-intelligent-document-processing-1.webp) ### A worked example: distributor billback statements into Salesforce Make it concrete with the document finance leaders raise most, because it carries real money risk: the distributor billback statement. A statement arrives by email from a distributor finance already works with. The intake trigger fires on the sender and subject. The extraction step reads the statement and returns the distributor name, the billing period, and an array of line items, each with a description, an amount, and a confidence score. The distributor name comes back clean. The period comes back clean. Two of the eleven line items come back soft, because the statement had a credit adjustment that overlapped the amount column on the page. **Find Record by Query** matches the distributor name to the existing Salesforce account, so nothing duplicates. **Find Child Records** pulls the related billback agreements and the prior periods already on the account. The workflow computes two derived signals: whether this billing period already posted (it has not) and whether the sum of the line items reconciles with the statement header within tolerance. The two soft line items fail the confidence check, so the workflow does not write. It packages the extracted fields, the per-field confidence, the reconciliation delta, and the source PDF. Then it posts a card to the finance channel: “Billback from \[distributor\], period clean, 2 of 11 line items below confidence, header reconciles within tolerance.” The reviewer opens the statement in the same click, confirms the two amounts against the page, and approves. The workflow resumes. **Create Record** writes the billback header against the matched account, **Create Child Records (with line item support)** writes the eleven lines in one request, and **Create Attachment** keeps the statement on the record. Every step is logged: which account was matched, which fields were soft, who approved them, when. If anyone asks in October why this billback posted the way it did and who signed off, the answer is in the run, not in someone’s memory. The other distributor statements that came in clean that morning never paused at all; they were records before anyone opened Slack. ![Slack review card posted by FlowRunner for a billback needing a decision](https://flowrunner.ai/images/blog/salesforce-intelligent-document-processing-2.webp) This is the same exception-routing spine that finance teams use for [parsed invoices flowing into the ERP without manual entry](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) and for [vendor documents flowing into NetSuite](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack). The system of record changes; the shape of the work does not. And it is adjacent to, but distinct from, keeping the records themselves clean: the companion piece on [treating Salesforce data hygiene as a control surface](https://flowrunner.ai/workflows/salesforce-data-hygiene) covers the intake-quality problem for leads and contacts, where the question is duplicate matching rather than document extraction. The [Salesforce plus Slack human-in-the-loop workflow](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack) documents the reviewer-side experience in detail. ### Edge cases and trade-offs worth naming No architecture is free, and the honest read includes where it stops. - **The confidence check is a filter, not a correctness oracle.** A field the extractor was sure about but got wrong (because a layout shifted invisibly and a credit landed in the amount column) writes without pausing. The defenses are a reconciliation check (does the header match the sum of the lines) and sampling (route every Nth clean document through review anyway). Neither is perfect; both are layered safety nets on top of the confidence branch. - **It only governs intake paths that route through it.** If someone keys a billback directly into Salesforce by hand, only the native Salesforce controls apply. The workflow polices the document sources wired to it, not every record in the org. - **It does not replace Salesforce’s native document AI or its native rules.** If Salesforce’s reader is your extractor, the workflow orchestrates it. Salesforce duplicate and validation rules keep running as the last-mile check on whatever the workflow writes. The orchestration layer adds an upstream check; it does not retire the ones inside the CRM. - **A human review gate adds latency.** Soft documents wait for a reviewer. For finance documents that carry payment risk, that is the right trade, because the document that needed judgment gets it before it posts. Where speed matters more than scrutiny, tune the threshold to escalate fewer borderline cases. - **The audit trail is a platform behavior, not a compliance certification.** FlowRunner records each step of each run, which gives finance a per-document trail of what was matched, what was soft, and who approved it. That is designed to meet common expectations for a record per document. It is not, on its own, a statement that the workflow is a certified control under a specific framework. Treat the log as evidence, and pair it with your own controls where an audit requires it. ![FlowRunner execution log filtered to the document-into-Salesforce workflow](https://flowrunner.ai/images/blog/salesforce-intelligent-document-processing-3.webp) The load-bearing decision in the whole build is the threshold and the business rules in the branch. Set the bar too low and the reviewer drowns in documents that did not need them. Set it too high and soft extractions post as records and surface as problems at month-end. That rule deserves real design time, and it belongs to finance, not to whoever wired the workflow. ### How to evaluate an IDP approach for Salesforce Four questions sort the decision faster than any feature comparison. 1. **Can you see and edit the workflow, or is the logic locked inside a vendor’s agent?** If the routing and the threshold live somewhere you cannot read, you have bought the black box finance leaders already distrust. The logic that decides what posts and what pauses should be visible and editable. 2. **Can a non-developer in finance change a routing rule?** When the distributor list changes, or a new document type arrives, or the tolerance needs tightening, the person who owns the numbers should be able to change the rule without filing a ticket. 3. **Does the exception path produce something an auditor would recognize as a control?** A per-document log of what was matched, what was soft, and who approved it is a control. A model’s chat transcript is not. Decide where the answer to “who approved this and why” will live before you need it. 4. **Does it work with the document sources you already use?** Documents arrive by email, in shared drives, through vendor portals. An approach that only reads what is already inside Salesforce has skipped the hardest part of the workflow, which is everything before the document reaches the CRM. ### The takeaway Salesforce can read the document. That was never the hard part, and the search results that lead with it are answering a question finance settled a while ago. The hard part is the workflow around the read: pulling the document from wherever it actually arrives, checking it against the records you already have, and putting the soft cases in front of a person before they become payable records. That work is coordination across the inbox, the extractor, the system of record, and the reviewer, and it is the part that grows as the document volume grows. Keep Salesforce as the system of record. Use its document AI as the reader if it fits your documents. Then put the orchestration layer above it: the layer that lets the clean documents flow into Salesforce on their own and routes the judgment cases to the people who own the judgment, with everything they need attached and a trail behind every decision. The records belong in the CRM. The work of deciding which documents become records belongs to the layer above it. ### Quick answers **Does Salesforce have intelligent document processing built in?** Yes. Salesforce ships document AI capabilities such as Intelligent Document Automation and Intelligent Document Reader that extract and classify data from documents inside the platform. They are real and worth using. What they do not own is the cross-system workflow: pulling documents from the inboxes and portals they arrive in, validating an extraction against existing Salesforce records, and routing a low-confidence result to a named reviewer before it becomes a record. **Can FlowRunner replace Salesforce Einstein or Agentforce for document processing?** No, and it is not meant to. FlowRunner is an orchestration layer that coordinates document sources, an extractor (which can be Salesforce’s own document AI), the Salesforce write, and a human reviewer. It sits above the extraction step and governs what happens around it. If Salesforce’s native document AI is the right extractor for your documents, FlowRunner orchestrates it rather than competing with it. **How does intelligent document processing avoid creating bad Salesforce records?** By putting a confidence check and a business-rule check between the extraction and the write. A clean, high-confidence extraction that matches an existing account creates or updates the record automatically. A low-confidence field, or one that violates a finance rule like a duplicate billing period, pauses the workflow and routes the document to a reviewer with the extracted fields and the source attached. Nothing lands in Salesforce until it clears the bar. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Salesforce Lead Assignment Rules: What They Do, Where They Stop Source: https://flowrunner.ai/blog/salesforce-lead-assignment-rules Articles May 25, 2026 Updated May 28, 2026 9 min read How Salesforce lead assignment rules work, the API and post-create gaps that quietly drop leads, and what to put around them so reps see the right inbound. ![A bald cartoon traffic cop in dusty blue directing little envelope figures down assigned streets, with one amber envelope wandering off the road that he has just noticed.](https://flowrunner.ai/images/blog/salesforce-lead-assignment-rules-hero.webp) Salesforce assignment rules failing quietly in production is the single most common reason a sales org’s lead routing looks fine on paper and feels broken at the rep level. The rule itself does exactly what it was designed to do. The breakdown is in the gap between what the rule handles and what every growing sales stack actually asks of its routing layer. ### What Salesforce lead assignment rules are Lead assignment rules are a native Salesforce feature that auto-assigns incoming Leads to a user or queue based on criteria you define. They are the default mechanism Salesforce expects you to use for inbound lead distribution. Each object (Leads, Cases) supports multiple defined rules, but only one rule can be active at a time. Inside the active rule, you create rule entries; each has its own criteria and assignee, and the first match wins. The rule fires when a Lead is created or edited with the _Assign using active assignment rule_ checkbox selected, when a Lead enters through [Web-to-Lead](https://help.salesforce.com/s/articleView?id=sf.leads_setting_web.htm&type=5), or when a Flow or Apex trigger explicitly invokes it. Cases have a parallel system; this article focuses on Leads. ### How a rule actually evaluates a lead Salesforce walks the rule entries top to bottom and stops on the first match. No scoring, no weighting. First match wins. Criteria are field-based (Country, State, LeadSource, Industry, custom fields) or formula-based. Each assignee is a single user or a queue; queues let multiple reps pull from a shared pool. Anything that matches nothing falls through to the [Default Lead Owner](https://help.salesforce.com/s/articleView?id=sf.leads_setting_default_lead_owner.htm&type=5), which acts as the safety net. A real rule for a small enterprise sales team might look like this: | Sort | Criteria | Assignee | | --- | --- | --- | | 1 | Existing\_Customer\_\_c equals TRUE | Account Manager queue | | 2 | Country in (United Kingdom, Ireland) | EMEA Inbound Queue | | 3 | Industry equals “Healthcare” AND NumberOfEmployees > 500 | Healthcare Enterprise Queue | | 4 | LeadSource equals “Partner Referral” | Partner Channel Queue | | 5 | (catchall) | Default Lead Owner: SDR Manager | Row 1 is the article’s thesis on a single line. Native rules only read fields already on the Lead at the moment they fire, so catching existing customers means a separate enrichment job has to populate `Existing_Customer__c` before the lead is created or edited. Sophisticated routing always depends on getting the right field there first. The bottom row is the catchall. ### Setting them up: the practical steps Available across the current Sales Cloud lineup (Starter, Pro Suite, Enterprise, Unlimited, Unlimited Plus, and Developer), per Salesforce’s [setup documentation](https://help.salesforce.com/s/articleView?id=service.creating_assignment_rules.htm&type=5). The interface is the same across editions. 1. **Find the setup.** Setup > Feature Settings > Marketing > Lead Assignment Rules. (Cases are under Service Setup.) 2. **Create the rule.** Name it for intent, not object (“Inbound Lead Routing v3” beats “Lead Rule”). 3. **Add rule entries with criteria and assignee.** Keep criteria specific; permissive entries absorb leads that should have hit a more specific entry below. 4. **Set the sort order.** First match wins, so order from most specific to most general. 5. **Activate the rule.** Only one rule per object can be active. Activating a new rule deactivates the previous one without warning. 6. **Expose the assignment checkbox where Leads get created.** Add _Assign using active assignment rule_ to the Lead page layout, or set it as default on the create paths your team uses. Without this, manual lead creation bypasses the rule entirely. 7. **Confirm the Default Lead Owner is set and active.** A misconfigured Default Lead Owner is one of the most common reasons leads disappear. 8. **Test with sample records before pointing real inbound at it.** Create test leads that hit each entry, plus a few that fall through to the default. [Salesforce’s documentation on setting up lead assignment rules](https://help.salesforce.com/s/articleView?id=service.creating_assignment_rules.htm&type=5) covers the click path. The piece worth internalizing is not the setup; it is the discipline of writing the routing logic down before you click anything. ### Common routing patterns teams build with them The shapes most growing sales orgs land on: - **Geographic territory routing** by Country or State, sometimes layered with a region custom field. - **Round-robin within a region** using queues plus a custom counter field and a Flow or Apex job. Round-robin is not native; you assemble it on top of queues. - **Routing by Lead Source** so partner and event leads land with the right team. - **Segmenting by company size or industry** to split named-account leads from SMB inbound. - **A holding queue for unknown or low-fit leads** so an SDR triages them before they consume AE capacity. None are exotic. All sit inside what assignment rules can express. The trouble starts when the routing decision depends on data the rule does not have at the moment it fires. ### Where assignment rules stop being enough Here is what most posts on lead routing will not say plainly: the problem is rarely the rule. The problem is the architectural assumption underneath it. Assignment rules assume the data they need is on the Lead record at the moment of create or edit. That holds for geography, source, and basic firmographics. It does not hold for: - **Rep capacity and workload balancing.** [Salesforce’s own documentation](https://help.salesforce.com/s/articleView?id=service.customize_leadrules.htm&type=5) is explicit that native rules do not include capacity logic. Round-robin requires custom fields and Flow or Apex. - **Enrichment that arrives after creation.** Firmographic, intent, and behavioral data land seconds to minutes after the lead is created. The rule has already fired. - **Lead scoring, SLA timers, and reassignment.** Scoring runs on a delay. SLA timers and reassignment (rep out, quota hit, lead aging) live outside the rule contract entirely. - **Cross-system handoff.** Slack notifications, mobile pings, calendar holds: none of that is the rule’s job, but all of it is the work. - **Records created via the API.** The [assignment rule header](https://developer.salesforce.com/docs/atlas.en-us.api.meta/api/sforce_api_header_assignmentruleheader.htm) must be set explicitly on the request, or the rule is skipped. In a modern stack that is most leads. Each of these can be patched with custom fields, Flow, Apex, or AppExchange tools. After two or three patches, the routing logic stops living in any one place and starts living in a half-documented mesh. The rule did not break; the layer above it never got built. This is a general property of record-based routing, not a Salesforce complaint. The same pattern appears in HubSpot, Dynamics, and every other CRM that ships a single-fire criteria engine. No native CRM rule will ever own the layer between systems, because no native CRM rule sees outside its own database at the moment of fire. The work between systems, and the human judgment between those systems, is where every growing sales stack actually lives. That seam is where an orchestration layer lives: a system above the systems of record that listens for what they emit, gathers context the rule did not have, and pulls a human in at the moments that need judgment. FlowRunner is built for that layer. ### Extending assignment rules with the rest of the stack Leave the native rule in place. Add a workflow layer that handles what it cannot: - **Enrich after creation, then re-route if the enrichment changes the answer.** The native rule owns the first assignment; the workflow layer owns the correction. - **Push the assigned owner and lead context into the channel reps already work in.** A structured message with key fields and an open-record link beats an email that gets buried. See the FlowRunner guide on [automating Salesforce lead handoffs into Slack](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack). - **Run an SLA check** that flags leads sitting unworked past a defined window and escalates with full context attached. - **Cross-reference duplicate detection** so the same lead is not assigned twice across forms, imports, and API paths. - **The same patterns apply to HubSpot.** The routing problem does not change with the CRM logo; see [routing leads with duplicate-detection in HubSpot](https://flowrunner.ai/workflows/connect-hubspot-with-slack) or [qualifying inbound leads without manual research](https://flowrunner.ai/workflows/orchestrate-google-forms-and-hubspot-and-slack). Run the math on what the manual version costs first. The framework in [calculate the ROI of automating lead routing](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) is the cleanest version of that conversation we have. ### When your assignment rule does not fire When a rule is configured and reps still see leads landing in the wrong place (or not at all), the cause is almost always one of these: - The rule is inactive, or activating a new rule deactivated the previous one without warning. - An earlier rule entry is too permissive and absorbs leads that should have hit a later, more specific entry. - The _Assign using active assignment rule_ checkbox is not exposed on the Lead page layout or not selected at create time. - The Lead was created via the [Salesforce API](https://developer.salesforce.com/docs/atlas.en-us.api.meta/api/sforce_api_header_assignmentruleheader.htm) without the assignment rule header on the request. - The Default Lead Owner is misconfigured or points to an inactive user; leads fall into a black hole instead of the safety net. - Post-create automation (a Flow, Process Builder, or Apex trigger) reassigns the owner immediately after the rule fires. - You are approaching the [3000 rule entries per object cap](https://help.salesforce.com/s/articleView?id=service.customize_leadrules.htm&type=5); the engine still works at the limit, but maintenance becomes its own problem long before you hit it. Work this list top to bottom. The first three account for most cases. ### A practical checklist before you ship a routing change - Write the routing logic down in plain language first. If you cannot describe it in two paragraphs, it will not be debuggable in six months. - Confirm the _Assign using active assignment rule_ flag is set wherever Leads are created: manual paths, Web-to-Lead, API imports, every integration that writes a Lead. - Set the Default Lead Owner to a real human or queue, not a placeholder admin user. - Define what triggers a reassignment and who owns that decision before you need it. - Plan how reps get notified and how SLAs get tracked outside Salesforce. Native assignment rules are a good engine for the slice of routing they were designed to do. Build the rest around them, so reps see the right leads at the right time, so API and after-the-fact-enrichment leads do not fall through the seams, and so the routing logic lives in a place you can still understand a year from now. ### Quick answers #### How many assignment rules can be active per object? One. Salesforce supports multiple defined rules per object, but only one is active at any time. Activating a new rule deactivates the previous one. #### Do lead assignment rules apply to existing leads, or only on create? Both, but only when the _Assign using active assignment rule_ checkbox is selected on the create or edit action. Existing leads are not silently re-evaluated. #### Why is my assignment rule not firing on API-created leads? The [assignment rule header](https://developer.salesforce.com/docs/atlas.en-us.api.meta/api/sforce_api_header_assignmentruleheader.htm) is not set by default. Any integration writing Leads through the API must set it explicitly, or the rule is skipped. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## SOX Compliance Checklist: The Control Activities That Actually Fail Audits Source: https://flowrunner.ai/blog/sox-compliance-checklist Articles May 26, 2026 Updated May 27, 2026 10 min read A SOX compliance checklist organized around the controls that actually fail audits: approval evidence, segregation of duties, and reconciliation trails. ![A bald cartoon inspector with three hairs sticking up holding a clipboard and a small flashlight, examining one amber-tagged box on a wall of stacked file boxes labeled with control names.](https://flowrunner.ai/images/blog/sox-compliance-checklist-hero.webp) A SOX checklist that lists every control your team already knows it needs is not the one that holds up at audit. The failures happen at execution, where approvals get rubber-stamped, reconciliations live in spreadsheets, and the audit trail is reconstructed during fieldwork from email threads and memory. A defensible checklist is organized around how each control activity produces verifiable evidence in the systems where the work actually runs. The framework below covers the same control families every Big Four SOX template lists. The difference is in what each line is asked to prove, and where the proof lives when the auditor asks for it. ### What SOX actually requires of finance operations Two sections of the statute do most of the work for finance. [Section 302](https://www.sec.gov/about/laws/soa2002.pdf) requires the CEO and CFO to personally certify, in each periodic report, that the disclosure controls and procedures are designed and operating effectively, that the financial statements fairly present the financial condition of the company, and that material weaknesses have been disclosed to the auditor and the audit committee. [Section 404](https://www.sec.gov/about/laws/soa2002.pdf) requires management to assess and report on the effectiveness of internal controls over financial reporting (ICFR), and requires the external auditor to attest to that assessment for non-smaller-reporting-company filers. The PCAOB standard that governs how auditors test ICFR is [AS 2201](https://pcaobus.org/oversight/standards/auditing-standards/details/AS2201). Most public companies structure their control environment around the [COSO Internal Control Integrated Framework](https://www.coso.org/guidance-on-ic). The SEC accepts other suitable frameworks, but COSO is the default and the framework auditors are most fluent in. Adopting COSO is not a legal requirement; it is the path of least friction. SOX applies to all SEC-registered public companies, with specific accommodations for smaller reporting companies and emerging growth companies. Treating it as a problem only for large filers is a common first-year misread. Private companies preparing for IPO, sitting under a public-company parent, or operating under acquisition or financing covenants often adopt SOX-style controls voluntarily because the buyers and lenders expect them. The checklist that follows is organized by control family. Each family lists the activities, then the evidence the activities must produce. The evidence question is where most teams lose the audit. ### Entity-level controls These are the controls that set the tone for everything else. Auditors look at them first because weakness here changes how they test the rest. - A documented control environment: code of conduct, ethics policy, whistleblower process, board and audit committee oversight, charter documents. - An annual risk assessment refreshed against the current business, with named owners for each identified risk. - Change management for financial systems, chart-of-accounts changes, and process changes that affect reporting. - Evidence of management review for key financial reports, with reviewer identity and date captured. - Documented escalation paths for control deficiencies, with remediation timelines tracked. Entity-level controls fail at audit when the policy document exists but the evidence of it operating does not. A code of conduct nobody signed, a whistleblower hotline nobody can name, a risk assessment dated three years ago. The fix is treating each control as a recurring activity with an artifact, not as a one-time policy. ### Access controls and segregation of duties This is the family that catches the most teams off guard, because the controls live across IT, finance, and HR and the evidence sits in multiple systems. - User access reviews on a stated cadence (quarterly is the common bar) for every in-scope financial system, with the reviewer’s identity and the date captured against each access list. - Separation of initiation, approval, and posting in the AP, revenue, and journal entry workflows. No single user can both create and approve a transaction in any in-scope cycle. - Privileged access logging in the ERP and adjacent systems, with periodic review of admin activity. - A documented joiner-mover-leaver process for finance system access, with evidence that access is revoked or modified within a stated window of the HR change. - Documented exceptions where segregation cannot be achieved (small teams, after-hours coverage), with a compensating control and management review. The honest read here, which most SOX guides will not say plainly, is that segregation of duties is the control most likely to be cited as a deficiency in companies under 200 people, because the team is too lean to fully separate the duties. The defensible posture is not pretending otherwise. It is documenting the exception explicitly, describing the compensating control, and capturing the management review that closes the loop. ### Approval controls with verifiable evidence This is the family where the audit-trail question gets concrete. - Approval workflows that capture named approver and timestamp on every approval, not just the final one in a chain. - Approval thresholds tied to a documented delegation of authority matrix, refreshed on a stated cadence. - Exception routing that requires explicit review when transactions fall outside the standard rule, instead of rubber-stamp approval through volume. - For automated approvals (system-driven posting under a threshold), documentation of the rule that fired and the inputs it evaluated. - Evidence retention for the entire approval cycle, not just the approved record. A real example. Bill approvals captured in email threads are reconstructable but expensive at audit. Bill approvals captured against the bill record inside the accounting system, with approver identity and timestamp stamped at the moment of approval, are not reconstructable because they were never lost. The first pattern survives audit by spending a controller’s October on it. The second pattern survives audit because the evidence is already there. A pattern that produces this kind of evidence at the moment of work, rather than reconstructing it later, is a workflow that posts an approval request into the channel approvers already use, captures the response against the bill record, and writes the approver identity and timestamp back to the system of record. See the FlowRunner workflow guide on [capturing every bill approval with named approver and timestamp into QuickBooks and Slack](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack) for the structural pattern. The same pattern shows up across ERPs. On Acumatica, the controls live in the AP approval cycle and the evidence in the bill record audit log; see [automated AP controls with reviewer identity captured](https://flowrunner.ai/workflows/automate-with-acumatica). The structural question, not the platform, is what the auditor cares about. If you are still defining the upstream piece, the underlying [PO approval workflow](https://flowrunner.ai/blog/po-approval-workflow) shapes most of what the approval evidence will look like at audit. ### Reconciliation and close-cycle controls The close cycle is where reconciliation evidence either exists or has to be reassembled. - Balance sheet reconciliations performed for every account on a stated cadence, with preparer and reviewer captured per account and per period. - Bank, merchant processor, and intercompany reconciliations with decision history retained for every reconciling item. - Journal entry approval before posting, with supporting documentation attached to the entry, not stored separately. - A close checklist with sign-offs at each milestone, dated, and tied to the named individual. - Quarterly and annual review of the reconciliation process itself, looking for accounts that consistently roll forward with unexplained items. The failure mode here is the reconciliation spreadsheet that has the right numbers but no preparer name, no review evidence, no decision log on the items that did not reconcile cleanly. An automated reconciliation pattern that records the preparer, the reviewer, and the per-item decision as the close runs produces evidence in the form auditors expect. The FlowRunner workflow on [reconciling Stripe payments against QuickBooks invoices with decision history retained per step](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack) is a concrete example. ### Preventive and detective controls in procure-to-pay The procure-to-pay cycle carries some of the highest fraud and error risk, which is why auditors look hard at the preventive controls. - Vendor master validation and duplicate detection before bill creation, with the validation evidence retained against the bill. - Three-way match (PO, receiver, invoice) where applicable, with mismatches routed for explicit review rather than waived. - Outlier detection on payment amounts, unusual vendor patterns, and round-dollar invoices, with exception decisions logged. - Bank account change verification for vendors, with the verification path documented for each change. - Documented segregation between vendor master maintenance, invoice entry, and payment release. Preventive controls that fire at the moment of the transaction and route exceptions for review produce a cleaner audit footprint than detective controls that catch the same issue weeks later in a journal entry reversal. The FlowRunner pattern on [vendor validation and duplicate checks before bill creation](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack) is the structural example for this control family. ### IT general controls that support the financial controls ITGCs are tested because the financial application controls are only reliable if the systems they run in are reliable. - Logical access management for in-scope systems, with provisioning, modification, and termination tied to documented approval. - Change management for the financial systems and adjacent integrations, with separation between developers and the production push. - Computer operations: backup procedures with periodic restore testing, batch job monitoring, incident response. - Audit logging retention aligned to the audit period, with retention controls preventing premature deletion. - SOC 1 or SOC 2 reports collected for material third-party systems that touch financial reporting, reviewed annually. ITGCs are where the audit asks how you know your system is doing what you say. Vendor SOC reports answer most of that question for hosted systems. For systems you operate, the answer is your own change management and access management evidence. ### Building the audit trail before the auditor asks Here is what most SOX guides will not say out loud. The cost of audit is mostly the cost of reconstruction. The substantive testing fee, the controller’s late nights in the period before fieldwork, the back-and-forth with the audit team on evidence requests: those add up because the evidence was not where it needed to be when the auditor asked for it. Auditors ask the same questions every year. Approver identity, timestamp, supporting documents, exception rationale, the system the activity happened in, the access list at the time it happened. If those answers are captured at the moment of the transaction, in the system of record, the evidence request is a query. If they are not, the evidence request is a project. The pattern that closes this gap is unglamorous. Capture approver identity, timestamp, and exception rationale at the moment of the transaction. Store the evidence against the record it relates to, in the system that owns the record. Route exceptions for explicit review rather than waiving them at volume. Retain the decision history alongside the data. This is where the discrete control activities stop being enough. SOX is not a problem any one system solves, because financial reporting touches the ERP, the AP system, the close tool, the bank, the merchant processor, the HR system, the access provisioning system, and the channels finance actually works in (email, Slack, the accounting platform). The control activities live inside those systems. The work between them, where approvals route, where reconciling items get explained, where exceptions get escalated, is where evidence quietly fails to get captured. No single system of record sees the seam between systems. An orchestration layer is the category that does: a layer above the systems of record that listens for what they emit, routes the human-judgment moments to the people who need to see them, and writes the resulting evidence back into the system that owns the record. FlowRunner is built for that layer. That positioning is not a SOX certification. SOX compliance is the issuing company’s responsibility, not a vendor’s, and no automation tool satisfies Section 404 by itself. The role a workflow layer plays is producing the evidence the company’s control activities are designed to produce, in the form auditors expect to receive it. ### Putting the checklist into practice For first-year SOX teams and teams refreshing a stale program, the priority order that holds up empirically is: 1. **Map each control activity to the system that owns the evidence.** Not the policy document, the system. If the evidence does not live somewhere queryable, that is the gap to close first. 2. **Start with the families that draw the most audit findings: approvals, access, and reconciliations.** These are where material weakness findings concentrate. 3. **Design for exceptions, not the happy path.** The control activity that holds up is the one that handles the unusual case explicitly, not the one that assumes everything fits the rule. 4. **Bring the external auditor in early on control design.** A 30-minute conversation in February is cheaper than a finding in October. 5. **Prioritize design over breadth.** A small number of controls designed to produce verifiable evidence is more defensible than a long list of controls operating on faith. A clean SOX checklist is a list of activities, each producing an artifact, each artifact living in a system where the auditor can find it without asking the controller to dig. That is the version of the checklist worth the time to build. ### Quick answers #### What does a SOX compliance checklist need to cover? At a minimum: entity-level controls, access and segregation of duties, approval evidence with named approver and timestamp, reconciliation trails, preventive and detective controls in procure-to-pay, and IT general controls supporting the financial systems. The activities are not the hard part; the evidence each activity produces is. #### Does SOX apply to private companies? SOX directly applies to SEC-registered public companies, including smaller reporting and emerging growth companies with specific accommodations. Private companies preparing for IPO, acquisition by a public company, or operating under financing or contractual covenants often adopt SOX-style controls voluntarily. #### What is the difference between SOX 302 and SOX 404? Section 302 requires officers to certify the disclosure controls and the accuracy of financial reports each period. Section 404 requires management to assess and attest to the effectiveness of internal controls over financial reporting, and the external auditor to test those controls per PCAOB [AS 2201](https://pcaobus.org/oversight/standards/auditing-standards/details/AS2201). By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Supplier Onboarding Best Practices, and the One Most Lists Get Wrong Source: https://flowrunner.ai/blog/supplier-onboarding-best-practices Articles May 28, 2026 Updated May 30, 2026 9 min read The most-repeated supplier onboarding best practice quietly creates the bypass problem. Here are the controls that survive contact with a busy team. ![A bald cartoon gatekeeper waving a LOW RISK supplier folder through a short sage-green lane while holding a thicker STRATEGIC folder at the entrance of a longer amber review lane, illustrating risk-tiered supplier onboarding.](https://flowrunner.ai/images/blog/supplier-onboarding-best-practices-hero.webp) The most-repeated supplier onboarding best practice is the one that quietly does the most damage: standardize one rigorous intake, collect every document up front, and run every vendor through the same gate. It reads like discipline. In a real procurement team it is the reason business owners stop calling procurement at all. When the office-furniture vendor has to clear the same gauntlet as the SaaS vendor with access to production data, someone in the business decides the gauntlet is the problem and routes around it. The best practice created the bypass. This article exists to argue that good supplier onboarding is not about collecting more, earlier. It is about scaling scrutiny to stakes, and building controls that survive contact with a busy team. The list of dos below is shaped by that distinction. If you want the underlying mechanics of the process itself, the [supplier onboarding process and where it actually breaks](https://flowrunner.ai/blog/supplier-onboarding-process) covers the step-by-step arc. This piece is about what to do and what to stop doing inside that arc. ![a two-lane onboarding diagram](https://flowrunner.ai/images/blog/supplier-onboarding-best-practices-1.webp) ### Best practice 1: tier the risk before you collect anything The best practice says collect a complete document set from every vendor. The better practice is to decide, at intake, which lane a vendor belongs in, and let the lane decide the document set. A low-risk, low-spend, no-data-access supplier needs a tax form, banking details, and a quick duplicate check. A strategic vendor that touches regulated data or carries six figures of annual spend needs the full treatment: sanctions screening, a security review, a negotiated contract, a data processing agreement, and ongoing monitoring. Forcing both down the same path does two bad things at once. It over-burdens the simple vendors, which trains the business to bypass you. And it under-serves the complex ones, because a process built to clear volume rarely has the patience for the cases that actually carry risk. Tiering is the practice that makes onboarding faster and safer in the same move. You speed up the many and you concentrate attention on the few that earn it. The intake decision is the highest-leverage moment in the entire process, and most lists bury it under a document checklist. ### Best practice 2: collect what the engagement needs, not everything you can name Over-collection masquerades as diligence. A vendor that will never see a customer record does not need a SOC 2 report on file. A one-time supplier does not need a five-year financial history. Every document you require is a document someone has to chase, store, verify, and re-verify at renewal, and every unnecessary one slows the vendor down for no reduction in risk. The defensible baseline for a US supplier is narrow: - A [W-9, or a W-8 series form for non-US suppliers](https://www.irs.gov/forms-pubs/about-form-w-9), for tax reporting - Banking details for remit-to, captured in a form rather than a free-text email - A certificate of insurance with the coverage levels and additional-insured language your contracts require Everything beyond that should be conditional on the tier. Diversity certifications when a customer or regulation requires reporting. Security certifications (SOC 2, ISO 27001, HIPAA) when the vendor handles data that makes them relevant. The rule is simple: a document earns its place on the checklist by mapping to a specific risk this vendor actually presents. If you cannot name the risk, drop the document. ### Best practice 3: validate against the ERP before you create the record, not after Here is where most duplicate-vendor messes are born. A business owner is impatient, a purchase order is waiting, so procurement creates the vendor record to unblock the PO with every intention of circling back to verify. The circling back does not happen. Three months later there are two records for the same legal entity with different tax IDs, and the AP team is paying invoices against both. The control that prevents this is sequence, not effort. Validate the submission against the ERP’s existing vendor list at intake, before any record is written. Match on legal entity and tax ID rather than display name, because the same supplier reliably shows up as Acme Inc., Acme LLC, and Acme Industries across three half-finished records. The pattern of [checking submitted vendor data against NetSuite before a record is created](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) is documented end to end, and the same shape runs against [Acumatica with the AP-side duplicate check before bill creation](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack). The point is architectural: the duplicate check has to query the system of record, and it has to happen before the write, or it does not happen at all. ### Best practice 4: make banking-change verification a step, not a policy Banking detail changes on existing vendors are a known fraud vector. The standard advice is to “verify banking changes,” which everyone agrees with and which fails the moment a convincing email arrives during a busy week. A policy that asks a human to remember to verify is not a control. It is a hope. The control is to make the verification a step the process will not skip. When a banking detail changes on an existing vendor, the request pauses. A callback to a known contact (not the contact on the change request) is required, and a named approver signs off before the ERP record updates. The verification is not optional and it is not dependent on whoever happened to open the email being suspicious that day. The difference between a policy and a control is whether the system enforces it when nobody is watching. ![a Slack-style approval card titled "Banking change requires verification](https://flowrunner.ai/images/blog/supplier-onboarding-best-practices-2.webp) ### Best practice 5: make the request’s state visible to everyone who touches it Onboarding stalls are almost never caused by one slow step. They are caused by nobody being able to see where the request is. Finance is waiting on legal. Legal thinks procurement is still collecting a document. Procurement assumed IT finished the security review last week. Three days evaporate before anyone notices nothing has moved, because the request’s state lives in five separate inboxes and no shared place. The best practice that addresses this is written, too softly, as “communicate status.” The actual control has two parts. One is a single observable view of where every in-flight request sits and how long it has been there. The other is a defined service level for each review step, so a stalled request surfaces itself instead of waiting to be discovered. Visibility is not a nicety layered on top of the controls. For a multi-party handoff, visibility is half the control, because a handoff you cannot see is a handoff you cannot govern. This is the point where the best-practice list runs into a wall that no single application solves, and it is worth naming plainly. The intake form lives in one tool. The duplicate check queries the ERP. The contract gets signed in DocuSign. The approvals happen in Slack or email. The renewal calendar lives somewhere else again. The “single source of truth” that every onboarding guide recommends does not exist as a product, because no one system holds all of those pieces. The truth is spread across the seams between systems, which is exactly where it goes missing. That seam is the category the problem actually points to. An orchestration layer is a system that sits above the systems of record. It listens for what each one emits: a form submission, a signed agreement, an approval response, an ERP write. It carries the context across each handoff, and it pulls a human in only at the moments that need judgment instead of every moment by default. It is not a sixth system that becomes a sixth seam. It is the layer that makes the five existing systems agree on what is happening to which vendor at which step. FlowRunner is built for that layer, and the same coordination idea is described from the buyer’s angle on the [supplier portal software](https://flowrunner.ai/solutions/supplier-portal-software) and [supplier relationship management software](https://flowrunner.ai/solutions/supplier-relationship-management-software) pages. The reason the “single source of truth” advice keeps failing is that teams keep looking for it inside a tool. It lives in the coordination across tools, and coordination is a layer, not an app. ### The anti-patterns to retire Several habits show up in nearly every team that has not reworked its onboarding, and each one is the shadow of a best practice done wrong. - **Documents collected by email and scattered across inboxes.** The W-9 in one reply, the COI forwarded from a broker, the banking confirmation three threads later. When an auditor asks where the COI was on the date of the first purchase order, the answer becomes inbox archaeology. Collect into a structured place, attached to the vendor, from the start. - **Verification that depends on one specific person.** Sanctions screening through one analyst, security review through one IT contact. When that person is out, the request sits, because the step was a habit, not a documented process with a defined backup. - **Records created before risk review finishes.** The workaround that creates duplicates and orphaned vendors who get paid before anyone produced their insurance. Treat the ERP write as the reward for completing the checks, not the move that unblocks the PO. - **An audit trail made of Slack reactions and remembered approvals.** A thumbs-up emoji is not a record of who approved a vendor at which payment term. If the answer to “who approved this” is a reconstruction from memory, you do not have an audit trail, you have a story. - **Treating onboarding and ongoing management as one program.** Onboarding ends when a vendor can transact. Everything after (COI expiration tracking, banking-change verification, contract renewal surfacing, scorecard data) is supplier management, a different job. When the handoff between them is implicit, expiration dates and renewal clauses go missing. The pattern under all five is the same one this article opened with. The work inside each step is rarely the problem. The continuity between steps is. Strengthen the handoffs and the anti-patterns disappear on their own, because most of them are just continuity failing somewhere along the chain. ### Where to start when everything needs fixing ![a single onboarding request card in a queue view, drawn in clean ink on a cream tile](https://flowrunner.ai/images/blog/supplier-onboarding-best-practices-3.webp) If two or more of those anti-patterns sound like your team, the question is not whether to rework onboarding. It is which seam to strengthen first, and the answer is almost always the intake-to-validation seam. Make it impossible to create a vendor record in the ERP until the documents are in, the approvals are captured, and the duplicate check has run. That single change pays compounding dividends. The audit trail gets easier because structured handoffs leave timestamps behind on their own. The renewal handoff gets easier because the documents now live where supplier management can see them. The bypass behavior fades because the formal path stops being slower than the workaround. A separate framework on [how to decide which onboarding steps are worth automating first](https://flowrunner.ai/blog/quick-guide-how-to-know-whats-worth-automating) handles the financial side of sequencing that work. The best practice worth adopting before any other is the one almost no list leads with: stop trying to make every vendor pass through the same rigorous front door, and start making the handoffs between your existing systems observable and enforced. Rigor applied uniformly creates bypass. Rigor applied to risk, with controls that hold when the team is busy, is the version that actually scales. ### Quick answers #### What is the most important supplier onboarding best practice? Risk tiering. Decide at intake how much scrutiny a vendor needs, then run a fast path for low-risk suppliers and a slow path for strategic or regulated ones. A single one-size process is the practice that quietly drives business owners to bypass procurement, because the fast vendors get punished with the slow vendors’ controls. #### What documents should you collect during supplier onboarding? For US suppliers, a W-9 (or a W-8 series form for non-US suppliers), banking details for remit-to, and a certificate of insurance with the right coverage and additional-insured language. Add diversity certifications and security certifications (SOC 2, ISO 27001, HIPAA) only when the engagement actually requires them. Collecting every document from every vendor is over-collection, not diligence. #### How do you prevent duplicate vendor records? Run the duplicate check against the ERP at intake, before the vendor record is created, not after. Most duplicates come from creating the record early to unblock a purchase order, then never reconciling. Match on legal entity and tax ID rather than display name, because the same supplier shows up as Acme Inc., Acme LLC, and Acme Industries across three records. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## The Supplier Onboarding Process, and Where It Actually Breaks Source: https://flowrunner.ai/blog/supplier-onboarding-process Articles May 27, 2026 Updated May 28, 2026 8 min read Supplier onboarding is a structured sequence of intake, verification, risk review, system setup, and activation. The breakage is rarely in any single step. It is in the seams between them. ![Two bald cartoon runners mid-relay-handoff with the baton suspended between their hands and supplier documents (W-9, COI, Banking) drifting behind them, illustrating where supplier onboarding actually breaks: at the handoff.](https://flowrunner.ai/images/blog/supplier-onboarding-process-hero.webp) Supplier onboarding rarely fails at a single step. It fails at the seams between them, where the document somebody collected last Tuesday loses its context by the time risk review needs it, and the approval that came back in an email thread never made it into the vendor record. The steps themselves are not the hard part. The handoffs are. That distinction sounds small. It is not. It changes what you are buying when you decide to formalize or rework supplier onboarding, and it changes which kind of tool actually closes the gap. ### What supplier onboarding actually means Supplier onboarding is the process of bringing a new vendor from selection to ready-to-transact status. It picks up where sourcing ends (a vendor has been chosen) and stops where ongoing supplier management begins (the vendor is in the system, can be paid, and is being tracked over time). The work in between is operational: collect what you need, verify what you collected, get the right approvals, set up the systems, and confirm everyone is ready. The cast of characters is consistent across mid-market companies: - **Procurement** owns the process and the policy - **Finance** owns the payment terms, tax forms, and the AP system setup - **Legal** owns the NDA, the master agreement, and the data privacy review where applicable - **IT or Security** owns the security and access review for any vendor that touches systems or data - **The requesting business owner** initiated the request and waits, with growing impatience, for the vendor to be usable Five parties, sometimes more, working on the same vendor at the same time. The process is structured. The choreography between the structured parts is where things go missing. ### The core steps in a supplier onboarding process The steps below are the standard arc. They show up in essentially every mid-market onboarding playbook, with minor wording differences. The variance between companies is not in the steps. It is in how clean the handoffs are. 1. **Intake and request.** A business owner needs a new supplier. Procurement captures scope, category, expected spend, and the business reason. This is the first decision point: which onboarding lane (low risk, standard, strategic, regulated) does this vendor enter? 2. **Information collection.** Legal name and entity type. Tax forms (a [W-9 for US suppliers or a W-8 series for non-US suppliers](https://www.irs.gov/forms-pubs/about-form-w-9)). Banking details for remit-to. Certificates of insurance with the required coverage and additional-insured language. Diversity certifications when relevant. Industry-specific certifications (PCI DSS, HIPAA, ISO 27001) when the engagement requires them. 3. **Verification and risk review.** Sanctions and watchlist screening (OFAC and equivalents). Credit check where the relationship carries credit risk. Security and data privacy review for any vendor that handles regulated data or accesses internal systems. The required depth of this step varies by risk tier, which is exactly why the intake decision in step one matters. 4. **Contract and terms.** NDA at minimum, master service agreement for ongoing relationships, statement of work for the first engagement, payment terms negotiated. Signatures captured through DocuSign or equivalent. The signed artifacts attach to the vendor record, not to an inbox thread. 5. **System setup.** Vendor record created in the ERP (NetSuite, Acumatica, SAP, QuickBooks Online) or the AP system. Payment method configured. Category, GL coding default, and approval routing assigned. Tax form indexed against the vendor record. The vendor is now technically able to be invoiced and paid. 6. **Activation and communication.** Procurement confirms readiness with both the supplier and the requesting business owner. The supplier knows the PO format, the invoicing email, and who to contact for questions. The business owner can issue the first PO. That is the structured part. Read end to end, it looks like a clean process. Walk it through a real procurement team and a different shape emerges. ### Where supplier onboarding typically breaks down Here is the part most onboarding guides will not say plainly: the structured steps are not where the time goes. The seams are. Five patterns show up in nearly every mid-market procurement team that has not formalized its onboarding flow. **Documents collected by email and stored across inboxes.** The W-9 arrived attached to a reply from the supplier’s controller. The COI came as a forwarded PDF from the broker. The banking detail confirmation sits in a separate thread three weeks later. Six months from now, when an auditor asks where the COI was on the date of the first PO, somebody is searching their inbox. **Verification steps that depend on a specific person.** Sanctions screening goes through one analyst. Security review goes through one person on the IT team. When that person is out, the request sits. There is no backup defined because the process was never documented as a process, just as a habit. **Approvals that stall because nobody can see where the request is.** Finance is waiting on legal. Legal thinks procurement is gathering one more document. Procurement assumed IT had finished the security review last week. Three days pass before anyone realizes nothing has moved. There is no shared view of the request’s state, only individual people’s inboxes. **Vendor records created in the ERP before risk review finishes.** The business owner is impatient. Procurement creates the vendor record so the first PO can be cut, with the intent to circle back on the outstanding documents. The outstanding documents do not get circled back to. Three months later the vendor is being paid and nobody can produce the COI because nobody chased it after the workaround. **No structured audit trail of who approved what and when.** The approvals exist as scattered email confirmations and Slack thumbs-up reactions. When someone asks “who approved this vendor at this payment term,” the answer is a reconstruction from memory and inbox archaeology. That is not an audit trail. That is a story. The pattern across all five is the same: the work inside each step is fine. The handoff between steps is where things lose continuity. The document, the approval, the verification, and the context that goes with them belong together. They get separated by tool, by team, and by time, and they have to be reassembled later, usually under deadline pressure. ### What a well-run onboarding process looks like A formalized supplier onboarding process is not a longer process. It is a process where the seams are explicit and the handoffs are observable. Four traits show up in the teams that have it working. **A single intake form that captures everything downstream.** Legal needs the entity type to decide which agreement template applies. Finance needs the tax classification to decide the tax form. IT needs the data-access scope to decide whether a security review is required. If those questions get asked at intake, the downstream teams know what they are receiving on day one. If they get asked later in five separate threads, the request takes three times as long to resolve. **Defined SLAs for each review step, with visibility into where the request currently sits.** Two-business-day finance review. Three-business-day legal review. Risk-tier-dependent security review. Everyone involved can see, at any time, which step is active and how long it has been there. The visibility itself is half the controls. **Risk tiering so the process scales with the stakes.** A one-time low-spend supplier for office furniture should not go through the same gauntlet as a strategic SaaS vendor with access to production data. A risk-tiered process gives the office-furniture vendor a fast path (intake, basic verification, system setup, done) and gives the SaaS vendor the slow path (full security review, contract negotiation, data processing agreement, ongoing monitoring) without forcing both through the same template. **A clean handoff from onboarding to ongoing supplier management.** When onboarding ends, the documents, approvals, payment terms, and contract references should land in a place where the team responsible for supplier management can see them. The COI’s expiration date should already be on a renewal calendar. The contract’s auto-renew clause should already be flagged. Treating onboarding as a self-contained project and supplier management as a separate program is how renewals get missed and expired insurance keeps invoicing. The honest read on what those four traits require is that they are not features of any single application. The intake form lives in one tool. The risk-tier decision routes through another. The SLAs run inside the team’s messaging channel. The system setup writes to the ERP. The handoff to ongoing management touches the AP system, the contract repository, and the renewal tracking. Five seams, minimum. The work is not inside any one of those tools. The work is the coordination across them. That is where an orchestration layer earns its place in the architecture. Not as a sixth tool that becomes a sixth seam, but as a system that sits above the others, listens for what each emits (a form submission, a signed agreement, an approval response, an ERP write), gathers the context each handoff needs, and pulls a human in at the moments that need judgment instead of every moment by default. This is the category the supplier-onboarding problem actually points to: orchestration as a service, a layer above the systems of record that owns the seams between them. FlowRunner is built for that layer, alongside [a coordination approach to supplier portals](https://flowrunner.ai/solutions/supplier-portal-software) and [SRM behavior operationalized across the ERP and AP systems](https://flowrunner.ai/solutions/supplier-relationship-management-software) the team already runs. The procurement application, the ERP, the contract repository, and the messaging channel keep doing what they do. The orchestration layer makes them agree on what is happening to which vendor at which step. ### Signals it is time to formalize or rework your process Most procurement teams know their onboarding process is imperfect long before they decide to do something about it. The decision to invest in a rework usually comes from one of these signals. - **Duplicate vendor records in the ERP.** Two records for the same legal entity with different tax IDs. Three records for the same supplier with different remit-to addresses. The duplicates are the symptom; the cause is intake skipping the ERP duplicate check. - **Inconsistent payment terms across vendors that should match.** Some are net-30, some are net-45, some are net-15 for no documented reason. The cause is approvals happening in email threads where the payment term gets negotiated but never gets captured in a way the AP system enforces. - **Business owners bypassing procurement.** When the formal onboarding path takes too long, business owners route requests through finance directly or convince a controller to set up the vendor as a one-time payee. The bypass is rational. The bypass is also the leading indicator that the formal path needs a rebuild, not a new policy memo. - **Audit findings about missing documents.** Missing tax forms. Expired insurance on active vendors. Vendors on the books who were never properly approved. These findings rarely come from one bad onboarding event. They come from twenty months of small seams leaking. - **Onboarding times that vary wildly.** The same kind of vendor takes four days when one analyst handles it and three weeks when a different analyst handles it. The variance is the signal. The variance is what a structured process eliminates. If two or more of those signals are present, the question is not whether to formalize the process. It is which seams to address first. The right starting point is usually the one that produces the most observable result: making the intake-to-validation seam structured, so vendor records do not get created in the ERP until the documents and approvals are in place and the duplicate check has happened. Once that seam is structured, the others get easier. The renewal handoff to supplier management gets easier because the documents are now where supplier management can see them. The audit trail gets easier because the structured handoffs leave timestamps and approvers behind on their own. The bypass behavior gets easier because the formal path is no longer slower than the workaround. Supplier onboarding is not a hard problem made of hard steps. It is a coordination problem made of routine steps with weak handoffs between them. Strengthen the handoffs and the process gets faster, cleaner, and more auditable at the same time. That is the work worth doing first. ### Quick answers #### What are the steps in a supplier onboarding process? Intake of the new-vendor request, document collection (tax forms, banking details, certificates of insurance, diversity certifications), verification and risk review (sanctions, credit, security where applicable), contract and terms (NDA, MSA, payment terms), system setup in the ERP or AP system, and activation with the supplier and the requesting team. Procurement, finance, legal, and IT all touch the flow; the business owner usually starts it. #### How long should supplier onboarding take? It depends on category and risk tier, and no honest answer exists as a single number. Low-risk, low-spend suppliers should move in days when the process is structured. Strategic or regulated suppliers should take longer because risk review is the work. The signal that something is wrong is variance, not absolute duration: when onboarding times swing wildly depending on who picks up the request, the process is unstructured, not slow. #### What is the difference between supplier onboarding and supplier management? Onboarding ends when a supplier is ready to transact. Supplier management is everything after that: document expiration tracking, banking change verification, scorecard data, contract renewal surfacing. The two are different jobs and usually owned by different parts of procurement. Treating them as one program is how renewal dates and expired COIs go missing. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Synchronize Block: Keeping Automations in Order Source: https://flowrunner.ai/blog/synchronize-block-keeping-automations-in-order Articles September 29, 2025 Updated May 26, 2026 5 min read When multiple branches run in parallel, the Synchronize block ensures downstream steps receive complete data before continuing. ![A hand-drawn editorial cartoon of three small streams meeting at a closed wooden gate in the center, with a single thicker stream continuing past the gate to the right.](https://flowrunner.ai/images/blog/synchronize-block-keeping-automations-in-order-hero.webp) FlowRunner’s Synchronize block is a workflow control feature designed to coordinate timing across parallel processes. It acts as a coordination checkpoint that halts execution until specified conditions are met. This is particularly useful when multiple branches are running in parallel or when steps depend on data prepared upstream. ### Practical Use Cases - **Data Pipelines:** Ensuring records are cleaned and validated before syncing to CRM systems - **Approvals:** Holding notifications until all required approvers sign off - **Multi-System Updates:** Running parallel updates across APIs while waiting for confirmation from each - **Batching:** Collecting SubFlow results for collective processing ### Benefits The block delivers three primary advantages: - **Accuracy:** Downstream steps receive complete data - **Consistency:** Uniform execution order every time - **Clarity:** Teams understand exactly when outputs finalize ### Implementation Add the block where coordination is needed, connect relevant branches, define release conditions, and continue building. The Synchronize block ensures your flows execute the same way every time, because automation should prioritize precision alongside speed. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## The Truth About Building AI Automations Source: https://flowrunner.ai/blog/the-truth-about-building-automations Articles September 17, 2025 Updated May 26, 2026 8 min read Your workflows don't fail because of tools. They fail because of how they're built. ![A bald cartoon carpenter with a hammer standing in his workshop, looking at a half-built terracotta tower with one amber beam leaning at the wrong angle, the moment the build met production.](https://flowrunner.ai/images/blog/the-truth-about-building-automations-hero.webp) Automations look great in demos. They run perfectly in test environments. But when they hit real business data, real edge cases, and real production loads, they break. The cost is real: lost prospects, missed financial records, and operations staff babysitting broken workflows. ### The Hidden Lifecycle of Fragile Automations There’s a recurring pattern. Initial workflows perform well during testing, but failures show up shortly after deployment. Common failure modes include: - Missing required data fields causing prospect loss - Incompatible file uploads triggering system crashes - Rate limiting causing silent record skipping These breakdowns come from insufficient safeguards. No retry mechanisms, no error management, no notification systems. Most automations are built for the happy path. Real-world complications don’t follow the happy path. ### Business Impact Across Roles Different stakeholders feel this differently: - **Leadership:** Capital allocated toward automations that don’t deliver - **Operations:** Extensive manual intervention required to repair workflows - **Consultants:** Difficulty demonstrating measurable business value ### Why Hacks Don’t Hold Up The core problem isn’t the tools. It’s the lack of engineering discipline. Professional software development incorporates error handling, monitoring systems, and resilience patterns that automation typically overlooks. ### A Framework for Building Automation That Lasts Five implementation stages can change this: 1. **Audit** - Identify workflows and cost inefficiencies 2. **Blueprint** - Design with safeguards and exception handling 3. **Patterns** - Deploy reusable structures 4. **Build** - Prioritize revenue impact 5. **Package** - Create scalable, reusable assets ### Engineering Principles That Hold Up Reliability requires foundational engineering principles established from project inception. You can’t bolt on resilience after the fact. You have to design for it from the start. ### Approaching Automation as a Science Sustainable automation requires systematic thinking rather than improvisation. Treat automation as engineering, not experimentation, and you transform brittle scripts into reliable, scalable business infrastructure. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## The Wait Block: Adding Smart Timing to Your Automations Source: https://flowrunner.ai/blog/the-wait-block-adding-smart-timing-to-your-automations Articles October 2, 2025 Updated May 26, 2026 5 min read The difference between a good workflow and a great one comes down to when things happen. ![An hourglass with rust-colored frame and olive sand falling mid-flow between two halves of a teal workflow sequence, the designed pause that the wait block places into the run.](https://flowrunner.ai/images/blog/the-wait-block-adding-smart-timing-to-your-automations-hero.webp) FlowRunner’s Wait block lets you introduce intentional pauses within workflows, ensuring automations execute at precisely the right moments rather than rushing through every step. ### Core Functionality The block operates in two modes: - **Simple Mode:** Fixed delays measured in seconds, minutes, hours, or days - **Advanced Mode:** Dynamic wait times calculated via the Expression Editor based on contextual data ### Why Timing Matters The difference between a good workflow and a great one comes down to _when_ things happen. Raw execution speed isn’t always the goal. Intelligent sequencing is. Practical applications include: - **Onboarding:** Delayed welcome messages that feel intentional, not robotic - **Payment Processing:** Pauses before fulfillment steps until transactions confirm - **Customer Support:** Staggered escalation notifications that allow resolution time before escalating - **Batch Operations:** Spacing API calls to prevent rate limiting ### Strategic Value The Wait block transforms raw execution speed into intelligent sequencing. Your workflows align with user experience requirements and operational dependencies rather than simply processing as fast as possible. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Transform Data Block: The Power Tool for Structuring Automation Data Source: https://flowrunner.ai/blog/transform-data-block-the-power-tool-for-structuring-automation-data Articles September 30, 2025 Updated May 26, 2026 6 min read Automation often receives data in suboptimal formats. The Transform Data block reshapes messy, unstructured data into clean, predictable outputs. ![A hand-cranked machine in muted forest green with a chaotic pile of amber paper scraps tumbling in on the left and a clean stack of dusty blue ordered sheets coming out on the right, the transformation rendered as a workshop mechanism.](https://flowrunner.ai/images/blog/transform-data-block-the-power-tool-for-structuring-automation-data-hero.webp) FlowRunner’s Transform Data block lets you reshape messy, unstructured data into clean, predictable outputs. Automation regularly receives data in suboptimal formats: nested JSON, inconsistent text, scattered values. This block provides the mechanisms to restructure it for reliable workflows. ### Core Functionality The block operates on multiple data types: - **Objects:** Map, switch, merge, or extract values from key/value structures - **Arrays:** Find extremes, filter, flatten, or iterate through lists - **Strings:** Check for text presence, extract substrings, concatenate values - **Dates/Times:** Format, calculate offsets, normalize values - **Logic:** Execute conditional operations inline without branching Configuration happens through the Expression Editor, where dynamic values from preceding blocks combine with static inputs. ### Why Transform Instead of Pass Through Rather than passively moving data, transformations actively shape it. This approach delivers cleaner outputs, simpler debugging via TestMonitor visualization, and downstream systems receiving appropriately formatted information. ### Key Operations Notable capabilities include JSON restructuring, multi-property mapping, string containment checks, list aggregation (finding maximums), and substring extraction by index. ### Practical Applications Real-world uses include code-to-label mapping, value aggregation, ID trimming, and keyword detection for conditional routing. ### Best Practices - Use multiple smaller transformations rather than complex single operations - Apply descriptive naming conventions - Design outputs with subsequent consumers in mind By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Understanding AI Automation vs. AI Agents Source: https://flowrunner.ai/blog/understanding-ai-automation-vs-ai-agents Articles September 25, 2025 Updated May 26, 2026 6 min read One of the most common mistakes companies make when implementing AI: treating automation and agents as the same thing. ![A bald cartoon man gesturing between two small mechanical creatures: a dusty blue wind-up toy on a fixed track on his left and a warm rust autonomous robot choosing between paths on his right, drawing the distinction between automation and agent.](https://flowrunner.ai/images/blog/understanding-ai-automation-vs-ai-agents-hero.webp) One of the most common mistakes companies make when implementing AI solutions is treating AI automation and AI agents as synonymous. They’re not. Understanding the distinction changes how you design, build, and scale your workflows. ### The Core Distinction Think of it as a driving analogy. Automation is the entire trip: the route, the schedule, the sequence of stops from start to finish. Agents are the drivers, specialized executors that make autonomous decisions within their domains, handling specific road conditions as they arise. - **Automation:** End-to-end orchestration of sequential processes - **Agents:** Autonomous decision-makers that handle specific task segments The fundamental principle: agents drive, automation coordinates. ### Business Implications Confusing these concepts creates real operational problems: - Assigning excessive responsibility to single agents - Building poorly coordinated agent networks - Overlooking workflow optimization opportunities ### FlowRunner’s Approach FlowRunner is a coordination layer. Think of it as the navigation system that orchestrates workflows while integrating agents through a visual interface. Agents do the driving. FlowRunner makes sure they’re on the right road, in the right order, at the right time. ### Why This Matters Understanding this distinction lets you design workflows that are efficient, reliable, and scalable. You stop overloading agents with coordination logic and start building systems where each component does what it’s best at. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## What Is Vendor Onboarding? A Working Definition for Procurement Teams Source: https://flowrunner.ai/blog/what-is-vendor-onboarding Articles May 28, 2026 Updated August 11, 2026 8 min read Vendor onboarding is collecting, verifying, and approving what you need to pay a new supplier. The checklist is easy. Keeping a vendor record true is the job. ![Cartoon museum docent beside a single pedestal holding one VENDOR RECORD card, with four frames on the wall behind meant to hold matching copies, two crooked and one empty, illustrating one record that must stay identical across many systems.](https://flowrunner.ai/images/blog/what-is-vendor-onboarding-hero.webp) Vendor onboarding gets defined as a checklist, and that definition is the reason it goes wrong. Collect the tax form, collect the banking details, collect the insurance certificate, create the record, done. The checklist is real and it matters, but it is the easy part of the job. The actual definition of vendor onboarding is harder and more useful: it is the work of producing one true vendor record and keeping it true across every system that will ever pay or reference that supplier, including the renewals that have not happened yet. Hold onto that framing, because it changes what you are actually buying when you decide to fix your onboarding process. If onboarding were a checklist, you would buy a form. It is not a checklist, so a form does not solve it. ### What vendor onboarding actually means Vendor onboarding is the end-to-end process that takes a third party from “we want to work with them” to “we can legally pay them and reference them as an approved supplier.” It begins with a request, generally from a business unit that needs the supplier, and it ends when the vendor exists as a complete, approved record in the systems that issue purchase orders and cut payments. Three boundaries make the definition precise, because the word “onboarding” gets stretched to cover work that belongs elsewhere. - **It is not sourcing.** Sourcing is the work of finding and selecting the supplier. By the time onboarding starts, the choice has already been made. Onboarding does not evaluate whether this is the right vendor. It makes a chosen vendor transactable. - **It is not ongoing vendor management.** Performance reviews, scorecards, renewal tracking, and the relationship over time all sit after onboarding. Onboarding hands the vendor off to that work. Where the handoff is clean, vendor management starts with full context. Where it is not, the renewals and expirations that should have been inherited quietly go missing. - **It applies to any third party you will pay, not only strategic suppliers.** The one-time contractor, the small SaaS subscription, and the multi-year manufacturing partner all need a vendor record before money can move. Companies that reserve “onboarding” for big suppliers end up with their highest-volume, lowest-scrutiny vendors set up informally, which is exactly where the messy records accumulate. One more scope note, because teams genuinely split on it. Some organizations fold risk and compliance review into onboarding as a single workstream. Others run risk review as a separate track that gates onboarding without being part of it. Both are defensible. What matters is that the seam between them is owned by someone, because an unowned seam is where a vendor gets activated before its risk review actually finished. ### The standard steps in a vendor onboarding process The arc below is the common structure across procurement organizations. The wording shifts company to company, but the shape holds. For a deeper walk through each step and the failure points inside it, the companion piece on [the supplier onboarding process and where it breaks](https://flowrunner.ai/blog/supplier-onboarding-process) goes further than this definition needs to. A standard process moves through five stages: 1. **Request intake.** A business unit needs a supplier. Procurement captures the scope, the category, the expected spend, and the business reason. This is the first real decision: which lane does the vendor enter, given its risk and dollar value? 2. **Documentation collection.** The supplier provides what the company needs on file. This is the part most people picture when they hear “vendor onboarding,” and it is covered in the next section. 3. **Verification.** Tax ID validation, sanctions and watchlist screening, bank account verification, and a security or data privacy review for vendors that touch systems or regulated data. The required depth scales with the lane chosen at intake. 4. **Internal approvals.** Procurement, finance, legal, and IT or security sign off on the parts they own, with the set of required approvers determined by category and spend. A low-spend office supplier needs fewer signatures than a vendor with access to customer data. 5. **Vendor record creation.** The approved supplier becomes a record in the ERP or accounting system (NetSuite, Acumatica, SAP, QuickBooks Online), with payment method, tax classification, and approval routing configured. That record then has to propagate to every downstream system that references vendors. That last clause is where the definition quietly gets hard, and the rest of this article keeps returning to it. ![A horizontal five-stage flow diagram reading left to right: "Request intake" then "Documentation" then "Verification" then "Approvals" then "Vendor record created"](https://flowrunner.ai/images/blog/what-is-vendor-onboarding-1.webp) #### The documentation, specifically The documentation set is the most concrete part of vendor onboarding, so it deserves a flat answer rather than a hedge. A standard documentation set includes: - **A tax form.** A [W-9 for US suppliers, or a W-8 series form for non-US suppliers](https://www.irs.gov/forms-pubs/about-form-w-9). This drives tax reporting and payment setup. - **Banking details.** The remit-to account information needed to actually pay the supplier. - **A certificate of insurance.** Proof of coverage at the required limits, with the correct additional-insured language where the engagement calls for it. - **A business license or registration** where the category or jurisdiction requires it. - **Category-specific attestations.** Security and privacy questionnaires for technology vendors, diversity certifications where relevant, and industry certifications such as PCI DSS, HIPAA, or ISO 27001 when the engagement touches regulated data. Treat that as a starting set, not a universal one. The exact list varies by category, spend, and jurisdiction, and the variation is the point: a structured onboarding process decides the document list at intake based on the lane, rather than asking for a fixed bundle from every supplier regardless of need. ### Who is involved and what each function owns Vendor onboarding looks like a procurement task and is mostly a coordination task across four or five functions that rarely sit in the same system. Each owns a different slice of the same vendor, at the same time. | Function | Owns | The question they answer | | --- | --- | --- | | Procurement | The process, the policy, and the vendor relationship | Is this vendor approved, and in which lane? | | Finance and AP | Banking, tax forms, payment terms, and setup in the AP or accounting system | How and when does this vendor get paid? | | Legal | Contract terms, NDAs, and regulatory exposure | What are we agreeing to, and what is the risk? | | IT and security | Access reviews and data handling assessment for technology vendors | Should this vendor touch our systems or data? | | Requesting business unit | The justification and the category context | Why do we need this vendor, and for what? | Read down that column of owners and the structural reality of onboarding becomes obvious. No single person sees the whole record at once. Procurement holds the relationship, finance holds the payment setup, legal holds the agreement, and IT holds the access decision. The vendor record they are collectively building is one object. The people building it are working from four different desks, in four different tools, on four different clocks. ### Where vendor onboarding breaks down Here is the part most “what is vendor onboarding” explainers will not say, because it complicates the tidy checklist: most teams already have a process. It just lives across email, spreadsheets, and an ERP form rather than in one structured flow. The breakage is not the absence of a process. It is the absence of continuity between the people running it. Four failure modes show up again and again, and all four are seams, not steps. **Documentation arrives piecemeal, with no single source of truth for what is outstanding.** The W-9 comes back as a reply to one email. The insurance certificate arrives forwarded from a broker. The banking confirmation lands in a separate thread a week later. Nobody can answer “what is still missing on this vendor” without reconstructing it from several inboxes, so the answer comes back wrong. **Approvals stall because routing is unclear or the next approver was never told it is their turn.** Finance is waiting on legal. Legal assumes procurement is still gathering a document. The security review has been sitting in one person’s queue for four days. No shared view of the request exists, so the delay is invisible until someone asks why the vendor still is not set up. ![A single Slack-style notification card on a dark background reading "New vendor approval needed: security review"](https://flowrunner.ai/images/blog/what-is-vendor-onboarding-2.webp) **Information collected during onboarding never fully propagates downstream.** This is the failure the working definition is built around. The vendor gets created in the ERP, but the procurement system, the AP tool, and the contract repository each hold a slightly different version. One has the old remit-to address. One never got the tax classification. The single true vendor record the company thinks it has is actually four near-copies that drift apart over time. **Renewals surface only when something fails.** The insurance certificate expires and nobody notices until a claim or an audit exposes it. The annually refreshed W-9 lapses. The compliance attestation goes stale. Onboarding treated these as one-time collection events instead of recurring obligations, so the expiration dates never landed anywhere that would surface them in time. The honest baseline here matters, and it is not the strawman version where vendors are set up by whoever opens the email first. Most teams have real people running a real sequence with genuine care. The sequence just loses continuity at every handoff, and the lost continuity is what later looks like a “broken process.” ### What good vendor onboarding looks like A well-run onboarding process is not a longer or more bureaucratic one. It is one where the seams are explicit and the handoffs are observable. Five traits separate it from the email-and-spreadsheet baseline. - **A single intake point** that captures the request once and routes it based on category, spend, and risk, so every downstream function knows what it is receiving on day one rather than discovering it through five follow-up threads. - **A defined documentation checklist with clear ownership** for each item, set by lane at intake, so “what is outstanding” has one authoritative answer at any moment. - **Approval routing that adapts to vendor type and dollar threshold** rather than forcing every supplier through one flat sequence. The office-furniture vendor takes the fast path; the data-processing vendor takes the full path. - **Verification steps that leave an audit trail** recording who approved what and when, as a byproduct of the process running, not as a reconstruction assembled later under audit pressure. - **System updates that fan out from one approved record** so the ERP, the AP system, the procurement tool, and the contract repository stay aligned to a single source of truth rather than drifting into near-copies. That last trait is the one that quietly contains the whole definition. “Keep the systems aligned to one record” sounds like a feature. It is not. It is coordination work that no single application in the stack actually owns, and naming that gap is the most useful thing this definition can do. ![A clean vendor-onboarding status board on a dark background](https://flowrunner.ai/images/blog/what-is-vendor-onboarding-3.webp) ### How automation fits in Once you accept that vendor onboarding is a continuity problem rather than a checklist, the role of automation gets specific. The job is not to replace the people who make the decisions. The decisions themselves, what documents a lane requires, what dollar threshold triggers which approver, what risk tier a vendor belongs in, are procurement and finance judgment calls and stay there. Automation’s job is the part between the decisions: moving the right document, with the context attached, to the right person, and keeping the one approved record consistent everywhere it has to live. This is the seam the question “what is vendor onboarding” actually points to once you follow it all the way down. The vendor record is a single object that has to exist identically in the ERP, the AP system, the procurement tool, and the contract repository. Onboarding is the act of creating that object correctly and propagating it without it drifting. No intake form owns that propagation. No ERP owns the steps that happen before the record reaches it. The work lives in the space between the systems, and the space between the systems is precisely what none of them was built to manage. That space is what an orchestration layer is for. It is a category of system that sits above the tools of record. It listens for what each one emits: a submitted intake form, a signed agreement, an approval response, an ERP write. It carries the context across each handoff, and it pulls a human in at the checkpoints that need judgment rather than at every step by default. The orchestration layer does not become a fifth place the vendor record lives. It is the thing that keeps the four places agreeing. [FlowRunner](https://flowrunner.ai) is built for that layer, with a human in the loop at the points where a person should decide and automation everywhere a person should not have to. You can see the shape of it in the reference pattern for [validating a new vendor against NetSuite before the record is written](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack) and the equivalent [duplicate-and-tax-ID check against Acumatica](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack), where the verification and the human approval happen before anything posts, not after. Keeping this section short is deliberate, because if you are searching “what is vendor onboarding” you are early, and the right next move is to understand the process before evaluating any tool. When you do reach the tooling question, the framing to carry with you is whether a candidate becomes another system the vendor record lives in, or a layer that keeps the existing systems aligned. The deeper treatments live in [the supplier onboarding best practices guide](https://flowrunner.ai/blog/supplier-onboarding-best-practices) and in how [supplier relationship management runs across the ERP, contract, and AP systems](https://flowrunner.ai/solutions/supplier-relationship-management-software) you already operate, alongside the role a [supplier-facing portal](https://flowrunner.ai/solutions/supplier-portal-software) plays when the gap is intake from the supplier side. The connectors most procurement teams ask about first ([NetSuite](https://flowrunner.ai/integrations/netsuite) for the vendor master, [DocuSign](https://flowrunner.ai/integrations/docusign) for agreements, [Slack](https://flowrunner.ai/integrations/slack) for approvals and exception routing) are the systems that have to stay aligned, which is the whole point. Vendor onboarding, defined honestly, is not the form you collect on day one. It is the discipline of producing one true vendor record and keeping it true through every approval, every system, and every renewal that follows. Define it that way and you will recognize the right fix when you see it, because it will be the one that owns the seams instead of adding another. ### Quick answers #### What is vendor onboarding in simple terms? It is the process of collecting, verifying, and approving everything needed to pay a new supplier, then creating that supplier’s record in your finance systems so the company can transact with them. It starts when someone requests a new vendor and ends when that vendor can be issued a purchase order and paid. #### What documents are required for vendor onboarding? A standard documentation set includes a tax form (a W-9 for US suppliers or a W-8 series form for non-US suppliers), banking details for payment, and a certificate of insurance. Category-specific items get added when the engagement requires them: security or privacy attestations for technology vendors, diversity certifications, industry certifications. The exact list varies by category, spend, and jurisdiction. #### What is the difference between vendor onboarding and supplier sourcing? Sourcing is choosing the supplier. Onboarding is everything that happens after the choice is made and before the supplier can be paid: documentation, verification, approvals, and system setup. Vendor performance management comes after onboarding. The three are sequential and owned by different parts of procurement. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## What You Can Automate Now With FlowRunner's Newest Integrations Source: https://flowrunner.ai/blog/what-you-can-automate-now-new-integrations Product July 16, 2026 Updated August 8, 2026 6 min read A wave of core integrations just landed. Here are eight real flows you can build today, from order-to-books to employee offboarding to RAG over your own docs, each with a human on the step that matters. ![A bald cartoon man calmly linking three rounded tiles, sage, blue, and a glowing amber center, into one clean automation flow with a single cable on an open cream background.](https://flowrunner.ai/images/blog/what-you-can-automate-now-new-integrations-hero.webp) A wave of core, high-demand integrations just landed in FlowRunner, and more are on the way. Each one is built and verified against the vendor’s own API, and now they all sit in the same runtime. Here are eight flows you can build with them today. Every tool named is a real FlowRunner connector, and in each flow the step that carries real consequence stays with a human. #### Sell, ship, and book the sale without a spreadsheet When a [Shopify](https://flowrunner.ai/integrations/shopify) order comes in, push it straight into fulfillment with [ShipBob](https://flowrunner.ai/integrations/shipbob) and email the customer a confirmation. When the [Stripe](https://flowrunner.ai/integrations/stripe) payment settles, record it in [QuickBooks Online](https://flowrunner.ai/integrations/quickbooks-online) as an invoice and a payment so the books match the store without a nightly export. The routine order flows straight through; an unusual refund or an amount that will not reconcile stops for a person. #### Onboard and offboard employees in one flow A new-hire record triggers [Okta](https://flowrunner.ai/integrations/okta) to create the user, add them to the right groups, assign their applications, and enroll an MFA factor, all in one pass. When someone leaves, the same connector suspends the account, revokes OAuth grants and refresh tokens, and clears active sessions in seconds. The reversible containment happens immediately. The permanent deletion of the account waits for IT to confirm. #### Answer questions from your own documents Upload your docs to [OpenAI](https://flowrunner.ai/integrations/openai-ai), index them into a vector store, and answer questions with file search grounded in your own content. Prefer to own the store? Keep a [Pinecone](https://flowrunner.ai/integrations/pinecone) knowledge base in sync as files arrive, embed each new document, and retrieve the right context at query time. Either way the agent answers from what you gave it, with citations, instead of guessing. #### Draft support replies and let a person hit send When a [Zendesk](https://flowrunner.ai/integrations/zendesk) ticket comes in urgent, alert the on-call channel in [Slack](https://flowrunner.ai/integrations/slack) with the subject, requester, and link. For the routine questions, have [Anthropic Claude](https://flowrunner.ai/integrations/anthropic-ai) draft a complete, on-brand reply from the ticket history and your knowledge base. The draft waits in the ticket for an agent to read, edit a line, and send. Nothing customer-facing goes out on the agent’s own judgment. #### Turn a signup into a full CRM record Run a new signup email through [Clearbit](https://flowrunner.ai/integrations/clearbit) to enrich it into a full person and company profile, then seed a fully populated contact in [HubSpot](https://flowrunner.ai/integrations/hubspot). Want to work inbound intent? Reveal the company behind a website visitor and find the decision-makers there with [Apollo](https://flowrunner.ai/integrations/apollo), so a raw visit becomes a routed, researched lead. #### Open incidents and run CI from where work already happens When an outage gets reported in a Slack channel, open a [ServiceNow](https://flowrunner.ai/integrations/servicenow) incident and post the number back to the channel so responders have the ticket without leaving chat. On the engineering side, a [GitHub](https://flowrunner.ai/integrations/github) push can trigger a CI/CD workflow, poll the run to completion, and report the result to Slack. The pipeline runs itself; a production deploy still waits for a named approver. #### Keep your warehouse in sync and pipe analytics to a sheet Read new records from [Airtable](https://flowrunner.ai/integrations/airtable) or [PostgreSQL](https://flowrunner.ai/integrations/postgresql) and load them into [Snowflake](https://flowrunner.ai/integrations/snowflake) with an INSERT or MERGE, so the warehouse tracks the operational store on every run. Run the reverse direction too: a Snowflake query results into a reporting sheet for the stakeholders who live in a spreadsheet. A bulk write or a change against a production table stops for the engineer who owns it. #### Generate and send invoices, sync the vendor bills Have [Sage Intacct](https://flowrunner.ai/integrations/sage-intacct) generate an invoice PDF and email it to the customer the moment it is ready. Coming the other way, read a batch of new vendor bills and record each one in the ledger, coded and ready. The payment itself, the step that moves money, waits for an approver, and the approver and timestamp land in the audit trail. #### Build one of these next None of these needed an engineer to wire an API. They are flows you compose from connectors that already speak to each vendor correctly, with a human on the step that matters. Browse the full library on the [integrations directory](https://flowrunner.ai/integrations), see what other teams build on the [team pages](https://flowrunner.ai/teams), and start with the one closest to a problem you have this week. This is the first in a regular roundup of what the newest integrations make possible. More soon. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## What You Can Automate Outside the US SaaS Default Source: https://flowrunner.ai/blog/what-you-can-automate-outside-the-us-saas-default Product August 10, 2026 Updated August 11, 2026 7 min read 203 new FlowRunner connectors landed, 55 built for a specific country. Seven flows on French, German, Brazilian, Czech, and Italian business software. ![A FlowRunner Integration Release card on a dark field: the headline What You Can Automate Now, the subhead Outside the US SaaS Default, an agentic flow glyph, the caption 203 new integrations from 20 countries, and a grid of connector logos including Axonaut, Sellsy, RAYNET, Agendor, Moskit, HERO Software, remberg, apaleo, and Upsales.](https://flowrunner.ai/images/blog/what-you-can-automate-outside-the-us-saas-default-hero.webp) An automation platform’s integration list is a map of where its customers are, and for most of this industry that map has one country on it. That is the honest read nobody in automation publishes: the long tail of “unsupported” business software is not obscure, it is just not American. A French agency runs Axonaut and Bleez. A Czech sales team runs RAYNET. A Brazilian inside-sales floor runs Meetime and Agendor. Those are not niche tools. They are the systems those businesses actually run on, and the usual path to automating them is a generic HTTP node, an API reference you may not read in your first language, and a weekend. 203 new connectors landed in FlowRunner this week, taking the catalog to 1,500 services. 55 of them name a specific country or region in their own documentation, across 20 countries: France, Germany, Brazil, Czechia, Poland, Slovakia, Italy, Switzerland, Denmark, Sweden, Ukraine, Israel, Romania, Mexico, India, Vietnam, and more. Every one is built and verified against that vendor’s own API, the same way as every connector before it. This is where the argument gets past the release note. A team in Lyon or Brno is not asking for a longer list. They are asking for the same thing a team in Austin asks for: a layer that sits above their systems of record, listens to what those systems emit, gathers the context the record does not carry, and pulls a person in at the moment something gets committed. That layer is a category, orchestration as a service, and its defining property is indifference to where a system happened to be built. FlowRunner is built for that layer, which is why a Slovak CRM lands in the same runtime, with the same audit trail and the same [human-in-the-loop](https://flowrunner.ai/concepts/human-in-the-loop) step, as Salesforce. Human-in-the-loop is an execution pattern where AI agents pause autonomously, assemble the relevant context and the decision choices available, route to a human via their preferred channel, and resume the moment the human responds. It does not get a local variant. One honest note before the flows, because release posts skip it and builders find out later. Most connectors in this wave are read and write, not event sources. Where a tool exposes no event stream, the flow runs on a schedule or a cursor poll, which for these workloads is the right shape anyway. Where a tool genuinely emits events, like Channex and apaleo below, the flow starts the moment something happens, and I say so. Here are seven flows you can build today. Every tool named is a real FlowRunner connector, and in each flow the step that carries real consequence stays with a human. #### Run a French quote-to-cash cycle without retyping it into the ledger Every morning, an agent lists opportunities in [Axonaut](https://flowrunner.ai/integrations/axonaut), picks up the ones marked won since yesterday, raises the matching quotation, and once the client accepts, creates the invoice and records the payment against it. On the [Sellsy](https://flowrunner.ai/integrations/sellsy) side the same cycle runs through estimates: the agent moves the estimate status as the deal progresses and keeps the company, contact, and opportunity records aligned. The accounting tail lands in [Bleez](https://flowrunner.ai/integrations/bleez) as a journal entry, with the source document uploaded into the dossier and the account balances read back for the reconciliation. The step that stops is Sellsy’s invoice validation. A validated French invoice is a legal document with a sequential number, and the connector exposes no way to un-validate one, so the agent assembles it and the person who owns the numbers presses validate. #### Convert a Czech lead without spawning a duplicate account Every hour, an agent pulls the new leads in [RAYNET](https://flowrunner.ai/integrations/raynet-crm-v2), checks each one against existing accounts so a returning customer does not arrive as a fresh name, and creates the follow-up task that keeps the activity trail honest. Converting the lead is what spawns an account, a contact, and a deal in a single move, and untangling that afterwards is manual work, so the agent presents the candidate matches side by side and a person decides. For the half of the team that lives in Outlook, the same run writes the deal and its history into [eWay-CRM](https://flowrunner.ai/integrations/eway-crm) as a journal entry against the company, so nobody has to switch tools to see what happened. #### Close the loop between a Brazilian call floor and the CRM Every evening, an agent pulls the day’s calls and prospecting activity from [Meetime](https://flowrunner.ai/integrations/meetime), reads each call summary, and writes the outcome into [Agendor](https://flowrunner.ai/integrations/agendor): upsert the person and the organization, move the deal to its new stage, create the next task. Prospects who went quiet get added back into the right cadence instead of dying in a rep’s notebook. [Moskit CRM](https://flowrunner.ai/integrations/moskitcrm) takes the same shape if that is your stack, with the summary filed as a deal note. The outbound WhatsApp message through [Yunique](https://flowrunner.ai/integrations/yunique) is where it stops. A message landing on a prospect’s personal number under a rep’s name is a relationship decision, so the agent drafts it with the full call history attached and the rep sends. #### Dispatch a German trades crew on evidence, not on a phone call Every fifteen minutes, an agent lists the open work requests in [remberg](https://flowrunner.ai/integrations/remberg-de), pulls the asset’s service history and its failure types, checks whether the parts are actually in inventory, and posts the whole packet to whoever schedules the week. Approving a work request commits a crew, a truck, and stock, so that approval stays with a person looking at a complete picture rather than a one-line ticket. On approval the agent opens the work order, and on the next pass it reads the forms the technician filed, files the site photos into the right [MemoMeister](https://flowrunner.ai/integrations/memomeister) project folder with labels applied, and saves the signed document against the job in [HERO Software](https://flowrunner.ai/integrations/hero-software). The documentation gets done because nobody had to remember to do it. #### Turn anonymous website traffic into a contact you are allowed to email Every morning, an agent reads the previous day’s intensive and returning visits from [Webmetic](https://flowrunner.ai/integrations/webmetic), enriches the company behind each one, and for the visits that match your profile creates or updates the contact in [4leads](https://flowrunner.ai/integrations/four-leads) with the firmographics written into custom fields and a tag applied for segmentation. What the agent does not do is start the campaign. Enrolling someone in a mail sequence is the step that needs a lawful basis under GDPR, so the agent reads the opt-in case, finds nothing, and stops. A person decides whether this signal is warm enough to send an opt-in request at all. Consent is a judgment about a relationship, not a checkbox an agent gets to tick on your behalf. #### Take a booking from any channel through to a self-check-in This is the flow that starts on a real event. [Channex](https://flowrunner.ai/integrations/channex) fires the moment a booking lands from a connected booking site. The agent creates the matching booking in [Chiavistello](https://flowrunner.ai/integrations/chiavistello) so the guest’s self-check-in is provisioned, raises a hosted payment request for the balance and the tourist tax, pushes updated availability and rates back across the channels so the room does not double-sell, and issues the fiscal receipt Italian hospitality requires once the payment clears. On the hotel side, [apaleo](https://flowrunner.ai/integrations/apaleo) drives the same shape from its own event stream: assign the unit, check the reservation in, post charges to the folio, add the payment. Then it is one in the morning and a guest is standing outside a door that will not open. The agent assembles the booking, the payment status, and the ID on file into one message, and a person decides whether to open that lock remotely. Opening a door for whoever is currently asking is the definition of a call you want a human on. #### Onboard a regulated client and let a named person approve them A new client signs. The agent creates the account and contact in [Xama](https://flowrunner.ai/integrations/xama-onboarding), initiates the onboarding and AML reports, and opens an [IdentityCheck](https://flowrunner.ai/integrations/identitycheck) verification that invites the client into a hosted capture session. Worth knowing before you build it: IdentityCheck returns its verdict to a webhook endpoint you configure in its own dashboard rather than to a FlowRunner trigger, so the flow polls for the captured media and picks up the decision from wherever you route it. When the risk assessment lands, the agent pulls it with the supporting documents attached. Moving that account to approved status is a regulated decision, so it waits for the person whose name is on the file. In Israel, [Surense](https://flowrunner.ai/integrations/surense) runs the same shape one step earlier: before a flow places business with an agent, it checks the licensee is valid and halts if the registry disagrees. #### Build one of these next Seven flows does not cover twenty countries. [Upsales](https://flowrunner.ai/integrations/upsales) in Sweden, [Livespace](https://flowrunner.ai/integrations/livespace-crm) in Poland, [FLOWii](https://flowrunner.ai/integrations/flowii) in Slovakia, [Uspacy](https://flowrunner.ai/integrations/uspacy) in Ukraine, [Bind ERP](https://flowrunner.ai/integrations/bind-erp) in Mexico, [Persat](https://flowrunner.ai/integrations/persat) across Latin America, and [BoondManager](https://flowrunner.ai/integrations/boondmanager) and [SingleCase](https://flowrunner.ai/integrations/singlecase) for consulting and legal practices all landed in the same wave, and every one of them composes the same way. Browse them on the [CRM and sales hub](https://flowrunner.ai/integrations/category/crm-sales), the [finance and accounting hub](https://flowrunner.ai/integrations/category/finance-accounting), or the full [integrations directory](https://flowrunner.ai/integrations), and see how builders wire them together on the [developers page](https://flowrunner.ai/teams/developers). The point of a wave like this is not that the catalog got bigger. It is that the question “does it support the software we actually run?” stops being the first thing you have to ask, and the flow you wanted to build becomes something you compose in an afternoon rather than something you argue for in a budget meeting. This is the fourth in our weekly roundup of what the newest integrations make possible. Last week was [570+ new integrations](https://flowrunner.ai/blog/what-you-can-automate-with-570-new-integrations), before that [30+ AI providers](https://flowrunner.ai/blog/what-you-can-automate-with-30-ai-providers), and before that [the new core connectors](https://flowrunner.ai/blog/what-you-can-automate-now-new-integrations). More soon. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## What You Can Automate With 30+ AI Providers on FlowRunner Source: https://flowrunner.ai/blog/what-you-can-automate-with-30-ai-providers Product July 23, 2026 Updated July 25, 2026 6 min read More than 30 AI providers are one connector each in FlowRunner, with your own keys and no markup. Here are seven flows you can build today, from cost-aware model routing to RAG over your own docs, each with a human on the step that matters. ![A FlowRunner Integration Release card on a dark field: the headline What You Can Automate Now, the subhead With 30+ AI Providers, Your Own Keys, an agentic flow glyph, and a grid of AI provider logos including Anthropic Claude, OpenAI, Gemini, Mistral, Azure OpenAI, AWS Bedrock, Cohere, Groq, and Pinecone.](https://flowrunner.ai/images/blog/what-you-can-automate-with-30-ai-providers-hero.webp) Every AI feature in FlowRunner runs on a model you choose. Not one bundled model, not one vendor’s credit meter. More than 30 AI providers are each a verified connector, from [Anthropic Claude](https://flowrunner.ai/integrations/anthropic-ai) and [OpenAI](https://flowrunner.ai/integrations/openai-ai) to [Gemini](https://flowrunner.ai/integrations/gemini-ai), [Mistral](https://flowrunner.ai/integrations/mistral-ai), [Cohere](https://flowrunner.ai/integrations/cohere), [Groq](https://flowrunner.ai/integrations/groq), and [xAI Grok](https://flowrunner.ai/integrations/xai-grok), alongside [AWS Bedrock](https://flowrunner.ai/integrations/aws-bedrock), [Azure OpenAI](https://flowrunner.ai/integrations/azure-openai), and [Google Vertex AI](https://flowrunner.ai/integrations/google-vertex-ai), plus speech, vision, document, and translation models and eight vector stores. You bring your own keys, pay the provider directly at your rate, and FlowRunner takes no markup on inference. The point is not the length of the list. It is what you can build when any model is one connector away and you compose them per job instead of per vendor. Here are seven flows you can build today. In each, the model does the work and a person owns the step that carries consequence. #### Send cheap models the easy work, save the frontier model for the hard part Most requests do not need your most expensive model. Have a fast, low-cost model like [Groq](https://flowrunner.ai/integrations/groq) or a small [OpenAI](https://flowrunner.ai/integrations/openai-ai) model triage every incoming item, and escalate only the genuinely hard cases to a frontier model like [Anthropic Claude](https://flowrunner.ai/integrations/anthropic-ai). Route it through [OpenRouter](https://flowrunner.ai/integrations/openrouter) if you want a single endpoint across providers. The routine classification flows straight through; the ambiguous edge case, the one the cheap model flagged as low-confidence, stops for a person. #### Cross-check a high-stakes call with a second model For a decision that matters, one model’s confidence is not enough. Run the same extraction or judgment through two independent providers, [OpenAI](https://flowrunner.ai/integrations/openai-ai) and [Anthropic Claude](https://flowrunner.ai/integrations/anthropic-ai), and compare. When they agree, proceed. When they disagree, that is precisely the signal to pull in a human, who sees both answers side by side and decides. You get the speed of automation on the easy majority and a genuine second opinion, then a person, on the rest. #### Keep AI on the provider your security team already approved BYOK is not only about price. Point the same agent at [Azure OpenAI](https://flowrunner.ai/integrations/azure-openai) under your Azure agreement, or [AWS Bedrock](https://flowrunner.ai/integrations/aws-bedrock) under your AWS contract, and your prompts and data ride on the processor and data-processing terms your security team already signed. No new vendor in the AI path, no markup on inference, and the model choice is a configuration rather than a rebuild. When the output feeds a regulated system, the write waits for a human. #### Answer from your own documents, any embedding model, any store Embed your documents with the provider you prefer, [Cohere](https://flowrunner.ai/integrations/cohere) or [OpenAI](https://flowrunner.ai/integrations/openai-ai), and keep them in the vector store you run: [Pinecone](https://flowrunner.ai/integrations/pinecone), [Qdrant](https://flowrunner.ai/integrations/qdrant), [Weaviate](https://flowrunner.ai/integrations/weaviate), or [pgvector](https://flowrunner.ai/integrations/pgvector). At query time, retrieve the right context and answer with [Anthropic Claude](https://flowrunner.ai/integrations/anthropic-ai), grounded in what you gave it, with citations. Swap the model or the store without rebuilding the flow. Anything the answer sets in motion downstream, a refund, a policy exception, a customer email, waits for a person. #### Turn a call into a transcript, action items, and a spoken reply Transcribe a recorded call with [AWS Transcribe](https://flowrunner.ai/integrations/aws-transcribe), summarize it and pull the action items with [Anthropic Claude](https://flowrunner.ai/integrations/anthropic-ai) or [OpenAI](https://flowrunner.ai/integrations/openai-ai), and file them where the team already works. For outbound, generate a natural voice message with [ElevenLabs](https://flowrunner.ai/integrations/elevenlabs). Speech in, text in the middle, speech out, and anything that actually reaches a customer waits for a person to approve it first. #### Read a document, classify it, and let a person post it Parse an incoming invoice, receipt, or form with [Mindee](https://flowrunner.ai/integrations/mindee) or [AWS Textract](https://flowrunner.ai/integrations/aws-textract), then have a model classify and route it, coding an invoice to the right account or flagging a contract clause for review. The parsing and the classification run on their own. The consequential post, the entry into the ledger or the CRM, waits for the person who owns those numbers, with the source document attached. #### Translate and localize, with a native speaker on anything public Translate content with [DeepL](https://flowrunner.ai/integrations/deepl), then have a model like [Mistral AI](https://flowrunner.ai/integrations/mistral-ai) or [Gemini AI](https://flowrunner.ai/integrations/gemini-ai) adapt the tone and localize the phrasing for the market. Machine translation handles the volume; a human who speaks the language reviews and approves anything that will be published or sent to a customer. Fast draft, human polish, and no unreviewed copy going out under your name. #### Build one of these next Every model here is a connector your agents can call as a tool, with your keys, at your rates, and no markup on inference. Pick the provider that fits the job, not the one you happened to standardize on. Browse the full set on the [AI integrations hub](https://flowrunner.ai/integrations/category/ai-llms), see how builders put them together on the [developers page](https://flowrunner.ai/teams/developers), and start with the flow closest to a problem you have this week. This is the second in our weekly roundup of what the newest integrations make possible. Last week was [the new core connectors](https://flowrunner.ai/blog/what-you-can-automate-now-new-integrations). More soon. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## What You Can Automate With 570+ New FlowRunner Integrations Source: https://flowrunner.ai/blog/what-you-can-automate-with-570-new-integrations Product July 31, 2026 Updated August 8, 2026 6 min read FlowRunner just added more than 570 new integrations, from CRM and identity to DevOps, payments, and voice AI. Here are seven flows you can build today, each with a human on the step that matters. ![A FlowRunner Integration Release card on a dark field: the headline What You Can Automate Now, the subhead With 570+ New Integrations, an agentic flow glyph, and a grid of new connector logos including Attio, Auth0, Datadog, Terraform Cloud, Square, and Botpress.](https://flowrunner.ai/images/blog/what-you-can-automate-with-570-new-integrations-hero.webp) More than 570 new integrations just landed in FlowRunner, taking the catalog past 900 verified connectors. Each one is built and verified against the vendor’s own API, the same way as every connector before it, and now they all sit in the same runtime as your existing flows. The number is not the point. The point is what a wave this size makes buildable in categories that had no coverage last month. Here are seven flows you can build today. Every tool named is a real FlowRunner connector, and in each flow the step that carries real consequence stays with a human. #### Keep your CRM clean without a weekly dedup project Every morning, pull the prior day’s new signups, enrich each one with a full profile from [People Data Labs](https://flowrunner.ai/integrations/people-data-labs), and upsert the record into [Attio](https://flowrunner.ai/integrations/attio), matched on email so a returning contact updates instead of duplicating. Attio adapts to your workspace schema at runtime, so the same flow works whether you track People and Companies or a set of custom objects. When two records plausibly describe the same account, the agent surfaces both side by side and a person decides which one survives. Merging is fast. Un-merging is not, so that call stays with a human. #### Catch access anomalies before they become incidents Every hour, an agent pulls the latest authentication logs from [Auth0](https://flowrunner.ai/integrations/auth0) and scans for the pattern that matters: a login from a new country, a spike in failed attempts, a token issued outside business hours. The routine logins get filed and forgotten. A genuine anomaly gets posted to the security channel in [Slack](https://flowrunner.ai/integrations/slack) with the user, the location, and the log entry attached, and a person decides whether to force a logout or leave it. Revoking a session is reversible. Assuming an anomaly is nothing and being wrong about it is not, so the agent flags and a human confirms. #### Turn a monitoring signal into a fix, not just a page Every few minutes, an agent checks [Datadog](https://flowrunner.ai/integrations/datadog) for monitors sitting in an alert state and pulls the recent metric history for context. For a known, low-risk pattern, like a queue depth that clears itself within a few cycles, the agent logs it and moves on. For anything that would normally page someone at 2am, it posts the monitor, the graph, and a proposed remediation to the on-call channel, and an engineer decides whether to run it. Muting a monitor or restarting a service is a judgment call about what is actually broken, and that judgment stays human. #### Ship infrastructure changes without surprises When a pull request merges to main on [GitHub](https://flowrunner.ai/integrations/github), kick off a [Terraform Cloud](https://flowrunner.ai/integrations/terraform-cloud) plan and post the resource diff back to the PR thread, then queue the matching [Azure DevOps](https://flowrunner.ai/integrations/azure-devops) pipeline so the application deploy is staged and ready. The plan runs itself, and everyone can see exactly what would change before it does. Applying that plan against production, and running the pipeline that ships it, both wait for a named engineer to approve, because a plan is a proposal and an apply is a commitment. #### Turn a sales call into a booked next step Every evening, an agent pulls the day’s completed calls from [Aircall](https://flowrunner.ai/integrations/aircall), reads the call notes and tags, and finds the ones marked as a live opportunity. For those, it drafts a follow-up meeting through [ScheduleOnce](https://flowrunner.ai/integrations/scheduleonce) with a set of proposed times and drops the draft in the rep’s inbox instead of sending it. The rep picks the times that actually work and the invite goes out under their name. An agent guessing at a prospect’s calendar and sending on a rep’s behalf is exactly the kind of thing that should cost a human one click first. #### Run the same order-to-books flow on the tools you actually use If you read our first roundup, you saw an order-to-books flow built on Shopify, Stripe, and QuickBooks Online. The same pattern now runs on a different toolchain. Every morning, an agent pulls the prior day’s settled payments from [Square](https://flowrunner.ai/integrations/square), matches each one to an order, and posts the corresponding entry in [FreeAgent](https://flowrunner.ai/integrations/freeagent) or [Fortnox](https://flowrunner.ai/integrations/fortnox), coded and ready. The books match the till without a nightly export either way. Anything that will not reconcile, an amount that is off by more than a rounding error, still waits for the person who owns the ledger. #### Give a chatbot an escalation path instead of a dead end Every few minutes, an agent checks [Botpress](https://flowrunner.ai/integrations/botpress) or [Dify](https://flowrunner.ai/integrations/dify) for conversations the bot flagged low-confidence or that contain a phrase like “talk to a person.” It pulls the full conversation history and posts it to the support channel in Slack with the customer’s context attached, so whoever picks it up is not starting from zero. The bot keeps answering everything it actually knows. The moment it does not, a person takes over with the whole exchange already in front of them, instead of the customer repeating themselves. #### Build one of these next None of these needed an engineer to wire an API. They are flows composed from connectors that already speak to each vendor correctly, spanning categories FlowRunner had thin or no coverage in a month ago, with a human on the step that matters. Browse the full library on the [integrations directory](https://flowrunner.ai/integrations), see how builders put them together on the [developers page](https://flowrunner.ai/teams/developers), and start with the one closest to a problem you have this week. This is the third in our weekly roundup of what the newest integrations make possible. Last week was [30+ AI providers](https://flowrunner.ai/blog/what-you-can-automate-with-30-ai-providers), and the week before that was [the new core connectors](https://flowrunner.ai/blog/what-you-can-automate-now-new-integrations). More soon. By [Mark Piller](https://flowrunner.ai/about), Founder of FlowRunner [Editorial policy](https://flowrunner.ai/editorial-policy) --- ## Workflow guides (566 guides) - [Workflow Guides](https://flowrunner.ai/workflows.md): Step-by-step guides for automating operations with FlowRunner integrations. - [Everything You Can Automate with Acumatica in FlowRunner](https://flowrunner.ai/workflows/automate-with-acumatica.md): Connect your Acumatica ERP to FlowRunner and automate the full accounts payable lifecycle: vendor validation, bill creation, duplicate detection, approvals, and daily AP reporting with human oversight on exceptions. - [ActiveCampaign + Google Sheets: Rows Become Synced Leads](https://flowrunner.ai/workflows/connect-activecampaign-with-google-sheets.md): Connect ActiveCampaign and Google Sheets so new rows sync into contacts automatically, with an AI agent that pauses for a human before a high-value deal is created. - [How to Connect ActiveCampaign to Shopify: Tag Every New Order](https://flowrunner.ai/workflows/connect-activecampaign-with-shopify.md): Connect Shopify orders to ActiveCampaign contacts and automations automatically, and run it as an AI agent that pauses for a sales manager before a high-value order becomes a deal. - [ActiveCampaign Typeform Integration: Answers Reviewed First](https://flowrunner.ai/workflows/connect-activecampaign-with-typeform.md): Connect ActiveCampaign and Typeform so an agent creates a personalized intake form and syncs the response back automatically, pausing for a human when the answers are sensitive. - [How to Connect Acuity Scheduling with Mailchimp Marketing (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-acuity-with-mailchimp-marketing.md): Connect Acuity Scheduling to Mailchimp Marketing so every new booking automatically adds or updates the subscriber, applies the right tag, and enrolls them in the right email journey, optionally as an AI agent that pauses for a human before sending to large lists. - [How to Connect Acumatica with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-acumatica-with-slack.md): Connect Acumatica and Slack so AP bills, exceptions, and approvals move automatically between your ERP and your team, with a human on every step that carries real consequence. - [Aircall Pipedrive Integration: Insight Cards Before You Say Hello](https://flowrunner.ai/workflows/connect-aircall-with-pipedrive.md): Connect Aircall and Pipedrive so every inbound call arrives with deal context on screen, and let an AI agent run the sync while a human approves any record merge. - [Can an AI Agent Route Escalated Aircall Calls to Slack?](https://flowrunner.ai/workflows/connect-aircall-with-slack.md): Connect Aircall and Slack so tagged calls post to your team automatically, or run it as an AI agent that decides which calls need a human and pauses for a decision in Slack. - [Connect Aircall to Zendesk: Every Complaint Call Becomes a Ticket](https://flowrunner.ai/workflows/connect-aircall-with-zendesk.md): Connect Aircall and Zendesk so a flagged call opens a ticket automatically, or run it as an AI agent that escalates repeat complaints to a support lead before priority changes. - [How to Connect Airtable with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-aircall.md): Connect Airtable records to Aircall contacts and insight cards so your base drives the phone system, optionally as an AI agent that syncs updates automatically and pauses for a human before Delete Contact removes anyone. - [How to Connect Airtable with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-attio.md): Connect Airtable bases to Attio records so operational data flows into your CRM automatically, with an AI agent that pauses for a human before it overwrites a rep-edited field or deletes a record. - [How to Connect Airtable with Bigin (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-bigin-by-zoho.md): Connect Airtable lead tables to Bigin by Zoho so captured leads become pipeline automatically, with an AI agent that pauses for a human before it deletes a duplicate record from the CRM. - [How to Connect Airtable with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-bland.md): Connect an Airtable call list to Bland AI phone outreach, optionally as an AI agent that validates every row, writes outcomes back to the base, and pauses for a human before any call is placed. - [How to Connect Airtable with Capsule CRM (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-capsule-crm.md): Connect Airtable intake tables to Capsule CRM records and opportunities, with an AI agent that dedupes every lead against the CRM and pauses for a human before Update Party overwrites data your team already trusts. - [How to Connect Airtable with CloudTalk (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-cloudtalk.md): Connect Airtable records to CloudTalk calls and SMS, optionally as an AI agent that works the outreach queue automatically and pauses for a human before a message goes out to an entire segment. - [How to Connect Airtable with Confluence (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-confluence.md): Connect Airtable records to Confluence pages, optionally as an AI agent that drafts and publishes documentation automatically and pauses for a human before Update Page overwrites a live published page. - [How to Connect Airtable with Copilot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-copilot.md): Connect Airtable records to the Copilot client portal, optionally as an AI agent that provisions companies and clients from your pipeline base and pauses for a human before a new client gets portal access or an invoice. - [How to Connect Airtable with Crisp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-crisp.md): Connect Airtable records to Crisp profiles and conversations, optionally as an AI agent that enriches chat profiles from your base automatically and pauses for a human before Send Message starts a proactive conversation with a customer. - [How to Connect Airtable with Custify (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-custify.md): Connect Airtable records to Custify accounts, optionally as an AI agent that keeps people, companies, and subscription data in sync and pauses for a human before Delete Company erases an account's customer success history. - [How to Connect Airtable with Delighted (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-delighted.md): Connect Airtable records to Delighted so a delivered project or closed order triggers an NPS survey, with an AI agent that checks survey history first and pauses for a human before surveying an account in the middle of an open issue. - [How to Connect Airtable with Dialpad (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-dialpad.md): Connect Airtable records to Dialpad so a status change sends the customer an SMS or queues a call, with an AI agent that drafts the message from record context and pauses for a human before texting anything sensitive. - [How to Connect Airtable with Elastic Email (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-elastic-email.md): Connect Airtable records to Elastic Email so record changes send transactional email and build clean lists, with an AI agent that pauses for a human before any send that reaches a whole segment at once. - [How to Connect Airtable with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-google-sheets.md): Connect Airtable and Google Sheets so new or updated records flow between them automatically, optionally as an AI agent that pauses for human judgment before writing disputed or malformed data. - [Airtable Notion Integration: Exceptions Logged, Decisions Kept](https://flowrunner.ai/workflows/connect-airtable-with-notion.md): Connect Airtable and Notion in FlowRunner so exception records flagged in Airtable become a searchable decision log in Notion automatically, with a human confirming the call before it's written down. - [How to Connect Airtable with PostgreSQL (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-postgresql.md): Connect Airtable and PostgreSQL in FlowRunner so records created or updated in Airtable automatically sync to your PostgreSQL database, optionally with an AI agent that reconciles conflicts and pauses for a human before any bulk write. - [How to Connect Airtable with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-airtable-with-slack.md): Connect Airtable and Slack so record changes trigger Slack messages automatically, or run it as an AI agent that reasons about exceptions and pauses for a human on the steps that carry real consequence. - [Send Anthropic Claude Contract Reviews to Google Sheets](https://flowrunner.ai/workflows/connect-anthropic-ai-with-google-sheets.md): Connect Anthropic Claude and Google Sheets so a new contract row triggers Claude to extract and cite key terms, with a person reviewing anything non-standard before it's logged. - [Anthropic Claude + Notion: Every Session Finding Filed for Review](https://flowrunner.ai/workflows/connect-anthropic-ai-with-notion.md): Connect Anthropic Claude to Notion so Managed Agents sessions log their findings automatically, with a human review gate before anything non-standard reaches Notion. - [Anthropic Claude + Pinecone: Documents Become Searchable Memory](https://flowrunner.ai/workflows/connect-anthropic-ai-with-pinecone.md): Connect Anthropic Claude and Pinecone so extracted document content becomes a searchable, self-updating knowledge base, run as an AI agent with a human approving anything that deletes memory. - [How to Connect Anthropic Claude to Slack for Agent Approvals](https://flowrunner.ai/workflows/connect-anthropic-ai-with-slack.md): Connect Anthropic Claude and Slack so a Managed Agents session posts its work and pauses for a human, in Slack, on the step that carries real consequence. - [How to Connect Anthropic Claude with Zendesk (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-anthropic-ai-with-zendesk.md): Connect Anthropic Claude and Zendesk in FlowRunner so an AI agent reads, classifies, and routes tickets automatically, then pauses for a human before any bulk change affects your whole queue. - [How to Connect Apollo.io with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-apollo-with-hubspot.md): Connect Apollo.io and HubSpot so every inbound lead arrives fully enriched, routed, and ready for outreach, optionally as an AI agent that pauses for human review before adding contacts to high-priority sequences. - [How to Connect Apollo.io with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-apollo-with-salesforce-pro.md): Connect Apollo.io to Salesforce Pro so enriched prospect data flows into your CRM automatically, with an AI agent that pauses for human approval before converting high-value leads. - [Asana + Google Sheets: Every Flagged Row Becomes a Tracked Task](https://flowrunner.ai/workflows/connect-asana-with-google-sheets.md): Connect Asana and Google Sheets so a new spreadsheet row becomes an assigned Asana task automatically, with an AI agent that pauses for a human on the rows that carry real risk. - [Attio Gmail Integration: Every Reply Becomes a Clean CRM Record](https://flowrunner.ai/workflows/connect-attio-with-gmail-service.md): Connect Attio and Gmail so every inbound email upserts the right Person and Company in Attio, with an AI agent that pauses for a human when a match is ambiguous. - [How to Connect Attio to Typeform: Route Submissions to Your CRM](https://flowrunner.ai/workflows/connect-attio-with-typeform.md): Connect Typeform submissions to Attio records automatically, and run it as an AI agent that pauses for a compliance reviewer before sensitive answers ever reach the CRM. - [How to Connect AWS Textract with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-aws-textract-with-quickbooks-online.md): Connect AWS Textract to QuickBooks Online so vendor invoice PDFs are automatically extracted, validated, and recorded as bills, with a human in the loop before any uncertain or high-value entry posts. - [How to Connect BambooHR to Google Sheets: Track Every Pay Change](https://flowrunner.ai/workflows/connect-bamboohr-with-google-sheets.md): Connect BambooHR to Google Sheets to keep a headcount tracker current automatically, or run it as an AI agent that pauses for a person before compensation or employment-status changes land in the sheet. - [How to Connect BambooHR with Okta (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-bamboohr-with-okta.md): Connect BambooHR's On Employee Changed trigger to Okta provisioning actions, optionally as an AI agent that runs onboarding automatically and pauses for a human before any identity write that carries real consequence. - [How to Connect BambooHR with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-bamboohr-with-slack.md): Connect BambooHR to Slack so every employee lifecycle event fires an automated Slack update, optionally with an AI agent that reasons over the change and pauses for a person before writing to compensation or employment status. - [How to Connect BigCommerce with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-bigcommerce-with-quickbooks-online.md): Connect BigCommerce to QuickBooks Online so every new order triggers an invoice or payment record automatically, with a FlowRunner AI agent that pauses for human approval before posting refunds or large transactions. - [Can an AI Agent Turn BigCommerce Orders Into Xero Invoices?](https://flowrunner.ai/workflows/connect-bigcommerce-with-xero.md): Connect BigCommerce and Xero so every new order becomes a Xero invoice automatically, with an AI agent that pauses for a human before voiding or refunding. - [Connect Google BigQuery to Google Sheets: Self-Updating KPIs](https://flowrunner.ai/workflows/connect-bigquery-with-google-sheets.md): Connect Google BigQuery and Google Sheets so validated sheet rows stream into the warehouse and query results write straight back into the sheet, with an AI agent that pauses for a person before it loads a bad row or drops a table. - [How to Connect BILL with NetSuite (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-billcom-with-netsuite.md): Connect BILL and NetSuite so every vendor invoice, payment, and failed charge moves between your AP system and ERP automatically, with a human on the consequential step. - [Send Bitbucket Pipeline Failures to Jira Issues Automatically](https://flowrunner.ai/workflows/connect-bitbucket-with-jira-issues.md): Connect Bitbucket and Jira Issues so failed pipeline runs become tracked, assigned Jira issues, with a human confirming the assignee on P1-level failures. - [Bitbucket + Slack: PR Opens, Team Notified, Merge Gated](https://flowrunner.ai/workflows/connect-bitbucket-with-slack.md): Connect Bitbucket pull requests and pipelines to Slack so reviewers see them the moment they open, with an AI agent that pauses for a human before any merge into main. - [Connect Bland AI to HubSpot: Every Call Becomes a CRM Field](https://flowrunner.ai/workflows/connect-bland-with-hubspot.md): Connect Bland AI to HubSpot so qualification calls write their own contact and deal fields, run as an AI agent that pauses for a person before any number dials. - [Box + Slack: External Shares Wait for Human Approval](https://flowrunner.ai/workflows/connect-box-with-slack.md): Connect Box and Slack so filed documents post automatically and external share requests pause for a human in Slack before the link goes out. - [Connect Brevo to Google Sheets: New Rows Become Contacts](https://flowrunner.ai/workflows/connect-brevo-with-google-sheets.md): Connect Brevo and Google Sheets so every new lead row becomes an enrolled Brevo contact, or run it as an AI agent that pauses for a human when a row fails validation. - [Send New WordPress Posts to Brevo Subscribers Automatically](https://flowrunner.ai/workflows/connect-brevo-with-wordpress.md): Connect Brevo and WordPress so every published post triggers an announcement email to the matching subscriber list, optionally as an AI agent that pauses for a human before a large send. - [How to Connect Brex to QuickBooks Online: Cards Reconciled Daily](https://flowrunner.ai/workflows/connect-brex-with-quickbooks-online.md): Connect Brex settled card and cash transactions to QuickBooks Online bills so reconciliation runs nightly instead of monthly, optionally as an AI agent that verifies the vendor and pauses for a finance lead before a large bill posts or a new virtual card gets issued. - [How to Connect Cal.com with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-cal-com-with-hubspot.md): Connect Cal.com and HubSpot so every new booking request creates or updates a HubSpot contact and deal automatically, with an AI agent that pauses for human judgment when a request scores borderline. - [Connect Cal.com to Zoom: Every Confirmed Booking Gets a Meeting](https://flowrunner.ai/workflows/connect-cal-com-with-zoom.md): Connect Cal.com to Zoom so every confirmed booking automatically becomes a Zoom meeting with a join link sent to the attendee, optionally as an AI agent that pauses for a human on borderline requests before any meeting is created. - [How to Connect Calendly with Accelo (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-accelo.md): Connect Calendly bookings to Accelo so every client meeting lands on the right company record with a prep task assigned, with an AI agent that pauses for a human before a client-facing meeting is canceled. - [How to Connect Calendly with Adobe Workfront (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-adobe-workfront.md): Connect Calendly bookings to Adobe Workfront projects and tasks, optionally as an AI agent that stands up standard projects on its own and pauses for a delivery lead before it reshuffles team deadlines. - [How to Connect Calendly with Aha! (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-aha-io.md): Connect Calendly customer calls to Aha! ideas and features, optionally as an AI agent that logs every request as an idea automatically and pauses for the product lead before anything is promoted onto the roadmap. - [How to Connect Calendly with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-aircall.md): Connect Calendly bookings to Aircall contacts and insight cards so every scheduled call starts with context, optionally as an AI agent that pauses for a human before a suspect booking is canceled on a real person's calendar. - [How to Connect Calendly with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-attio.md): Connect Calendly bookings to Attio records so every meeting builds CRM history automatically, with an AI agent that pauses for a human before it cancels a scheduled event on a customer's calendar. - [How to Connect Calendly with Basecamp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-basecamp3.md): Connect Calendly bookings to Basecamp projects so client kickoffs build their own workspace automatically, with an AI agent that pauses for a human before a canceled booking trashes project work. - [How to Connect Calendly with Bigin by Zoho (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-bigin-by-zoho.md): Connect Calendly bookings to Bigin deals, optionally as an AI agent that builds pipeline from every meeting, chases no-shows with a fresh link, and pauses for a human before a deal is marked lost. - [How to Connect Calendly with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-bland.md): Connect Calendly no-shows and new bookings to Bland AI phone calls, with an AI agent that drafts the recovery call and a fresh scheduling link, then pauses for a human before Send Call dials the invitee. - [How to Connect Calendly with Canny (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-canny.md): Connect Calendly feedback calls to Canny's roadmap board, with an AI agent that matches each customer request to existing posts and pauses for a human before Create Post publishes anything to the public board. - [How to Connect Calendly with Capsule CRM (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-capsule-crm.md): Connect Calendly bookings, cancellations, and no-shows to Capsule CRM records, with an AI agent that builds the pipeline automatically and pauses for a human before a deal is marked lost or a customer meeting is canceled. - [How to Connect Calendly with Clio Manage (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-clio-manage.md): Connect Calendly consultation bookings to Clio Manage contacts, matters, and time entries, with an AI agent that runs the intake automatically and pauses for an attorney's conflict check before a matter is opened. - [How to Connect Calendly with CloudTalk (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-cloudtalk.md): Connect Calendly bookings to CloudTalk SMS and calls, optionally as an AI agent that confirms and rebooks meetings automatically and pauses for a human before it contacts a no-show or cancels a booking. - [How to Connect Calendly with Shortcut (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-clubhouse.md): Connect Calendly bookings and routing forms to Shortcut (formerly Clubhouse) stories, optionally as an AI agent that files and updates work automatically and pauses for a human before anything lands in the active iteration. - [How to Connect Calendly with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-google-sheets.md): Connect Calendly to Google Sheets so every booking, cancellation, or no-show lands as a row automatically, then add an AI agent that reasons about the data and pauses for a human on the steps that carry real consequence. - [How to Connect Calendly with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-hubspot.md): Connect Calendly to HubSpot so every new booking, cancellation, or no-show automatically updates your CRM deals and contacts, with an AI agent that pauses for a human on high-value exceptions. - [How to Connect Calendly with Notion (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-notion.md): Connect Calendly to Notion so every booking, no-show, and cancellation creates or updates a Notion page automatically, with a human on the step that carries real consequence. - [How to Connect Calendly with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-salesforce-pro.md): Connect Calendly booking events to Salesforce Pro CRM records automatically, and optionally run it as an AI agent that pauses for human approval before irreversible actions like lead conversion. - [How to Connect Calendly with Slack Using FlowRunner](https://flowrunner.ai/workflows/connect-calendly-with-slack.md): Stop losing high-value prospects to missed follow-ups and silent cancellations. Connect Calendly scheduling with Slack notifications and human decisions through FlowRunner. - [How to Connect Calendly with Zoom (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-calendly-with-zoom.md): Connect Calendly and Zoom so every booking automatically creates the Zoom meeting, registers the attendee, and sends their personal join link, with an AI agent that pauses for human judgment on the steps that carry real consequence. - [Capsule CRM Google Sheets Integration: Rows That Become Pipeline](https://flowrunner.ai/workflows/connect-capsule-crm-with-google-sheets.md): Connect Capsule CRM and Google Sheets so every new lead row becomes a Capsule person and opportunity automatically, with an AI agent that pauses for a human before it merges a match or deletes a record with open pipeline attached. - [Connect Capsule CRM to Mailchimp Marketing: Deals Fuel Campaigns](https://flowrunner.ai/workflows/connect-capsule-crm-with-mailchimp-marketing.md): Connect Capsule CRM to Mailchimp Marketing so every closed-won opportunity becomes a synced subscriber and tagged segment, with an AI agent that pauses for a human before any erasure request touches both systems. - [How to Connect Chargebee with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-chargebee-with-quickbooks-online.md): Connect Chargebee and QuickBooks Online so subscription invoices and payments sync to your books automatically, with an AI agent that pauses for a human before posting disputed charges or issuing credits above threshold. - [How to Connect Chargebee to Xero: Every Invoice Synced](https://flowrunner.ai/workflows/connect-chargebee-with-xero.md): Connect Chargebee subscription billing to Xero's ledger, optionally as an AI agent that mirrors clean invoices automatically and pauses for a named approver before an invoice gets voided. - [How to Connect CircleCI with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-circleci-with-slack.md): Connect CircleCI and Slack so build results, failures, and deploy gates land in the right Slack channel automatically, with an AI agent that pauses for an engineer before shipping to production. - [How to Connect Clearbit with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-clearbit-with-hubspot.md): Connect Clearbit enrichment to HubSpot CRM using FlowRunner: enrich inbound signups automatically, create complete contact and company records, and pause for a human before any low-confidence match triggers an outreach campaign. - [ClickUp + GitHub: Tasks Become Issues, Merges Log Back](https://flowrunner.ai/workflows/connect-clickup-with-github.md): Connect ClickUp and GitHub so triage tasks become GitHub issues and merges update the ClickUp task automatically, with an AI agent that pauses for a human before a risky merge. - [How to Connect ClickUp with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-clickup-with-google-sheets.md): Connect ClickUp and Google Sheets so new tasks, status changes, and triage decisions flow into your spreadsheets automatically, optionally as an AI agent that pauses for human judgment on the ambiguous calls. - [How to Connect Close CRM with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-closecrm-with-slack.md): Connect Close CRM and Slack so lead status changes, replied emails, and bulk CRM actions surface immediately in Slack, optionally as an AI agent that pauses for human judgment on bulk edits and late-stage deal routing. - [CloudTalk + HubSpot: Dial Qualified Leads While They're Warm](https://flowrunner.ai/workflows/connect-cloudtalk-with-hubspot.md): Connect CloudTalk and HubSpot so a qualified lead gets synced, matched, and dialed within seconds, run as an AI agent that pauses for a person only when a contact match is ambiguous. - [Log CloudTalk Call Outcomes in Pipedrive Automatically](https://flowrunner.ai/workflows/connect-cloudtalk-with-pipedrive.md): Connect CloudTalk and Pipedrive so qualified deals trigger an agent-first call and the outcome writes back to the deal automatically, with a human approving any contact merge. - [How to Connect Cohere with Pinecone (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-cohere-with-pinecone.md): Connect Cohere's embeddings and reranking to a Pinecone vector index through a FlowRunner workflow that runs autonomously and pauses for human judgment before any write that cannot be undone. - [Connect Confluence to Slack: Stale Pages Flagged, Not Deleted](https://flowrunner.ai/workflows/connect-confluence-with-slack.md): Connect Confluence and Slack so stale wiki pages get flagged for review automatically, with an AI agent that pauses for the documentation owner before anything is permanently removed. - [Constant Contact Eventbrite Integration: Attendees to Subscribers](https://flowrunner.ai/workflows/connect-constant-contact-with-eventbrite.md): Connect Constant Contact and Eventbrite so every event attendee becomes a subscriber automatically, with an AI agent that pauses for a human before the campaign sends. - [Copper + Google Sheets: Every Lead Becomes a Tracked Deal](https://flowrunner.ai/workflows/connect-copper-with-google-sheets.md): Connect Copper and Google Sheets so a new spreadsheet row becomes a qualified Copper lead automatically, with an AI agent that pauses for a sales manager before converting strategic accounts. - [How to Connect Copper with Mailchimp Marketing (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-copper-with-mailchimp-marketing.md): Connect Copper CRM to Mailchimp Marketing so new leads and converted contacts are added to the right audience automatically, with a human checkpoint before any campaign sends to a large list. - [Datadog + PagerDuty: Every Alert Pages With Evidence Attached](https://flowrunner.ai/workflows/connect-datadog-with-pagerduty.md): Connect Datadog and PagerDuty so confirmed alerts page on-call with correlated logs and metrics attached, and an AI agent can run the flow while a human acknowledges before any recovery step. - [Datadog + Slack: Incidents Declared Before the Pager Stops](https://flowrunner.ai/workflows/connect-datadog-with-slack.md): Connect Datadog to Slack so incident evidence and downtime approvals land in the same channel, run it as an AI agent that pauses for a human before it silences any monitor. - [How to Connect Deel with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-deel-with-quickbooks-online.md): Connect Deel contractor and employee payment events to QuickBooks Online automatically, with an AI agent that pauses for human review before recording large or disputed charges. - [Route Deel HR Exceptions to Slack for Approval](https://flowrunner.ai/workflows/connect-deel-with-slack.md): Connect Deel and Slack so contract, time-off, and termination events post to the right channel automatically, with an AI agent pausing for human approval on the moments that carry weight. - [Log Deepgram Call Transcripts in Notion Automatically](https://flowrunner.ai/workflows/connect-deepgram-with-notion.md): Connect Deepgram call transcripts to Notion pages automatically, and run it as an AI agent that pauses for a CS lead before a flagged call gets its save play decided. - [Send Delighted Detractor Alerts to Slack Automatically](https://flowrunner.ai/workflows/connect-delighted-with-slack.md): Connect Delighted and Slack so low NPS scores reach the account owner as an interactive Slack alert, optionally run as an AI agent that pauses for a human on the save call. - [Dialpad + HubSpot: Every Sales Call Becomes CRM History](https://flowrunner.ai/workflows/connect-dialpad-with-hubspot.md): Connect Dialpad to HubSpot so every sales call becomes a logged summary and follow-up task on the contact record, optionally as an AI agent that drafts the follow-up text and pauses for the rep's approval before it reaches the customer. - [Sync Dialpad Call Transcripts to Salesforce Pro Leads](https://flowrunner.ai/workflows/connect-dialpad-with-salesforce-pro.md): Connect Dialpad and Salesforce Pro so every sales call transcript lands on the right lead automatically, and let an AI agent decide when a transcript signals a lead is ready to convert, pausing for manager approval before that irreversible step. - [How to Connect Discord with Notion (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-discord-with-notion.md): Connect Discord and Notion so messages, threads, and moderation decisions flow automatically into your team's knowledge base, with an AI agent that pauses for a human before any consequential action. - [How to Connect DocuSign with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-aircall.md): Connect DocuSign envelope events to Aircall contacts and insight cards so reps see contract status on every call, optionally as an AI agent that pauses for the account owner before a declined contract is voided and reissued. - [How to Connect DocuSign with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-attio.md): Connect DocuSign envelopes to Attio records so signed contracts update your CRM automatically, with an AI agent that pauses for a named human before it marks a deal closed-won on mismatched terms. - [How to Connect DocuSign with Bigin (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-bigin-by-zoho.md): Connect DocuSign envelopes to Bigin by Zoho so signatures move your pipeline automatically, with an AI agent that pauses for a human before it voids a live envelope sitting in a customer's inbox. - [How to Connect DocuSign with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-bland.md): Connect DocuSign envelope events to Bland AI phone outreach, optionally as an AI agent that chases unsigned contracts by email first and pauses for a human before any phone call reaches a signer. - [How to Connect DocuSign with Capsule CRM (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-capsule-crm.md): Connect DocuSign envelope events to Capsule CRM opportunities and tasks, with an AI agent that closes won deals automatically and pauses for a human before Resend Envelope re-approaches a customer or a deal is marked lost. - [How to Connect DocuSign with Cin7 Omni (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-cin7.md): Connect DocuSign signed agreements to Cin7 Omni customers and sales orders, with an AI agent that onboards accounts automatically and pauses for a human before a signed order commits real inventory. - [How to Connect DocuSign with Cloudinary (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-cloudinary.md): Connect DocuSign completed envelopes to Cloudinary asset storage, optionally as an AI agent that archives and tags every executed contract automatically and pauses for a human before any signed document is deleted. - [How to Connect DocuSign with CloudTalk (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-cloudtalk.md): Connect DocuSign envelopes to CloudTalk calls and SMS, optionally as an AI agent that chases stalled signatures automatically and pauses for a human before it contacts a signer who declined. - [How to Connect DocuSign with Copilot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-copilot.md): Connect DocuSign envelope events to the Copilot client portal, optionally as an AI agent that onboards the signed client automatically and pauses for a human before Create Invoice sends the first bill. - [How to Connect DocuSign with Crisp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-crisp.md): Connect DocuSign envelope events to Crisp conversations, optionally as an AI agent that keeps signers informed in chat and pauses for a human before Send Message reaches a customer who just declined to sign. - [How to Connect DocuSign with Custify (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-custify.md): Connect DocuSign envelope events to Custify customer success records, optionally as an AI agent that logs signings and routes declines to CSMs, and pauses for a human before Create or Update Subscription rewrites an account's revenue data. - [How to Connect DocuSign with Cin7 Core (DEAR) (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-dear-inventory.md): Connect signed DocuSign agreements to Cin7 Core (DEAR) so a completed envelope becomes a customer and a sales order, with an AI agent that checks stock first and pauses for a human before an order allocates inventory. - [How to Connect DocuSign with Deskera (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-deskera.md): Connect signed DocuSign envelopes to Deskera Books so an executed contract becomes a contact, a sales order, and an invoice, with an AI agent that pauses for a human before the first invoice reaches the customer. - [How to Connect DocuSign with Google Drive (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-google-drive-api.md): Connect DocuSign and Google Drive so every completed envelope is automatically downloaded and filed in the right Drive folder, optionally as an AI agent that pauses for human approval before granting external access. - [How to Connect DocuSign with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-hubspot.md): Connect DocuSign and HubSpot so every completed envelope automatically updates the deal stage, creates the billing record, and routes exceptions to a human before anything posts. - [How to Connect DocuSign with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-docusign-with-salesforce-pro.md): Connect DocuSign and Salesforce Pro on FlowRunner so a signed envelope triggers instant CRM updates, record creation, and optional human approval before irreversible steps. - [DocuSign Slack Integration: Declined Contracts Ping the Right Human](https://flowrunner.ai/workflows/connect-docusign-with-slack.md): Connect DocuSign and Slack so completed and declined envelopes post automatically, or run it as an AI agent that escalates high-value declines to a sales manager and waits for a decision. - [How to Connect Dropbox with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-dropbox-service-with-google-sheets.md): Connect Dropbox and Google Sheets so every new file uploaded to a watched folder is logged as a row in your spreadsheet automatically, with an AI agent and a human in the loop for exceptions. - [Connect Dropbox to Slack: Route File Exceptions for Review](https://flowrunner.ai/workflows/connect-dropbox-service-with-slack.md): Connect Dropbox and Slack so new file uploads trigger a Slack summary automatically, and run it as an AI agent that pauses in Slack when a parsed file needs a human decision. - [How to Connect Dropcontact with Pipedrive (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-dropcontact-with-pipedrive.md): Connect Dropcontact and Pipedrive to enrich B2B contacts with verified emails and company data, then write clean records into your CRM automatically, with a human in the loop before any bulk outreach goes out. - [Connect EasyPost to Shopify: Verified Addresses, Automatic Labels](https://flowrunner.ai/workflows/connect-easypost-with-shopify.md): Connect EasyPost and Shopify so a new order buys a verified shipping label automatically, with the option to run it as an AI agent that pauses for a person when an address fails verification. - [Can an AI Agent Draft Eventbrite Events from Google Sheets Rows?](https://flowrunner.ai/workflows/connect-eventbrite-with-google-sheets.md): Connect Eventbrite to Google Sheets so a new event request row builds a draft Eventbrite event automatically, with a human confirming the details before it goes live and attendee and order data synced back for reporting. - [How to Connect Eventbrite with Mailchimp Marketing (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-eventbrite-with-mailchimp-marketing.md): Connect Eventbrite to Mailchimp Marketing so that new event registrants are automatically added to the right audience, tagged by event, and enrolled in follow-up sequences, with a human confirmation step before any campaign deploys to a large list. - [Expensify QuickBooks Online Integration: Reports Become Bills](https://flowrunner.ai/workflows/connect-expensify-with-quickbooks-online.md): Connect Expensify's approved expense reports to QuickBooks Online bills so reimbursements post themselves, optionally as an AI agent that verifies the vendor and pauses for a human before any bill over $10,000 goes out. - [How to Connect Facebook Lead Ads with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-facebook-lead-ads-with-hubspot.md): Connect Facebook Lead Ads to HubSpot so every form submission becomes a CRM contact in under a minute, optionally as an AI agent that deduplicates, routes, and pauses for a human before acting on ambiguous data. - [How to Connect Facebook Lead Ads with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-facebook-lead-ads-with-salesforce-pro.md): Use FlowRunner to connect Facebook Lead Ads and Salesforce Pro so every form submission becomes a scored, routed Salesforce lead automatically, with a human approval step before irreversible conversions. - [How to Connect Formstack with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-formstack-with-salesforce-pro.md): Connect Formstack to Salesforce Pro so every new submission creates or updates the right CRM record, with an AI agent that pauses for human approval before converting a lead. - [FreshBooks + Slack: Invoices Posted, Credit Notes Approved First](https://flowrunner.ai/workflows/connect-freshbooks-with-slack.md): Connect FreshBooks and Slack so invoices and payments post to your finance channel automatically, with an AI agent that pauses for a controller before a credit note or void goes through. - [Freshdesk + Google Sheets: Every Urgent Ticket Tracked](https://flowrunner.ai/workflows/connect-freshdesk-with-google-sheets.md): Connect Freshdesk and Google Sheets so urgent tickets land in a live triage sheet automatically, with a human approving any reply before it reaches a customer. - [How to Connect Freshdesk with Jira Issues (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-freshdesk-with-jira-issues.md): Connect Freshdesk and Jira Issues so that new support tickets automatically create tracked engineering issues, with an AI agent that pauses for human assignment on high-priority cases. - [How to Connect Freshdesk with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-freshdesk-with-slack.md): Connect Freshdesk and Slack so new support tickets post to your team channel automatically, or run it as an AI agent that triages tickets, drafts replies, and pauses for a human before anything reaches the customer. - [How to Connect Freshworks CRM with Mailchimp Marketing (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-freshworks-crm-with-mailchimp-marketing.md): Connect Freshworks CRM and Mailchimp Marketing so new contacts and deals automatically sync to the right audience, with an optional AI agent that pauses for human approval before sending to large lists. - [Front HubSpot Integration: Leads Synced, Duplicates Held for Review](https://flowrunner.ai/workflows/connect-front-service-with-hubspot.md): Connect Front and HubSpot so every new inbound conversation becomes a deduplicated HubSpot contact and deal, with an AI agent that pauses for a human on duplicate matches and outbound replies. - [How to Connect Front with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-front-service-with-slack.md): Connect Front's shared inbox to Slack so new inbound messages trigger Slack alerts and agent-drafted replies pause for human approval before anything reaches the customer. - [Gemini AI Google Sheets Integration: Extraction, Not Retyping](https://flowrunner.ai/workflows/connect-gemini-ai-with-google-sheets.md): Connect Gemini AI and Google Sheets so a new row triggers document extraction, and run it as an AI agent that pauses for a human on invoices that don't add up. - [How to Connect Gemini AI with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gemini-ai-with-slack.md): Connect Gemini AI and Slack so documents are analyzed, classified, and routed automatically, with a human decision requested in Slack whenever the agent finds something it should not decide alone. - [GitHub Google Sheets Integration: PR Tracker That Holds the Merge](https://flowrunner.ai/workflows/connect-github-with-google-sheets.md): Connect GitHub and Google Sheets to log pull requests and releases automatically, and run the same connection as an AI agent that pauses for a human before a risky merge ships. - [How to Connect GitHub with Jira Issues (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-github-with-jira-issues.md): Connect GitHub events to Jira Issues so new issues, pull requests, and releases automatically create or update Jira tickets, optionally as an AI agent that pauses for human judgment before assigning high-priority bugs. - [Send GitHub Issues to Linear Without Duplicate Tickets](https://flowrunner.ai/workflows/connect-github-with-linear.md): Connect GitHub and Linear so new issues sync automatically, with an AI agent that checks for duplicates and pauses for a human on ambiguous matches. - [Connect GitHub to Notion: Every PR Decision Logged](https://flowrunner.ai/workflows/connect-github-with-notion.md): Connect GitHub and Notion so pull request and release events write decision logs and release pages automatically, with a human approving the merges that matter. - [How to Connect GitHub with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-github-with-slack.md): Connect GitHub events to Slack channels so pull requests, releases, and exceptions surface automatically, with an AI agent that pauses for human judgment before merging or shipping. - [Can an AI Agent Keep GitLab and Jira Issues in Sync?](https://flowrunner.ai/workflows/connect-gitlab-with-jira-issues.md): Connect GitLab and Jira Issues so a monitoring alert or customer escalation files a linked issue in both, with an AI agent that pauses for a named engineer before any production merge. - [How to Connect GitLab with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gitlab-with-slack.md): Connect GitLab and Slack so issues, merge requests, and pipeline events post to your team's channels automatically, with an AI agent that pauses for a human before any code touches production. - [How to Connect Gmail with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gmail-service-with-google-sheets.md): Connect Gmail and Google Sheets on FlowRunner to log emails, attachments, and form responses automatically, or run the connection as an AI agent that pauses for human judgment on the steps that carry real consequence. - [How to Connect Gmail with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gmail-service-with-hubspot.md): Connect Gmail and HubSpot so inbound emails create contacts, open deals, and update your CRM automatically, with a human on the steps that carry real consequence. - [How to Connect Gmail with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gmail-service-with-salesforce-pro.md): Connect Gmail and Salesforce Pro so inbound emails automatically create leads, convert contacts, and log communications, with an AI agent that pauses for human judgment before irreversible CRM writes. - [How to Connect Gmail with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gmail-service-with-slack.md): Connect Gmail and Slack so new emails, attachments, and threads trigger Slack messages automatically, optionally as an AI agent that pauses for human judgment on the emails that carry real consequence. - [How to Connect Gmail with Trello (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gmail-service-with-trello.md): Connect Gmail and Trello so new emails automatically create and update Trello cards, optionally as an AI agent that reasons about each message and pauses for human judgment on exceptions. - [GoCardless QuickBooks Online Integration: Payouts to the Penny](https://flowrunner.ai/workflows/connect-gocardless-with-quickbooks-online.md): Connect GoCardless and QuickBooks Online so every payout reconciles to its line items automatically, with an AI agent that pauses for human review before any refund or mismatched entry posts to the books. - [How to Connect GoCardless with Xero (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gocardless-with-xero.md): Connect GoCardless bank-to-bank payments to Xero accounting with a FlowRunner flow that records collections automatically and pauses for human approval before issuing refunds or voiding invoices. - [Connect GoHighLevel to Google Sheets: Clean Rows Become Leads](https://flowrunner.ai/workflows/connect-gohighlevel-with-google-sheets.md): Connect GoHighLevel and Google Sheets so every tracked row becomes a qualified contact and opportunity automatically, with an AI agent that pauses for a person when a row is missing what it needs. - [GoHighLevel + Stripe: Every Closed Deal Billed Automatically](https://flowrunner.ai/workflows/connect-gohighlevel-with-stripe.md): Connect GoHighLevel to Stripe so a closed deal provisions a Stripe customer and invoice automatically, with an AI agent pausing for approval on large or first-time transactions. - [Sync Gong Call Insights to Salesforce Pro Automatically](https://flowrunner.ai/workflows/connect-gong-with-salesforce-pro.md): Connect Gong and Salesforce Pro so every call recap lands on the right record automatically, with an AI agent that pauses for a sales manager before it converts a lead. - [Gong + Slack: Every Call Recapped in the Team Channel](https://flowrunner.ai/workflows/connect-gong-with-slack.md): Connect Gong and Slack so every finished call lands as a recap in the team channel, and run it as an AI agent that pauses for a compliance owner before any GDPR erasure goes through. - [How to Connect Google Ads with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-ads-with-google-sheets.md): Connect Google Ads and Google Sheets so campaign performance writes to your spreadsheet automatically, with an AI agent that pauses for human approval before pausing any live campaign. - [How to Connect Google Calendar with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-calendar-with-slack.md): Connect Google Calendar and Slack so new events trigger instant channel notifications or direct messages, optionally run by an AI agent that pauses for a human before calendar changes affect multiple attendees. - [Log New Google Drive Files in Google Sheets Automatically](https://flowrunner.ai/workflows/connect-google-drive-api-with-google-sheets.md): Connect Google Drive and Google Sheets so every new file that lands in a watched folder is logged as a tracking row automatically, with an AI agent and a human in the loop on external sharing requests. - [How to Connect Google Drive with OpenAI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-drive-api-with-openai-ai.md): Connect Google Drive and OpenAI in FlowRunner so files landing in Drive trigger AI analysis, generation, or extraction automatically, with a human reviewing output before it leaves the flow. - [How to Connect Google Drive with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-drive-api-with-slack.md): Connect Google Drive and Slack so file events trigger instant channel updates, optionally as an AI agent that pauses for a human before granting external access or acting on sensitive documents. - [How to Connect Google Forms with Accelo (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-accelo.md): Connect Google Forms client requests to Accelo so every submission lands on the right client and job, with an AI agent that files routine requests automatically and pauses for a human before out-of-scope work becomes a scheduled task. - [How to Connect Google Forms with Adobe Workfront (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-adobe-workfront.md): Turn Google Forms work requests into Adobe Workfront tasks and projects, optionally as an AI agent that triages every submission and pauses for a traffic manager before a full project commits team capacity. - [How to Connect Google Forms with Aha! (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-aha-io.md): Turn Google Forms feedback submissions into deduplicated Aha! ideas, optionally as an AI agent that clusters raw responses by meaning and pauses for the product lead before any cluster is promoted onto the roadmap with Create Feature. - [How to Connect Google Forms with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-aircall.md): Turn Google Forms callback requests into a staged, routed Aircall queue, optionally as an AI agent that triages every submission and pauses for a human before Update Contact overwrites an existing contact's phone and email set. - [How to Connect Google Forms with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-attio.md): Turn Google Forms submissions into clean Attio records and pipeline entries, with an agent that classifies each response and pauses for a human before Delete Record removes a suspected duplicate. - [How to Connect Google Forms with AWeber (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-aweber.md): Move Google Forms signups onto the right AWeber list automatically, with an agent that screens junk submissions, tags real subscribers by interest, and pauses for a marketer before Create Broadcast reaches the list. - [How to Connect Google Forms with Amazon Redshift (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-aws-redshift.md): Stream Google Forms responses into Amazon Redshift for analysis, with an agent that detects form edits before they corrupt the table and pauses for an engineer before any schema migration runs. - [How to Connect Google Forms with Basecamp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-basecamp3.md): Turn Google Forms intake responses into Basecamp to-dos and briefs automatically, with an AI agent that pauses for a human before Create Project spins up a whole new project and pulls a team in. - [How to Connect Google Forms with beehiiv (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-beehiiv.md): Connect Google Forms signups to your beehiiv newsletter so clean responses become subscribers automatically, with a human approving any reactivation of a reader who previously unsubscribed. - [How to Connect Google Forms with Bigin (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-bigin-by-zoho.md): Connect Google Forms responses to Bigin contacts and deals, optionally as an AI agent that creates clean pipeline entries and pauses for a human before Update Contact overwrites data you already trust. - [How to Connect Google Forms with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-bland.md): Connect Google Forms responses to Bland AI phone calls, optionally as an AI agent that drafts a callback per respondent and pauses for a human before Send Call dials anyone who submitted the form. - [How to Connect Google Forms with Braze (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-braze.md): Connect Google Forms responses to Braze profiles and campaigns, optionally as an AI agent that updates attributes and fires welcome journeys on its own and pauses for a human before Delete Users erases a profile forever. - [How to Connect Google Forms with Campaign Monitor (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-campaign-monitor.md): Sync Google Forms responses into Campaign Monitor subscriber lists automatically, with an AI agent that maps custom fields and consent, and pauses for a human before a previously unsubscribed address is re-added. - [How to Connect Google Forms with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-forms-with-slack.md): Connect Google Forms to Slack so every form submission posts a real-time Slack notification, with optional AI-agent routing that pauses for a human on low scores, vendor exceptions, or any response that carries real consequence. - [How to Connect Google Sheets with Mailchimp Marketing (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-sheets-with-mailchimp-marketing.md): Connect Google Sheets to Mailchimp Marketing so every new row automatically adds or updates the subscriber, applies the right tag, and enrolls them in the matching email journey, optionally as an AI agent that pauses for a human before sending campaigns to large lists. - [How to Connect Google Sheets with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-google-sheets-with-slack.md): Connect Google Sheets and Slack so a new or updated row triggers a Slack message automatically, or run it as an AI agent that reads the data, decides what to escalate, and pauses for a human on the step that carries real consequence. - [How to Connect Gravity Forms with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-gravity-forms-with-hubspot.md): Connect Gravity Forms to HubSpot so every WordPress form submission creates a contact and deal automatically, with an AI agent that pauses for human review on the high-value leads validation flags. - [Help Scout + HubSpot: Every Ticket Linked to Its Deal](https://flowrunner.ai/workflows/connect-help-scout-with-hubspot.md): Connect Help Scout conversations to HubSpot contacts and deals automatically, and run it as an AI agent that pauses for a person before a support case touches an active deal. - [Send Help Scout Escalations to Jira Issues Automatically](https://flowrunner.ai/workflows/connect-help-scout-with-jira-issues.md): Connect Help Scout and Jira Issues so escalated conversations become tracked engineering issues automatically, or run it as an AI agent that pauses for a team lead before assigning a P1. - [How to Connect Help Scout with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-help-scout-with-slack.md): Connect Help Scout and Slack so every new support conversation triggers a Slack notification, optionally running as an AI agent that triages, drafts replies, and pauses for a human before anything reaches the customer. - [HeyGen + YouTube: Campaign Video Rendered, Reviewed, Then Published](https://flowrunner.ai/workflows/connect-heygen-with-youtube.md): Connect HeyGen and YouTube so finished campaign video renders upload straight to a channel, optionally as an AI agent that pauses for a human before anything using a cloned voice goes public. - [How to Connect HubSpot with Accelo (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-accelo.md): Connect HubSpot closed-won deals to Accelo so every new client gets a company record, contacts, and a kickoff checklist, with an AI agent that runs the handoff automatically and pauses for a human before creating a client record that looks like a duplicate. - [How to Connect HubSpot with Adobe Acrobat Sign (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-adobe-sign.md): Connect HubSpot deals to Adobe Acrobat Sign so contracts go out the moment a deal reaches the contract stage, with an AI agent that prepares every agreement automatically and pauses for a human before a binding document reaches the customer. - [How to Connect HubSpot with Adobe Workfront (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-adobe-workfront.md): Connect closed-won HubSpot deals to Adobe Workfront projects, optionally as an AI agent that builds the delivery plan automatically and pauses for a delivery lead before project dates become client commitments. - [How to Connect HubSpot with Aha! (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-aha-io.md): Route feature requests and deal-blocking gaps from HubSpot into Aha! as ideas, optionally as an AI agent that dedupes and files them automatically and pauses for a product lead before anything is promoted onto the roadmap. - [How to Connect HubSpot with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-aircall.md): Sync Aircall calls and contacts with HubSpot so every call lands on a CRM record, optionally as an AI agent that logs and tags calls automatically and pauses for a human before Delete Contact permanently removes a record during dedupe. - [How to Connect HubSpot with AssemblyAI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-assembly-ai.md): Turn call recordings into speaker-labeled transcripts with AssemblyAI and log them against the right HubSpot contact and deal, with an agent that pauses for a rep before it changes a deal stage based on what it heard. - [How to Connect HubSpot with AWeber (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-aweber.md): Keep AWeber lists in step with HubSpot lifecycle stages automatically, with an agent that segments subscribers by CRM context and pauses for a marketer before Create Broadcast sends anything to a list. - [How to Connect HubSpot with Amazon Redshift (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-aws-redshift.md): Load HubSpot contacts, companies, and deals into Amazon Redshift on a schedule, with an agent that handles schema drift and pauses for an engineer before any destructive SQL like a TRUNCATE or table rebuild runs. - [How to Connect HubSpot with Base64.ai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-base64-ai.md): Connect Base64.ai document extraction to HubSpot so scanned order forms and contracts update contacts, companies, and deals automatically, with a human approving any low-confidence write before it touches the deal record. - [How to Connect HubSpot with Basecamp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-basecamp3.md): Connect HubSpot deals to Basecamp so closed-won deals become fully scaffolded delivery projects, with a human approving the irreversible step of trashing a project when a deal falls through. - [How to Connect HubSpot with beehiiv (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-beehiiv.md): Connect HubSpot contacts to your beehiiv newsletter so the CRM and the audience stay in step, with a human approving every permanent subscriber deletion before it runs. - [How to Connect HubSpot with Bettermode (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-bettermode.md): Connect your Bettermode community to HubSpot so members become CRM contacts with their activity on the record, with a human moderator approving every irreversible post deletion before it runs. - [How to Connect HubSpot with BigMarker (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-bigmarker.md): Connect HubSpot contacts to BigMarker webinars, optionally as an AI agent that builds the registrant list and pauses for a human before Register Person sends confirmation emails to a whole segment. - [Turn Google Sheets Rows Into HubSpot Contacts and Deals](https://flowrunner.ai/workflows/connect-hubspot-with-google-sheets.md): Connect HubSpot and Google Sheets so a new row creates a matching contact and deal automatically, with an AI agent that pauses for a person when two records could be the same prospect. - [How to Connect HubSpot with Jira Issues (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-jira-issues.md): Connect HubSpot and Jira Issues so deals, customer escalations, and CRM events automatically create, update, and transition issues, with a human on the step that carries real consequence. - [Connect HubSpot to Notion: Deals Documented Automatically](https://flowrunner.ai/workflows/connect-hubspot-with-notion.md): Connect HubSpot and Notion so every deal gets a documentation page automatically, with an AI agent that pauses for human review before anything goes external. - [How to Connect HubSpot with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-salesforce-pro.md): Connect HubSpot and Salesforce Pro so contacts, deals, and lead conversions stay in sync across both CRMs automatically, with an AI agent that pauses for human approval before irreversible Salesforce actions. - [How to Connect HubSpot with Slack Using FlowRunner](https://flowrunner.ai/workflows/connect-hubspot-with-slack.md): Route inbound leads to your CRM in seconds, catch duplicates before they multiply, and give your sales team deal updates where they already work. - [How to Connect HubSpot with Tally (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hubspot-with-tally.md): Connect Tally form submissions to HubSpot so every response becomes a contact, company, and deal, with an AI agent that pauses for a human before it overwrites an existing contact a sales rep already owns. - [How to Connect Hunter.io with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-hunter-with-hubspot.md): Connect Hunter.io and HubSpot so FlowRunner agents discover and verify email addresses, then create contacts and deals in your CRM automatically, pausing for human approval before any found list enters a live campaign. - [Sync Insightly Deals to Mailchimp Marketing Automatically](https://flowrunner.ai/workflows/connect-insightly-with-mailchimp-marketing.md): Connect Insightly and Mailchimp Marketing so a won deal syncs the contact and tag automatically, with an AI agent that holds large campaign sends for a human's approval. - [Log Instantly Email Replies in Google Sheets Automatically](https://flowrunner.ai/workflows/connect-instantly-with-google-sheets.md): Connect Instantly and Google Sheets so every campaign reply is classified and logged automatically, with an AI agent that pauses for a sales manager on ambiguous replies. - [Instantly + HubSpot: Positive Replies Become Deals](https://flowrunner.ai/workflows/connect-instantly-with-hubspot.md): Connect Instantly and HubSpot so every campaign reply routes itself, and run it as an AI agent that pauses for a sales manager on ambiguous replies. - [Intercom Google Sheets Integration: Log Chats, Flag the Hard Ones](https://flowrunner.ai/workflows/connect-intercom-with-google-sheets.md): Connect Intercom and Google Sheets so every conversation is logged automatically, with an AI agent that drafts routine replies and pauses for a person on the angry or ambiguous ones. - [How to Connect Intercom with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-intercom-with-hubspot.md): Connect Intercom and HubSpot so every new conversation, contact, and ticket flows into your CRM automatically, with an AI agent that pauses for human judgment before touching a record that carries real consequence. - [Connect Intercom to Jira Issues: Escalations Tracked in Seconds](https://flowrunner.ai/workflows/connect-intercom-with-jira-issues.md): Connect Intercom and Jira Issues so a support conversation becomes a tracked engineering ticket automatically, with an AI agent that pauses for a human before assigning a P1. - [How to Connect Intercom with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-intercom-with-slack.md): Connect Intercom and Slack so new conversations, tickets, and contacts route to the right Slack channel automatically, with a human in the loop on messages that need judgment. - [Invoice Ninja Google Sheets Integration: Rows Become Invoices](https://flowrunner.ai/workflows/connect-invoice-ninja-with-google-sheets.md): Connect Invoice Ninja and Google Sheets so a new billing request row creates the client and sends the invoice automatically, with an AI agent pausing for a person before an oversized or unfamiliar invoice goes out. - [Connect Jasper to WordPress: On-Brand Drafts, Staged for Approval](https://flowrunner.ai/workflows/connect-jasper-ai-with-wordpress.md): Connect Jasper and WordPress so grounded, on-brand drafts stage automatically as WordPress posts, with an AI agent that pauses for an editor before anything publishes. - [How to Connect Jira with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-aircall.md): Turn Aircall support calls into Jira issues automatically, with an AI agent that files routine reports on its own and pauses for a human before a critical escalation pages the on-call engineer. - [How to Connect Jira Issues with BigMarker (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-bigmarker.md): Run webinar operations out of Jira: approved webinar tickets become BigMarker conferences automatically, and an AI agent pauses for a human before Update Conference changes a live event that registrants already signed up for. - [How to Connect Jira with ClickMeeting (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-clickmeeting.md): Turn Jira tickets into scheduled ClickMeeting sessions automatically, with an AI agent that manages registrations and pauses for a human before Update Conference or Delete Conference touches a session people signed up for. - [How to Connect Jira with Confluence (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-confluence.md): Turn resolved Jira tickets into Confluence documentation automatically, with an AI agent that drafts postmortems and release notes and pauses for a human before Update Page rewrites a runbook the whole company relies on. - [How to Connect Jira with Delighted (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-delighted.md): Route Delighted detractor feedback into Jira issues automatically, with an AI agent that triages verbatims into the right project and pauses for a human before Unsubscribe Person or Delete Person permanently changes a customer's survey record. - [How to Connect Jira Issues with Elastic Email (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-elastic-email.md): Connect Jira Issues to Elastic Email so resolved tickets turn into customer notifications automatically, with an AI agent that drafts every message and pauses for a human before a bulk announcement reaches a full contact list. - [How to Connect Jira Issues with Everhour (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-everhour.md): Connect Jira Issues to Everhour so closed tickets become logged time in the right project, with an AI agent that mirrors tasks automatically and pauses for a human before billable hours land on a client project. - [How to Connect Jira Issues with Fathom (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-fathom.md): Connect Fathom meeting recordings to Jira Issues so action items from recorded calls become tracked tickets, with an AI agent that drafts each issue and pauses for a human before a batch of work lands in the team's backlog. - [How to Connect Jira Issues with Formbricks (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-formbricks.md): Connect Formbricks survey responses to Jira Issues so real user feedback becomes tracked engineering work, with an AI agent that separates bugs from noise and pauses for a human before severe reports jump the sprint queue. - [How to Connect Jira Issues with Frame.io (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-frame-io.md): Connect Frame.io review comments to Jira Issues so timestamped video feedback becomes tracked production work, with an AI agent that clusters notes into tickets and pauses for a human before client change requests enter the sprint as free work. - [How to Connect Jira Issues with Fulcrum (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-fulcrum.md): Connect Fulcrum field inspection records to Jira Issues so defects found on site become tracked work orders with photos attached, with an AI agent that files routine findings and pauses for a human before closing out a field record as resolved. - [How to Connect Jira with Hubstaff (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-hubstaff.md): Reconcile Jira issue status with Hubstaff tracked time automatically, with an AI agent that comments the evidence on each issue and pauses for a project lead before it moves or reassigns anyone's work. - [How to Connect Jira with iAuditor (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-iauditor.md): Turn failed iAuditor inspection items into tracked Jira issues with the inspection PDF attached, and let an AI agent triage severity while a person confirms every safety finding before it is closed. - [How to Connect Jira Issues with Tally (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jira-issues-with-tally.md): Connect Tally form submissions to Jira so every request becomes a triaged, assigned issue, with an AI agent that classifies each submission and pauses for a human before a duplicate ticket is deleted. - [How to Connect Jotform with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jotform-with-google-sheets.md): Connect Jotform to Google Sheets so every form submission lands as a row automatically, then let a FlowRunner AI agent validate the data, route exceptions to a human, and keep your sheet clean. - [How to Connect Jotform with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-jotform-with-hubspot.md): Connect Jotform to HubSpot so every form submission automatically becomes a validated contact and deal in your CRM, with a human in the loop before any incomplete or duplicate record is created. - [JustCall HubSpot Integration: Calls Logged, Duplicates Reviewed](https://flowrunner.ai/workflows/connect-justcall-with-hubspot.md): Connect JustCall to HubSpot so every call logs its outcome in the CRM automatically, with an AI agent that pauses for a human when a contact match is ambiguous. - [How to Connect JustCall to Pipedrive: Log Every Call on the Deal](https://flowrunner.ai/workflows/connect-justcall-with-pipedrive.md): Connect JustCall and Pipedrive so every call and follow-up text lands on the right deal, and let an AI agent run the sync while a human approves the first text sent from the team's number. - [How to Connect Keap with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-keap-with-quickbooks-online.md): Connect Keap to QuickBooks Online so every new order, contact, or opportunity triggers the right accounting action automatically, with a human on any transaction that carries real financial consequence. - [Sync Google Sheets Contacts to Klaviyo Profiles Automatically](https://flowrunner.ai/workflows/connect-klaviyo-with-google-sheets.md): Connect Google Sheets and Klaviyo so new contact rows upsert Klaviyo profiles and record consent automatically, with a human confirming any row where consent is missing or unclear. - [Can an AI Agent Sync Shopify to Klaviyo and Hold Refunds?](https://flowrunner.ai/workflows/connect-klaviyo-with-shopify.md): Connect Shopify orders to Klaviyo profiles and events automatically, and run it as an AI agent that pauses for a human before issuing a refund. - [Klaviyo + Typeform: Every Response Becomes a Subscribed Profile](https://flowrunner.ai/workflows/connect-klaviyo-with-typeform.md): Connect Klaviyo and Typeform so every form submission upserts a profile and records consent, with an AI agent holding sensitive answers for human review before anything reaches Klaviyo. - [Connect Klaviyo to WooCommerce: Orders Fuel the Next Campaign](https://flowrunner.ai/workflows/connect-klaviyo-with-woocommerce.md): Connect Klaviyo and WooCommerce so every order updates the buyer's profile and revenue attribution in minutes, with a human confirming the audience before any campaign sends. - [Connect Lemlist to HubSpot: Replies Become Deals](https://flowrunner.ai/workflows/connect-lemlist-with-hubspot.md): Connect Lemlist and HubSpot so an engaged prospect's reply creates or updates a HubSpot contact and deal automatically, with an AI agent pausing for a rep before an interested lead moves off the sequence. - [How to Connect LinkedIn with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-linkedin-with-google-sheets.md): Connect LinkedIn and Google Sheets so new spreadsheet rows trigger LinkedIn posts or post data flows back to Sheets, optionally as an AI agent that pauses for human approval before anything publishes. - [How to Connect Magento 2 with NetSuite (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-magento-with-netsuite.md): Connect Magento 2 and NetSuite so every new order flows into your ERP automatically, optionally as an AI agent that creates the sales order, invoices it, and pauses for a human before posting anything a finance team cannot easily undo. - [How to Connect Mailchimp Marketing with Bettermode (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-bettermode.md): Sync Bettermode community members into your Mailchimp Marketing audience and recap campaigns back to the community, with an AI agent that pauses for a human before Send Campaign reaches the whole list. - [How to Connect Mailchimp Marketing with Bluesky (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-bluesky.md): Turn Mailchimp campaign performance into Bluesky posts automatically, with an AI agent that drafts the social copy and pauses for a marketing lead before anything publishes under the brand handle. - [How to Connect Mailchimp with Circle (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-circle-so.md): Sync new Circle community members into Mailchimp with tags and segments, and let an AI agent run the loop that pauses for a human before Send Campaign emails the whole community. - [How to Connect Mailchimp with Delighted (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-delighted.md): Segment Mailchimp by real NPS sentiment from Delighted, with an AI agent that tags promoters and detractors automatically and pauses for a human before Send Campaign puts a referral ask in front of anyone. - [How to Connect Mailchimp Marketing with FirstPromoter (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-firstpromoter.md): Connect FirstPromoter to Mailchimp Marketing so affiliates flow into a tagged, segmented audience automatically, with an AI agent that keeps promoter data in step and pauses for a human before a campaign reaches your whole partner list. - [How to Connect Mailchimp Marketing with Formbricks (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-formbricks.md): Connect Formbricks survey responses to Mailchimp Marketing so feedback reshapes your segments automatically, with an AI agent that tags members by what they actually said and pauses for a human before a follow-up campaign reaches unhappy customers. - [How to Connect Mailchimp Marketing with Fulcrum (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-fulcrum.md): Connect Fulcrum field data collection to Mailchimp Marketing so contacts gathered on site flow into your audience automatically, with an AI agent that verifies consent per record and pauses for a human before ambiguous sign-ups join the list. - [How to Connect Mailchimp Marketing with Hootsuite (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-hootsuite.md): Turn Mailchimp campaign results into scheduled Hootsuite posts, optionally run by an AI agent that drafts the social version of every winning email and pauses for a human before anything publishes under a brand profile. - [How to Connect Mailchimp Marketing with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-hubspot.md): Connect Mailchimp Marketing and HubSpot so every CRM lifecycle change updates your email audiences in real time, with an AI agent that pauses for human approval before deploying campaigns to large lists. - [How to Connect Mailchimp Marketing with iAuditor (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-iauditor.md): Turn iAuditor inspection results into targeted Mailchimp briefings for the right site managers, with an AI agent that segments by inspection outcome and pauses for an operations lead before any campaign reaches the network. - [How to Connect Mailchimp Marketing with Instagram Business (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-instagram-business.md): Run email and Instagram as one loop: campaign click data shapes what gets published, media insights shape the next send, and an AI agent pauses for a person before anything posts to the brand feed. - [How to Connect Mailchimp Marketing with LinkedIn Ads (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-linkedin-ads-campaign-mgmt.md): Let Mailchimp engagement data shape your LinkedIn ad campaigns, with an AI agent that proposes campaigns and budget changes from click evidence and pauses for a named approver before any action that spends money. - [How to Connect Mailchimp Marketing with Mastodon (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-mastodon.md): Bring your Mailchimp campaign content to the fediverse the way Mastodon expects, with an AI agent that listens before it posts and pauses for a person before Post Status or Boost Status speaks for the brand. - [How to Connect Mailchimp Marketing with Meta Ads (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-meta-ads.md): Sync Mailchimp audiences into Meta Ads custom audiences on a schedule, optionally as an AI agent that builds campaigns paused and waits for a named human before anything starts to spend. - [How to Connect Mailchimp Marketing with Tally (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mailchimp-marketing-with-tally.md): Connect Tally form submissions to Mailchimp so every signup lands in the right audience with the right tags, with an AI agent that segments contacts automatically and pauses for a human before a campaign is sent or a member is removed. - [Connect Marketo to Salesforce Pro: Sync Leads, Gate Conversions](https://flowrunner.ai/workflows/connect-marketo-with-salesforce-pro.md): Connect Marketo and Salesforce Pro so new leads sync into the CRM automatically, with an AI agent that pauses for a sales manager before an irreversible lead conversion. - [How to Connect Microsoft Teams with Jira Issues (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-microsoft-teams-with-jira-issues.md): Connect Microsoft Teams and Jira Issues so that exceptions flagged in Teams automatically become tracked, assigned Jira issues, with a human on the consequential step. - [How to Connect Mindee with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mindee-with-quickbooks-online.md): Connect Mindee's document extraction to QuickBooks Online so vendor invoices and receipts move from image to verified bill automatically, with a human in the loop before uncertain data posts to your books. - [Can Mistral AI Flag Contract Clauses for Slack Review?](https://flowrunner.ai/workflows/connect-mistral-ai-with-slack.md): Connect Mistral AI and Slack so an agent reads and summarizes contracts, then pauses in Slack for a human when it finds a clause that isn't standard. - [Mollie + QuickBooks Online: Every Payout Posted, Refunds Gated](https://flowrunner.ai/workflows/connect-mollie-with-quickbooks-online.md): Connect Mollie and QuickBooks Online so settlements post to your books automatically, with an AI agent that pauses for human approval before a refund or mandate revocation goes through. - [Can an AI Agent Keep Monday.com and Google Sheets in Sync?](https://flowrunner.ai/workflows/connect-monday-with-google-sheets.md): Connect Monday.com and Google Sheets so sheet rows update board items and board status changes log back to the sheet, with a human approving any bulk pricing or status change. - [Connect Monday.com to HubSpot: Board Items Become CRM Deals](https://flowrunner.ai/workflows/connect-monday-with-hubspot.md): Connect Monday.com and HubSpot so a status change on a board item opens or updates a deal automatically, as an AI agent that pauses for a CRM admin when it finds a possible duplicate contact. - [How to Connect MongoDB with Google BigQuery (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mongodb-with-bigquery.md): Connect MongoDB and Google BigQuery so aggregated document data flows into your warehouse automatically, with an AI agent that pauses for human approval before any irreversible delete. - [MongoDB + Slack: Daily Rollups Posted, Deletes Paused for a Human](https://flowrunner.ai/workflows/connect-mongodb-with-slack.md): Connect MongoDB and Slack so a scheduled rollup posts to your team automatically, and run it as an AI agent that reads the collection, reasons about risk, and pauses for a human before any bulk delete or collection drop. - [How to Connect MySQL with Google BigQuery (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-mysql-with-bigquery.md): Connect MySQL to Google BigQuery with FlowRunner: sync rows from your operational database into the warehouse, run analytical queries, and let an AI agent pause for a human before any write that cannot be undone. - [Update MySQL from Google Sheets, With Approval on Bulk Writes](https://flowrunner.ai/workflows/connect-mysql-with-google-sheets.md): Connect MySQL and Google Sheets so ops teams submit bulk update requests from a spreadsheet, with an AI agent that runs the write and pauses for approval before touching a production table. - [How to Connect NetSuite with Adobe Acrobat Sign (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-adobe-sign.md): Send NetSuite sales orders out for e-signature with Adobe Acrobat Sign, run by an AI agent that stages and tracks agreements, archives the signed PDF and audit trail, and pauses for a human before a high-value or non-standard order goes out for signature. - [How to Connect NetSuite with AfterShip (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-aftership.md): Put every shipped NetSuite sales order under AfterShip tracking automatically, with an AI agent that watches for stalled shipments and pauses for a human before a replacement order ships product twice. - [How to Connect NetSuite with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-aircall.md): Put live NetSuite order and invoice context on screen during Aircall collections and support calls, with an AI agent that preps every call on its own and pauses for finance before a phone dispute changes an invoice. - [How to Connect NetSuite with Base64.ai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-base64-ai.md): Turn scanned customer purchase orders into NetSuite sales orders with Base64.ai extraction, run by an AI agent that matches customers and items automatically and pauses for a named approver before Create Sales Order books the order. - [How to Connect NetSuite with Braintree (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-braintree.md): Reconcile settled Braintree transactions into NetSuite customers, invoices, and payments automatically, with an AI agent that flags mismatches and pauses for a finance owner before a refund executes or a ledger record is corrected. - [How to Connect NetSuite with Canva (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-canva.md): Connect NetSuite records to Canva brand templates so price lists, customer one-pagers, and sales reports render themselves, with an AI agent that pauses for a human before a document carrying live pricing is exported for customers. - [How to Connect NetSuite with CloudConvert (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-cloudconvert.md): Connect NetSuite to CloudConvert so invoice and sales order documents are standardized, merged into audit packets, and archived automatically, with an AI agent that flags duplicate invoices and pauses for the controller before Delete Invoice touches the ERP. - [How to Connect NetSuite with Cloudinary (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-cloudinary.md): Keep Cloudinary product media in step with the NetSuite item master automatically, with an AI agent that tags and transforms assets on its own and pauses for a human before Destroy Asset permanently deletes anything. - [How to Connect NetSuite with DocuSeal (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-docuseal.md): Send vendor agreements and order confirmations for signature the moment NetSuite records are ready, with an AI agent that prefills every DocuSeal submission and pauses for a human before non-standard terms go out for signature. - [How to Connect NetSuite with Egnyte (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-egnyte.md): File NetSuite invoices, orders, and vendor records into a governed Egnyte folder structure automatically, with an AI agent that keeps the filing consistent and pauses for a human before Create Shared Link exposes a financial document outside the company. - [How to Connect NetSuite with Files.com (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-files-com.md): Connect NetSuite to Files.com so invoices, statements, and ERP exports land in governed folders automatically, with an AI agent that stages every file and pauses for a human before a share bundle exposes financial documents outside the company. - [How to Connect NetSuite with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-netsuite-with-google-sheets.md): Connect NetSuite and Google Sheets so financial records, sales orders, and payment data flow automatically between your ERP and your spreadsheets, optionally as an AI agent that pauses for human approval on consequential writes. - [How to Connect Notion with Accelo (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-accelo.md): Turn rows in a Notion tracker into tasks on the right Accelo client jobs, run by an AI agent that matches work to jobs and pauses for a human before billable work lands on a paused, closed, or mismatched engagement. - [How to Connect Notion with Adobe Workfront (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-adobe-workfront.md): Turn requests in a Notion database into Adobe Workfront tasks and projects, run by an AI agent that routes work into existing projects and pauses for a human before a request becomes a brand-new resourced project. - [How to Connect Notion with Aha! (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-aha-io.md): Sync customer feedback from a Notion database into Aha! as ideas automatically, with an AI agent that dedupes and files on its own and pauses for a product lead before anything is promoted to a roadmap feature. - [How to Connect Notion with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-aircall.md): Log every Aircall conversation into a Notion account database automatically, with an AI agent that reads call notes for churn signals and pauses for a human before an account is flagged at risk and the save play begins. - [How to Connect Notion with Amazon Redshift (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-aws-redshift.md): Sync Notion database rows into Amazon Redshift on a schedule, optionally as an AI agent that loads changed rows automatically and pauses for a human before any ALTER TABLE or destructive SQL touches the warehouse. - [How to Connect Notion with Basecamp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-basecamp3.md): Turn Notion database rows into Basecamp projects, to-do lists, and schedules automatically, with an AI agent that builds the workspace on its own and pauses for a human before Trash Project tears anything down. - [How to Connect Notion with Canny (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-canny.md): Keep a Notion roadmap database in step with your Canny feedback board, with an AI agent that syncs votes and statuses automatically and pauses for a product manager before a public comment goes out to every voter. - [How to Connect Notion with Caspio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-caspio.md): Mirror Caspio application records into a Notion workspace and write team decisions back, with an AI agent that previews every bulk operation and pauses for the operations owner before Update Table Records or Delete Table Records touches production data. - [How to Connect Notion with Clio Manage (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-clio-manage.md): Connect a Notion intake database to Clio Manage so approved intakes become contacts and matters automatically, with an AI agent that runs the conflict screen and pauses for the supervising attorney before Create Matter opens a billable file. - [How to Connect Notion with Google Calendar (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-google-calendar.md): Connect Notion and Google Calendar so meetings create documentation automatically, prep pages link to upcoming events, and a human confirms any change that affects multiple attendees. - [Connect Notion to Google Sheets: Every Row Becomes a Page](https://flowrunner.ai/workflows/connect-notion-with-google-sheets.md): Connect Notion and Google Sheets so every tracked row becomes a reviewed Notion record, run as a plain sync or as an AI agent that pauses for a person before a page goes external. - [How to Connect Notion with Keboola (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-keboola.md): Land Notion databases in Keboola Storage on a nightly schedule, with an AI agent that detects schema drift before it corrupts a table and pauses for a data owner before any destructive import or production job run. - [How to Connect Notion with Kintone (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-kintone.md): Keep Kintone app records and a Notion database in two-way sync, with an AI agent that reconciles conflicting edits and pauses for the record owner before Delete Records or a bulk Update Records touches the system of record. - [How to Connect Notion with Lark Base (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-larksuitebase.md): Mirror a shared tracker between Notion and Lark Base so each team works in its own suite, with an AI agent that reviews what crosses the boundary and pauses for an owner before records are shared or deleted. - [How to Connect Notion with Ninox (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-ninox.md): Mirror Ninox records into a Notion database on a schedule, optionally as an AI agent that reconciles both sides and pauses for a human before any delete or bulk overwrite touches the operational database. - [How to Connect Notion with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-slack.md): Connect Notion and Slack so that page creation, database updates, and document events in Notion trigger structured Slack messages automatically, with an AI agent that pauses for human review on the steps that carry real consequence. - [How to Connect Notion with Xata (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-notion-with-xata.md): Surface Xata serverless Postgres data in Notion on a schedule, optionally as an AI agent that applies workspace edits back to production and pauses for a human before any SQL write, replace, or delete. - [How to Connect Odoo with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-odoo-with-google-sheets.md): Connect Odoo and Google Sheets so order and contact records sync automatically, optionally as an AI agent that pauses for human approval before posting high-value commitments. - [How to Connect Okta with ServiceNow (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-okta-with-servicenow.md): Connect Okta's System Log trigger to ServiceNow incidents and change requests in FlowRunner, optionally as an AI agent that suspends the user, files the record, and pauses for a human before any irreversible action. - [How to Connect Okta with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-okta-with-slack.md): Connect Okta and Slack so security events, user provisioning changes, and access decisions surface in Slack automatically, with an AI agent that pauses for human judgment before any irreversible identity action. - [How to Connect OpenAI with ElevenLabs (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-openai-ai-with-elevenlabs.md): Connect OpenAI's text generation and moderation to ElevenLabs voice synthesis in a FlowRunner workflow, then optionally run it as an AI agent that pauses for a human before publishing branded audio. - [OpenAI + Google Sheets: Every New Row Gets a Reviewed Draft](https://flowrunner.ai/workflows/connect-openai-ai-with-google-sheets.md): Connect OpenAI and Google Sheets so every new row gets classified and drafted a reply automatically, with an AI agent that pauses for a human before a complaint or flagged draft goes out. - [OpenAI Notion Integration: Every Call Becomes a Reviewed Page](https://flowrunner.ai/workflows/connect-openai-ai-with-notion.md): Connect OpenAI and Notion so every sales call recording becomes a documented Notion page automatically, with an AI agent that pauses for a human before a recap reaches a client. - [How to Connect OpenAI with Pinecone (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-openai-ai-with-pinecone.md): Connect OpenAI embeddings and generation to Pinecone vector storage in a FlowRunner workflow, optionally as an AI agent that pauses for human approval before any irreversible memory change. - [How to Connect OpenAI with Qdrant (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-openai-ai-with-qdrant.md): Connect OpenAI's embedding and generation APIs to Qdrant's vector database in FlowRunner, optionally as an AI agent that pauses for human approval before any bulk write rewrites your collection at scale. - [Send OpenAI Drafts to Slack for Human Approval](https://flowrunner.ai/workflows/connect-openai-ai-with-slack.md): Connect OpenAI and Slack so generated replies, answers, and content route into a Slack approval card, with the AI agent inserting a human on any draft that reaches a customer or touches sensitive material. - [How to Connect OpenAI with Weaviate (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-openai-ai-with-weaviate.md): Connect OpenAI embeddings and generation to Weaviate vector search in FlowRunner, optionally as an AI agent that pauses for human approval before any batch delete removes data. - [How to Connect Outlook with Accelo (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-accelo.md): Turn client emails in Outlook into tracked Accelo tasks and logged activities, run by an AI agent that files each request against the right job and pauses for a human before any reply that commits your team reaches the client. - [How to Connect Outlook with Adobe Workfront (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-adobe-workfront.md): Turn client request emails in Outlook into Adobe Workfront tasks automatically, with an AI agent that routes routine work on its own and pauses for a human before a new project is created. - [How to Connect Outlook with Aha! (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-aha-io.md): Turn product feedback arriving in Outlook into Aha! ideas automatically, with an AI agent that files requests on its own and pauses for a product lead before an exec email reshapes an existing roadmap feature. - [How to Connect Outlook with AssemblyAI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-assembly-ai.md): Transcribe and summarize audio arriving in Outlook with AssemblyAI automatically, with an AI agent that processes recordings on its own and pauses for a human before a generated summary is emailed outside the org. - [How to Connect Outlook with Amazon Redshift (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-aws-redshift.md): Turn Amazon Redshift queries into Outlook email reports on a schedule, optionally as an AI agent that drafts the digest, flags anomalies in the numbers, and pauses for a human before Send Draft Email delivers it. - [How to Connect Outlook with Base64.ai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-base64-ai.md): Route documents arriving in an Outlook inbox through Base64.ai extraction automatically, with an AI agent that forwards clean results downstream and pauses for a human reviewer on low-confidence fields before anything enters the processing queue. - [How to Connect Outlook with Basecamp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-basecamp3.md): Turn client emails in a shared Outlook inbox into tracked Basecamp to-dos and comments automatically, with an AI agent that files the work on its own and pauses for a project lead before Reply to Email commits a date or scope to the client. - [How to Connect Outlook with BigMarker (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-bigmarker.md): Turn webinar requests sitting in an Outlook inbox into BigMarker registrations automatically, with an AI agent that drafts every confirmation and pauses for a human before an email leaves the building. - [How to Connect Outlook with Botpress (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-botpress.md): Route a shared Outlook inbox through your deployed Botpress chatbot, with an AI agent that judges whether the answer actually resolves the question and holds every outbound reply as a draft until a person approves the send. - [How to Connect Outlook with Braintree (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-braintree.md): Turn refund requests in a shared Outlook inbox into verified Braintree refunds or voids, with an AI agent that checks the transaction first and pauses for a finance owner before any money moves. - [How to Connect Outlook with Canny (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-canny.md): Connect an Outlook feedback inbox to Canny boards, optionally as an AI agent that dedupes and logs feedback automatically and pauses for a human before a new post goes on the public board. - [How to Connect Outlook with Caspio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-caspio.md): Turn emails in a shared Outlook inbox into structured Caspio records, with an AI agent that extracts the fields, validates them against the live schema, and pauses for a person before it overwrites an existing production record or replies to the sender. - [How to Connect Outlook with Chatbase (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-chatbase.md): Connect an Outlook support inbox to a Chatbase agent that drafts answers from your knowledge base, with a human approving every reply before Reply to Email sends it under your domain. - [How to Connect Outlook with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-hubspot.md): Connect Outlook and HubSpot so inbound emails automatically create or update HubSpot contacts and deals, optionally running as a FlowRunner AI agent that pauses for a human before writing to your CRM on ambiguous or high-stakes records. - [How to Connect Outlook with Tally (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outlook-with-tally.md): Connect Tally form submissions to Outlook so every inquiry gets a contact record, a drafted reply, and a calendar slot, with an AI agent that answers routine requests automatically and pauses for a human before a reply that commits the business goes out. - [How to Connect Outreach with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-outreach-with-salesforce-pro.md): Connect Outreach to Salesforce Pro in FlowRunner so prospect events trigger CRM actions automatically, with an AI agent that pauses for human approval before converting a lead or enrolling a strategic account. - [How to Connect Paddle with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-paddle-with-quickbooks-online.md): Connect Paddle billing events to QuickBooks Online automatically so every subscription, transaction, and refund lands in your books without manual entry, optionally as an AI agent that pauses for human approval before money moves. - [PagerDuty Jira Issues Integration: Every Incident Becomes a Ticket](https://flowrunner.ai/workflows/connect-pagerduty-with-jira-issues.md): Connect PagerDuty and Jira Issues so every triggered incident becomes a fully contextualized Jira issue, with an AI agent that confirms the assignee with a human before it touches a P1. - [How to Connect PagerDuty with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-pagerduty-with-slack.md): Connect PagerDuty's On New Triggered Incident trigger to Slack's Send Message to Channel action so your on-call channel gets full incident context the moment a service pages, optionally as an AI agent that pauses for human acknowledgement before any recovery step runs. - [PandaDoc + HubSpot: Closed-Won Deals Become Signed Contracts](https://flowrunner.ai/workflows/connect-pandadoc-with-hubspot.md): Connect PandaDoc and HubSpot so a closed-won deal builds its own contract from the record, with an AI agent that pauses for a human before the document ever reaches the client. - [Send PandaDoc Drafts to Slack for Approval Before Signing](https://flowrunner.ai/workflows/connect-pandadoc-with-slack.md): Connect PandaDoc and Slack so every contract draft is generated, posted for approval, and sent for signature automatically, optionally run by an AI agent that pauses for a human before the document reaches the client. - [How to Connect Parseur with AfterShip (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-aftership.md): Connect Parseur-extracted carrier emails to AfterShip trackings, optionally as an AI agent that registers clean shipments automatically and pauses for a human before an ambiguous tracking number starts sending updates to a customer. - [How to Connect Parseur with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-aircall.md): Connect Parseur-extracted lead emails to Aircall contacts and insight cards so reps call new inquiries in minutes, optionally as an AI agent that pauses for a human before an uncertain parse puts a wrong number on a rep's dial list. - [How to Connect Parseur with AssemblyAI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-assembly-ai.md): Connect Parseur-parsed emails to AssemblyAI transcription so recordings that arrive in a mailbox become searchable transcripts and summaries, optionally as an AI agent that pauses for a human before Delete Transcript and Delete Document purge anything for good. - [How to Connect Parseur with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-attio.md): Connect Parseur's parsed documents to Attio records so inbound emails and PDFs become CRM data automatically, with an AI agent that pauses for a human before a shaky parse overwrites an existing record. - [How to Connect Parseur with Bigin by Zoho (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-bigin-by-zoho.md): Connect Parseur document extraction to Bigin, optionally as an AI agent that turns parsed inquiries into contacts and deals and pauses for a human before overwriting an existing CRM record. - [How to Connect Parseur with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-bland.md): Connect Parseur's parsed emails and documents to Bland AI phone calls, with an AI agent that drafts the callback from the extracted fields and pauses for a human before Send Call dials the person who wrote in. - [How to Connect Parseur with Botpress (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-botpress.md): Connect Parseur's parsed documents to Botpress tables so your chat assistant answers from live data, with an AI agent that diffs every update and pauses for a human before Upsert Table Rows changes what the assistant tells customers. - [How to Connect Parseur with Capsule CRM (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-capsule-crm.md): Connect Parseur document extraction to Capsule CRM so inbound emails and PDFs become people, organisations, and opportunities, with an AI agent that pauses for a human before it overwrites an existing record on an uncertain match. - [How to Connect Parseur with Chatbase (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-chatbase.md): Connect Parseur document extraction to Chatbase so your chatbot's knowledge stays current with your documents, with an AI agent that assembles the retraining text and pauses for a human before Update Chatbot Data replaces what a live chatbot knows. - [How to Connect Parseur with ChatBot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-chatbot.md): Connect Parseur document extraction to ChatBot so parsed FAQs and guides train your published stories, with an AI agent that stages the training and pauses for a human before new content changes what a live story tells customers. - [How to Connect Parseur with Cin7 Omni (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-cin7.md): Connect Parseur to Cin7 Omni so emailed purchase orders become sales orders without rekeying, with an AI agent that resolves SKUs and pauses for a human before an uncertain line item commits inventory. - [How to Connect Parseur with Clarifai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-clarifai.md): Connect Parseur document extraction to Clarifai computer vision so inbound documents and their images are parsed, classified, and indexed together, with an AI agent that pauses for a human before low-confidence predictions become labeled training data. - [How to Connect Parseur with Clipdrop (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-parseur-with-clipdrop.md): Connect Parseur document extraction to Clipdrop image processing, optionally as an AI agent that cleans product imagery automatically and pauses for a human before a generated image replaces a real one. - [How to Connect Parseur with Slack Using FlowRunner](https://flowrunner.ai/workflows/connect-parseur-with-slack.md): Stop copying invoice data by hand and chasing approvals over email. Connect Parseur document extraction with Slack notifications and human decisions through FlowRunner. - [Connect PayPal to FreshBooks: Payments Recorded to the Right Invoice](https://flowrunner.ai/workflows/connect-paypal-with-freshbooks.md): Connect PayPal and FreshBooks so every captured payment is recorded against the right invoice automatically, with an AI agent that pauses for a controller before refunding or creating a credit note. - [How to Connect PayPal with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-paypal-with-quickbooks-online.md): Connect PayPal and QuickBooks Online so every payment, invoice, and payout syncs to your books automatically, with an AI agent that pauses for human review on the transactions that carry real financial weight. - [How to Connect PayPal with Xero (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-paypal-with-xero.md): Connect PayPal and Xero so every captured payment is recorded in Xero automatically, with an AI agent that pauses for a human before releasing batch payouts or voiding invoices. - [Perplexity Notion Integration: Cited Research, Reviewed Pages](https://flowrunner.ai/workflows/connect-perplexity-with-notion.md): Connect Perplexity and Notion in FlowRunner so Sonar-grounded, cited research briefs land as reviewed Notion pages automatically, with a human confirming sources before anything goes into the team's knowledge base. - [How to Connect Perplexity with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-perplexity-with-slack.md): Connect Perplexity and Slack in FlowRunner so Sonar-grounded research briefs post automatically to your team channel, with a human confirming sources before any answer drives a decision. - [Personio + Slack: Onboard New Hires With Compliance Sign-Off](https://flowrunner.ai/workflows/connect-personio-with-slack.md): Connect Personio and Slack so new hires are provisioned and welcomed automatically, with an AI agent that pauses in Slack for a human before a compliance step or an employment change runs. - [How to Connect Pipedrive with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-pipedrive-with-google-sheets.md): Connect Pipedrive and Google Sheets so every new or updated deal, person, or organization writes to your tracker automatically, with an AI agent that pauses for a human on any row that fails validation or triggers a destructive change. - [Sync Pipedrive Deals to Mailchimp Marketing Audiences](https://flowrunner.ai/workflows/connect-pipedrive-with-mailchimp-marketing.md): Connect Pipedrive and Mailchimp Marketing so a deal stage change syncs the contact into the right audience, with an AI agent that pauses for approval before large campaign sends. - [How to Connect Pipedrive with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-pipedrive-with-slack.md): Connect Pipedrive and Slack so deal events automatically surface in your team's channels, optionally running as an AI agent that pauses for human judgment before merging records or restructuring your pipeline. - [The Pipedrive Typeform Integration That Turns Forms Into Deals](https://flowrunner.ai/workflows/connect-pipedrive-with-typeform.md): Connect Pipedrive and Typeform so every form submission becomes a deduped Pipedrive deal, with an AI agent that pauses for compliance review on sensitive responses. - [How to Connect PostgreSQL with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-postgresql-with-google-sheets.md): Connect PostgreSQL and Google Sheets so database rows sync to a live sheet automatically, then optionally run the connection as an AI agent that pauses for human approval before any bulk write. - [Can an AI Agent Pause in Slack Before a PostgreSQL Write?](https://flowrunner.ai/workflows/connect-postgresql-with-slack.md): Connect PostgreSQL and Slack so database reconciliation runs automatically and posts a summary to your team, then optionally run it as an AI agent that pauses in Slack for approval before any risky database write. - [How to Connect PostgreSQL with Snowflake (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-postgresql-with-snowflake.md): Sync PostgreSQL records into Snowflake automatically with a FlowRunner workflow that runs autonomously and pauses for human approval before any production write. - [How to Connect QuickBooks Online with Adobe Acrobat Sign (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-adobe-sign.md): Connect QuickBooks Online estimates to Adobe Acrobat Sign so every engagement gets signed before it gets billed, with an AI agent that turns estimates into signature requests automatically and pauses for a human before binding terms reach the client. - [How to Connect QuickBooks Online with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-aircall.md): Turn overdue QuickBooks Online invoices into a prepared Aircall collections queue and log every call outcome against the books, optionally as an AI agent that pauses for the controller before Void Invoice erases a receivable. - [How to Connect QuickBooks Online with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-attio.md): Sync invoice balances and payment status from QuickBooks Online into Attio records, with an agent that flags overdue accounts and pauses for the account owner before a payment reminder reaches the customer. - [How to Connect QuickBooks Online with Base64.ai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-base64-ai.md): Connect Base64.ai invoice extraction to QuickBooks Online so vendor invoices become bills without re-keying, with a human approving any low-confidence or unusual payable before Create Bill posts it to the books. - [How to Connect QuickBooks Online with Bigin (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-bigin-by-zoho.md): Connect QuickBooks Online invoices and payments to Bigin deals, optionally as an AI agent that drafts the invoice when a deal closes and pauses for a human before Send Invoice reaches the customer. - [How to Connect QuickBooks Online with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-bland.md): Connect QuickBooks Online overdue invoices to Bland AI phone calls, optionally as an AI agent that prepares each payment-reminder call and pauses for a human before Send Call dials a customer. - [How to Connect QuickBooks Online with Braintree (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-braintree.md): Connect Braintree transactions to QuickBooks Online payments for automatic reconciliation, optionally as an AI agent that books settled charges on its own and pauses for a human before Refund Transaction moves money back out. - [How to Connect QuickBooks Online with Canva (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-canva.md): Turn QuickBooks Online financials into branded Canva reports automatically, with an AI agent that autofills the template from live figures and pauses for finance sign-off before the polished PDF is exported. - [How to Connect QuickBooks Online with Cin7 Omni (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-cin7.md): Turn Cin7 Omni sales orders into QuickBooks Online invoices and purchase orders into vendor bills, optionally as an AI agent that pauses for a human before a bill that does not reconcile commits the company to pay. - [How to Connect QuickBooks Online with CloudConvert (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-cloudconvert.md): Turn QuickBooks Online invoices into merged, optimized close packets with CloudConvert, run by an AI agent that reconciles the documents and pauses for a human before the packet is finalized and intermediate files are purged. - [How to Connect QuickBooks Online with CloudTalk (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-cloudtalk.md): Turn overdue QuickBooks Online invoices into CloudTalk reminders and collection calls, run by an AI agent that segments accounts and pauses for a human before any message or call reaches a customer about money. - [How to Connect QuickBooks Online with Crisp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-crisp.md): Answer billing questions in Crisp chat with live QuickBooks Online invoice data, run by an AI agent that resolves routine questions itself and pauses for a human before quoting figures on a disputed account or re-issuing an invoice. - [How to Connect QuickBooks Online with Cin7 Core (DEAR) (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-quickbooks-online-with-dear-inventory.md): Turn Cin7 Core (DEAR) sales into QuickBooks Online invoices and recorded payments, optionally as an AI agent that verifies every invoice against the sale and pauses for a human before Send Invoice puts a wrong number in front of a customer. - [How to Connect QuickBooks Online with Slack Using FlowRunner](https://flowrunner.ai/workflows/connect-quickbooks-online-with-slack.md): Route financial exceptions, post payment summaries, and deliver weekly P&L digests by connecting QuickBooks Online with Slack through FlowRunner. - [How to Connect QuickBooks Online with Stripe Using FlowRunner](https://flowrunner.ai/workflows/connect-quickbooks-online-with-stripe.md): Eliminate manual payment recording and AR reconciliation by connecting Stripe payment events directly to QuickBooks Online through FlowRunner. - [Ramp + QuickBooks Online: Vendor Bills Synced, Duplicates Caught](https://flowrunner.ai/workflows/connect-ramp-service-with-quickbooks-online.md): Connect Ramp bills to QuickBooks Online automatically, with an AI agent that catches duplicates and pauses for an accounting manager before posting a large or unfamiliar bill. - [Route Ramp Reimbursements to Slack for Approval](https://flowrunner.ai/workflows/connect-ramp-service-with-slack.md): Connect Ramp and Slack so reimbursements and bills route to the right approver automatically, with an AI agent that pauses for a human before any money moves. - [Sync Razorpay Payments to QuickBooks Online Automatically](https://flowrunner.ai/workflows/connect-razorpay-with-quickbooks-online.md): Connect Razorpay and QuickBooks Online so every captured payment posts as a paid invoice automatically, and an AI agent holds refunds for a named approver before money moves. - [Connect Recruitee to Slack: Disqualify Only After Sign-Off](https://flowrunner.ai/workflows/connect-recruitee-with-slack.md): Connect Recruitee and Slack so new applications route and screen automatically, and run it as an AI agent that pauses in Slack before any candidate is disqualified. - [RingCentral HubSpot Integration: Every Call Logged to the Deal](https://flowrunner.ai/workflows/connect-ringcentral-with-hubspot.md): Connect RingCentral and HubSpot so stalled deals get a real phone call instead of another unread email, run as an AI agent that pauses for a sales manager before it gives up on a deal or keeps dialing blind. - [Connect RingCentral to Slack: Confirm Every Fax Before It Sends](https://flowrunner.ai/workflows/connect-ringcentral-with-slack.md): Connect RingCentral and Slack so a shared document routes straight to a fax, or run it as an AI agent that stages the send and pauses in Slack for a human to confirm the number first. - [Can an AI Agent Route Rossum Invoices Into QuickBooks Bills?](https://flowrunner.ai/workflows/connect-rossum-elis-with-quickbooks-online.md): Connect Rossum and QuickBooks Online so extracted invoices post as matched bills automatically, with a human reviewing low-confidence fields and bills over $10,000. - [Connect Rossum to Xero: Invoices Become PO-Matched Bills](https://flowrunner.ai/workflows/connect-rossum-elis-with-xero.md): Connect Rossum and Xero so extracted invoice data becomes a matched Xero bill automatically, with an AI agent that pauses for an AP specialist on low-confidence fields or PO mismatches. - [How to Connect Amazon S3 / Cloud Storage with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-s3-with-slack.md): Connect Amazon S3 to Slack so file uploads automatically notify your team, and optionally run the flow as an AI agent that pauses for human approval before bulk deletions or sensitive file operations. - [How to Connect Salesforce Pro with Accelo (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-accelo.md): Connect Salesforce closed-won opportunities to Accelo so every new engagement gets a client record, contacts, and kickoff tasks, with an AI agent that runs the handoff automatically and pauses for a human before the client-facing kickoff email goes out. - [How to Connect Salesforce Pro with Adobe Acrobat Sign (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-adobe-sign.md): Connect Salesforce opportunities to Adobe Acrobat Sign so contracts go out when a deal reaches the contract stage and the signed PDF files itself back onto the record, with an AI agent that pauses for a human before any binding document reaches the customer. - [How to Connect Salesforce Pro with Adobe Workfront (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-adobe-workfront.md): Connect closed-won Salesforce opportunities to Adobe Workfront projects, optionally as an AI agent that builds the delivery plan automatically and pauses for a human before the client-facing kickoff email leaves through Send Email. - [How to Connect Salesforce Pro with Aha! (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-aha-io.md): Pipe feature gaps from Salesforce opportunities into Aha! as ideas with pipeline dollars attached, optionally as an AI agent that builds the revenue case per gap and pauses for the product lead before Create Feature commits the roadmap. - [How to Connect Salesforce Pro with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-aircall.md): Log every Aircall conversation in Salesforce and capture unknown callers as leads, optionally as an AI agent that reads each call and pauses for the rep before Convert Lead to Contact restructures the CRM around a judgment call. - [How to Connect Salesforce Pro with AssemblyAI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-assembly-ai.md): Transcribe call recordings with AssemblyAI and file them against the right Salesforce lead or contact, with an agent that logs notes automatically and pauses for a human before the irreversible Convert Lead to Contact step. - [How to Connect Salesforce Pro with AWeber (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-aweber.md): Sync Salesforce campaign members into the matching AWeber lists with consent checked on every contact, and an agent that pauses for a marketer before Create Broadcast reaches a list. - [How to Connect Salesforce Pro with Amazon Redshift (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-aws-redshift.md): Extract Salesforce records into Amazon Redshift with SOQL-driven queries on a nightly schedule, and let an agent write warehouse-computed scores back to the CRM, pausing for an admin before any mass Update Record batch touches Salesforce. - [How to Connect Salesforce Pro with Base64.ai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-base64-ai.md): Connect Base64.ai to Salesforce so scanned IDs, applications, and signed agreements become verified leads and records, with a human confirming any borderline identity match before Convert Lead to Contact runs. - [How to Connect Salesforce Pro with Basecamp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-basecamp3.md): Connect Salesforce opportunities to Basecamp so every closed-won opportunity becomes a scaffolded delivery project, with a human approving the client-facing kickoff email before Send Email reaches the customer. - [How to Connect Salesforce Pro with beehiiv (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-beehiiv.md): Connect Salesforce campaigns to your beehiiv newsletter so campaign members become attributed subscribers, with a human approving every journey enrollment before a sequence of emails starts landing in customer inboxes. - [How to Connect Salesforce Pro with Bettermode (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-bettermode.md): Connect your Bettermode community to Salesforce so high-intent members become leads with their posts on the record, with a human approving anything published in the company's name before Create Post goes live. - [How to Connect Salesforce Pro with BigMarker (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-bigmarker.md): Connect BigMarker webinar registrants to Salesforce leads and campaigns, optionally as an AI agent that drafts personalized follow-up and pauses for a human before Send Email reaches an attendee. - [Salesforce Pro Google Sheets Integration: Rows to Vetted Leads](https://flowrunner.ai/workflows/connect-salesforce-pro-with-google-sheets.md): Connect Salesforce Pro and Google Sheets so every new sheet row becomes a scored Salesforce lead, optionally as an AI agent that pauses for a sales manager before converting the lead to a contact and opportunity. - [How to Connect Salesforce Pro with Slack Using FlowRunner](https://flowrunner.ai/workflows/connect-salesforce-pro-with-slack.md): Stop losing leads to slow qualification and untracked conversions. Connect Salesforce Pro with Slack through FlowRunner to score leads automatically, convert with manager approval, and keep your revenue team aligned. - [How to Connect Salesforce with Tally (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-salesforce-pro-with-tally.md): Connect Tally form submissions to Salesforce so every response becomes a lead with campaign attribution, with an AI agent that pauses for a human before Convert Lead to Contact makes an irreversible change to the pipeline. - [The Salesloft + Salesforce Pro Integration That Checks First](https://flowrunner.ai/workflows/connect-salesloft-with-salesforce-pro.md): Connect Salesloft to Salesforce Pro so new leads land as CRM records and only reach outreach cadences after an AI agent checks for open opportunities and a person confirms the send. - [Can an AI Agent Vet Google Sheets Leads Before SendGrid Sends?](https://flowrunner.ai/workflows/connect-sendgrid-with-google-sheets.md): Connect SendGrid and Google Sheets so new rows become validated, deduplicated sends, with an AI agent that pauses for a human before a large segment goes out. - [How to Connect Sentry with Jira Issues (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-sentry-with-jira-issues.md): Connect Sentry error monitoring to Jira Issues so new critical errors automatically become tracked, assigned tickets, with a human in the loop for permanent deletes and high-priority assignments. - [Connect Sentry to Linear: Every Error Becomes One Tracked Issue](https://flowrunner.ai/workflows/connect-sentry-with-linear.md): Connect Sentry and Linear so new errors become tracked issues automatically, with an AI agent that checks for duplicates first and pauses for a triage lead when a match is unclear. - [How to Connect Sentry with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-sentry-with-slack.md): Connect Sentry to Slack so error alerts and triage decisions move through one channel, optionally as an AI agent that resolves issues by rule and pauses for human confirmation on any delete that cannot be undone. - [How to Connect ServiceNow with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-servicenow-with-slack.md): Connect ServiceNow to Slack so incidents and change requests fire Slack alerts automatically, optionally with an AI agent that pauses for human approval on changes that touch production. - [How to Connect SharePoint to Slack for Approved File Sharing](https://flowrunner.ai/workflows/connect-sharepoint-with-slack.md): Connect SharePoint to Slack so every new file gets tracked and announced automatically, then let an AI agent request approval in Slack before it creates an external sharing link. - [How to Connect ShipBob with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-aircall.md): Connect ShipBob shipment events to Aircall contacts and insight cards so support answers 'where is my order' calls with live fulfillment status, optionally as an AI agent that pauses for a human before a stuck order is canceled and reshipped. - [How to Connect ShipBob with Base64.ai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-base64-ai.md): Connect ShipBob fulfillment events to Base64.ai document intelligence so shipment exceptions and delivery disputes resolve on evidence, with an AI agent that pauses for a human before a replacement order spends real money. - [How to Connect ShipBob with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-bland.md): Connect ShipBob shipment exceptions to Bland AI phone calls, with an AI agent that diagnoses the fulfillment problem, proposes a fix, and pauses for a human before Send Call dials the customer. - [How to Connect ShipBob with Cin7 Omni (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-cin7.md): Connect ShipBob fulfillment events to Cin7 Omni order and stock records, with an AI agent that reconciles shipments automatically and pauses for a human before a reshipment spends money or a cancellation kills a live order. - [How to Connect ShipBob with CloudConvert (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-cloudconvert.md): Connect ShipBob shipments to CloudConvert document processing, optionally as an AI agent that builds carrier claim packets automatically and pauses for a human before an order is canceled or a return is created. - [How to Connect ShipBob with Cloudinary (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-cloudinary.md): Connect ShipBob shipments and returns to Cloudinary asset storage, optionally as an AI agent that files fulfillment evidence automatically and pauses for a human before Create Return sets reverse logistics moving. - [How to Connect ShipBob with CloudTalk (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-cloudtalk.md): Connect ShipBob shipment events to CloudTalk SMS and calls, optionally as an AI agent that tells customers about delivery problems before they notice and pauses for a human before an order is canceled. - [How to Connect ShipBob with Crisp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-crisp.md): Connect ShipBob fulfillment events to Crisp conversations, optionally as an AI agent that keeps customers informed of shipping progress and pauses for a human before Send Message delivers bad news about a shipment exception. - [How to Connect ShipBob with Cin7 Core (DEAR) (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-dear-inventory.md): Connect ShipBob fulfillment events to Cin7 Core (DEAR) so shipped orders stay reconciled with the ERP, with an AI agent that spots inventory drift and pauses for a human before a replenishment order commits inbound freight. - [How to Connect ShipBob with Dialpad (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-dialpad.md): Connect ShipBob shipment events to Dialpad so customers get texted about shipping, delivery, and exceptions, with an AI agent that triages each exception and pauses for a human before the bad-news message goes out. - [How to Connect ShipBob with Egnyte (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-egnyte.md): Connect ShipBob fulfillment events to Egnyte so every shipment, return, and receiving order builds its own document trail, with an AI agent that pauses for a human before a shared link exposes fulfillment records outside the company. - [How to Connect ShipBob with Exact Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-exact-online.md): Connect ShipBob fulfillment events to Exact Online so shipped orders become sales invoices in the ERP, with an AI agent that verifies what actually shipped and pauses for a human before an invoice bills a customer for it. - [How to Connect ShipBob with Facebook Messenger (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-facebook-messenger.md): Connect ShipBob shipment events to Facebook Messenger so customers get order updates in the thread where they already talk to you, with an AI agent that reads the conversation first and pauses for a human before messaging anyone whose thread is already unhappy. - [How to Connect ShipBob with Files.com (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shipbob-with-files-com.md): Connect ShipBob shipment events to Files.com so every order, exception, and return lands as an archived record, with an AI agent that files documents automatically and pauses for a human before anything is deleted or shared outside the company. - [Send ShipBob Shipment Exceptions to Slack Automatically](https://flowrunner.ai/workflows/connect-shipbob-with-slack.md): Connect ShipBob and Slack so shipment exceptions post to your ops channel automatically, with an AI agent that pauses for a human on high-value carrier issues. - [Shippo Shopify Integration: Rate-Shopped Labels, Every Order](https://flowrunner.ai/workflows/connect-shippo-with-shopify.md): Connect Shopify orders to Shippo so every label gets bought against the best carrier rate automatically, and run it as an AI agent that pauses for ops on a failed or returned high-value shipment. - [Can an AI Agent Route Failed Shippo Deliveries to Slack?](https://flowrunner.ai/workflows/connect-shippo-with-slack.md): Connect Shippo and Slack so failed or returned shipments post straight to your team, and let an AI agent decide when a human needs to approve the reship or refund. - [ShipStation + Google Sheets: Every Shipment Logged Automatically](https://flowrunner.ai/workflows/connect-shipstation-with-google-sheets.md): Connect ShipStation and Google Sheets so every shipped order writes itself to a tracking sheet, with an AI agent pausing for ops before it buys a costly label. - [Shopify Airtable Integration: Flagged Orders Tracked to Resolution](https://flowrunner.ai/workflows/connect-shopify-with-airtable.md): Connect Shopify and Airtable to log flagged orders and disputes as a tracked exception queue, or run it as an AI agent that pauses for a named approver before a refund goes out. - [How to Connect Shopify with Mailchimp Marketing (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shopify-with-mailchimp-marketing.md): Connect Shopify to Mailchimp Marketing so new customers and orders instantly update your audience, with an AI agent that applies tags, syncs behavior, and pauses for human approval before sending campaigns to large lists. - [How to Connect Shopify with NetSuite (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shopify-with-netsuite.md): Connect Shopify and NetSuite so every new order triggers customer creation, sales order, invoice, and payment recording in your ERP automatically, with a human in the loop on orders that carry real consequence. - [Can an AI Agent Log Shopify Order Decisions in Notion?](https://flowrunner.ai/workflows/connect-shopify-with-notion.md): Connect Shopify's On New Order trigger to Notion's database actions, then run the same flow as an AI agent that pauses for a person before a flagged order ships. - [How to Connect Shopify with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shopify-with-quickbooks-online.md): Connect Shopify orders and refunds to QuickBooks Online, optionally as an AI agent that records the sale automatically and pauses for a human before a refund posts. - [Connect Shopify to Sage Accounting: Orders Become Invoices](https://flowrunner.ai/workflows/connect-shopify-with-sage-accounting.md): Connect Shopify orders and payouts to Sage Accounting, optionally as an AI agent that mirrors every sale automatically and pauses for the bookkeeper before a receipt posts. - [How to Connect Shopify with ShipBob (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shopify-with-shipbob.md): Connect Shopify's On New Order trigger to ShipBob's Create Order action so every confirmed sale goes to fulfillment automatically, with a human on exceptions that carry real consequence. - [How to Connect Shopify with ShipStation (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shopify-with-shipstation.md): Connect Shopify's On New Order trigger to ShipStation's Create Shipment Label action, optionally as an AI agent that rate-shops carriers and pauses for a human before buying a costly label. - [How to Connect Shopify with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shopify-with-slack.md): Connect Shopify's On New Order trigger to Slack's Send Message to Channel action so your team sees every order, refund, and dispute the moment it happens, optionally as an AI agent that pauses for a human before moving money. - [How to Connect Shopify with Xero (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-shopify-with-xero.md): Connect Shopify's On New Order trigger to Xero invoice and payment actions, then optionally run it as an AI agent that pauses for a human before posting a disputed charge or issuing a refund. - [Log Shopify Orders as Zoho Books Invoices Automatically](https://flowrunner.ai/workflows/connect-shopify-with-zoho-books.md): Connect Shopify order and refund events to Zoho Books invoices and credit notes, optionally as an AI agent that records the sale automatically and pauses for a human before a refund is credited back. - [How to Connect Slack with Accelo (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-accelo.md): Turn Slack mentions in client channels into tracked Accelo tasks and logged activities, with an AI agent that pauses for a human before it creates a new company record. - [How to Connect Slack with Adobe Acrobat Sign (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-adobe-sign.md): Send contracts for e-signature straight from Slack and track them to completion, with an AI agent that pauses for a named human before Send Agreement puts a legal document in front of a counterparty. - [How to Connect Slack with Adobe Workfront (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-adobe-workfront.md): Connect Slack requests to Adobe Workfront tasks, optionally as an AI agent that files the work automatically and pauses for an operations lead before a new project is created. - [How to Connect Slack with AfterShip (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-aftership.md): Connect Slack to AfterShip so tracking numbers posted in a channel become monitored shipments with status replies, and an AI agent pauses for a human before Delete Tracking wipes a shipment's history. - [How to Connect Slack with Aha! (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-aha-io.md): Connect Slack feedback to the Aha! ideas portal automatically, with an AI agent that dedupes and files ideas on its own and pauses for a product lead before Create Feature puts anything on the roadmap. - [How to Connect Slack with Asana (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-asana.md): Connect Slack and Asana so messages and decisions flow into tracked tasks automatically. Optionally run it as a FlowRunner AI agent that creates, assigns, and escalates Asana tasks from Slack activity, with a human in the loop on the steps that carry real consequence. - [How to Connect Slack with AssemblyAI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-assembly-ai.md): Drop a recording in Slack and get a transcript, summary, and searchable text back from AssemblyAI, with an AI agent that pauses for a human before an unredacted or sensitive transcript reaches a shared channel. - [How to Connect Slack with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-attio.md): Turn Slack messages into Attio records with an emoji reaction, optionally as an AI agent that upserts people and companies automatically and pauses for a human before it overwrites existing CRM data. - [How to Connect Slack with Amazon Redshift (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-aws-redshift.md): Ask data questions in Slack and get answers from Amazon Redshift, with an AI agent that runs read queries on its own and pauses for a data engineer before any statement that writes, deletes, or alters. - [How to Connect Slack with Azure Service Bus (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-azure-service-bus.md): Operate Azure Service Bus queues from Slack, with an AI agent that receives and triages messages with peek-lock and pauses for an engineer before it completes a suspect message or deletes a queue. - [How to Connect Slack with Backendless (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-backendless.md): Operate your Backendless backend from Slack and route backend events into your channels, with an AI agent that queries and fixes records on request and pauses for a developer before any bulk delete or bulk update runs. - [How to Connect Slack with Base64.ai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-base64-ai.md): Scan documents shared in Slack with Base64.ai and post the extracted fields back to the thread, with an AI agent that accepts high-confidence extractions on its own and pauses for a human before low-confidence data or a borderline identity match is treated as verified. - [How to Connect Slack with Basecamp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-basecamp3.md): Connect Slack conversations to Basecamp to-dos automatically, with an AI agent that files and assigns tracked work on its own and pauses for a named approver before Trash Project removes anything. - [How to Connect Slack with Baserow (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-baserow.md): Turn Slack requests into Baserow rows and keep your tracker current from the channel, with an AI agent that reads, creates, and updates rows on its own and pauses for a human before any batch update or batch delete touches the table. - [How to Connect Slack with Bettermode (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-bettermode.md): Run your Bettermode community from Slack: unanswered questions surface in your channel, drafted replies and announcements wait for a human before Create Post or Create Comment reaches members, and Delete Post never runs without a named approval. - [How to Connect Slack with Bigin (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-bigin-by-zoho.md): Keep your Bigin pipeline current from Slack: sales chatter becomes contacts, deals, and tasks automatically, and an AI agent pauses for the deal owner before Update Deal moves anything into Closed Won or Closed Lost. - [How to Connect Slack with BigMarker (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-bigmarker.md): Run BigMarker webinar operations from Slack: registrations, reminders, and attendance digests flow automatically, and an AI agent pauses for the event owner before Update Conference reschedules a webinar that already has registrants. - [How to Connect Slack with Bluesky (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-bluesky.md): Connect Slack drafts and approvals to Bluesky publishing, optionally as an AI agent that formats and schedules posts automatically and pauses for a human before anything public goes out. - [How to Connect Slack with Botpress (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-botpress.md): Connect Slack to Botpress conversations so escalations reach your team in-channel, with an orchestrating agent that drafts the customer reply and pauses for a human before Send Chat Message delivers it into the live conversation. - [How to Connect Slack with Braintree (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-braintree.md): Connect Slack to Braintree so refund requests are researched and executed in-channel, with an AI agent that assembles the transaction evidence and pauses for a named approver before Refund Transaction or Void Transaction moves money. - [How to Connect Slack with Canny (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-canny.md): Connect Slack feedback to Canny boards so customer requests become tracked, voteable posts, with an AI agent that dedupes against the board and pauses for a human before anything publishes where customers can see it. - [How to Connect Slack with Canva (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-canva.md): Connect Slack requests to Canva brand templates so designs are autofilled and exported on demand, with an AI agent that fills the template automatically and pauses for a human before anything with a price, date, or claim is exported and circulated. - [How to Connect Slack with Capsule CRM (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-capsule-crm.md): Connect Slack deal conversations to Capsule CRM so the pipeline stays true to what sales actually said, with an AI agent that logs tasks and contacts automatically and pauses for a human before Update Opportunity rewrites a deal's value or milestone. - [How to Connect Slack with Caspio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-caspio.md): Connect Slack to your Caspio database so teammates query and update production app data in-channel, with an AI agent that answers reads instantly and pauses for a human, with criteria and affected row count shown, before any bulk update or delete runs. - [How to Connect Slack with Chatbase (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-chatbase.md): Connect Slack to Chatbase so docs changes retrain your customer-facing assistant and captured leads reach your team, with an orchestrating agent that pauses for a human before Update Chatbot Data changes what the live assistant tells customers. - [How to Connect Slack with ChatBot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-chatbot.md): Connect Slack to ChatBot.com so unmatched customer phrases become reviewed training improvements, with an orchestrating agent that clusters the misses and pauses for a human before Train Phrases changes how the production story answers. - [How to Connect Slack with ClickUp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-clickup.md): Connect Slack and ClickUp so task events trigger Slack messages automatically, or run the whole loop as a FlowRunner AI agent that pauses for human judgment on the calls that carry weight. - [How to Connect Slack with Jira Issues (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-jira-issues.md): Connect Slack and Jira Issues so that messages and exceptions in Slack automatically create, update, and route Jira issues, with a human on every consequential assignment or escalation. - [How to Connect Slack with Linear (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-linear.md): Connect Slack and Linear so that alerts, decisions, and status updates flow between your team's communication hub and issue backlog, with an AI agent that searches before it creates and pauses for a human on ambiguous calls. - [How to Connect Slack with Monday.com (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-monday.md): Connect Slack and Monday.com so that Monday.com item changes post to Slack automatically, with an optional AI agent that routes exceptions to humans before writing bulk updates. - [How to Connect Slack with Notion (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-notion.md): Connect Slack and Notion so messages, decisions, and exceptions automatically become searchable Notion records. Optionally run by an AI agent that pauses for a human before writing anything consequential. - [How to Connect Slack with Stripe Using FlowRunner](https://flowrunner.ai/workflows/connect-slack-with-stripe.md): Route Stripe payment failures, refund approvals, and revenue alerts through Slack with intelligent human oversight at every decision point. - [How to Connect Slack with Trello (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-slack-with-trello.md): Connect Slack and Trello so messages and reactions trigger card creation, movement, and updates automatically, then upgrade to an AI agent that handles exceptions and pauses for a human before any destructive action. - [How to Connect Smartsheet to Slack for Exception Approvals](https://flowrunner.ai/workflows/connect-smartsheet-with-slack.md): Connect Smartsheet and Slack so row changes and stalled intakes route to Slack for a quick approve or escalate, with the decision written back to the row automatically. - [How to Connect Snowflake with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-snowflake-with-google-sheets.md): Use FlowRunner to connect Snowflake and Google Sheets so query results flow directly into spreadsheets on a schedule, with an AI agent that pauses for a human before writing back to a production warehouse. - [Snowflake Slack Integration: Warehouse Writes Wait for a Yes](https://flowrunner.ai/workflows/connect-snowflake-with-slack.md): Connect Snowflake and Slack so query results post to a channel automatically, and any MERGE, INSERT, or DDL against a production database pauses for a human decision in Slack. - [Log Square Payouts in Google Sheets Automatically](https://flowrunner.ai/workflows/connect-square-with-google-sheets.md): Connect Square and Google Sheets so every payout reconciles into a workbook automatically, with a human called in only when the entries don't tie out. - [Connect Square to QuickBooks Online: Payouts Reconciled Daily](https://flowrunner.ai/workflows/connect-square-with-quickbooks-online.md): Connect Square and QuickBooks Online so every payout reconciles into matching invoices automatically, with an AI agent that pauses for a person when an entry doesn't tie out or a void tops $10,000. - [Square + Xero: Every Payout Reconciled Before It Posts](https://flowrunner.ai/workflows/connect-square-with-xero.md): Connect Square payouts to Xero's ledger, optionally as an AI agent that reconciles clean deposits automatically and pauses for a human before a refund becomes a credit note. - [How to Connect Squarespace to Mailchimp Marketing: Tag Every Buyer](https://flowrunner.ai/workflows/connect-squarespace-with-mailchimp-marketing.md): Connect Squarespace to Mailchimp Marketing so every new order adds or updates the subscriber and applies a purchase tag automatically, optionally as an AI agent that pauses for a human before a large campaign sends. - [How to Connect Squarespace with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-squarespace-with-quickbooks-online.md): Connect Squarespace to QuickBooks Online so every new order creates a QuickBooks invoice automatically, with a FlowRunner AI agent pausing for human review on disputed charges, refunds, or amounts that don't match expectations. - [How to Connect Stripe with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-aircall.md): Route Stripe payment events into Aircall so reps see billing context on every call, with an AI agent that pauses for a human before Create Refund or Cancel Subscription moves money or ends revenue. - [How to Connect Stripe with Airtable (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-airtable.md): Connect Stripe payment events to Airtable records automatically, and optionally run the connection as an AI agent that pauses for a human before processing refunds, disputed charges, or failed subscriptions. - [How to Connect Stripe with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-attio.md): Sync Stripe payment and subscription events into Attio records and lists, with an AI agent that keeps the CRM current on its own and pauses for a named human before it refunds a charge or cancels a subscription. - [How to Connect Stripe with Bigin (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-bigin-by-zoho.md): Connect Stripe payment events to Bigin by Zoho so paid checkouts move deals and update contacts automatically, with an AI agent that pauses for a human before a refund or subscription cancellation. - [How to Connect Stripe with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-bland.md): Connect Stripe payment failures to Bland AI outbound phone calls, with an AI agent that prepares the recovery call and pauses for a named human before Send Call dials a real customer. - [How to Connect Stripe with Capsule CRM (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-capsule-crm.md): Connect Stripe payments to Capsule CRM so opportunities advance through their milestones automatically, with an AI agent that opens cases for disputes and pauses for a human before a refund is issued. - [How to Connect Stripe with Cin7 Omni (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-cin7.md): Connect Stripe checkouts to Cin7 Omni sales orders with stock verified before booking, and an AI agent that pauses for a human before a refund goes out on an order inventory cannot cover. - [How to Connect Stripe with CloudTalk (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-cloudtalk.md): Connect Stripe payment events to CloudTalk so failed renewals become prepared outreach, with an AI agent that checks call history first and pauses for a human before a billing SMS reaches a customer's phone. - [How to Connect Stripe with Copilot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-copilot.md): Connect Stripe payments to Copilot so every paying customer becomes a portal client automatically, with an AI agent that pauses for a human before a client-visible invoice posts or a portal invite goes out on a mismatched record. - [How to Connect Stripe with Crisp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-crisp.md): Connect Stripe billing events to Crisp so payment failures and subscription changes reach customers as timely chat messages, with an AI agent that pauses for a human before a sensitive billing message reaches a customer. - [How to Connect Stripe with Custify (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-custify.md): Connect Stripe billing events to Custify so customer success sees revenue health in real time, with an AI agent that opens CSM tasks automatically and pauses for a human before a subscription is canceled in Stripe. - [How to Connect Stripe with DEAR Inventory (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-dear-inventory.md): Connect Stripe payments to DEAR Inventory so paid orders become sales records with stock checked first, with an AI agent that pauses for a human before refunding a customer when availability falls short. - [How to Connect Stripe with Deskera (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-deskera.md): Connect Stripe payments to Deskera so every charge becomes a posted invoice and payment record, with an AI agent that pauses for a human before a credit note reverses revenue in the books. - [How to Connect Stripe with Dialpad (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-dialpad.md): Connect Stripe billing events to Dialpad so failed payments trigger timely text outreach, with an AI agent that pauses for a human before a payment reminder reaches a customer's phone. - [Record Stripe Payments in FreshBooks Automatically](https://flowrunner.ai/workflows/connect-stripe-with-freshbooks.md): Connect Stripe and FreshBooks so every successful charge is matched to its invoice and recorded automatically, with a human called in only when the amounts don't line up. - [Connect Stripe to Google Sheets: A Ledger That Flags Refunds](https://flowrunner.ai/workflows/connect-stripe-with-google-sheets.md): Connect Stripe and Google Sheets to log every payment automatically, or run it as an AI agent that pauses for a human before a refund over $500 or a dispute gets written into the books as final. - [How to Connect Stripe with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-hubspot.md): Connect Stripe and HubSpot so payment events automatically update your CRM, optionally running as an AI agent that pauses for human review before posting disputed charges or processing large refunds. - [Stripe + Invoice Ninja: Payments Applied, Big Invoices Reviewed](https://flowrunner.ai/workflows/connect-stripe-with-invoice-ninja.md): Connect Stripe to Invoice Ninja so confirmed charges record and apply to the right invoice automatically, and run it as an AI agent that pauses for a person before an oversized or mismatched invoice sends. - [How to Connect Stripe with NetSuite (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-netsuite.md): Connect Stripe and NetSuite so payment events automatically record in your ERP, optionally as an AI agent that pauses for human judgment before posting disputed charges or large transactions. - [How to Connect Stripe with Sage Intacct (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-sage-intacct.md): Connect Stripe payment events to Sage Intacct's ledger so every charge, invoice, and refund posts automatically, with a human in the loop before money moves on exceptions. - [How to Connect Stripe with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-salesforce-pro.md): Connect Stripe payment events to Salesforce Pro CRM records automatically, and optionally run the connection as an AI agent that pauses for human approval on refunds, large transactions, and lead conversions. - [How to Connect Stripe to Wave: Payments Become Approved Invoices](https://flowrunner.ai/workflows/connect-stripe-with-wave.md): Connect Stripe and Wave so every successful charge drafts a matching Wave invoice automatically, with a bookkeeper on the seam whenever the amount, customer, or business entity does not line up. - [How to Connect Stripe with Xero (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-stripe-with-xero.md): Connect Stripe and Xero so every payment, invoice, and reconciliation event flows automatically, optionally as an AI agent that pauses for a human before voiding an invoice or processing a large refund. - [Stripe Zoho Books Integration: Payments Reconciled Live](https://flowrunner.ai/workflows/connect-stripe-with-zoho-books.md): Connect Stripe and Zoho Books so every successful payment closes the matching invoice automatically, optionally as an AI agent that pauses for a human before a refund goes out. - [Connect Supabase to Google Sheets: Sync Records, Flag Anomalies](https://flowrunner.ai/workflows/connect-supabase-with-google-sheets.md): Connect Supabase to Google Sheets so database events write straight into a live ops sheet, and run it as an AI agent that pauses on anomalous records before they sync. - [How to Connect Supabase with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-supabase-with-slack.md): Connect Supabase record events to Slack messages automatically, or run the connection as an AI agent that pauses for a human before propagating anomalous data downstream. - [How to Connect SurveyMonkey with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-surveymonkey-with-hubspot.md): Connect SurveyMonkey and HubSpot so survey responses automatically create or update contacts, flag detractors, and route exceptions to your team before any automated follow-up fires. - [How to Connect Telegram with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-telegram-with-google-sheets.md): Use FlowRunner to connect Telegram messages to Google Sheets rows automatically, with an AI agent that validates data and pauses for a human before writing bad records into systems of record. - [Sync Todoist Task Deadlines to Google Calendar Automatically](https://flowrunner.ai/workflows/connect-todoist-with-google-calendar.md): Connect Todoist and Google Calendar so tasks with due dates get blocked time on the calendar automatically, and a human confirms before a task is assigned to a teammate whose calendar it will affect. - [Todoist Slack Integration: Ask Before You Assign the Task](https://flowrunner.ai/workflows/connect-todoist-with-slack.md): Connect Todoist and Slack so channel requests become tasks automatically, or run it as an AI agent that proposes the right owner and pauses for a lead to confirm the assignment in Slack. - [Turn Google Sheets Rows into Trello Cards Automatically](https://flowrunner.ai/workflows/connect-trello-with-google-sheets.md): Connect Trello and Google Sheets so new sheet rows become Trello cards automatically, with an AI agent that pauses for a human when a row is missing required data. - [How to Connect Twilio with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-twilio-with-google-sheets.md): Connect Twilio SMS and voice activity to Google Sheets automatically, optionally as an AI agent that logs every message, escalates exceptions, and pauses for human approval before placing a call. - [How to Connect Twilio with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-twilio-with-slack.md): Connect Twilio and Slack so incoming SMS and voice triggers route directly to Slack channels, optionally as an AI agent that pauses for human sign-off before escalating to a voice call. - [How to Connect Typeform with Accelo (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-accelo.md): Turn Typeform intake submissions into Accelo contacts, tasks, and client records, run by an AI agent that matches each response to existing accounts and pauses for a human before a new company record enters the PSA. - [How to Connect Typeform with Adobe Workfront (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-adobe-workfront.md): Turn Typeform request submissions into Adobe Workfront tasks automatically, with an AI agent that files routine work on its own and pauses for a human before changing a live project's scope or deadline. - [How to Connect Typeform with Aha! (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-aha-io.md): Turn Typeform feature-request survey responses into Aha! ideas automatically, with an AI agent that clusters signal on its own and pauses for a human before Delete Responses purges the source data. - [How to Connect Typeform with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-aircall.md): Turn Typeform callback requests into prepared Aircall calls automatically, with an AI agent that builds contacts and briefing cards on its own and pauses for a human before Delete Contact removes a record during dedupe. - [How to Connect Typeform with AWeber (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-aweber.md): Pull new Typeform submissions into AWeber automatically, optionally as an AI agent that adds clean signups on its own and pauses for a human when consent is unclear before anyone lands on the list. - [How to Connect Typeform with Amazon Redshift (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-aws-redshift.md): Load Typeform responses into Amazon Redshift on a schedule, optionally as an AI agent that verifies every batch by row count and pauses for a human before Delete Responses clears the source data. - [How to Connect Typeform with Basecamp (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-basecamp3.md): Turn Typeform intake submissions into fully built Basecamp projects with task plans and kickoffs, run by an AI agent that stages everything automatically and pauses for the project lead before the kickoff is booked and the team is notified. - [How to Connect Typeform with beehiiv (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-beehiiv.md): Turn Typeform signups into beehiiv subscribers with UTM attribution and automation enrollment, run by an AI agent that adds clean opt-ins automatically and pauses for a human before reactivating anyone who previously unsubscribed. - [How to Connect Typeform with Braze (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-braze.md): Turn Typeform responses into Braze profile updates and triggered campaigns, with an AI agent that reads each response, respects the consent it contains, and pauses for a marketer before any send reaches more than one person. - [How to Connect Typeform with Campaign Monitor (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-campaign-monitor.md): Enroll Typeform respondents into Campaign Monitor lists automatically, with an AI agent that checks each address's subscription state first and pauses for the list owner before any add that would override a recorded opt-out. - [How to Connect Typeform with Canny (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-canny.md): Connect Typeform survey responses to Canny boards, optionally as an AI agent that turns open-text answers into deduped votes and comments and pauses for a human before a new post reaches the public board. - [How to Connect Typeform with Caspio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-caspio.md): Move Typeform submissions into Caspio tables as validated records, with an AI agent that deduplicates against existing rows and pauses for a person before merging records or purging source responses with Delete Responses. - [How to Connect Typeform with CleverReach (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-cleverreach.md): Move Typeform respondents into CleverReach groups automatically, with an AI agent that checks consent on every response and pauses for a human before Add Receiver puts anyone on a mailing list. - [How to Connect Typeform with Clio Manage (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-clio-manage.md): Connect Typeform intake questionnaires to Clio Manage so completed forms become contacts and matters, with an AI agent that reads the free-text answers, screens for conflicts, and pauses for the supervising attorney before Create Matter runs. - [How to Connect Typeform with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-google-sheets.md): Connect Typeform to Google Sheets so every new form submission lands as a row automatically, optionally with an AI agent that validates the data and pauses for a human before pushing incomplete or sensitive entries downstream. - [How to Connect Typeform with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-hubspot.md): Connect Typeform and HubSpot so every form submission creates or updates a HubSpot contact and deal automatically, optionally running as an AI agent that pauses for human review before writing duplicate or sensitive records. - [How to Connect Typeform with Salesforce Pro (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-typeform-with-salesforce-pro.md): Connect Typeform to Salesforce Pro so every form submission creates or updates the right lead, contact, or record automatically, with a human in the loop on the conversions that are irreversible. - [Can Vapi Call HubSpot Leads and Write the Outcome Back Itself?](https://flowrunner.ai/workflows/connect-vapi-with-hubspot.md): Connect Vapi and HubSpot so outbound calls dial from deal data and log their own transcript and outcome, with a human approving the list before anyone's phone rings. - [Wave + Google Sheets: Sheet Rows Become Reviewed Invoices](https://flowrunner.ai/workflows/connect-wave-with-google-sheets.md): Connect Wave and Google Sheets so a billing request row drafts a Wave invoice and Wave's transactions flow back into a reconciliation sheet, with an AI agent that pauses for a bookkeeper before an unusual invoice goes out. - [How to Connect Webflow with Airtable (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-webflow-with-airtable.md): Connect Webflow and Airtable so that form submissions, order events, and CMS content changes flow automatically into Airtable records, optionally as an AI agent that pauses for human sign-off before publishing. - [Webflow + HubSpot: Every Form Lead Lands in the CRM Deduped](https://flowrunner.ai/workflows/connect-webflow-with-hubspot.md): Connect Webflow and HubSpot so every form submission becomes a deduplicated, routed HubSpot contact, with an AI agent pausing for a CRM admin when it finds a conflicting record. - [Send Webflow Leads to Slack, Approve Every Publish](https://flowrunner.ai/workflows/connect-webflow-with-slack.md): Connect Webflow to Slack so new form leads post instantly, and run it as an AI agent that pauses for a marketing manager's approval before a content change goes live. - [How to Connect WhatsApp Business with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-whatsapp-with-hubspot.md): Connect WhatsApp Business to HubSpot so inbound messages create contacts and deals automatically, with an AI agent that pauses for human judgment on ambiguous records or high-stakes actions. - [Wise + Xero: Approved Bills Paid Abroad and Reconciled](https://flowrunner.ai/workflows/connect-wise-with-xero.md): Connect Wise and Xero so an approved vendor bill becomes a priced, funded international transfer, with a human approving the release and Xero marked paid automatically. - [Connect WooCommerce to Google Sheets: Approved Storewide Pricing](https://flowrunner.ai/workflows/connect-woocommerce-with-google-sheets.md): Connect WooCommerce and Google Sheets so a finance-owned price sheet updates the storefront catalog automatically, with a person approving the batch before it goes live storewide. - [How to Connect WooCommerce with QuickBooks Online (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-woocommerce-with-quickbooks-online.md): Connect WooCommerce and QuickBooks Online so every order, payment, and refund flows into your books automatically, with a human in the loop on the transactions that carry real consequence. - [How to Connect WooCommerce with ShipStation (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-woocommerce-with-shipstation.md): Connect WooCommerce and ShipStation so every new order triggers automatic rate shopping, label purchase, and customer notification, optionally as an AI agent that pauses for a human before buying a high-cost label. - [How to Connect WooCommerce with Xero (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-woocommerce-with-xero.md): Connect WooCommerce to Xero so every new order automatically creates an invoice and records payment in your books, optionally as an AI agent that pauses for a human before posting a disputed charge or voiding a record. - [WooCommerce Zoho Books Integration: Invoiced Fast, Refunds Held](https://flowrunner.ai/workflows/connect-woocommerce-with-zoho-books.md): Connect WooCommerce and Zoho Books so every order becomes an invoice on arrival, and run it as an AI agent that pauses for a human before a refund clears. - [How to Connect WordPress with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-wordpress-with-slack.md): Connect WordPress and Slack so every new published post sends a Slack message automatically, or run the connection as an AI agent that stages drafts, routes editorial approvals in Slack, and holds publishing for a human. - [How to Connect Wufoo with Mailchimp Marketing (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-wufoo-with-mailchimp-marketing.md): Connect Wufoo form submissions to Mailchimp Marketing audiences in real time, optionally as a FlowRunner AI agent that resolves field IDs, tags subscribers, and pauses for human review before adding flagged contacts. - [How to Connect Xero with Adobe Sign (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-adobe-sign.md): Connect Xero invoices to Adobe Sign so engagement letters and payment authorizations go out for signature the moment the billing exists, with an AI agent that pauses for a human before a signature request reaches a client. - [How to Connect Xero with Aircall (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-aircall.md): Connect Xero invoices and payments to Aircall contacts and insight cards so reps see the account balance on every call, optionally as an AI agent that pauses for the AR owner before any customer gets flagged for a collections conversation. - [How to Connect Xero with Attio (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-attio.md): Connect Xero invoices and payments to Attio records so account health lives in your CRM, with an AI agent that pauses for a human before a payment reminder reaches a customer. - [How to Connect Xero with Base64.ai (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-base64-ai.md): Connect Base64.ai document extraction to Xero so scanned supplier invoices become payable bills automatically, with an AI agent that pauses for a human before any payment leaves the account. - [How to Connect Xero with Bigin by Zoho (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-bigin-by-zoho.md): Connect Xero payments and invoices to Bigin deals, optionally as an AI agent that keeps the pipeline in step with the books and pauses for a human before a payment reminder reaches a customer. - [How to Connect Xero with Bland AI (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-bland.md): Connect Xero's AR aging to Bland AI outbound phone calls, with an AI agent that builds the daily collections call list and pauses for a named human before Send Call dials a customer. - [How to Connect Xero with Braintree (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-braintree.md): Connect Xero invoices to Braintree charges and refunds, with an AI agent that reconciles both ledgers and pauses for a named human before Charge Payment Method or Refund Transaction moves real money. - [How to Connect Xero with Canva (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-canva.md): Connect Xero's financial reports to Canva brand templates, with an AI agent that autofills the monthly report pack and pauses for a controller's sign-off before Export Design and Wait produces the files anyone sees. - [How to Connect Xero with Capsule CRM (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-capsule-crm.md): Connect Xero payments, invoices, and credit notes to Capsule CRM opportunities and tasks, optionally as an AI agent that updates the pipeline automatically and pauses for a human before a permanent record merge or a credit hold. - [How to Connect Xero with Cin7 Omni (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-cin7.md): Turn Xero invoices into Cin7 Omni sales orders with a stock check in between, optionally as an AI agent that pauses for a human before a credit note or void touches the books when inventory cannot cover the order. - [How to Connect Xero with CloudConvert (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-cloudconvert.md): Connect Xero invoices and credit notes to CloudConvert document processing, optionally as an AI agent that assembles billing packets automatically and pauses for a human before an invoice is emailed to the customer. - [How to Connect Xero with CloudTalk (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-cloudtalk.md): Connect Xero receivables to CloudTalk calls and SMS, optionally as an AI agent that runs payment reminders automatically and pauses for a human before a collections message reaches a sensitive account. - [How to Connect Xero with Copilot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-xero-with-copilot.md): Connect Xero invoices and payments to the Copilot client portal, optionally as an AI agent that mirrors the books into the portal and pauses for a human before a client-visible invoice goes live. - [How to Connect YouTube with Google Sheets (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-youtube-with-google-sheets.md): Connect YouTube to Google Sheets so every new video, comment, or subscriber event is logged automatically, and run it as an AI agent with a human on the steps that reach your audience. - [Can an AI Agent Route Zendesk Sell Exceptions to Slack?](https://flowrunner.ai/workflows/connect-zendesk-sell-with-slack.md): Connect Zendesk Sell and Slack so deal exceptions post to Slack automatically, with an AI agent that pauses for a rep's approval before deleting or merging any deal. - [Can an AI Agent Route Zendesk Tickets Into Google Sheets?](https://flowrunner.ai/workflows/connect-zendesk-with-google-sheets.md): Connect Zendesk and Google Sheets so every ticket logs itself, with an AI agent that pauses for a lead before it applies a bulk reassignment. - [How to Connect Zendesk with HubSpot (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-zendesk-with-hubspot.md): Connect Zendesk and HubSpot so every new support ticket creates or updates the matching HubSpot contact and deal automatically, with an AI agent that pauses for human review when a ticket signals a deal at risk. - [How to Connect Zendesk with Jira Issues (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-zendesk-with-jira-issues.md): Connect Zendesk and Jira Issues so support tickets automatically become tracked engineering issues, optionally run by an AI agent that pauses for a human before assigning a high-priority bug. - [How to Connect Zendesk with Linear (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-zendesk-with-linear.md): Connect Zendesk and Linear so new support tickets automatically become tracked engineering issues, with a FlowRunner AI agent that deduplicates, routes, and pauses for a human before creating anything that could clutter the backlog. - [Connect Zendesk to Notion: Every Escalation Logged for Review](https://flowrunner.ai/workflows/connect-zendesk-with-notion.md): Connect Zendesk and Notion so every urgent ticket and bulk reassignment lands in a Notion decision log, and run it as an AI agent that pauses for a support lead before the batch moves. - [How to Connect Zendesk with Slack (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-zendesk-with-slack.md): Connect Zendesk and Slack so every new ticket routes to the right Slack channel instantly, with an optional AI agent that triages, escalates, and pauses for a human before bulk moves go through. - [Send Zoho Books Bill Approvals to Slack Automatically](https://flowrunner.ai/workflows/connect-zoho-books-with-slack.md): Connect Zoho Books and Slack so a vendor bill posts for approval automatically, and let an AI agent run the connection with a controller in the loop on the bills that are actually risky. - [How to Connect Zoho CRM to Google Sheets: Sync Closed Deals](https://flowrunner.ai/workflows/connect-zoho-crm-with-google-sheets.md): Connect Zoho CRM and Google Sheets so closed deals log to a revenue tracker automatically, with an AI agent that pauses for a sales manager on strategic-account conversions. - [How to Connect Zoho CRM with Mailchimp Marketing (With or Without an AI Agent)](https://flowrunner.ai/workflows/connect-zoho-crm-with-mailchimp-marketing.md): Connect Zoho CRM and Mailchimp Marketing so every new lead, tag, and lifecycle change syncs automatically, with an AI agent that pauses for a human before sending to large audiences. - [Zoho CRM + Slack: New Leads Convert, Sales Gets Notified](https://flowrunner.ai/workflows/connect-zoho-crm-with-slack.md): Connect Zoho CRM and Slack so qualified leads convert automatically and sales hears about it instantly, with a manager approval step for strategic accounts. - [Zoho Inventory Shopify Integration: Reorders Get Human Approval](https://flowrunner.ai/workflows/connect-zoho-inventory-with-shopify.md): Connect Zoho Inventory and Shopify so every checkout creates a sales order and syncs stock across both systems, with a human approving any reorder above a spend threshold before the purchase order goes out. - [Zoho Recruit Gmail Integration: Offers Wait for Approval](https://flowrunner.ai/workflows/connect-zoho-recruit-with-gmail-service.md): Connect Zoho Recruit and Gmail so interview details send the moment a slot is booked, with an AI agent that drafts offer emails and pauses for a hiring manager before any offer goes out. - [Extract Data from PDF to Excel: The Capability Stack Behind a Workflow Finance Will Actually Trust](https://flowrunner.ai/workflows/extract-data-from-pdf-to-excel.md): The converter category answers the wrong question. Here is the capability stack a finance team needs to land PDF data in Excel without a reviewer regretting the run. - [Preparing Sales Reps Before Every Meeting Automatically](https://flowrunner.ai/workflows/orchestrate-calendly-and-hubspot-and-slack.md): A FlowRunner workflow coordinates scheduling, CRM, and team communication so every sales rep walks into every meeting prepared, with zero manual work. - [Turning a Signed Contract into an Invoice in Minutes](https://flowrunner.ai/workflows/orchestrate-docusign-and-hubspot-and-quickbooks-online-and-slack.md): How a FlowRunner workflow takes a signed contract and automatically updates your CRM, creates an invoice, and notifies your team, with human oversight on exceptions. - [Qualifying Inbound Leads Without Manual Research](https://flowrunner.ai/workflows/orchestrate-google-forms-and-hubspot-and-slack.md): One FlowRunner workflow coordinates Google Forms, HubSpot, and Slack to qualify inbound leads, enrich CRM records, and route exceptions to your team for the decisions that matter. - [How to Automate Invoice Processing from Email to Payment](https://flowrunner.ai/workflows/orchestrate-mailbox-and-parseur-and-quickbooks-online-and-slack.md): One FlowRunner workflow monitors your AP inbox, extracts invoice data, validates vendors and amounts in QuickBooks, creates bills, and routes exceptions to your team in Slack. - [Processing Vendor Documents into NetSuite Without Manual Entry](https://flowrunner.ai/workflows/orchestrate-netsuite-and-parseur-and-slack.md): How a FlowRunner workflow extracts vendor invoices, validates them against ERP data, and records bills in NetSuite with human oversight on exceptions. - [Eliminating Manual Data Entry from Documents to ERP](https://flowrunner.ai/workflows/orchestrate-parseur-and-acumatica-and-slack.md): How one FlowRunner workflow extracts invoice data from documents, validates vendors and catches duplicates in your ERP, and routes exceptions to humans in Slack for resolution. - [Catching Fulfillment Errors Before They Reach Customers](https://flowrunner.ai/workflows/orchestrate-shipbob-and-quickbooks-online-and-slack.md): A FlowRunner workflow monitors shipment exceptions in real time, cross-references order data with financial records, resolves routine issues, and escalates high-value cases to your ops team with context from every system involved. - [Reconciling Payments Against Invoices Automatically Across Your Revenue Stack](https://flowrunner.ai/workflows/orchestrate-stripe-and-quickbooks-online-and-slack.md): How a single FlowRunner workflow reconciles every Stripe payment against QuickBooks invoices in real time, escalates mismatches through Slack, and keeps your books accurate without manual data entry. - [PDF Data Extraction to Excel: A Workflow Pattern Finance Can Actually Trust](https://flowrunner.ai/workflows/pdf-data-extraction-to-excel.md): A field-level extraction pattern that lands recurring PDFs as Excel-ready rows, with a reviewer in the loop for low-confidence fields and format drift. - [Ramp and Sage Intacct: Where the Native Integration Stops and the Exceptions Begin](https://flowrunner.ai/workflows/ramp-sage-intacct-integration-exception-handling.md): Ramp's native Sage Intacct integration syncs coded transactions on a schedule. The cleanup work between sync and clean books is where an orchestration layer earns its keep. - [Salesforce Data Hygiene as a Control Surface, Not a Cleanup Project](https://flowrunner.ai/workflows/salesforce-data-hygiene.md): Salesforce hygiene degrades because most teams run it as a quarterly project. The fix is to enforce it inline, on every write, and route ambiguous matches to a human. - [Salesforce Leads vs Opportunities: The Conversion Gate Is the Decision](https://flowrunner.ai/workflows/salesforce-leads-vs-opportunities.md): Leads and Opportunities are different data models, and the conversion between them is the handoff that owns forecast accuracy. Most teams treat it as a click. --- ## Integration categories (24 categories) - [AI & LLMs Integrations for AI Agents](https://flowrunner.ai/integrations/category/ai-llms.md): 154 AI & LLMs integrations for AI agents and workflows. The full-surface AI provider layer for your agents. FlowRunner ships the complete current API for each major model provider, not a thin chat wrapper, so an agent can generate, reason, extract, transcribe, and moderate through the provider you c - [Vector Stores & AI Infra Integrations for AI Agents](https://flowrunner.ai/integrations/category/vector-stores-ai-infra.md): 10 Vector Stores & AI Infra integrations for AI agents and workflows. The memory and retrieval layer that gives your agents long-term recall. Connect every major vector store and RAG backend so an agent can embed, search, and remember across runs, and pause for a human before it erases memory the te - [Databases & Warehouses Integrations for AI Agents](https://flowrunner.ai/integrations/category/databases-warehouses.md): 54 Databases & Warehouses integrations for AI agents and workflows. Every operational database and analytics warehouse, one runtime. Agents read and write your data as a governed step in a flow, and stop for a human before the writes that carry risk, like a bulk delete or a change against a producti - [Developer & Infrastructure Integrations for AI Agents](https://flowrunner.ai/integrations/category/developer-infrastructure.md): 152 Developer & Infrastructure integrations for AI agents and workflows. The build, deploy, and observability stack as agent tools. Source control, CI/CD, monitoring, and edge infrastructure an agent can operate, with a human on the moves that touch production, like a release, a merge, or a config c - [Identity & Security Integrations for AI Agents](https://flowrunner.ai/integrations/category/identity-security.md): 39 Identity & Security integrations for AI agents and workflows. The complete SOC and identity stack in one runtime. Provisioning, threat intelligence, case management, and posture scoring, wired so an agent handles the routine investigation and a human owns the consequential action, like a deprovis - [CRM & Sales Integrations for AI Agents](https://flowrunner.ai/integrations/category/crm-sales.md): 162 CRM & Sales integrations for AI agents and workflows. Lead to revenue on autopilot. Enrich, route, sequence, and sync across every CRM and sales system, with a human in the loop on the moves that commit real pipeline, like converting a strategic lead or creating a high-value deal. - [Marketing & Social Integrations for AI Agents](https://flowrunner.ai/integrations/category/marketing-social.md): 90 Marketing & Social integrations for AI agents and workflows. Campaigns, ads, and social presence as agent workflows. An agent drafts, schedules, and reports, and a human approves anything that reaches an audience or spends budget. - [Communication & Messaging Integrations for AI Agents](https://flowrunner.ai/integrations/category/communication-messaging.md): 205 Communication & Messaging integrations for AI agents and workflows. The channels where your team already works, and where an agent asks for a decision. Every chat, SMS, voice, and push platform as a native human-in-the-loop surface, so the question lands where the responsible person is already l - [Voice & Telephony Integrations for AI Agents](https://flowrunner.ai/integrations/category/voice-telephony.md): 47 Voice & Telephony integrations for AI agents and workflows. Cloud phone systems, contact centers, and AI voice agents your flows can operate. Agents place the routine call, transcribe what was said, and attach it to the right record, then put a person on the line the moment the conversation stops - [Email & Marketing Integrations for AI Agents](https://flowrunner.ai/integrations/category/email-marketing.md): 143 Email & Marketing integrations for AI agents and workflows. Transactional and lifecycle email as agent tools. Agents send the one-to-one mail automatically and hold for a human before anything reaches a segment. - [Loyalty & Referral Integrations for AI Agents](https://flowrunner.ai/integrations/category/loyalty-referral.md): 21 Loyalty & Referral integrations for AI agents and workflows. Loyalty programs, wallet passes, referral engines, and reward platforms as agent tools. Agents award points and credit referrals off verified events in the systems where they actually happened, and hold anything that spends real money f - [Helpdesk & ITSM Integrations for AI Agents](https://flowrunner.ai/integrations/category/helpdesk-itsm.md): 27 Helpdesk & ITSM integrations for AI agents and workflows. The full service-desk and ITSM stack. Triage, route, and resolve across every helpdesk, with a human approving the changes that affect SLAs and production, like a change-request or a bulk reassignment. - [Finance & Accounting Integrations for AI Agents](https://flowrunner.ai/integrations/category/finance-accounting.md): 197 Finance & Accounting integrations for AI agents and workflows. Orders, payments, and books, connected. Agents run the routine finance work and stop for a human before money moves, with the approver and timestamp captured in the audit trail. - [E-commerce Integrations for AI Agents](https://flowrunner.ai/integrations/category/e-commerce.md): 88 E-commerce integrations for AI agents and workflows. The storefront and order lifecycle as agent workflows. Agents handle the routine order and catalog work, and a human owns anything a store cannot take back, like a refund above a threshold or a cancellation. - [Logistics & Fulfillment Integrations for AI Agents](https://flowrunner.ai/integrations/category/logistics-fulfillment.md): 41 Logistics & Fulfillment integrations for AI agents and workflows. Shipping, carriers, and fulfillment as agent tools. Agents shop rates, buy labels, and track shipments, with a human gate on the ones that carry unexpected cost or risk. - [Project Management & Productivity Integrations for AI Agents](https://flowrunner.ai/integrations/category/project-management-productivity.md): 232 Project Management & Productivity integrations for AI agents and workflows. The boards, docs, tasks, and time tracking your team runs on. An agent captures the routine and pulls a person in for the judgment calls, like assigning work or a sweeping bulk update. - [Documents & Forms Integrations for AI Agents](https://flowrunner.ai/integrations/category/documents-forms.md): 178 Documents & Forms integrations for AI agents and workflows. Generate, parse, and route documents and form submissions. An agent stages the paperwork, and a human approves anything that goes out to a client or customer. - [CMS & Content Integrations for AI Agents](https://flowrunner.ai/integrations/category/cms-content.md): 21 CMS & Content integrations for AI agents and workflows. Every headless and traditional CMS as agent tools. Agents draft and stage content automatically, and a human owns the publish that makes it live. - [Media & Video Integrations for AI Agents](https://flowrunner.ai/integrations/category/media-video.md): 19 Media & Video integrations for AI agents and workflows. Video hosting, video and audio generation, podcast platforms, and media catalogs as agent tools. Agents render, upload, tag, and transcribe media automatically, and hold for a person before anything is published to an audience or removed fro - [Storage Integrations for AI Agents](https://flowrunner.ai/integrations/category/storage.md): 30 Storage integrations for AI agents and workflows. Cloud storage and file operations as agent tools. Agents move, read, and organize files, with a human gate on external shares and destructive writes. - [HR Integrations for AI Agents](https://flowrunner.ai/integrations/category/hr.md): 43 HR integrations for AI agents and workflows. The HR and recruiting stack as agent workflows. Agents handle the routine records, and a human signs off on the writes that touch pay, employment status, or a candidate's standing. - [Education & LMS Integrations for AI Agents](https://flowrunner.ai/integrations/category/education-lms.md): 28 Education & LMS integrations for AI agents and workflows. The learning stack as agent tools. Course platforms, corporate LMS, and school management systems where agents handle enrollment, progress tracking, and credential workflows, and a human approves the writes that touch a learner record or r - [Analytics & Data Integrations for AI Agents](https://flowrunner.ai/integrations/category/analytics-data.md): 91 Analytics & Data integrations for AI agents and workflows. Web analytics and live data feeds as agent inputs. The read is safe to automate; a human owns the decision the data drives, like pausing spend or rescheduling on a forecast. - [Utilities & Personal Integrations for AI Agents](https://flowrunner.ai/integrations/category/utilities-personal.md): 70 Utilities & Personal integrations for AI agents and workflows. The long tail of utility and personal APIs your workflows reach for, from URL shorteners and bookmarks to smart-home control, each with an honest human gate on the actions that carry consequence. --- ## Integrations (2142 services) Each page documents the service's triggers and actions as AI agent capabilities in FlowRunner. - [Integrations](https://flowrunner.ai/integrations.md): Connect AI agents to the tools your team already uses. 2142 integrations across finance, CRM, communication, security, data, and more. - [3dcart Integration](https://flowrunner.ai/integrations/a3dcart.md): 3dcart, now sold as Shift4Shop, is a hosted storefront platform with a mature REST API. Agents sync products, stock and prices from an ERP, import orders from other channels, manage customers, promotions and returns, and react to store webhooks in real time. - [Abby Integration](https://flowrunner.ai/integrations/abby.md): Abby is the French invoicing and bookkeeping app for sole traders and micro-entrepreneurs. Agents manage business and individual clients, the product catalog, invoices, estimates, and credit notes, and keep the accounting registers current from inside a workflow. - [Abyssale Integration](https://flowrunner.ai/integrations/abyssale.md): Abyssale is a creative automation platform that renders images, videos, GIFs, HTML5 banners, and print-ready PDFs from templates. Agents generate on-brand assets in every format at once, build dynamic image URLs, and pull finished files or ZIP exports into the rest of a workflow. - [AcademyOcean Integration](https://flowrunner.ai/integrations/academyocean.md): Connect AI agents to AcademyOcean, an LMS for employee and customer training. Agents invite and manage learners, assign them to teams, track course progress and certificates, and pull engagement data so training operations run without manual roster upkeep. - [Accelo Integration](https://flowrunner.ai/integrations/accelo.md): Connect AI agents to Accelo, the professional services automation platform. Agents manage companies and contacts, look up jobs, create tasks, and log activities so client work stays tracked end to end. - [Access Charity CRM Integration](https://flowrunner.ai/integrations/access-charity-crm.md): Connect AI agents to Access Charity CRM (formerly thankQ), the fundraising CRM from The Access Group. Agents record donations and pledges, keep constituent records current, log communications, and track memberships, events, and Gift Aid declarations. - [Action Network Integration](https://flowrunner.ai/integrations/action-network.md): Connect AI agents to Action Network, an organizing and advocacy platform for progressive campaigns. Agents manage activists, events, tags, and action pages, upsert people by email, and read email message performance via an OSDI-compliant API key. - [ActiveCampaign Integration](https://flowrunner.ai/integrations/activecampaign.md): Connect AI agents to ActiveCampaign. Agents sync contacts, apply tags, manage lists and custom fields, move deals through pipelines, enroll contacts in automations, and react to marketing events in real time. - [ActiveMerge Integration](https://flowrunner.ai/integrations/activemerge.md): ActiveMerge fills document templates with your data and returns a finished PDF, Word, or EPUB file. Agents generate contracts and letters from records, merge several files into one, render images from templates, and send the result straight out for signature. - [ActiveTrail Integration](https://flowrunner.ai/integrations/activetrail.md): ActiveTrail is an Israeli email, SMS, and marketing automation platform. Agents manage contacts, groups, and mailing lists, launch email and SMS campaigns, send transactional messages, and pull campaign reports by outcome. - [Acuity Scheduling Integration](https://flowrunner.ai/integrations/acuity.md): Book, reschedule, and cancel appointments, check open availability, and keep clients, appointment types, and intake forms in sync through the Acuity Scheduling API. - [Acuity PPM Integration](https://flowrunner.ai/integrations/acuity-ppm.md): Acuity PPM is project portfolio management for PMOs that track projects, proposals and status reports across a portfolio. Agents pull whole-portfolio snapshots in one call, then read and update individual projects, tasks, status reports and proposals. - [Acumatica Integration](https://flowrunner.ai/integrations/acumatica.md): Connect AI agents directly to your Acumatica ERP. Agents manage vendors, process bills end-to-end, check duplicate entries, retrieve GL account data, and monitor AP balances. - [AcyMailing Integration](https://flowrunner.ai/integrations/acymailing.md): AcyMailing is the newsletter extension for Joomla and WordPress sites. Agents manage users, lists, and subscriptions, send emails and transactional messages, run campaigns and follow-up sequences, and read statistics from your own site. - [Adalo Integration](https://flowrunner.ai/integrations/adalo.md): Read and write records in your Adalo app collections through the Adalo Collections API. Authenticates with a Bearer API key scoped to a specific Adalo App ID. - [AddressVerify Integration](https://flowrunner.ai/integrations/addressverify-app.md): Validate United States residential addresses and identify the property behind them in a single call. Agents confirm deliverability, classify property type, and read estimated home value before an address reaches a CRM or a shipping label. - [Adobe Creative Cloud Libraries Integration](https://flowrunner.ai/integrations/adobe-cc-libraries.md): Adobe Creative Cloud Libraries hold the colors, styles, images, and graphics a design team reuses across every Adobe app. Agents create and rename libraries, upload assets, update element metadata, search, and keep shared brand libraries current without opening Photoshop or Illustrator. - [Adobe Acrobat Sign Integration](https://flowrunner.ai/integrations/adobe-sign.md): Send documents for e-signature with Adobe Acrobat Sign. Agents upload files, send and track agreements, chase signers with reminders, pull completed form data, and download signed PDFs and audit trails. - [Adobe Workfront Integration](https://flowrunner.ai/integrations/adobe-workfront.md): Manage Adobe Workfront work at scale with AI agents. Agents search and update projects, tasks, and issues, look up users and documents, and create new work items from signals in other systems. - [Affinity Integration](https://flowrunner.ai/integrations/affinity.md): Connect AI agents to Affinity, the relationship intelligence CRM. Agents manage lists, people, organizations, opportunities, custom fields, and notes, and react to Affinity events in real time. - [AfterShip Integration](https://flowrunner.ai/integrations/aftership.md): Track shipments with AI agents through AfterShip. Agents create and update trackings, fetch the last checkpoint, detect the right courier from a tracking number, and monitor delivery status across supported carriers. - [Agendor Integration](https://flowrunner.ai/integrations/agendor.md): Connect AI agents to Agendor, a sales CRM built for Brazilian small and medium businesses. Agents create organizations and people, move deals through funnels, assign tasks, and read the categories, sectors, and lead origins those records depend on. - [Agent.ai Integration](https://flowrunner.ai/integrations/agent-ai.md): Connect AI agents to the Agent.ai Actions API for model calls, web research, and company or person enrichment. Agents run published Agent.ai agents as a step inside a larger FlowRunner workflow. - [Agile CRM Integration](https://flowrunner.ai/integrations/agile-crm.md): Connect AI agents to Agile CRM. Agents create and update contacts, companies, deals, tasks, and notes, tag records for routing, and keep sales pipelines current from your flows. - [AgilePlace Integration](https://flowrunner.ai/integrations/agileplace.md): Planview AgilePlace, formerly LeanKit, runs enterprise Kanban boards that connect strategy to delivery. Agents create and move cards, manage lanes and boards, log comments and attachments, and react to board events as they happen. - [Aha! Integration](https://flowrunner.ai/integrations/aha-io.md): Connect AI agents to Aha!, the product roadmap platform. Agents create and update features, capture ideas, and read products, releases, and initiatives to keep roadmaps in sync with customer feedback. - [AI Image Generator Integration](https://flowrunner.ai/integrations/ai-image-generator.md): Generate images from text prompts with OpenAI DALL-E. FlowRunner agents produce visuals on demand and pass the hosted URL straight into downstream flow steps. - [AI Scraper by Parsera Integration](https://flowrunner.ai/integrations/ai-scraper-by-parsera.md): Extract structured data from any web page with Parsera's LLM-powered scraper. Agents describe the fields they want in plain language and receive typed records back, with proxy and geolocation control for sites that resist automated access. - [AI Vision Integration](https://flowrunner.ai/integrations/ai-vision.md): Analyze images through a single connector that routes to ten vision providers, including OpenAI, Anthropic, Google, and Mistral. Describe scenes, read text, and answer questions about any image inside a flow. - [AI21 Labs Integration](https://flowrunner.ai/integrations/ai21-labs.md): Call AI21's Jamba models from your flows. Agents run chat completions with tool calling and JSON mode, ask one-shot questions, and answer questions grounded in supplied documents using Jamba's 256K token context window. - [AidaForm Integration](https://flowrunner.ai/integrations/aidaform.md): AidaForm is a form, survey, quiz and landing page builder from Bonn, Germany. Agents list the forms in an account, pull submissions with their answers, render a submission to PDF, Excel, CSV or JSON, and delete entries when a retention rule says so. - [Aidbase Integration](https://flowrunner.ai/integrations/aidbase.md): Connect AI agents to Aidbase, the AI support platform for SaaS startups. Agents query chatbots, add knowledge sources such as crawled sites, transcripts, and uploaded documents, set ticket status and priority, and poll connected email inboxes. - [aiKonnector Integration](https://flowrunner.ai/integrations/aikonnector.md): Push an item into an aiKonnector approval or review queue and read the decision back. Agents use it as an external checkpoint when a review already lives in aiKonnector rather than in the flow itself. - [Aimfox Integration](https://flowrunner.ai/integrations/aimfox.md): Run LinkedIn outreach campaigns through Aimfox, covering audiences, connected accounts, sending limits, leads, labels, templates, and the unified inbox. Agents build target lists, launch sequences, and route replies to the right owner. - [AI/ML API Integration](https://flowrunner.ai/integrations/aimlapi.md): Reach the AI/ML API catalog of more than 800 models through one FlowRunner connection. Agents run chat completions, image generation, and embeddings without a separate account per model provider. - [Aircall Integration](https://flowrunner.ai/integrations/aircall.md): Connect AI agents to Aircall, your cloud phone system. Agents log comments and tags on calls, transfer calls, control recordings, surface insight cards during live conversations, and manage contacts, numbers, teams, and webhooks. - [Airmeet Integration](https://flowrunner.ai/integrations/airmeet.md): Work Airmeet events with AI agents. Agents list events and sessions, pull attendee lists, and register attendees so virtual event logistics run without manual roster work. - [AirMenu Integration](https://flowrunner.ai/integrations/airmenu.md): Connect AI agents to AirMenu, a digital menu and ordering platform for restaurants and hospitality venues. Agents publish menu changes, send and flag orders, run promotions and redeem codes, keep stock counts accurate, and configure which payment methods a venue offers. - [Airparser Integration](https://flowrunner.ai/integrations/airparser.md): Extract structured data from emails, PDFs, scans, and spreadsheets using Airparser's large language model parsers. Agents send a document to an inbox, receive typed fields back, and act on parsed results as soon as they land. - [Airtable Integration](https://flowrunner.ai/integrations/airtable.md): Use Airtable as a living operational database for your AI agents. 2 record triggers kick off workflows when data changes. 17 actions cover full CRUD across bases, tables, records, and comments. - [Airtop Integration](https://flowrunner.ai/integrations/airtop.md): Drive real, stateful cloud browsers from your agents: open pages, scrape or query their content in natural language, and click, type, and fill forms on live sites. Built for browser tasks that have no API. - [AIVOOV Integration](https://flowrunner.ai/integrations/aivoov.md): Turn text into natural speech with AIVOOV, a single interface over more than 1500 voices and 140 languages. Agents generate voiceovers, IVR prompts, and audio notifications as a step in a workflow. - [Alegra Integration](https://flowrunner.ai/integrations/alegra.md): Alegra is cloud accounting and electronic invoicing for Latin America and Spain. Agents create contacts and items, issue sales invoices through the electronic stamping flow, raise quotes and credit notes, and record supplier bills and payments. - [Algebras AI Integration](https://flowrunner.ai/integrations/algebras.md): Translate content into more than 300 languages with Algebras AI, enforcing your own glossary so product and brand terms stay fixed. Agents translate records in bulk and score translation quality before anything is published. - [Algolia Integration](https://flowrunner.ai/integrations/algolia-com.md): Keep Algolia search indices fresh with AI agents. Agents add, update, and delete objects, run searches, execute batch operations and delete-by-query cleanups, and manage index settings so results always reflect live data. - [ALITEO Integration](https://flowrunner.ai/integrations/aliteo.md): Connect AI agents to ALITEO, the KARAT Software project, task, and service desk platform. Agents create and update tasks, run rich task filters, manage service requests, and keep delivery boards current. - [All-Images.ai Integration](https://flowrunner.ai/integrations/all-images-ai.md): Search a royalty-free AI stock library and generate new Midjourney-backed images with commercial licensing through All-Images.ai. Agents source or create artwork for posts, listings, and campaigns without a manual design round trip. - [AllMySMS Integration](https://flowrunner.ai/integrations/allmysms.md): Connect AI agents to AllMySMS, a French SMS gateway for single and bulk campaign messaging. Agents send SMS and voice campaigns, manage contact lists and the opt-out blacklist, and pull delivery reports and replies so outreach to French customers stays tracked and compliant. - [Allthings Integration](https://flowrunner.ai/integrations/allthings.md): Allthings is the tenant engagement and building operations platform for property managers. Agents work the property, unit, and tenant hierarchy, open and update service tickets, post articles and pinboard content, manage bookable assets and bookings, and export data on demand. - [AlphaInsider Integration](https://flowrunner.ai/integrations/alphainsider.md): AlphaInsider is a social trading marketplace where strategies are published, followed, and wired to real brokers. Agents search and monitor strategies, track positions and orders, place trades from allocation targets, and read bot, invoice, and payout activity. - [Altoviz Integration](https://flowrunner.ai/integrations/altoviz.md): Altoviz is a French small business management platform for TPE and PME. Agents manage customers, suppliers, and the product catalog, create and send quotes and invoices, and react to webhook events as the books change. - [AltTextLab Integration](https://flowrunner.ai/integrations/alttextlab.md): Generate accessibility alt text for images with AltTextLab. Agents describe uploaded media automatically so published pages, product listings, and email campaigns stay accessible without a manual writing pass. - [Alvanda Integration](https://flowrunner.ai/integrations/alvanda.md): Alvanda is process management where the work of an organization is written as procedures made of steps, owned by positions on an org chart rather than named people. Agents read and update procedures, steps, positions and assignments so documented process and real work stay in sync. - [Amabis Data Quality Integration](https://flowrunner.ai/integrations/amabis-data-quality.md): Clean, correct, and validate email addresses, phone numbers, postal addresses, and person or company identities with Amabis. Agents run contact records through real-time quality checks before they reach a CRM or a send list. - [Amazing Marvin Integration](https://flowrunner.ai/integrations/amazing-marvin.md): Amazing Marvin is the customizable personal task manager built around strategies. Agents create and schedule tasks, projects, and events, read the day's plan and what is due, log time, score habits, track goals and reward points, and set reminders. - [Amazon Creators API Integration](https://flowrunner.ai/integrations/amazon-pa-api-v5.md): The Amazon Creators API is the live successor to Product Advertising API 5.0 for Amazon Associates. Agents search the Amazon catalog, pull titles, prices, images and availability for ASINs, and watch listings for price and stock changes. - [Amazon Seller Central Integration](https://flowrunner.ai/integrations/amazon-seller-central.md): Run your Amazon storefront with AI agents through the Selling Partner API. Agents pull orders and order items, confirm shipments, manage catalog listings, check inventory, request and download reports, and track financial events. - [AmeriCommerce (Cart.com) Integration](https://flowrunner.ai/integrations/americommerce.md): AmeriCommerce by Cart.com is a multi-store online storefront platform. Agents push orders into a warehouse or ERP, write shipments and tracking back, keep stock in step by SKU, and manage the catalog, customers and checkout pipeline. - [Amplitude Integration](https://flowrunner.ai/integrations/amplitude.md): Connect AI agents to Amplitude. Agents track events and identify users through the ingestion APIs, query segmentation, funnels, retention, and revenue LTV, manage cohorts including uploads and downloads, and govern the event taxonomy. - [Amwork Integration](https://flowrunner.ai/integrations/amwork.md): Amwork is a modular business workspace that combines CRM, boards, tasks, documents, and email in one account. Agents create and update CRM entities with custom fields, move records through board stages, assign tasks with subtasks and comments, and read the reporting dashboards. - [Anabix Integration](https://flowrunner.ai/integrations/anabix.md): Connect AI agents to Anabix, a Czech CRM for small and mid-sized businesses. Agents manage contacts, companies, deals, and tasks, log activities, and segment marketing lists against custom field definitions. - [Anaplan Integration](https://flowrunner.ai/integrations/anaplan.md): Connect AI agents to Anaplan, the enterprise planning and FP&A platform. Agents discover workspaces and models, read cell data from saved views, add and update list members, and run and monitor import, export, and process tasks through the Integration API. - [Anchor Integration](https://flowrunner.ai/integrations/anchor.md): Anchor is an autonomous billing and collections platform built on client agreements. Agents draft and publish proposals, track the agreements they become, and watch invoicing and collections run from those agreements. - [Anchor Browser Integration](https://flowrunner.ai/integrations/anchorbrowser.md): Anchor Browser gives AI agents a cloud browser with sessions, profiles, and identities. Agents open sessions, navigate and click, run plain-language web tasks, capture screenshots and PDFs, and react when a task completes or a human intervention is requested. - [AnnounceKit Integration](https://flowrunner.ai/integrations/announcekit.md): Publish multi-language changelog posts, run an ideas board, and read reader sentiment through the AnnounceKit GraphQL API. Agents turn shipped work into release notes and route feature requests to the right product owner. - [anny Integration](https://flowrunner.ai/integrations/anny.md): Connect AI agents to anny, a booking platform for rooms, workspaces, equipment, and event locations. Agents check resource availability, create and update bookings, manage customers, and cancel orders so shared spaces stay reserved without manual coordination. - [Anthropic Claude Integration](https://flowrunner.ai/integrations/anthropic-ai.md): Run Anthropic's Claude models across the full API surface: text, vision, PDF analysis with citations, tool use, structured JSON output, Message Batches, the Files API, and Managed Agents. Bring your own Anthropic key. - [Any.do Workspace Integration](https://flowrunner.ai/integrations/anydo-workspace.md): Manage team tasks in Any.do Workspace. Agents list, create, update, and delete tasks and browse the categories they belong to, turning inbound requests from forms, email, or chat into tracked work. - [Anymail Finder Integration](https://flowrunner.ai/integrations/anymailfinder.md): Find and verify B2B email addresses with Anymail Finder, covering person, decision maker, and company-wide searches. Agents resolve a contact from a name and domain, then confirm deliverability before the first send. - [Anysite Web Data Integration](https://flowrunner.ai/integrations/anysite.md): Pull LinkedIn profiles, companies, employees, posts, and job listings through the Anysite web data extraction platform. Agents build prospect lists and monitor hiring or headcount signals as buying triggers. - [Apaleo Integration](https://flowrunner.ai/integrations/apaleo.md): Connect AI agents to apaleo, the API-first cloud property management system for hotels. Agents check reservations in and out, cancel or mark them no-show, post charges and payments to folios, raise and settle invoices, and read rate plans, services, and unit availability across properties. - [Apifonica Integration](https://flowrunner.ai/integrations/apifonica.md): Send messages, place calls, and manage numbers on the Apifonica voice and SMS platform. Agents reach customers on the channel that fits the moment and provision numbers as campaigns require them. - [APIFRAME Integration](https://flowrunner.ai/integrations/apiframe.md): Reach Midjourney, Kling, Sora, Veo, Flux, Seedream, Suno, and more than 80 other image, video, and music models through the single APIFRAME v2 API. Agents generate media assets without managing an account per model. - [Apify Integration](https://flowrunner.ai/integrations/apify.md): Run Apify Actors and saved scraping tasks from your flows. Agents launch web scrapes on demand, poll or abort runs, read dataset items and key-value store records, and append structured results for downstream steps. - [Apify AI Website Crawler Integration](https://flowrunner.ai/integrations/apify-ai-website-crawler.md): Apify's Website Content Crawler turns an entire site into clean Markdown built for LLM, RAG, and vector-database pipelines. Workflows start crawls with page and spend limits, wait for the run to finish, and pull the results into a knowledge base. - [Apify LLM Scraper Integration](https://flowrunner.ai/integrations/apify-llm-scraper.md): Apify LLM Scraper runs the open-source Crawl4AI engine on Apify. Agents scrape a page or a whole site into Markdown, structured JSON by CSS or XPath selectors, or whatever an LLM prompt asks for, then collect the results when the run succeeds. - [AITable Integration](https://flowrunner.ai/integrations/apitable.md): AITable (formerly APITable) is an Airtable-style database with datasheets, views, spaces and a built-in AI agent. Agents read, create, update and delete records, manage fields, datasheets, members, teams and roles, and chat with the AITable AI agent from inside a flow. - [APITemplate.io Integration](https://flowrunner.ai/integrations/apitemplate.md): Generate PDFs and images from reusable templates or raw HTML with APITemplate.io, returning hosted download URLs. Agents render invoices, receipts, certificates, and marketing assets from record data as a step in a workflow. - [APIVerve Integration](https://flowrunner.ai/integrations/apiverve.md): Reach a curated set of APIVerve's utility APIs behind one key, covering conversion, validation, lookup, and generation tasks. Agents fill small gaps in a workflow without standing up a service for each one. - [Apple Push Notifications Integration](https://flowrunner.ai/integrations/apn.md): Apple Push Notifications (APNs) delivers alerts to your own iOS, iPadOS, macOS, watchOS, and tvOS apps. Workflows send alert, background, VoIP, and Live Activity pushes to one device or many, signed with your .p8 key. - [Apolearn Integration](https://flowrunner.ai/integrations/apolearn.md): Connect AI agents to Apolearn, a French learning management system. Agents create and manage users, courses, and cohorts, control enrollment, and read usage statistics so training rosters stay current across your instance. - [Apollo.io Integration](https://flowrunner.ai/integrations/apollo.md): Enrich leads and build prospect lists from a 275M+ contact database. Agents search people and companies, enrich contact data, find job postings, and add qualified leads to outreach sequences. - [Apple Map Links Integration](https://flowrunner.ai/integrations/apple-map-links.md): Build Apple Maps deep links entirely offline, with no API key, account, or outbound call. Agents attach directions, search links, and place lookups to messages, tickets, and dispatch records. - [AppSheet Integration](https://flowrunner.ai/integrations/appsheet.md): Read and write the data behind your Google AppSheet apps. Agents add, edit, delete, and find rows in any table or slice, honoring the app's security filters and firing the bots you have configured for each action. - [ApptiveGrid Integration](https://flowrunner.ai/integrations/apptivegrid.md): ApptiveGrid is a German no-code database and form builder built on a hypermedia API. Agents list spaces and grids, create and update entities, submit and read forms, and follow the links the server returns to reach any resource. - [Atlas (AQX) Integration](https://flowrunner.ai/integrations/aqx.md): Run AI voice and messaging campaigns through an Atlas (AQX) account, covering outbound, inbound, voice bubble, and Messages campaigns. Agents launch and monitor conversations at campaign scale and read results back into the flow. - [Arbox Integration](https://flowrunner.ai/integrations/arbox.md): Connect AI agents to Arbox, management software for gyms, studios, and fitness businesses. Agents enroll members, schedule classes, chase leads, send messages and digital forms for signature, and pull the built-in attendance and revenue reports. - [ArcGIS Field Maps Integration](https://flowrunner.ai/integrations/arcgis-field-maps.md): Connect AI agents to ArcGIS Field Maps, Esri's mobile data collection app backed by hosted feature layers. Agents query and edit feature records, attach photos to field observations, and read layer metadata for downstream processing. - [Arduino IoT Cloud Integration](https://flowrunner.ai/integrations/arduino.md): Connect AI agents to the Arduino IoT Cloud, where connected devices publish and receive properties. Agents read sensor properties, write control values back to devices, list registered things, and pull time series history for analysis. - [Arlo Training Integration](https://flowrunner.ai/integrations/arlo.md): Connect AI agents to Arlo, a training and course management platform. Agents search published events, courses, and online activities and look up presenters and time zones so upcoming training catalogs feed into downstream planning and outreach flows. - [Asaas Integration](https://flowrunner.ai/integrations/asaas.md): Asaas is a Brazilian payment institution and billing platform covering Pix, boleto, and cards. Agents create customers and charges, run subscriptions and payment links, handle refunds, splits, transfers, and bill payments, and react to charge events as they happen. - [Asana Integration](https://flowrunner.ai/integrations/asana.md): Turn operational workflows into tracked Asana work items. Agents create tasks and projects, organize sections, attach files, log comments, and find users and templates. - [Askara Integration](https://flowrunner.ai/integrations/askara.md): Connect AI agents to Askara, the medical practice platform for French healthcare providers. Agents read and write patient records, manage professional contacts and organizations, and retrieve clinical documents and notes. - [Aspose Cloud Integration](https://flowrunner.ai/integrations/aspose.md): Convert Word documents, PDFs, spreadsheets, and presentations between formats and merge Word files through the Aspose Cloud document APIs. Agents drop format conversion into any flow, turning inbound DOCX into PDF or XLSX into CSV without a desktop app. - [ASPSMS Integration](https://flowrunner.ai/integrations/aspsms.md): Connect AI agents to ASPSMS, a Swiss SMS gateway operated by VADIAN.NET. Agents send text and token SMS, verify two-factor codes, check credits, and manage originators and vouchers so notification and verification flows run through one Swiss gateway. - [AssemblyAI Integration](https://flowrunner.ai/integrations/assembly-ai.md): Transcribe recorded audio and video with AssemblyAI. Agents upload media, run speech-to-text with speaker labels, PII redaction, sentiment, chapters, and entity detection, export SRT or VTT subtitles, and run LLM prompts over finished transcripts. - [Astara Connect Integration](https://flowrunner.ai/integrations/astara-connect.md): Connect AI agents to Astara Connect, the connected-vehicle telematics platform from mobility group Astara. Agents track vehicles and trackers, create and delete points of interest, read telemetry, and set tracking schedules per fleet. - [Atera Integration](https://flowrunner.ai/integrations/atera.md): Connect AI agents to Atera, the RMM and PSA platform for MSPs and IT teams. Agents open and route support tickets, manage customers and contacts, and surface device monitoring alerts for escalation. - [AttachmentAV Integration](https://flowrunner.ai/integrations/attachmentav.md): Connect AI agents to attachmentAV, a malware scanning service for files and attachments. Agents scan uploads for viruses, trojans, and ransomware before a flow moves them anywhere, and check remaining account quota. - [Attentive Integration](https://flowrunner.ai/integrations/attentive.md): Feed Attentive's SMS and email marketing engine from your flows. Agents subscribe and unsubscribe shoppers, stream purchase, add-to-cart, and custom events to trigger journeys, and enrich profiles with custom attributes for segmentation. - [Attio Integration](https://flowrunner.ai/integrations/attio.md): Connect AI agents to Attio, the data-driven CRM. Agents upsert and query records across People, Companies, and custom objects, manage lists and pipeline entries, and log notes, tasks, and comments against any record. - [Auchan Integration](https://flowrunner.ai/integrations/auchan.md): Auchan runs its marketplace on Mirakl, and this connector covers the seller side end to end. Agents accept orders inside the acceptance window, ship and refund them, push stock and price imports, answer customer messages, and reconcile payouts and accounting documents. - [Audiense Insights Integration](https://flowrunner.ai/integrations/audiense-insights.md): Connect AI agents to Audiense Insights, a social audience intelligence platform. Agents read audience reports and their status, pull segment insights, influencer lists, and baseline comparisons, and read the content those audiences engage with, all across a read-only surface. - [Aunoa Integration](https://flowrunner.ai/integrations/aunoa.md): Send approved WhatsApp and RCS message templates and pull dashboard reporting from Aunoa, the Spanish conversational AI platform. Agents deliver templated customer messages and read engagement results back. - [Aurora Solar Integration](https://flowrunner.ai/integrations/aurora-solar.md): Connect AI agents to Aurora Solar, the design and sales platform for residential and commercial solar. Agents create projects and design requests, generate proposals, apply pricing and financing options, and register the webhook subscriptions Aurora notifies. - [Autentique Integration](https://flowrunner.ai/integrations/autentique.md): Autentique is a Brazilian electronic signature platform that reaches signers by email, SMS, or WhatsApp. Agents upload a PDF, name who signs and how they prove identity, organize documents into folders, and track every signature to completion. - [Auth0 Integration](https://flowrunner.ai/integrations/auth0.md): Connect AI agents to your Auth0 tenant through the Management API. Agents provision, update, and remove users, assign roles, inspect connections and applications, and pull authentication logs for security review. - [Authvia Integration](https://flowrunner.ai/integrations/authvia.md): Authvia is a text-to-pay platform that collects payments through conversational messaging. Agents manage customers and their payment methods, launch business processes and blueprints, create transactions and bulk jobs, and act on webhook events in real time. - [AutoContent API Integration](https://flowrunner.ai/integrations/auto-content-api.md): Turn websites, YouTube videos, PDFs, or a bare topic into AI podcasts, videos, infographics, slide decks, and quizzes with the AutoContent API. Agents repurpose one source into several published formats in a single pass. - [Autocalls AI Integration](https://flowrunner.ai/integrations/autocalls-ai.md): Place outbound phone calls with an Autocalls.ai voice agent and run lead-calling campaigns end to end. Agents dial, qualify, and hand off to a person when the conversation needs judgment. - [Automation Anywhere Integration](https://flowrunner.ai/integrations/automation-anywhere.md): Drive Automation Anywhere's Control Room from FlowRunner. Agents deploy TaskBots to Bot Runners, monitor executions to completion, feed work items into WLM queues, and browse the bot repository, bridging enterprise RPA with the rest of your stack. - [Autovit Integration](https://flowrunner.ai/integrations/autovit.md): Connect AI agents to Autovit, the Romanian car classifieds marketplace. Agents publish and update dealer listings, apply promotions to boost visibility, and read the make, model, and body-type taxonomy behind each advert. - [Avalara AvaTax Integration](https://flowrunner.ai/integrations/avalara-avatax.md): Calculate jurisdiction-accurate sales and use tax with Avalara AvaTax. Agents compute tax on multi-line transactions, validate and normalize customer addresses, and look up companies and tax codes before invoices go out. - [AvatarTalk AI Integration](https://flowrunner.ai/integrations/avatartalk-ai.md): Generate talking-avatar videos from text in 17 languages and drop a live avatar agent into a LiveKit call with AvatarTalk. Agents produce personalized video responses without a studio or a recording session. - [Avaza Integration](https://flowrunner.ai/integrations/avaza.md): Avaza is an all-in-one suite for project management, resource scheduling, timesheets, expenses, quoting and invoicing. Agents create projects and tasks, log time and expenses, issue estimates and invoices, and start flows the moment Avaza records change. - [Avisa API Integration](https://flowrunner.ai/integrations/avisaapi.md): Connect AI agents to Avisa API, a Brazilian gateway that links a WhatsApp account to your automations. Agents send text, media, and interactive messages, manage groups and communities, validate numbers, and react to inbound messages so WhatsApp conversations run inside governed workflows. - [Avochato Integration](https://flowrunner.ai/integrations/avochato.md): Connect AI agents to Avochato, a business texting inbox for SMS, MMS, and live chat. Agents send and schedule messages, manage contacts and tags, assign and update tickets, and publish broadcasts so shared inboxes stay responsive without manual triage. - [Award Force Integration](https://flowrunner.ai/integrations/award-force.md): Connect AI agents to Award Force, a platform for running awards, grants, scholarships, and competitions. Agents register entrants and entries, create categories and chapters, assign judges, and read the rounds, score sets, and leaderboards judging produces. - [AWeber Integration](https://flowrunner.ai/integrations/aweber.md): Grow AWeber email lists with AI agents. Agents look up accounts and lists, add and retrieve subscribers, and create broadcast messages so signups from any system land on the right list. - [awork Integration](https://flowrunner.ai/integrations/awork.md): awork is project management for agencies and teams, built around projects, tasks and time tracking with a light CRM attached. Agents open projects from templates, create and complete tasks, book time, and start flows in real time when work is created, moved or finished. - [AWS Certificate Manager Integration](https://flowrunner.ai/integrations/aws-acm.md): Connect AI agents to AWS Certificate Manager. Agents request and describe SSL/TLS certificates, retrieve DNS validation records, export PEM bodies, tag and renew certificates, and inventory posture across a region. - [AWS Bedrock Integration](https://flowrunner.ai/integrations/aws-bedrock.md): Run inference against Amazon Bedrock foundation models from Anthropic, Amazon, Meta, Mistral, Cohere, and Stability AI through one connector, and discover which models your account can call. - [AWS Cognito Integration](https://flowrunner.ai/integrations/aws-cognito.md): Connect AI agents to Amazon Cognito User Pools. Agents provision and de-provision users, manage passwords and attributes, run role-based group membership, and inspect user pools and app clients. - [AWS Comprehend Integration](https://flowrunner.ai/integrations/aws-comprehend.md): Run Amazon Comprehend natural-language processing on any text in a flow: sentiment, entities, key phrases, dominant language, and syntax, with native AWS Signature V4 signing. - [AWS Elastic Load Balancing Integration](https://flowrunner.ai/integrations/aws-elb.md): Manage and monitor Amazon Elastic Load Balancing (ELBv2): Application, Network, and Gateway Load Balancers, target groups, listeners, rules, and tags, with target-health monitoring. Authenticates to AWS with hand-rolled Signature Version 4. - [AWS IAM Integration](https://flowrunner.ai/integrations/aws-iam.md): Connect AI agents to AWS Identity and Access Management. Agents provision users and access keys, attach and detach managed policies, manage group membership, rotate credentials, and audit account posture. - [AWS KMS Integration](https://flowrunner.ai/integrations/aws-kms.md): Encrypt, decrypt, and sign data with AWS Key Management Service. Agents protect small payloads, generate data keys for envelope encryption, produce digital signatures with asymmetric keys, and list, describe, and create KMS keys, all via native SigV4 signing. - [AWS Marketplace Entitlements Integration](https://flowrunner.ai/integrations/aws-marketplace-entitlements.md): AWS Marketplace Entitlements covers the seller-side APIs a SaaS or container product needs at runtime. Workflows resolve a registration token to a customer, check what they are entitled to, and meter usage for billing. - [AWS End User Messaging SMS Integration](https://flowrunner.ai/integrations/aws-mms.md): Connect AI agents to AWS End User Messaging SMS, the AWS service formerly branded Amazon Pinpoint SMS. Agents send SMS, MMS, and voice messages and manage phone numbers, sender IDs, opt-out lists, and configuration sets so messaging infrastructure stays under programmatic control. - [Amazon Redshift Integration](https://flowrunner.ai/integrations/aws-redshift.md): Run SQL against an Amazon Redshift cluster or serverless workgroup through the Data API, no persistent connection required. Agents execute statements and batches asynchronously, poll for completion, fetch typed result rows, and explore databases, schemas, and table columns. - [AWS Rekognition Integration](https://flowrunner.ai/integrations/aws-rekognition.md): Analyze images with Amazon Rekognition: detect objects, text, faces, unsafe content, celebrities, and protective equipment, and manage face collections for search and comparison. - [AWS Textract Integration](https://flowrunner.ai/integrations/aws-textract.md): Extract text, forms, tables, and query answers from documents with Amazon Textract. Runs synchronous OCR for single pages and asynchronous jobs for multi-page PDFs in S3. - [AWS Transcribe Integration](https://flowrunner.ai/integrations/aws-transcribe.md): Convert speech in audio and video into text with Amazon Transcribe. Start and manage asynchronous S3-based transcription jobs and custom vocabularies from inside a flow. - [Axelor Integration](https://flowrunner.ai/integrations/axelor.md): Connect AI agents to an Axelor Open Suite instance, the open source ERP and CRM platform. Agents read and write records across sales, invoicing, projects, and partners, invoke Axelor actions, and inspect model metadata at runtime. - [Axiom Integration](https://flowrunner.ai/integrations/axiom.md): Axiom is an event and log analytics platform. Agents ingest events, run APL queries, manage datasets, monitors, notifiers, and views, and drop annotations on deploys so incidents line up with changes. - [Axonaut Integration](https://flowrunner.ai/integrations/axonaut.md): Connect AI agents to Axonaut, a French CRM and ERP for small businesses. Agents manage companies and contacts, open opportunities, raise quotations and invoices as independent documents, and read the expenses and finance records behind an account. - [Azure AI Foundry Integration](https://flowrunner.ai/integrations/azure-ai-foundry.md): Run inference against models deployed in Azure AI Foundry. Agents generate chat completions and structured output, create text and image embeddings for RAG pipelines, and inspect deployments through one unified endpoint across OpenAI, Mistral, Llama, and other providers. - [Azure AI Search Integration](https://flowrunner.ai/integrations/azure-ai-search.md): Managed cloud search and retrieval for agent memory. Manage indexes, ingest documents, and run keyword, vector, and hybrid search to ground RAG pipelines in your own content. - [Azure Blob Storage Integration](https://flowrunner.ai/integrations/azure-blob-storage.md): Manage Azure Blob Storage containers and blobs from FlowRunner agents: list, create, upload, download, copy, snapshot, and manage metadata, with hand-rolled Shared Key request signing. - [Azure Cosmos DB Integration](https://flowrunner.ai/integrations/azure-cosmos-db.md): Connect AI agents to Azure Cosmos DB through the Core (SQL) API. Agents manage databases and containers and create, read, query, replace, upsert, and delete documents. - [Azure DevOps Integration](https://flowrunner.ai/integrations/azure-devops.md): Connect AI agents to Azure DevOps. Agents create and update work items, run WIQL queries, manage Git repositories and pull requests, queue builds and pipeline runs, and track sprint iterations across a single organization. - [Azure OpenAI Integration](https://flowrunner.ai/integrations/azure-openai.md): Run OpenAI models hosted in your own Azure tenant: chat and reasoning, embeddings, DALL-E and gpt-image-1 image generation, and Whisper and TTS audio, all through Azure deployments. - [Azure Service Bus Integration](https://flowrunner.ai/integrations/azure-service-bus.md): Publish and consume messages on Azure Service Bus. Agents send to queues and topics, receive reliably with peek-lock, complete or abandon deliveries, read topic subscriptions, and provision queues and topics in the namespace. - [Azure Table Storage Integration](https://flowrunner.ai/integrations/azure-table-storage.md): Connect AI agents to Azure Table Storage, the NoSQL key/attribute store in Azure Storage. Agents manage tables and entities keyed by PartitionKey and RowKey with OData filter queries. - [B2Cor CRM Integration](https://flowrunner.ai/integrations/b2cor-crm.md): Connect AI agents to B2Cor, a Brazilian CRM built for insurance brokers. Agents capture leads, move them through sales funnels, and read the lead origins, task types, and users those records depend on. - [Backendless Integration](https://flowrunner.ai/integrations/backendless.md): Full Backendless platform integration with 20 triggers and 13 actions. Agents react to database events, file operations, user registrations, push notifications, and timers. - [BambooHR Integration](https://flowrunner.ai/integrations/bamboohr.md): Automate the BambooHR employee lifecycle from FlowRunner agents: profiles, time off, time tracking, recruiting, training, goals, reports, files, and change events over OAuth2. - [Bambuser Integration](https://flowrunner.ai/integrations/bambuser.md): Bambuser powers live video shopping with hosted and recorded broadcasts. Agents list and update broadcasts, create clips, tag archives, fetch download links, sign player URLs, and pick up new broadcasts on a polling schedule. - [Bannerbear Integration](https://flowrunner.ai/integrations/bannerbear.md): Generate branded images, videos, collections, animated GIFs, and website screenshots from Bannerbear templates. Renders run asynchronously against the Bannerbear API v2. - [Bannerbite Integration](https://flowrunner.ai/integrations/bannerbite.md): Bannerbite renders video and image variants from After Effects templates. Agents pick a project and bite, substitute text, colors, images, and footage by scene name, and kick off renders for personalized ads and social creative at scale. - [Base64.ai Integration](https://flowrunner.ai/integrations/base64-ai.md): Extract structured data from documents with Base64.ai. Agents scan IDs, invoices, receipts, and forms into labeled fields with confidence scores, detect and match faces and signatures for identity verification, and redact sensitive fields before storage. - [Basecamp Integration](https://flowrunner.ai/integrations/basecamp.md): Basecamp is 37signals' project management and team communication tool. Agents post messages, create and update to-dos, manage card tables, docs, files, schedules and Campfire chat, and start flows from Basecamp events across every project. - [Basecamp Integration](https://flowrunner.ai/integrations/basecamp3.md): Connect AI agents to Basecamp. Agents create and update projects, manage to-do lists and to-dos, post messages, comments, and Campfire lines, add schedule entries, and maintain documents so project chatter turns into tracked work. - [BaseLinker Integration](https://flowrunner.ai/integrations/baselinker.md): BaseLinker, now branded Base, is the Polish multichannel order management platform that pulls orders in from marketplaces and stores. Agents move orders through statuses, issue invoices and receipts, book courier shipments and labels, and keep catalog stock and prices in step across warehouses. - [Baserow Integration](https://flowrunner.ai/integrations/baserow.md): Connect AI agents to Baserow, the open-source no-code database. Agents read and write rows and manage databases, tables, and fields on Baserow Cloud or a self-hosted instance. - [Basin Integration](https://flowrunner.ai/integrations/basin.md): Basin is a no-code form backend: point an HTML form at a Basin endpoint and it handles storage, spam filtering and notifications. Agents read and manage submissions, forms and projects, submit entries, and react to every new submission through a signed webhook. - [Uniqode Integration](https://flowrunner.ai/integrations/beaconstac.md): Create, style, track, and retire QR codes with the Uniqode QR Code API. Agents mint codes for campaigns and assets, then read scan data to see which placements actually worked. - [Beamer Integration](https://flowrunner.ai/integrations/beamer.md): Publish product changelog posts, run an ideas board, and collect NPS with Beamer, covering posts, comments, reactions, and segments. Agents turn shipped work into announcements and route feedback to the right team. - [Beatoven.ai Integration](https://flowrunner.ai/integrations/beatoven-ai.md): Generate royalty-free, commercially licensed background music from a text prompt with Beatoven.ai. Agents score videos, podcasts, and product demos without a licensing negotiation. - [Bcon Integration](https://flowrunner.ai/integrations/becon.md): Bcon is a non-custodial crypto payment gateway that settles directly into your own wallet. Agents create tracked payment addresses, check balances and transaction history, list stores and supported tokens, and convert token prices into fiat. - [Beds24 Integration](https://flowrunner.ai/integrations/beds24.md): Connect AI agents to Beds24, a vacation rental channel manager and property management system. Agents create and modify bookings, push availability and rates to connected channels, and reconcile property and account records. - [beehiiv Integration](https://flowrunner.ai/integrations/beehiiv.md): Grow and manage a beehiiv newsletter from your flows. Agents create and update subscribers with UTM attribution, adjust tiers and custom fields, enroll readers into automation journeys, and pull publications, posts, and segments for reporting. - [Beeminder Integration](https://flowrunner.ai/integrations/beeminder.md): Connect over a personal auth token to manage Beeminder goals and datapoints, refresh goal graphs, read your profile, and issue charges through the Beeminder API v1. - [BeLazy Integration](https://flowrunner.ai/integrations/belazy.md): Connect AI agents to BeLazy, integration middleware for the translation industry supply chain. Agents watch vendor portals for new opportunities, accept or dismiss them, add deliverables to projects, and stage files in a workspace. - [Best Buy Integration](https://flowrunner.ai/integrations/best-buy.md): The Best Buy Developer APIs are a read-only window onto Best Buy's retail catalog, store network and in-store availability. Agents watch SKUs and categories for price drops, check availability near a store, pull recommendations and open box offers, and keep a local product feed in sync. - [BetterContact Integration](https://flowrunner.ai/integrations/bettercontact.md): Run waterfall B2B contact enrichment and lead search through BetterContact, chaining multiple data providers until a record resolves. Agents fill missing emails and phone numbers before a sequence starts. - [Bettermode Integration](https://flowrunner.ai/integrations/bettermode.md): Manage a Bettermode community through its GraphQL API. Agents publish and edit posts, add comments and reactions, read members and spaces, manage tags, and reach any uncovered part of the schema with a generic GraphQL action. - [Betty Blocks Integration](https://flowrunner.ai/integrations/betty-blocks.md): Betty Blocks is a low-code application platform, and its data API exposes each application's records over GraphQL. Workflows list and read records, describe models, and run raw queries against the app's schema. - [bexio Integration](https://flowrunner.ai/integrations/bexio.md): bexio is Swiss SMB business software covering the quote to order to invoice chain. Agents manage contacts and items, run projects with time tracking, record supplier bills and expenses, and post banking payments and double entry bookkeeping through a single OAuth2 connection. - [Big Cartel Integration](https://flowrunner.ai/integrations/big-cartel.md): Read a Big Cartel storefront from your flows. Agents pull the product catalog, categories, orders, and account settings to drive fulfillment, bookkeeping, and catalog syncs in downstream systems. - [BigCommerce Integration](https://flowrunner.ai/integrations/bigcommerce.md): Manage your BigCommerce store catalog, customers, orders, carts, and pricing from automated flows. Agents add products and variants, adjust inventory, manage customers and orders, and react in real time to order and inventory events. - [Bigin Integration](https://flowrunner.ai/integrations/bigin-by-zoho.md): Connect AI agents to Bigin by Zoho. Agents create and update contacts, companies, deals, and tasks, search records in any module, and add notes to keep small-business pipelines moving. - [BigMailer Integration](https://flowrunner.ai/integrations/bigmailer.md): BigMailer is a multi-brand email marketing platform that sends through your own Amazon SES or other delivery connection. Agents manage brands, contacts, lists, fields, and segments, build bulk, RSS, and transactional campaigns, and maintain suppression lists. - [BigMarker Integration](https://flowrunner.ai/integrations/bigmarker.md): Run BigMarker webinars from your flows. Agents create and update conferences, register attendees, keep registration data in sync, and browse channels to route events to the right webinar hub. - [BigML Integration](https://flowrunner.ai/integrations/bigml.md): Train and run machine learning models with BigML, covering sources, datasets, models, predictions, batch scoring, and evaluation. Agents score records against a trained model as a step inside a business workflow. - [Google BigQuery Integration](https://flowrunner.ai/integrations/bigquery.md): Connect AI agents to Google BigQuery. Agents run GoogleSQL with named parameters, stream rows into tables, read table data without query cost, and manage datasets and tables. - [BILL Integration](https://flowrunner.ai/integrations/billcom.md): Automate accounts payable and accounts receivable end-to-end. 7 real-time event triggers fire on bill creation, invoice updates, and payment events. Full CRUD for vendors, bills, customers, and invoices. - [Billit Integration](https://flowrunner.ai/integrations/billit.md): Billit is a Belgian invoicing platform and Peppol access point that reaches national e-invoicing networks across Europe and beyond. Agents issue and send invoices, receive incoming e-invoices, and route documents to the right country channel without leaving the workflow. - [Billomat Integration](https://flowrunner.ai/integrations/billomat.md): Billomat is German online invoicing and accounting. Agents manage clients, suppliers, and articles, issue invoices with items and payments, send estimates, create credit notes, and log incoming supplier invoices. - [Billplz Integration](https://flowrunner.ai/integrations/billplz.md): Billplz is a Malaysian payment gateway for bills, collections, and FPX bank transfers. Agents create and track bills, manage collections and payment methods, run V5 payment orders and auto-deduct agreements, and check receipt delivery. - [Billsby Integration](https://flowrunner.ai/integrations/billsby.md): Billsby is a subscription billing platform for recurring revenue businesses. Agents create customers and subscriptions, change plans and add-ons, issue one-time charges, refund invoices and credit notes, and manage products, plans, and feature tags. - [Binance Integration](https://flowrunner.ai/integrations/binance.md): Track Binance Spot markets and manage trades. Agents read live prices, 24-hour stats, candlesticks, and order books without a key, and with signed credentials check account balances and place, query, or cancel Spot orders. - [Bind ERP Integration](https://flowrunner.ai/integrations/bind-erp.md): Connect AI agents to Bind ERP, a Mexican cloud ERP for small and medium businesses. Agents raise sales orders and quotes, maintain the product catalog, issue purchase orders, and keep inventory and contact records aligned. - [BippyBox Integration](https://flowrunner.ai/integrations/bippybox.md): Connect AI agents to BippyBox, an internet-connected hardware box that plays a sound on demand. Agents trigger the box when a flow reaches a moment that deserves attention in the physical room, and read its account status. - [Birdeye Integration](https://flowrunner.ai/integrations/birdeye.md): Manage your Birdeye reputation pipeline. Agents pull reviews collected across Google, Facebook, and other connected sites, send review requests over email or SMS after a purchase or appointment, and keep the customer list in sync. - [BirdSend Integration](https://flowrunner.ai/integrations/birdsend.md): BirdSend is email marketing built for creators. Agents create and tag contacts, manage sequences and broadcasts, set automation rules and conversion goals, and connect signup forms to the rest of your workflow. - [Bitbucket Integration](https://flowrunner.ai/integrations/bitbucket.md): Manage Bitbucket Cloud repositories, issues, pull requests, source files, branches, and pipelines from your flows over the Bitbucket Cloud REST API 2.0. - [BitcoinApi.io Integration](https://flowrunner.ai/integrations/bitcoin-api.md): BitcoinApi.io writes records onto the Bitcoin blockchain as timestamped, tamper-proof logs. Agents write log entries and JSON records on chain, read them back with proof of existence, generate wallets, and check global chain statistics. - [Bitly Integration](https://flowrunner.ai/integrations/bitly.md): Connect over a Bitly access token to create and manage Bitlinks, pull click analytics, generate QR codes, and inspect groups and organizations through the Bitly API. - [Bitskout Integration](https://flowrunner.ai/integrations/bitskout.md): Extract data from documents and text with Bitskout, using your own configured plugins or prebuilt extractors for invoices, purchase orders, and bills. Agents turn inbound paperwork into typed fields ready for an accounting system. - [Bitwarden Integration](https://flowrunner.ai/integrations/bitwarden.md): Manage your Bitwarden organization with AI agents. Invite, update, and remove members, organize collections and groups, review enterprise policies, and pull the event log for audit trails through the Bitwarden Public API. - [Biyo POS Integration](https://flowrunner.ai/integrations/biyopos.md): Connect AI agents to Biyo POS, a cloud point of sale platform for retail and restaurants. Agents sync the catalog and modifiers across stores, update order status and customer, raise purchase orders to suppliers, complete stock movements, and pull inventory and sales reports. - [BizMachine Integration](https://flowrunner.ai/integrations/bizmachine.md): Pull Czech and Slovak B2B company data from BizMachine, covering registry records, contacts, financials, and ownership. Agents qualify accounts against real registry data rather than a scraped profile. - [Bland AI Integration](https://flowrunner.ai/integrations/bland.md): Dispatch autonomous AI phone agents with Bland AI. Agents place real outbound calls from a plain-language objective or a pre-built pathway, then collect transcripts, recordings, summaries, and structured answers, with stop controls for in-flight calls. - [Bleez Integration](https://flowrunner.ai/integrations/bleez.md): Connect AI agents to Bleez, a French cloud accounting and invoicing platform. Agents issue and track invoices, post journal entries, and upload documents into the dossier for processing. - [Blink Integration](https://flowrunner.ai/integrations/blink.md): Send feed posts, read forms, and provision users on Blink, the frontline employee experience app. Agents reach deskless staff where they already are and keep the roster in step with the HR system of record. - [BlockSurvey Integration](https://flowrunner.ai/integrations/blocksurvey.md): BlockSurvey is a privacy-first survey and form platform that encrypts responses in the respondent's browser. Agents manage surveys and encrypted audience contacts, and keep contact lists in step with the rest of your stack. - [Blogger Integration](https://flowrunner.ai/integrations/blogger.md): Publish to Blogger with AI agents. Agents create, update, publish, and revert posts, manage pages, moderate comments including spam handling, and read page view stats across every blog on the account. - [Blooio Integration](https://flowrunner.ai/integrations/blooio.md): Connect AI agents to Blooio, a business messaging platform that puts iMessage, SMS, RCS, and WhatsApp behind one API. Agents send messages across channels with automatic fallback, manage contacts and group chats, look up phone numbers, and react to inbound events so conversations run on the channel each customer actually uses. - [Bloom Growth Integration](https://flowrunner.ai/integrations/bloom-growth.md): Bloom Growth, formerly Traction Tools, runs the meeting and accountability rhythm of teams on the EOS / Traction methodology. Agents manage rocks, to-dos, issues, scorecards and meeting agendas so weekly Level 10 meetings reflect what actually happened. - [Blotato Integration](https://flowrunner.ai/integrations/blotato.md): Connect AI agents to Blotato, an AI social content creation and cross-posting tool. Agents host media from a URL, publish a post to one connected account per call, and read billing and account status before a campaign runs. - [Blue Integration](https://flowrunner.ai/integrations/blue.md): Blue is project and work management built around records on workspaces, with custom fields and views per project. Agents create and update records, manage projects, tags and assignees, and keep Blue in step with the systems where work originates. - [Bluebarry Integration](https://flowrunner.ai/integrations/bluebarry.md): Bluebarry powers product recommendation quizzes, onsite search, landing pages and popups for ecommerce stores. Agents read the product catalog and quiz results, act on customer profiles and conversion events, manage campaigns and popups, and subscribe to webhooks. - [Bluesky Integration](https://flowrunner.ai/integrations/bluesky.md): Publish and manage content on Bluesky over the AT Protocol. Agents create posts with automatic link, mention, and hashtag facets, reply, quote, repost, and like, search feeds and threads, and maintain the social graph with follows and mutes. - [Bluestone PIM Integration](https://flowrunner.ai/integrations/bluestone-api-public.md): Bluestone PIM is a Norwegian product information management platform. Agents create and enrich products, maintain the catalog tree and attribute definitions, manage the Media Bank, score data completeness, and publish approved content to downstream channels. - [Boldem Integration](https://flowrunner.ai/integrations/boldem.md): Boldem is a Czech customer data and marketing automation platform. Agents manage contacts, mailing lists, and consent, send mass and transactional email and SMS, run automations, and sync carts, orders, and coupons for ecommerce campaigns. - [Bolna Integration](https://flowrunner.ai/integrations/bolna.md): Place outbound calls from a Bolna conversational voice agent and read the transcript and outcome back. Agents run phone conversations at scale and escalate to a person when the caller needs one. - [BoloForms Integration](https://flowrunner.ai/integrations/boloforms.md): BoloForms, through its BoloSign product, sends templates out for signature and collects the responses. Agents fill a template's placeholders from a record, send it to the right signers with roles assigned, and read back who has responded. - [Bolt IoT Integration](https://flowrunner.ai/integrations/bolt-iot.md): Connect AI agents to the Bolt IoT cloud platform for connected hardware. Agents read analog and digital GPIO values, write control signals to pins, exchange serial data, and check device online status. - [Bolta Integration](https://flowrunner.ai/integrations/bolta.md): Bolta is the Korean electronic tax invoice service that files with the National Tax Service on your behalf. Agents issue tax invoices, generate the correcting invoices Korean law requires, and handle reverse invoices initiated by the buyer. - [Bonjoro Integration](https://flowrunner.ai/integrations/bonjoro.md): Queue personal video messages for your team to record and send with Bonjoro. Agents create a greet task when a signup, renewal, or milestone deserves a human face rather than another templated email. - [Bonusly Integration](https://flowrunner.ai/integrations/bonusly.md): Connect AI agents to Bonusly, the employee recognition and rewards platform. Agents give point-based bonuses when milestones land in other systems, create and update users from onboarding flows, and report on redemptions and the reward catalog. - [Bookafy Integration](https://flowrunner.ai/integrations/bookafy.md): Connect AI agents to Bookafy, an online appointment scheduling platform. Agents create and update appointments, look up available time slots, and manage customers and staff so booking changes happen without back-and-forth email. - [BookFunnel Integration](https://flowrunner.ai/integrations/bookfunnel.md): Connect AI agents to BookFunnel, which delivers ebooks and audiobooks to readers for authors and publishers. Agents queue download links to buyers and read the books and account records behind each send, though no action confirms a link was received. - [Bookoly Integration](https://flowrunner.ai/integrations/bookoly.md): Connect AI agents to Bookoly, an automated video creation platform. Agents generate speech from a script, assemble video and sound, and produce transcripts and subtitles as a single render pipeline. - [Boomerangme Integration](https://flowrunner.ai/integrations/boomerangme.md): Issue and manage digital wallet loyalty cards with Boomerangme, covering customers, Apple and Google Wallet passes, stamps, and points. Agents enrol members and push balance updates straight to the phone. - [BoondManager Integration](https://flowrunner.ai/integrations/boondmanager.md): Connect AI agents to BoondManager, a French ERP for consulting, staffing, and IT services firms. Agents create candidates, resources, companies, and contacts, open opportunities and positionings, and read the projects, deliveries, orders, and invoices each engagement produces. - [Booqable Integration](https://flowrunner.ai/integrations/booqable.md): Connect AI agents to Booqable, rental business management software. Agents create and reschedule rental orders, check equipment availability, move orders through pickup and return, and generate the paperwork each stage needs. - [Botcake Integration](https://flowrunner.ai/integrations/botcake.md): Automate the Botcake Messenger marketing platform, reading customers and custom fields, sending flows and ad-hoc content, and managing keyword rules. Agents run Meta messaging campaigns and react to subscriber behavior. - [BotDistrikt Integration](https://flowrunner.ai/integrations/botdistrikt.md): Read and reply to BotDistrikt bot users across WhatsApp, SMS, and messaging apps, manage user attributes, and schedule sends. Agents handle routine conversation and hand the thread to a person when it stops being routine. - [BotPenguin Integration](https://flowrunner.ai/integrations/botpenguin.md): Search and import BotPenguin inbox contacts, read bot chats, update captured leads, and send WhatsApp Cloud messages. Agents keep chatbot-captured leads flowing into the systems that act on them. - [Botpress Integration](https://flowrunner.ai/integrations/botpress.md): Connect AI agents to Botpress, the chatbot and agent platform. Agents converse with deployed bots through the Chat API, manage runtime conversations, users, and events, and read and write structured data in Botpress Tables and Files. - [Botsheets Integration](https://flowrunner.ai/integrations/botsheets.md): Talk to Botsheets chatbots from a flow. Botsheets bots answer from a Google Sheet and write collected data back into it, so agents can drive a sheet-backed conversation without a separate database. - [Botsify Integration](https://flowrunner.ai/integrations/botsify.md): Drive Botsify chatbots from a flow across web, Messenger, WhatsApp, Telegram, Instagram, and SMS. Agents read conversations, send messages and stories, and manage the users behind them. - [BotStar Integration](https://flowrunner.ai/integrations/botstar.md): Manage BotStar chatbots from a flow, publishing drafts, editing bot and user attributes, and maintaining CMS content collections. Agents keep conversational content current and send Facebook Messenger messages on cue. - [Bouncer Integration](https://flowrunner.ai/integrations/bouncer.md): Verify email addresses and clean lists with Bouncer, validating single addresses in real time and large lists in batch. Agents screen addresses before a send so bounce rates never decide your sender reputation for you. - [Box Integration](https://flowrunner.ai/integrations/box.md): Connect AI agents to Box cloud storage. Agents manage the full file and folder lifecycle, control sharing and collaborators, apply metadata, and react to item changes in real time. - [Braintree Integration](https://flowrunner.ai/integrations/braintree.md): Process payments through Braintree, the PayPal-owned gateway. Agents charge vaulted payment methods or client nonces, capture and void authorizations, issue full or partial refunds, and search transactions for reconciliation via the GraphQL API. - [Braintrust Integration](https://flowrunner.ai/integrations/braintrust.md): Run LLM evaluations and record experiment results with Braintrust, logging production traces alongside them. Agents score model output against a dataset before a prompt change reaches customers. - [BranchIS Integration](https://flowrunner.ai/integrations/branchis.md): Connect AI agents to BranchIS, AI project scheduling and resource management built on the APUtime engine. Agents open projects and apply templates, raise tickets with an owner, track time, and read workspace insights and cash flow. - [Brandfetch Integration](https://flowrunner.ai/integrations/brandfetch.md): Look up brand assets and company firmographics by domain, brand ID, or name. Agents pull logos, color palettes, fonts, links, and company data to enrich records or auto-brand generated documents. - [Brasil API Integration](https://flowrunner.ai/integrations/brasil-api.md): Query Brazilian public data through Brasil API, a free and keyless service covering postal codes, company registry records, banks, and more. Agents validate and enrich Brazilian records without a vendor contract. - [Braze Integration](https://flowrunner.ai/integrations/braze.md): Connect AI agents to Braze, the cross-channel customer engagement platform. Agents track user attributes, events, and purchases, fire API-triggered campaigns and Canvases, send messages across email, push, and SMS, and manage subscription group state. - [Breeze Integration](https://flowrunner.ai/integrations/breeze.md): Breeze is project management built around boards of cards with time tracking built in. Agents create and move cards, assign work, add comments, log time and read board status so project boards reflect what is happening elsewhere. - [Brevo Integration](https://flowrunner.ai/integrations/brevo.md): All-in-one marketing and CRM platform. Agents send transactional email and SMS, manage contacts and lists, and track deals, companies, tasks, and notes across the sales pipeline. - [Brex Integration](https://flowrunner.ai/integrations/brex.md): Connect AI agents to Brex corporate cards and spend management. Agents monitor card and cash transactions, issue virtual cards with spend limits, check account balances, and pull expenses into approval and reconciliation workflows. - [Bright Data Integration](https://flowrunner.ai/integrations/brightdata.md): Collect web data at scale through Bright Data. Agents trigger Web Scraper dataset collections and download result snapshots, fetch bot-protected pages through the Web Unlocker, and inspect the proxy and unlocker zones on the account. - [Brightflag Integration](https://flowrunner.ai/integrations/brightflag.md): Connect AI agents to Brightflag, an AI legal spend management platform. Agents open matters and budgets, create and assign allocations, update invoice payment status, and read vendor invoices and accrual configuration. - [Brosix Integration](https://flowrunner.ai/integrations/brosix.md): Send notifications into Brosix team chat through the Notifications API. Agents deliver alerts and workflow updates to the channel a team already watches. - [Browse AI Integration](https://flowrunner.ai/integrations/browse-ai.md): Run no-code web scraping with Browse AI. Agents discover trained robots, run them against target URLs one at a time or in bulk, poll tasks for extracted lists and texts, and set up monitors that watch pages for changes. - [BrowserAct Integration](https://flowrunner.ai/integrations/browser-act.md): BrowserAct is an AI browser automation and scraping platform. Workflows run the automations you built in the BrowserAct dashboard or the vendor's templates, follow each task to its result, and pause, resume, or cancel runs. - [Browserflow Integration](https://flowrunner.ai/integrations/browserflow-io.md): Browserflow automates browser tasks, most often on LinkedIn. Workflows start a flow you built in Browserflow and pass values into it, with results arriving through the flow's own webhook. - [Bubble Integration](https://flowrunner.ai/integrations/bubble.md): Connect FlowRunner to a Bubble app Data API and Workflow API using a bearer API token. Read, search, create, modify, replace, delete, and bulk-import database records, and trigger backend workflows against the Live or Development branch. - [BugHerd Integration](https://flowrunner.ai/integrations/bugherd.md): BugHerd is a visual website feedback and bug tracker. Agents create and update tasks, move them across columns, add comments and attachments, manage projects and members, and react to new tasks and comments in real time. - [BulkGate Integration](https://flowrunner.ai/integrations/bulkgate.md): Connect AI agents to BulkGate, a Czech multichannel messaging platform covering SMS, Viber, RCS, and WhatsApp. Agents send transactional and promotional messages, verify one-time passwords, and manage contacts and blacklists so customer notifications reach every channel from one workflow. - [Bullet Integration](https://flowrunner.ai/integrations/bullet.md): Connect AI agents to Bullet, the platform that turns Notion pages into a membership website. Agents add, update, and remove members and manage their segment access so gated content stays in sync with your customer records. - [Bullhorn Integration](https://flowrunner.ai/integrations/bullhorn-api.md): Connect AI agents to Bullhorn, the staffing industry ATS and CRM. Agents search and retrieve candidates, create candidate records from lead sources, look up open job orders, and run ad-hoc queries against any Bullhorn entity over the REST session. - [Kudosity (Burst SMS) Integration](https://flowrunner.ai/integrations/burst-sms.md): Connect AI agents to Kudosity, the Australian business messaging platform formerly known as Burst SMS. Agents send and schedule SMS, manage contact lists, lease virtual numbers, and pull delivery and reply reports so campaigns and transactional texts stay measurable. - [BurstyAI Integration](https://flowrunner.ai/integrations/burstyai.md): Run BurstyAI no-code SEO, content, and outreach workflows from a flow through the Workflow API. Agents kick off a content or backlink task and collect the finished result without leaving FlowRunner. - [Businesslogic Integration](https://flowrunner.ai/integrations/businesslogic.md): Businesslogic publishes an Excel workbook as a JSON REST API. Workflows read the model's input and output schema, validate their inputs against it, and execute the spreadsheet as a callable calculation. - [Businessmap Integration](https://flowrunner.ai/integrations/businessmap.md): Businessmap, formerly Kanbanize, is portfolio kanban with workspaces, boards, cards, time tracking and measurable outcomes. Agents create and move cards, update custom fields, log time and roll up outcomes across boards and workspaces. - [Byteplant Integration](https://flowrunner.ai/integrations/byteplant.md): Validate postal addresses, email addresses, and phone numbers with Byteplant's three verification services in one connector. Agents standardize and correct contact data before it reaches a CRM or a shipping system. - [Caflou Integration](https://flowrunner.ai/integrations/caflou.md): Caflou is a Czech business management suite that runs CRM, projects, time, invoicing and cashflow in one place. Agents manage companies and contacts, projects, milestones and tasks, log time, issue invoices and record payments and expenses without leaving the flow. - [Cal.com Integration](https://flowrunner.ai/integrations/cal-com.md): Manage Cal.com bookings, event types, availability slots, and schedules, confirm or decline bookings that need approval, and react in real time when Cal.com events fire. - [CalendarLink Integration](https://flowrunner.ai/integrations/calendarlink.md): Connect AI agents to CalendarLink, a platform for add-to-calendar links, RSVPs, and subscription calendars. Agents create events, list registrations, manage collections, and look up contacts so event promotion and RSVP tracking run without manual upkeep. - [Calendly Integration](https://flowrunner.ai/integrations/calendly.md): Automate scheduling workflows end-to-end. 5 Calendly triggers fire on bookings, cancellations, no-shows, and form submissions. Agents create one-off meetings and personalized scheduling links. - [Calley Integration](https://flowrunner.ai/integrations/calley.md): Feed and monitor Calley auto-dialer campaigns, pushing contacts into calling lists individually or in bulk. Agents build the day's call list from live pipeline data and read call outcomes back. - [CallHippo Integration](https://flowrunner.ai/integrations/callhippo.md): Manage users, contacts, call logs, and dialler campaigns on the CallHippo cloud phone system. Agents provision seats, log call activity against the right record, and keep outbound campaigns supplied with contacts. - [Callin Integration](https://flowrunner.ai/integrations/callin.md): Create and configure AI voice agents on Callin, setting prompt, model, voice, language, and call direction. Agents retrieve call conversations and transcripts and act on what was actually said. - [Callingly Integration](https://flowrunner.ai/integrations/callingly.md): Turn inbound leads into instant phone calls with Callingly, triggering callbacks to new leads and sending personalized SMS. Agents connect a rep to a lead while intent is still live rather than the next business day. - [CallRail Integration](https://flowrunner.ai/integrations/callrail.md): Connect AI agents to CallRail call tracking and marketing attribution. Agents pull tracked calls, form submissions, and text conversations, tie each lead to the campaign and keyword that drove it, update tags, values, and lead status, and manage companies and tracking numbers. - [CallTrackingMetrics Integration](https://flowrunner.ai/integrations/calltrackingmetrics.md): Connect AI agents to CallTrackingMetrics, a call tracking and contact center platform. Agents read call records and recordings, attribute calls to tracking numbers and sources, record call sales, sync contacts, and read agent activity. - [CallX Integration](https://flowrunner.ai/integrations/callx.md): Start outbound AI phone calls through a CallX account and search the call log history. Agents place a call, read the outcome, and pass the transcript to whatever comes next in the flow. - [Cambrion Integration](https://flowrunner.ai/integrations/cambrion.md): Run a pipeline over media or text with Cambrion, the German AI document data platform, to produce a structured observation. Agents describe the extraction they want in plain English and receive typed results back. - [Campaign Cleaner Integration](https://flowrunner.ai/integrations/campaign-cleaner.md): Campaign Cleaner sanitizes and analyzes email campaign HTML before it goes out. Agents submit campaigns for cleanup and correction, read the analysis and PDF report, run inbox placement tests, and watch the credit balance. - [Campaign Monitor Integration](https://flowrunner.ai/integrations/campaign-monitor.md): Manage Campaign Monitor audiences from your flows. Agents sync signups into subscriber lists with custom fields and consent tracking, update and unsubscribe subscribers as records change elsewhere, provision new lists, and review sent and draft campaigns. - [Candu Integration](https://flowrunner.ai/integrations/candu.md): Connect AI agents to Candu, the no-code in-app UI content builder. Agents identify users and groups and track custom events so Candu can segment audiences and target the right in-app content, checklists, and announcements. - [Canny Integration](https://flowrunner.ai/integrations/canny.md): Manage user feedback in Canny from your flows. Agents create feature requests on any board, browse and search posts, add comments when linked tickets resolve, tally votes, and resolve users by email before acting on their behalf. - [Canva Integration](https://flowrunner.ai/integrations/canva.md): Automate Canva design workflows end to end. Agents create and list designs, export finished work to PDF, PNG, or PPTX with variants that wait for completion, upload assets to the media library, organize folders, and autofill brand templates with structured data. - [Canvas LMS Integration](https://flowrunner.ai/integrations/canvas-lms.md): Connect AI agents to Canvas LMS by Instructure. Agents provision courses and assignments, enroll students automatically, sync rosters and grades to reporting systems, look up users by name or email, and monitor submissions to drive follow-up. - [Canvasflare Integration](https://flowrunner.ai/integrations/canvasflare.md): Turn HTML and CSS into images through Canvasflare's headless Chrome rendering API, or capture a screenshot of a live URL. Agents produce social cards, receipts, and report visuals without running a browser. - [Capri AI Integration](https://flowrunner.ai/integrations/capri.md): Connect AI agents to Capri AI, an AI receptionist that answers customer questions and books appointments. Agents read booked appointments and caller context and pick up the work that follows the call. - [Capsule CRM Integration](https://flowrunner.ai/integrations/capsule-crm.md): Connect AI agents to Capsule CRM. Agents sync people and organizations from other systems, create and advance sales opportunities through pipelines, open and close cases, manage tasks, and read pipelines, milestones, tags, and custom fields. - [Captain Data Integration](https://flowrunner.ai/integrations/captain-data.md): Extract LinkedIn and Sales Navigator data through the Captain Data External API, resolving people and companies by name or URL. Agents enrich a prospect list and pull the signals that decide whether an account is worth working. - [CaptainBook Integration](https://flowrunner.ai/integrations/captainbook.md): Connect AI agents to CaptainBook, a booking system for tours and activities. Agents manage products and availability, cancel and refund bookings, and resend confirmations so tour operations keep moving without manual admin work. - [Captivate Integration](https://flowrunner.ai/integrations/captivate.md): Captivate is a podcast hosting platform with shows, episodes, media, RSS feeds, and download analytics. Agents create and schedule episodes, upload media and artwork, pull analytics by show or date range, and react when a new episode publishes. - [Carbone Integration](https://flowrunner.ai/integrations/carbone.md): Generate documents from templates with Carbone. Agents upload DOCX, XLSX, or HTML templates tagged with placeholders, merge JSON data into them, convert the result to PDF and other formats, and save finished files to FlowRunner file storage. - [Cargoboard Integration](https://flowrunner.ai/integrations/cargoboard.md): Connect AI agents to Cargoboard, a digital freight forwarder for groupage, LTL, and FTL shipments. Agents place quotations, book and cancel orders, print shipment labels, and pull tracking and invoice data so freight moves without manual portal work. - [Cargus Integration](https://flowrunner.ai/integrations/cargus-romania.md): Connect AI agents to Cargus, a Romanian courier and parcel delivery service. Agents calculate shipping prices, create and print waybills, track shipments, and reconcile cash on delivery and invoices so Romanian parcel operations run without manual data entry. - [Carix Integration](https://flowrunner.ai/integrations/carix.md): Connect AI agents to Carix, a dealer vehicle management system from German vendor HT Automotive. Agents read the dealer's vehicle inventory and resolve the dealer-specific field set at runtime, feeding listings and stock reports downstream. The API is pull-only, with no write operations. - [Carrefour Integration](https://flowrunner.ai/integrations/carrefour.md): Carrefour's marketplace runs on Mirakl, and this connector covers the seller surface for each country instance. Agents accept and ship orders, handle returns and refunds, push offer, price and stock imports, message customers, and pull seller accounting documents. - [CarsXE Integration](https://flowrunner.ai/integrations/carsxe-api.md): CarsXE is a vehicle data API covering VIN decoding, license plate lookups, market values, history reports, recalls, lien and theft records and vehicle imagery. Agents decode a VIN or plate, pull specs, value a vehicle and check recall history in one step. - [Caspio Integration](https://flowrunner.ai/integrations/caspio.md): Read and write data in your Caspio online database. Agents list tables and views, query records with filters, and insert, update, or delete rows to keep Caspio-built applications in sync with the rest of your stack. - [Celoxis Integration](https://flowrunner.ai/integrations/celoxis.md): Celoxis is a project and portfolio management suite for planning, tracking and reporting on projects at scale. Agents create projects and tasks, post task updates, log time and expenses, manage users and read and write the account's custom apps. - [Centiment Integration](https://flowrunner.ai/integrations/centiment.md): Centiment pairs a survey builder with a profiled respondent panel spanning 100+ countries. Agents create and launch surveys, target audiences, pull responses and completion stats, and react when a new response lands. - [Cerebras Integration](https://flowrunner.ai/integrations/cerebras-ai.md): Run chat and text completions on Cerebras, the wafer-scale inference platform for open-weight models. Agents call the OpenAI-compatible API for high-throughput generation, use tool calling and structured outputs, and discover available models. - [Certifier Integration](https://flowrunner.ai/integrations/certifier.md): Connect AI agents to Certifier, a digital certificate and badge platform. Agents create credentials from templates, issue and send them to recipients, and revoke them when needed so credential delivery follows course completions automatically. - [Cflow Integration](https://flowrunner.ai/integrations/cflow.md): Cflow by Cavintek is a no-code approval workflow and business process automation platform. Agents submit and update records, move them through workflow stages, post comments, clarifications, and attachments, pull reports, and administer users, roles, and team access. - [Channels Integration](https://flowrunner.ai/integrations/channels.md): Manage calls, contacts, and do-not-call lists on the Channels phone system. Agents log call activity against the right record and keep suppression lists honored before any outbound dial. - [Channex Integration](https://flowrunner.ai/integrations/channex.md): Connect AI agents to Channex, a hotel channel manager that syncs availability and rates across booking sites. Agents push rate and availability updates, read bookings as they land, and keep room types and rate plans consistent across channels. - [Chargebee Integration](https://flowrunner.ai/integrations/chargebee.md): Connect AI agents to Chargebee subscription billing. Agents create customers and item-based subscriptions, issue and collect invoices, run the pause, resume, cancel, and reactivate lifecycle, sync the product catalog, and generate hosted checkout pages through the Chargebee API v2. - [ChargeOver Integration](https://flowrunner.ai/integrations/chargeover.md): ChargeOver is a recurring billing and invoicing platform. Agents manage customers and subscriptions, issue invoices and quotes, record transactions, and push usage for metered billing. - [Chartly Integration](https://flowrunner.ai/integrations/chartimagegenerator.md): Render chart images from data without a browser or charting library, using a full Chart.js configuration or a quick preset. Agents attach a real chart to an email, a Slack message, or a PDF report. - [ChartMogul Integration](https://flowrunner.ai/integrations/chartmogul.md): Pull subscription analytics from ChartMogul into your flows. Agents read MRR, ARR, churn, and LTV metrics, manage customers, plans, and invoices in Import API data sources, push subscription events, and alert when revenue metrics cross thresholds. - [Chat ON Desk Integration](https://flowrunner.ai/integrations/chat-on-desk.md): Send WhatsApp and SMS messages, verify phone numbers, and schedule reminders through a Chat ON Desk account. Agents run templated customer messaging and confirm a number is reachable before the first send. - [Chatbase Integration](https://flowrunner.ai/integrations/chatbase.md): Operate Chatbase AI chatbots from your flows. Agents send grounded chat messages, create and retrain chatbots from raw text, refresh training knowledge when documentation changes, and pull captured conversations and leads into CRM and reporting systems. - [ChatBot Integration](https://flowrunner.ai/integrations/chatbot.md): Drive ChatBot.com conversations and bot operations from your flows. Agents send messages to published stories, train bots on unmatched phrases and custom text, manage entities, segments, and webhooks, and pull chat archives and analytics reports. - [Chat Data Integration](https://flowrunner.ai/integrations/chatdata.md): Build, train, and operate AI chatbots with Chat Data, training on text, URLs, and documents. Agents keep a bot's knowledge current and react when a conversation crosses into territory a person should handle. - [Chatfuel Integration](https://flowrunner.ai/integrations/chatfuel.md): Re-engage Chatfuel subscribers from your flows. Agents push existing bot blocks and flows to individual users on Messenger, Instagram, and WhatsApp, deliver updates outside the 24-hour window with approved message tags, and sync subscriber attributes from other systems. - [ChatGuru Integration](https://flowrunner.ai/integrations/chatguru.md): Automate WhatsApp conversations in a ChatGuru account, registering chats and sending text and file messages. Agents keep chat metadata and custom fields aligned with the CRM behind them. - [ChatNode Integration](https://flowrunner.ai/integrations/chatnode.md): Send queries to a trained ChatNode agent and receive grounded answers with their source URLs and documents. Agents answer from your own corpus and cite where each answer came from. - [ChatPDF Integration](https://flowrunner.ai/integrations/chatpdf.md): Ask questions about PDF documents and get answers grounded in their contents, with optional page-level citations. Agents add a PDF once and query it as many times as the workflow needs. - [Chatra Integration](https://flowrunner.ai/integrations/chatra.md): Connect AI agents to Chatra, a live chat widget for websites. Agents read visitor conversations, push proactive messages when a flow detects intent, and keep client records attached to each thread. - [Chatsistant Integration](https://flowrunner.ai/integrations/chatsistant.md): Create and operate multi-agent AI chatbots with Chatsistant, training them on URLs, files, and question-and-answer pairs. Agents keep the knowledge base fresh and route conversations between specialized bots. - [Chatsonic Integration](https://flowrunner.ai/integrations/chatsonic.md): Send prompts to Chatsonic, Writesonic's assistant, and receive its reply with the supporting context. Agents draft and rewrite marketing copy inside a workflow that already has the customer record. - [Chatvolt AI Integration](https://flowrunner.ai/integrations/chatvolt-ai.md): Operate Chatvolt AI agents from FlowRunner, building vector datastores from files, web pages, and question-and-answer pairs. Agents query a grounded knowledge base and keep it current as source material changes. - [Chatwork Integration](https://flowrunner.ai/integrations/chatwork.md): Automate Chatwork, Japan's leading business chat platform, across messages, chats, members, tasks, and files. Agents post updates, assign chat tasks, and pull the thread context a decision needs. - [Checkbook.io Integration](https://flowrunner.ai/integrations/checkbook.md): Send and collect money through Checkbook.io without a bank portal: digital payments the recipient claims and routes themselves, paper checks printed and mailed, multi-party payments each party endorses, invoices, and recurring subscriptions. Agents can stage a payment for approval before any funds move, manage saved payees and every payment-account type, and pull attachments and printable checks down as PDFs. - [Checkr Integration](https://flowrunner.ai/integrations/checkr.md): Automate background checks with Checkr. Agents create candidates when hires advance in an ATS, send secure screening invitations, order reports directly when candidate data is on file, poll report status, and look up available screening packages. - [Checkvist Integration](https://flowrunner.ai/integrations/checkvist.md): Checkvist is the keyboard-driven outliner and task list. Agents create checklists, add and update items with tags, due dates, priorities, assignees, and repeats, attach notes, and import or export lists as text. - [CherryIN Integration](https://flowrunner.ai/integrations/cherryin.md): Reach CherryIN's multi-provider model catalog through one OpenAI-compatible connection. Agents switch models without rewriting the flow or adding another credential. - [Chiavistello Integration](https://flowrunner.ai/integrations/chiavistello.md): Connect AI agents to Chiavistello, an Italian self-check-in and smart-lock platform. Agents issue and revoke door access tied to a booking, request payment from guests through a hosted page, and issue the fiscal receipts Italian hospitality requires. - [Chroma Integration](https://flowrunner.ai/integrations/chroma.md): Open-source AI-native vector database for agent memory. Manage collections and records, and run nearest-neighbor similarity search against Chroma Cloud or a self-hosted server for RAG retrieval. - [ChytryStart Integration](https://flowrunner.ai/integrations/chytrystart.md): Turn a Czech name into the grammatical forms needed to address someone properly, including vocative case and gender detection. Agents personalize Czech-language messages without getting the grammar wrong. - [Cin7 Omni Integration](https://flowrunner.ai/integrations/cin7.md): Connect AI agents to Cin7 Omni inventory and order management. Agents read products, stock levels, sales and purchase orders, and branches, create sales orders from storefront events, and onboard customers and suppliers as contacts. - [Circle Integration](https://flowrunner.ai/integrations/circle-so.md): Run your Circle community from your flows. Agents create and update members from CRM or checkout events, cross-post announcements into spaces, apply member tags for segmentation, send invitations, publish events, and revoke access when subscriptions lapse. - [Circleback Integration](https://flowrunner.ai/integrations/circleback.md): Read and update meetings captured by Circleback across Zoom, Google Meet, Microsoft Teams, and Slack, covering notes, action items, transcripts, and tags. Agents turn a meeting into tracked work the moment it ends. - [CircleCI Integration](https://flowrunner.ai/integrations/circleci.md): Trigger and inspect CircleCI pipelines, workflows, and jobs, and manage project environment variables from your flows over the CircleCI API v2. - [Clarifai Integration](https://flowrunner.ai/integrations/clarifai.md): Run computer vision and AI predictions with Clarifai. Agents classify images, video, audio, and text against models and multi-step workflows, manage inputs stored in apps, search media by visual similarity or concept, and browse available models. - [Clay Integration](https://flowrunner.ai/integrations/clay.md): Push records into Clay for GTM enrichment. Agents post leads, signups, and CRM contacts as JSON rows to a table's webhook source so Clay's enrichment columns run against each record, with a generic passthrough for any other Clay endpoint. - [Mesh Integration](https://flowrunner.ai/integrations/clay-earth.md): Connect AI agents to Mesh (formerly Clay), a personal relationship manager that keeps track of who you owe a reply. Agents enrich contacts, organize them into groups, and surface the relationships that have gone quiet. - [Cleanvoice Integration](https://flowrunner.ai/integrations/cleanvoice.md): Cleanvoice AI is an automated audio and podcast editor that removes filler words, silences, stutters, and mouth sounds, then normalizes loudness. Agents submit audio from a URL, apply a saved editing template, and pick up the cleaned file, transcript, and chapter summaries when the edit completes. - [Clearbit Integration](https://flowrunner.ai/integrations/clearbit.md): Turn an email, domain, or IP address into a rich person and company profile: employment, social, firmographics, and technology stack. Agents enrich inbound signups, de-anonymize website visitors, and build prospect lists. - [Clearout Integration](https://flowrunner.ai/integrations/clearout.md): Clearout verifies, discovers, and enriches email addresses in real time. Agents run instant and bulk verification, validate names, find business emails, run reverse lookups by address, domain, or LinkedIn profile, and resolve MX and WHOIS records. - [Clepher Integration](https://flowrunner.ai/integrations/clever-messenger.md): Automate Clepher Messenger bots, looking up subscribers, dispatching flows and cards, and managing tags and custom fields. Agents keep Messenger audiences segmented and react to capture-tool submissions. - [CleverReach Integration](https://flowrunner.ai/integrations/cleverreach.md): Connect AI agents to CleverReach. Agents manage recipient groups, add, update, and remove receivers, and read mailings and signup forms so email lists stay current without manual imports. - [ClickFunnels Integration](https://flowrunner.ai/integrations/click-funnels.md): Connect AI agents to ClickFunnels Classic (1.0). Agents add contacts, browse contact attribute definitions, and inventory the funnels and funnel steps in your account to keep leads flowing into downstream systems. - [ClickFunnels 2.0 Integration](https://flowrunner.ai/integrations/click-funnels-2.md): Connect AI agents to ClickFunnels 2.0 workspaces. Agents create and update contacts, apply tags, and read products, orders, courses, and tags to sync funnel activity with your CRM and reporting stack. - [Click and Sign Integration](https://flowrunner.ai/integrations/clickandsign.md): Click&Sign is the electronic signature product of Lleida.net, a Spanish certified-communications provider. Agents start signature processes under a stored configuration, attach documents and signers, run bulk sends, and download the signed evidence. - [Clickatell Integration](https://flowrunner.ai/integrations/clickatell.md): Send SMS and WhatsApp messages worldwide through the Clickatell One API. Agents deliver one-time passwords, order confirmations, and bulk broadcasts, then track delivery with per-message status checks. - [ClickMeeting Integration](https://flowrunner.ai/integrations/clickmeeting.md): Run ClickMeeting webinars and online meetings from your flows. Agents create, update, and delete conference rooms, register attendees, and pull past session statistics for reporting. - [ClickSend Integration](https://flowrunner.ai/integrations/clicksend.md): Multi-channel communications including SMS, MMS, voice, fax, letter, and postcard. Agents reach contacts through 6 distinct channels. 19 actions cover contact management and multi-channel message delivery. - [Clicksign Integration](https://flowrunner.ai/integrations/clicksign.md): Clicksign is a Brazilian electronic signature platform built around the envelope. Agents assemble documents, signers, and identity requirements in a draft, activate it to send everything at once, and manage templates, folders, and webhook subscriptions. - [ClickUp Integration](https://flowrunner.ai/integrations/clickup.md): Automate ClickUp tasks, lists, folders, and time tracking across the workspace hierarchy. Agents triage new tasks, set priority and assignees, log progress, and pull a person in on the ambiguous calls. - [Cliengo Integration](https://flowrunner.ai/integrations/cliengo.md): Automate the Cliengo chatbot and CRM platform, managing contacts and conversations and replying across web and WhatsApp. Agents send approved WhatsApp templates and keep lead records current as conversations progress. - [Clientjoy Integration](https://flowrunner.ai/integrations/clientjoy.md): Connect AI agents to Clientjoy, an agency CRM covering leads through invoices. Agents qualify leads into customers, send proposals, raise invoices and line items, and read the forms a client submitted. - [Cliniko Integration](https://flowrunner.ai/integrations/cliniko.md): Connect AI agents to Cliniko, the practice management system for healthcare clinics. Agents manage patient records, book and reschedule appointments, check practitioner availability, and read invoices, with the regional shard detected from the key itself. - [Clio Manage Integration](https://flowrunner.ai/integrations/clio-manage.md): Connect AI agents to Clio Manage, the cloud legal practice management platform. Agents open matters, create contacts, log billable time entries, and monitor activities and tasks across the practice. - [Clipdrop Integration](https://flowrunner.ai/integrations/clipdrop.md): Edit and generate images with Clipdrop by Stability AI. Agents remove and replace backgrounds, clean up and upscale photos, erase text, create product photography, and generate images from text or sketches, saving every result to FlowRunner file storage. - [Clockify Integration](https://flowrunner.ai/integrations/clockify.md): Track time in Clockify from your flows: log entries, start and stop timers, keep projects and clients in sync, and generate summary reports. Agents track routine work and route billing totals to a person before they become an invoice. - [Clockodo Integration](https://flowrunner.ai/integrations/clockodo.md): Clockodo is the German time tracking and workforce management platform. Agents start and stop the clock, create and report on time entries, manage customers, projects, and services, and handle absences, target hours, overtime, and work time change requests. - [Close CRM Integration](https://flowrunner.ai/integrations/closecrm.md): Connect AI agents to Close CRM. Agents manage leads, contacts, opportunities, activities (calls, emails, SMS, meetings), tasks, sequences, bulk actions, and webhooks via OAuth2. - [Cloud BOT Integration](https://flowrunner.ai/integrations/cloud-bot.md): Run Cloud BOT browser-automation bots from a flow. Cloud BOT records browser steps and replays them in the cloud, so agents can reach systems that expose no API at all. - [CloudConvert Integration](https://flowrunner.ai/integrations/cloudconvert.md): Convert files between 200+ formats through CloudConvert. Agents convert documents, merge files to PDF, capture websites, optimize and archive files, extract metadata, and manage multi-step conversion jobs end to end. - [Cloudflare Integration](https://flowrunner.ai/integrations/cloudflare.md): Manage Cloudflare zones, DNS records, cache, security rulesets, and Workers KV storage from your flows over the Cloudflare API v4. - [Cloudinary Integration](https://flowrunner.ai/integrations/cloudinary.md): Manage a Cloudinary media library from your flows. Agents upload images, videos, and raw files with signed requests, tag, search, and organize assets, generate transformation and delivery URLs, and monitor account usage. - [Cloudmersive Integration](https://flowrunner.ai/integrations/cloudmersive.md): Use Cloudmersive's utility API suite for document conversion, validation, virus scanning, and data transformation. Agents handle the file and data chores a workflow hits between the systems that matter. - [Cloudonix Integration](https://flowrunner.ai/integrations/cloudonix.md): Manage calls, voice applications, numbers, subscribers, and trunks on Cloudonix. Agents provision telephony infrastructure and route calls as part of a larger operational workflow. - [Cloudpress Integration](https://flowrunner.ai/integrations/cloudpress.md): Connect AI agents to Cloudpress, which exports content from Google Docs and Notion to a connected CMS. Agents start exports, convert docs to HTML or Markdown, and monitor job status so publishing to WordPress, Webflow, or Contentful happens without copy and paste. - [CloudTalk Integration](https://flowrunner.ai/integrations/cloudtalk.md): Connect AI agents to CloudTalk, the cloud call center and business phone system. Agents place click-to-call calls, send SMS, manage contacts, and pull call history, agents, campaigns, and statistics for reporting and routing. - [Clover Integration](https://flowrunner.ai/integrations/clover-pos.md): Connect AI agents to Clover, the Fiserv point-of-sale platform. Agents read and create orders, inventory items, and customers, list payments, and pull merchant settings to automate retail and restaurant back-office work. - [Cloze Integration](https://flowrunner.ai/integrations/cloze.md): Connect AI agents to Cloze, a relationship-focused CRM that unifies contacts and every communication with them. Agents create people and companies, attach them to projects, and read the full timeline of messages behind a relationship. - [Shortcut (Clubhouse) Integration](https://flowrunner.ai/integrations/clubhouse.md): Connect AI agents to Shortcut (formerly Clubhouse), the project management platform for software teams. Agents search, create, and update stories, organize epics and iterations, and read workspace projects, workflows, and members. - [Club Planner Integration](https://flowrunner.ai/integrations/clubplanner.md): Connect AI agents to Club Planner, fitness club management software from Belgium. Agents add members and prospects, sell subscriptions and credits, book reservations and calendar items, check members in and out, and ring up point of sale items. - [CmonSite Integration](https://flowrunner.ai/integrations/cmonsite.md): Connect AI agents to CmonSite, the French website and online store builder. Agents create, update, list, and delete blog articles so site content stays fresh without manual publishing. - [CNPJa Integration](https://flowrunner.ai/integrations/cnpja.md): CNPJa is a Brazilian company data API over the CNPJ register, the Receita Federal, state tax registrations, Simples Nacional and SUFRAMA. Agents look up establishments and companies, search people, verify federal and state registrations and resolve postcodes. - [Coassemble Integration](https://flowrunner.ai/integrations/coassemble.md): Connect AI agents to Coassemble, an online employee training platform. Agents manage courses and collections, update learners and clients, and read tracking data so training progress reporting stays hands-off. - [Cockpit Integration](https://flowrunner.ai/integrations/cockpit.md): Connect AI agents to Cockpit CMS through the v2 Content API. Agents read and write content items, singletons, and assets, query with MongoDB-style filters, and stage or remove entries as part of an editorial pipeline. - [Coda Integration](https://flowrunner.ai/integrations/coda.md): Read and write structured Coda docs from your flows: rows, tables, columns, formulas, controls, buttons, and pages. Agents sync rows without duplicates and pull a person in before deleting rows others depend on. - [CodeGPT Integration](https://flowrunner.ai/integrations/codegpt.md): Build and run AI coding agents with CodeGPT, managing agents and loading documents into their knowledge base. Agents review, explain, and generate code grounded in your own repositories and docs. - [CodeQR Integration](https://flowrunner.ai/integrations/codeqr.md): Generate and restyle dynamic QR codes, repoint short links, and read scan analytics with CodeQR. Agents mint trackable codes per campaign and retarget them without reprinting anything. - [CodeREADr Integration](https://flowrunner.ai/integrations/codereadr.md): Connect AI agents to CodeREADr for barcode scanning, validation, and scan records. Agents manage scanning services and validation databases so a scan at the door or the dock resolves against live data. - [Cognito Forms Integration](https://flowrunner.ai/integrations/cognitoforms.md): Work with Cognito Forms submissions from your flows. Agents list forms, inspect field schemas, and create, update, filter, and delete entries with OData-style paging and queries. - [Cohere Integration](https://flowrunner.ai/integrations/cohere.md): Connect to Cohere's enterprise language AI: chat and reasoning with the Command family, embeddings, best-in-class reranking for retrieval, classification, and dataset management. - [CoinGecko Integration](https://flowrunner.ai/integrations/coingecko.md): Access live and historical cryptocurrency market data from the CoinGecko API v3: current prices, market charts, OHLC candles, trending coins, exchange and category data, and global market statistics. - [CoinMarketCap Integration](https://flowrunner.ai/integrations/coinmarketcap.md): Pull live cryptocurrency market data from the CoinMarketCap Pro API. Agents fetch price quotes, ranked listings, coin metadata, and global market metrics, and convert amounts between currencies at live rates. - [Colete Online Integration](https://flowrunner.ai/integrations/colete-online.md): Connect AI agents to Colete Online, a Romanian shipping aggregator that books all major local couriers through one API. Agents compare shipping prices, create orders, retrieve AWB labels, and check order status and account balance so multi-courier shipping runs from a single flow. - [800.com Integration](https://flowrunner.ai/integrations/com-800.md): Manage toll-free numbers, call tracking, messaging, and analytics on 800.com. Agents attribute inbound calls to the campaign that produced them and route the follow-up work. - [Comeet Integration](https://flowrunner.ai/integrations/comeet.md): Connect AI agents to Comeet (now Spark Hire Recruit), a collaborative recruiting applicant tracking system. Agents create positions, add candidates, and find duplicate profiles so the recruiting pipeline stays clean and current. - [CometAPI Integration](https://flowrunner.ai/integrations/cometapi.md): Reach CometAPI's multi-provider model catalog through one FlowRunner connection. Agents compare or fail over between models without a second credential or a second connector. - [ComPDF AI Integration](https://flowrunner.ai/integrations/comidp.md): ComPDF AI is the document-understanding side of ComPDF's Cloud API. Agents extract named values and table columns from invoices and forms, run OCR over scans, detect tables and stamps, and straighten photographed pages before filing them. - [CommCare Integration](https://flowrunner.ai/integrations/commcare.md): CommCare HQ is Dimagi's mobile data collection platform for frontline field programs that work offline. Agents read and write cases, pull form submissions and attachments, manage mobile workers, locations and lookup tables, and react in real time when a form is submitted or a case changes. - [CommerceHQ Integration](https://flowrunner.ai/integrations/commercehq.md): CommerceHQ is a hosted ecommerce platform for direct-to-consumer stores. Agents poll new orders into an ERP or warehouse, create fulfillments and shipments with tracking, refund orders, manage products, collections and customers, and search across store data. - [CompanyHub Integration](https://flowrunner.ai/integrations/companyhub.md): Connect AI agents to CompanyHub, a customizable CRM for small and medium businesses. Agents create and update contacts, companies, and deals across custom record types without a schema migration. - [ComPDFKit PDF Converter Integration](https://flowrunner.ai/integrations/compdfkit-pdf-converter.md): ComPDFKit PDF Converter turns PDFs into Word, Excel, PowerPoint, HTML, CSV, JSON, Markdown, or images and converts those formats back. Agents make scans searchable with OCR, pull PDF data into structured output, and hand editable files to the people who need them. - [ComPDFKit PDF Editor Integration](https://flowrunner.ai/integrations/compdfkit-pdf-editor.md): ComPDFKit PDF Editor edits PDFs through the ComPDF Cloud API without opening a file. Agents merge, split, rotate, compress, and watermark documents, add or remove pages, compare two versions, and convert to PDF/A for archiving. - [Confluence Integration](https://flowrunner.ai/integrations/confluence.md): Connect AI agents to Confluence Cloud. Agents create and update pages and blog posts, manage comments, labels, and attachments, and search content with CQL across spaces. - [Connverz Integration](https://flowrunner.ai/integrations/connverz.md): Work with a Connverz omnichannel messaging account on the Convrs platform, sending WhatsApp template messages and SMS. Agents manage conversations across channels and keep contact records aligned. - [Constant Contact Integration](https://flowrunner.ai/integrations/constant-contact.md): Run Constant Contact email marketing from your flows. Agents upsert contacts, manage lists, tags, custom fields, and segments, and build, schedule, test, and report on email campaigns. - [Contact Form 7 Integration](https://flowrunner.ai/integrations/contact-form-seven.md): Submit and inspect Contact Form 7 forms on any WordPress site. Agents list forms, read field definitions and mail settings, and submit entries that run through the plugin's normal validation, mail, and spam checks. - [ContactShip Integration](https://flowrunner.ai/integrations/contactship.md): Place and track outbound AI phone calls with ContactShip agents, and manage the contacts, tags, and phone numbers behind them. Agents run calling programs and read outcomes back into the pipeline. - [Contacts+ Integration](https://flowrunner.ai/integrations/contactsplus.md): Connect AI agents to Contacts+, a shared contact management platform from FullContact. Agents create and dedupe contacts, apply tags, share address books across teams, and manage webhook subscriptions for downstream systems. - [Content Snare Integration](https://flowrunner.ai/integrations/content-snare.md): Content Snare collects content, documents, and answers from clients through structured requests. Agents create clients and requests, fill and review the pages, sections, and fields inside them, approve or reject submissions, apply templates, and react to request events in real time. - [ContentStudio Integration](https://flowrunner.ai/integrations/content-studio.md): Schedule and publish social media content through ContentStudio. Agents create, list, and publish posts across connected channels and enumerate the workspaces and channels available to your account. - [Contentful Integration](https://flowrunner.ai/integrations/contentful.md): Connect AI agents to Contentful, the API-first headless CMS. Agents create, update, publish, archive, and delete entries and assets, manage content types and locales, and read published content through the cached Delivery API. - [Contractbook Integration](https://flowrunner.ai/integrations/contractbook.md): Manage contract lifecycle in Contractbook, covering documents, templates, spaces and automations. Agents draft from templates and move contracts through their stages, and stop for a person before anything is sent for signature. - [Convercus Integration](https://flowrunner.ai/integrations/convercus-loyalty.md): Run enterprise loyalty programmes with Convercus, enrolling accounts, managing cards, and posting point transactions. Agents award and redeem points from the systems where the qualifying event actually happened. - [ConvertAPI Integration](https://flowrunner.ai/integrations/convertapi.md): Convert files between hundreds of format pairs with ConvertAPI. Agents turn documents into PDF and Office formats, render live web pages to PDF, and check remaining conversion quota before large batches. - [Convertio Integration](https://flowrunner.ai/integrations/convertio.md): Convert files between more than 300 formats with Convertio, covering documents, images, audio, video, ebooks, presentations, spreadsheets, archives, and CAD. Agents normalize whatever arrives into the format the next system accepts. - [Copicake Integration](https://flowrunner.ai/integrations/copicake.md): Render saved image templates through the Copicake API with your own text, images, QR codes, and shape colors. Agents produce on-brand visuals per record without a designer in the loop. - [Copilot Integration](https://flowrunner.ai/integrations/copilot.md): Manage a Copilot client portal from your flows. Agents onboard clients and companies, create and list invoices, read contracts, and look up internal team members to keep service operations in sync. - [Copper Integration](https://flowrunner.ai/integrations/copper.md): Connect AI agents to Copper CRM. Agents manage people, companies, leads, opportunities, tasks, and activities, convert qualified leads, and advance deals through the pipeline. - [Copy.ai Integration](https://flowrunner.ai/integrations/copy-ai.md): Run Copy.ai Workflows from your flows. Agents trigger GTM and copywriting workflows with per-record inputs, wait for or poll run results, and manage completion webhooks for downstream automation. - [coreBOS Integration](https://flowrunner.ai/integrations/corebos.md): Connect AI agents to a self-hosted coreBOS instance, the open source ERP and CRM fork of Vtiger. Agents read and write any module record, resolve relationships between them, and inspect module metadata to work against a customized schema. - [Cornerstone OnDemand Integration](https://flowrunner.ai/integrations/cornerstone.md): Read employee and learning data from Cornerstone OnDemand. Agents look up employees, pull training transcripts, and browse the learning object catalog to drive compliance tracking and training follow-ups. - [Corsizio Integration](https://flowrunner.ai/integrations/corsizio.md): Connect AI agents to Corsizio, an online event and class registration platform. Agents read events, attendee registrations, and account records, so a flow can act on new signups in the systems downstream of Corsizio. - [Cortex Integration](https://flowrunner.ai/integrations/cortex.md): Connect AI agents to Cortex, the observable analysis and active-response engine from the TheHive Project. Agents run analyzers to enrich IOCs, collect structured reports and verdicts, and fire responders to block, notify, or ticket. - [Corymbus Integration](https://flowrunner.ai/integrations/corymbus.md): Connect AI agents to Corymbus, a French CRM for small businesses, freelancers, and independents. Agents create contacts and accounts, open opportunities, and keep the pipeline current without leaving the flow. - [Couchdrop Integration](https://flowrunner.ai/integrations/couchdrop.md): Couchdrop is a cloud SFTP and managed file transfer platform that mounts S3, SFTP, Dropbox, SharePoint, Box, and Azure into one virtual filesystem. Agents move and share files, provision scoped SFTP users and groups, manage automations, and react the moment a partner drops a file. - [Coupa Integration](https://flowrunner.ai/integrations/coupa.md): Connect AI agents to Coupa business spend management. Agents read requisitions and purchase orders, create and list supplier invoices, onboard suppliers, and monitor procure-to-pay records by status. - [Coveo Integration](https://flowrunner.ai/integrations/coveo.md): Put Coveo enterprise search behind your AI agents. Agents run relevance-ranked queries and query suggestions, manage content sources, and push, update, or delete documents in Push sources for indexing. - [Coze Integration](https://flowrunner.ai/integrations/coze.md): Connect flows to Coze, ByteDance's AI agent and chatbot platform. Agents chat with published bots, manage conversations and messages, run workflows and chatflows, upload files, and maintain retrieval knowledge bases. - [cPanel Integration](https://flowrunner.ai/integrations/cpanel.md): Automate cPanel hosting administration from your flows. Agents provision and remove email accounts, list databases, domains, and FTP accounts, monitor disk usage, and call any UAPI or WHM function through generic escape hatches. - [CPF/CNPJ Integration](https://flowrunner.ai/integrations/cpf-cnpj.md): Validate and enrich Brazilian tax identification numbers in real time, querying CPF for individuals and CNPJ for companies against Receita Federal. Agents confirm an identity before onboarding or invoicing. - [CrateDB Integration](https://flowrunner.ai/integrations/cratedb.md): Connect AI agents to CrateDB over its SQL-over-HTTP endpoint. Agents run parameterized SQL and high-throughput batched writes against distributed time-series and IoT data. - [Creatify Integration](https://flowrunner.ai/integrations/creatify-ai.md): Generate AI video ads and avatar videos with Creatify. Agents turn product URLs and scripts into rendered videos, create text-to-speech voiceovers, shorts, and template renders, and save finished media to FlowRunner file storage. - [Creatomate Integration](https://flowrunner.ai/integrations/creatomate.md): Render finished MP4s, GIFs, and images from Creatomate templates by filling in dynamic text, media, and colors. Agents start renders from a template or raw source JSON, poll render status, and browse templates to batch-produce personalized video and image variations. - [Crewmeister Integration](https://flowrunner.ai/integrations/crewmeister.md): Connect AI agents to Crewmeister, a workforce management tool for time tracking, absences, and shift planning. Agents record time stamps, create and approve absences, and plan shifts so payroll and staffing data stay accurate without manual entry. - [Crisp Integration](https://flowrunner.ai/integrations/crisp.md): Connect AI agents to Crisp, the customer messaging platform. Agents list and read chat conversations, post operator replies into the visitor's widget in real time, and create or update contact profiles on your Crisp website. - [CRM Messaging Integration](https://flowrunner.ai/integrations/crm-messaging.md): Connect AI agents to CRM Messaging, a messaging layer that adds SMS, WhatsApp, and voice to a CRM. Agents send SMS and WhatsApp messages, manage contacts, read message history, and place voice calls so customer outreach runs from one workflow. - [CRM Whats Pro Integration](https://flowrunner.ai/integrations/crm-whats-pro.md): Connect AI agents to CRM Whats Pro, a Brazilian WhatsApp CRM built on a WhatsApp Web browser extension. Agents send text and media messages, verify numbers on WhatsApp, and read chats and conversation history so WhatsApp outreach and follow-up run without manual copying. - [CareCloud Integration](https://flowrunner.ai/integrations/crmcarecloud.md): Connect AI agents to the CareCloud Customer Data Platform, a Czech loyalty and engagement suite. Agents enroll customers, issue cards, assign points, apply vouchers, move customers between segments, and read the rewards catalog behind each redemption. - [Crossbeam Integration](https://flowrunner.ai/integrations/crossbeam.md): Surface where your accounts overlap with your partners' in Crossbeam. Agents read partners, populations, account-mapping reports and their overlapping accounts, and collaboration threads to route co-selling signals into your CRM or sales channels. - [Crowdin Integration](https://flowrunner.ai/integrations/crowdin.md): Drive Crowdin localization from your workflows. Agents create projects, push source files and strings, track per-language translation and approval progress, and build and download translated packages for release pipelines. - [CrowdPower Integration](https://flowrunner.ai/integrations/crowdpower.md): Feed customer identification and behavioral data into CrowdPower through the Beacon API. Agents keep lifecycle campaigns triggered by what customers actually did rather than by a nightly export. - [Crypto Tools Integration](https://flowrunner.ai/integrations/crypto-tools.md): Run cryptography utilities entirely inside FlowRunner with no API key, vendor account, or outbound call. Agents hash and checksum data, sign and verify HMACs, encrypt field values, and generate tokens as a workflow step. - [CS-Cart Integration](https://flowrunner.ai/integrations/cs-cart.md): CS-Cart and CS-Cart Multi-Vendor are self-hosted store and marketplace platforms. Agents keep the catalog in step with a PIM or ERP, watch and update orders and shipments, manage customers and vendors, recover abandoned carts, and adjust taxes, shipping and payment settings. - [Curator.io Integration](https://flowrunner.ai/integrations/curator-app.md): Connect AI agents to Curator.io, a social media feed aggregation platform. Agents pull posts from connected sources, set feed moderation mode and manage sources, and inject custom posts alongside the aggregated stream. - [Currency Converter Integration](https://flowrunner.ai/integrations/currency-converter.md): Convert money between currencies and retrieve published exchange rates, with no account or API key required. Agents normalize multi-currency amounts before they reach an accounting or reporting system. - [Custify Integration](https://flowrunner.ai/integrations/custify.md): Connect AI agents to Custify, the customer success platform. Agents sync people and companies, track product events, keep subscription and revenue data current, log NPS responses, and manage customer success tasks to power health scores and playbooks. - [Custom JS Integration](https://flowrunner.ai/integrations/custom-js.md): CustomJS bundles PDF generation, screenshots, web scraping, hosted pages, and server-side JavaScript behind one API key. Workflows convert HTML or Markdown to PDF, merge and compress PDFs, scrape or screenshot pages, run stored functions, and publish hosted pages. - [Customer.io Integration](https://flowrunner.ai/integrations/customerio.md): Connect AI agents to Customer.io. Agents identify people, track behavioral events, curate manual segments, send transactional email, trigger broadcasts, and pull customer attributes and campaign metrics. - [Customerly Integration](https://flowrunner.ai/integrations/customerly.md): Connect AI agents to Customerly, a live chat and customer lifecycle messaging platform. Agents sync users and leads, tag and associate them with companies, read conversations, and publish knowledge base content. - [Cutt.ly Integration](https://flowrunner.ai/integrations/cuttly.md): Shorten, brand, and measure links with Cutt.ly across both its personal and team API surfaces. Agents mint a tracked link per send and read click data back to see what actually landed. - [Cyfe Integration](https://flowrunner.ai/integrations/cyfe.md): Connect AI agents to Cyfe, an all-in-one KPI dashboard platform. Agents push metrics from anywhere in a flow onto a Cyfe widget so a dashboard reflects live operational state. - [D7 Networks Integration](https://flowrunner.ai/integrations/d7-sms.md): Connect AI agents to D7 Networks, a multichannel business messaging gateway based in Dubai. Agents send SMS, WhatsApp, Viber, and Slack messages, verify OTPs, and look up numbers and delivery status so global customer messaging runs through one API. - [Daktela Integration](https://flowrunner.ai/integrations/daktela.md): Connect AI agents to Daktela, an omnichannel contact center and helpdesk platform. Agents open and resolve tickets, keep contact records in sync, create accounts, and read the activities, queues, and directory behind each conversation. - [Darujme.cz Integration](https://flowrunner.ai/integrations/darujme-cz.md): Darujme.cz is the Czech donation platform non-profits collect through. Agents read pledges and transactions, update pledge custom fields, pull project statistics, and read peer-to-peer fundraising promotions. - [Data24-7 Integration](https://flowrunner.ai/integrations/data24-7.md): Look up real-time phone, carrier, identity, and contact data through the Data247 API. Agents confirm line type and SMS reachability before a message is sent and resolve identity before a record is trusted. - [Databox Integration](https://flowrunner.ai/integrations/databox.md): Push live KPIs into Databox dashboards from any workflow. Agents send metric values through the Push API, list metric keys, and audit or purge recent pushes to keep Custom data sources clean. - [Databricks Integration](https://flowrunner.ai/integrations/databricks.md): Connect AI agents to the Databricks REST API. Agents run SQL on a warehouse, orchestrate jobs, manage clusters, browse Unity Catalog, and inspect DBFS and workspace paths. - [Datadog Integration](https://flowrunner.ai/integrations/datadog.md): Operate Datadog from your workflows: agents post events, submit and query metrics, ship and search logs, manage monitors and downtimes, declare incidents, and work with dashboards, SLOs, hosts, users, Synthetics, and notebooks across all Datadog sites. - [DataForms.io Integration](https://flowrunner.ai/integrations/dataforms-io.md): DataForms.io is a no-code form builder built around reusable templates. Agents create templates and fields, spin up live forms from them, read the entries those forms collect, and react to each new entry as it arrives. - [DataForSEO Integration](https://flowrunner.ai/integrations/dataforseo.md): Pull SEO, SERP, and keyword research data into AI workflows. Agents run organic and maps searches across Google and Bing, score keyword difficulty in bulk, expand seed terms, and surface ranking keywords for any domain. - [Datagma Integration](https://flowrunner.ai/integrations/datagma.md): Enrich B2B contacts, find email addresses, and look up phone numbers with Datagma. Agents complete a prospect record from a partial signal before it enters the sequence. - [DataMerge AI Integration](https://flowrunner.ai/integrations/datamerge.md): Enrich companies and contacts through the DataMerge Company and Contact API, pulling firmographics from a domain and walking corporate hierarchies. Agents qualify an account against structure, not just a logo. - [DataRobot Integration](https://flowrunner.ai/integrations/datarobot.md): Score records against DataRobot MLOps deployments in real time or as batch prediction jobs. Agents also manage modeling projects, browse the Leaderboard to compare candidate models, and ingest datasets into the AI Catalog. - [Datelist Integration](https://flowrunner.ai/integrations/datelist.md): Connect AI agents to Datelist, an online booking and appointment scheduling tool. Agents list calendars and products, update or cancel booked slots, and register webhooks so new bookings start a flow the moment they arrive. - [Daxium-Air Integration](https://flowrunner.ai/integrations/daxium-air.md): Daxium-Air is a French no-code platform for mobile forms and field data capture. Agents read and write field submissions, manage form structures and reference lists, generate mail merge reports, schedule field tasks, move attachments in and out of file storage, and administer users and groups. - [Cin7 Core (DEAR) Integration](https://flowrunner.ai/integrations/dear-inventory.md): Connect AI agents to Cin7 Core (formerly DEAR Inventory). Agents check real-time stock availability across warehouses, read and search the product catalog, create sales orders from storefront events, audit purchase orders, and manage customers and suppliers. - [DeBounce Integration](https://flowrunner.ai/integrations/debounce.md): DeBounce validates email addresses and cleans lists. Agents verify single addresses, check for disposable domains, run reverse email lookups, upload bulk lists and download the results, and track balance and usage. - [Decathlon Integration](https://flowrunner.ai/integrations/decathlon.md): Connect AI agents to Decathlon's public sports intelligence APIs. Agents look up sports, resolve places where they are played, and read the referential data behind Decathlon's catalog. - [Deel Integration](https://flowrunner.ai/integrations/deel.md): Connect AI agents to Deel for global HR, payroll, contracting, and ATS workflows. Agents onboard workers, manage time off, drive recruiting, and sync people data via OAuth2. - [DeepAI Integration](https://flowrunner.ai/integrations/deep-ai.md): Generate, edit, and enhance images through DeepAI's single-call REST endpoints. Agents produce and clean up visual assets inside a workflow that already knows what the asset is for. - [Deep-Image.ai Integration](https://flowrunner.ai/integrations/deep-image-ai.md): Upscale, enhance, and edit image backgrounds with Deep-Image.ai, including RAW formats such as NEF and CR2. Agents prepare product and listing photography without a manual editing pass. - [Deepgram Integration](https://flowrunner.ai/integrations/deepgram.md): Give agents Deepgram voice AI: transcribe recorded audio with diarization, summarization, and sentiment, generate speech with Aura voices, and analyze text for topics and intent. Also manages Deepgram projects, API keys, usage, and billing. - [DeepInfra Integration](https://flowrunner.ai/integrations/deepinfra.md): Run open models on DeepInfra without hosting GPUs. Agents chat with Llama, Qwen, DeepSeek, and Mistral through the OpenAI-compatible API, create embeddings for RAG, generate images with FLUX or SDXL, and reach any hosted model via the native inference endpoint. - [DeepL Integration](https://flowrunner.ai/integrations/deepl.md): Translate text and full documents with DeepL neural machine translation, improve writing with DeepL Write, and manage glossaries to keep terminology consistent across languages. - [DeepSeek Integration](https://flowrunner.ai/integrations/deepseek.md): Run DeepSeek large language models with extended reasoning, tool and function calling, JSON output, and prefix completion. Fast, low-cost inference with an OpenAI-compatible interface. - [Deezer Integration](https://flowrunner.ai/integrations/deezer.md): Deezer is a music streaming catalog with albums, artists, tracks, playlists, charts, radios, and podcasts. Agents search the catalog, read charts and editorial picks, and manage the connected account's playlists and favorites. - [Deftform Integration](https://flowrunner.ai/integrations/deftform.md): Deftform is a lightweight online form builder positioned against Typeform and Jotform. Agents enumerate forms and fields, pull submissions with their answers, record submissions from a custom front end, render a submission to PDF, and adjust form settings and notifications without opening the editor. - [Delesign Integration](https://flowrunner.ai/integrations/delesign.md): Connect AI agents to Delesign, a subscription graphic design service with a managed request queue. Agents submit design briefs as projects, send messages to designers, and retrieve finished files. - [Delighted Integration](https://flowrunner.ai/integrations/delighted.md): Work with Delighted NPS, CSAT, and customer experience surveys. Agents sync survey responses, add or update the people you survey, honor unsubscribes, and pull aggregate metrics like your Net Promoter Score with promoter and detractor counts. - [Demio Integration](https://flowrunner.ai/integrations/demio.md): Register participants into Demio webinar sessions, look up scheduled session dates, and pull per-session participant and attendance data for follow-up and reporting. - [Demodesk Integration](https://flowrunner.ai/integrations/demodesk.md): Connect AI agents to Demodesk, the sales meeting and coaching platform. Agents list recordings, retrieve transcripts, summaries, and coaching scorecards, and react when new recordings land so meeting insights flow straight into follow-up workflows. - [Deputy Integration](https://flowrunner.ai/integrations/deputy.md): Manage Deputy workforce scheduling from your flows. Agents create employees and scheduled shifts, and pull rosters, timesheets, leave, and locations to sync new hires, publish schedules, and feed reporting. - [Deskera Integration](https://flowrunner.ai/integrations/deskera.md): Run Deskera Books accounting end to end. Agents manage contacts, products, sales invoices, quotations, purchase bills, and payments, post journal entries, and keep the chart of accounts consistent across your finance stack. - [Detecting-AI Integration](https://flowrunner.ai/integrations/detecting-ai.md): Screen writing for AI-generated content and plagiarism, and humanize text, through the Detecting-AI API. Agents check submitted or outsourced copy before it is published under your name. - [Detrack Integration](https://flowrunner.ai/integrations/detrack.md): Connect AI agents to Detrack, a last-mile delivery management platform. Agents create and update delivery jobs, capture proof of delivery, and plan optimized driver routes so dispatch teams keep every delivery on schedule. - [DeutschlandGPT Integration](https://flowrunner.ai/integrations/deutschlandgpt.md): Reach DeutschlandGPT, the GDPR-compliant German AI platform, where one key gives flows access to hosted models. Agents run German-language generation under a data residency posture that legal has already accepted. - [Dext Integration](https://flowrunner.ai/integrations/dext.md): Read practice and client data from Dext Precision. Agents pull client rosters with data health scores and alert levels, track bookkeeping activity trends, and reach any Precision endpoint through a generic authenticated call. - [DHL Integration](https://flowrunner.ai/integrations/dhl.md): Track shipments across every DHL business unit and find DHL service points, post offices, and parcel lockers by address or geo-coordinate. - [Diabolocom Integration](https://flowrunner.ai/integrations/diabolocom.md): Manage users, queues, campaigns, and live statistics on the Diabolocom contact center platform. Agents staff queues against real demand and read live performance without a supervisor watching a wallboard. - [DialoX Integration](https://flowrunner.ai/integrations/dialox.md): Drive conversational AI agents built on DialoX, steering conversations across WhatsApp, web, phone, and email. Agents hand context into a bot and take the thread back when a person needs to decide. - [Dialpad Integration](https://flowrunner.ai/integrations/dialpad.md): Run Dialpad from your workflows: agents place outbound calls, send SMS and MMS, provision users and phone numbers, sync contacts, retrieve AI call transcripts and usage statistics, and register webhooks for call and SMS events. - [Diffbot Integration](https://flowrunner.ai/integrations/diffbot.md): Turn any public web page into structured data with Diffbot. Agents extract clean articles, products, and discussions from URLs, auto-detect page types, query the Knowledge Graph with DQL, and enrich people and companies from a name, URL, or email. - [Dify Integration](https://flowrunner.ai/integrations/dify.md): Invoke your Dify LLM apps from FlowRunner. Agents send chat and completion messages, run and monitor Dify workflows, manage conversations and annotation replies, upload files, and convert between audio and text on Dify Cloud or self-hosted deployments. - [Digiclose Integration](https://flowrunner.ai/integrations/digiclose.md): Connect AI agents to Digiclose, a phone sales and sales automation platform. Agents create contacts, move them through pipelines, log transactions, assign tasks, and manage team and product records. - [DigiSign Integration](https://flowrunner.ai/integrations/digisign.md): DigiSign is a Czech electronic signature platform with qualified signatures, Bank iD, and SMS authentication. Agents build envelopes from documents and recipients, check they are sendable, send them in one step, and react to envelope events as they happen. - [Digistore24 Integration](https://flowrunner.ai/integrations/digistore.md): Read your Digistore24 sales operation. Agents list purchases, orders, products, buyers, and financial transactions to sync new sales into a CRM, notify your team on purchases, and reconcile payments, refunds, and chargebacks. - [DigitalPilot Integration](https://flowrunner.ai/integrations/digitalpilot.md): Connect AI agents to DigitalPilot, a cookieless B2B marketing attribution platform. Agents read attribution reports and the leads its on-site forms captured, maintain target account lists and form definitions, and manage outbound webhook subscriptions. - [Diigo Integration](https://flowrunner.ai/integrations/diigo.md): Diigo is the social bookmarking tool. Agents save bookmarks with tags and descriptions and read back a user's bookmark list, including any annotations already attached, for research and archiving flows. - [Disciple.Tools Integration](https://flowrunner.ai/integrations/disciple-tools.md): Connect AI agents to a self-hosted Disciple.Tools site, the open source WordPress CRM for ministry and outreach. Agents create contacts, groups, and records, post comments, manage who a record is shared with, and read the activity log behind each change. - [Discko Integration](https://flowrunner.ai/integrations/discko.md): Connect AI agents to Discko, an AI lead qualification widget embedded in your site. Agents read qualified leads together with the qualification context the widget captured, and update each lead's status as it moves to won or lost. - [Discord Integration](https://flowrunner.ai/integrations/discord.md): Give agents a presence in your Discord server. Post and manage messages, open threads, react, DM members, assign roles, and relay through webhooks over the Discord REST API. - [Discourse Integration](https://flowrunner.ai/integrations/discourse.md): Create and manage topics and posts on a Discourse forum, search content, manage users and private messages, and browse categories and tags via the Discourse REST API. - [Disparo Pro Integration](https://flowrunner.ai/integrations/disparo-pro.md): Connect AI agents to Disparo Pro, a Brazilian bulk SMS gateway. Agents send single and bulk SMS, check delivery status, retrieve inbound replies, and read the credit balance so Brazilian SMS campaigns stay monitored end to end. - [Disqus Integration](https://flowrunner.ai/integrations/disqus.md): Read and manage forums, threads, posts (comments), and users on the Disqus commenting platform. List and moderate comments, create threads and posts, and look up user activity via the Disqus API 3.0. - [dlvr.it Integration](https://flowrunner.ai/integrations/dlvrit.md): Connect AI agents to dlvr.it, a social media automation service that publishes content from feeds. Agents publish to connected accounts and read the publishing routes and account state behind each one. - [DNC Solution Integration](https://flowrunner.ai/integrations/dnc-solution.md): Connect AI agents to DNCScrub from Contact Center Compliance, the do-not-call compliance service. Agents scrub phone lists against federal, state, and internal DNC registries, check reassigned numbers, and evaluate whether a time is legal to call before a dialer runs. - [Doc Crafter Integration](https://flowrunner.ai/integrations/doc-crafter.md): Doc Crafter is a stateless document toolbox. Agents turn web pages and spreadsheets into PDFs, read text out of PDFs and photos, generate and decode QR codes and barcodes, and merge or split PDF files in single calls. - [Docparser Integration](https://flowrunner.ai/integrations/docparser.md): Extract structured fields from PDFs with Docparser. Agents upload documents to a parser by URL, poll processing status, and retrieve parsed invoice, receipt, and purchase order data as JSON to route into accounting or CRM systems. - [DocuGenerate Integration](https://flowrunner.ai/integrations/docugenerate.md): Connect AI agents to DocuGenerate, which merges structured data into Word and OpenDocument templates to produce finished documents. Agents manage templates, generate Word or PDF output from JSON or spreadsheet data, and download the results so paperwork gets produced and filed automatically. - [Document360 Integration](https://flowrunner.ai/integrations/document360.md): Author and organize Document360 knowledge bases programmatically. Agents create, update, and delete articles, route them into the right workspace and category, and reference Drive folders to keep self-service help centers current. - [Documentero Integration](https://flowrunner.ai/integrations/documentero.md): Documentero fills Word and Excel templates with your data and returns a finished DOCX or PDF. Agents generate contracts, quotes, and certificates from records and file the result straight into storage. - [DocuPanda Integration](https://flowrunner.ai/integrations/docupanda.md): DocuPanda turns documents into structured data. Agents parse files, classify them, extract fields against a JSON schema you define, route results to a human reviewer when needed, and react when a job finishes. - [DocuSeal Integration](https://flowrunner.ai/integrations/docuseal.md): Run document e-signing on DocuSeal. Agents create reusable templates from PDF or HTML, send documents out for signature, pre-fill submitter fields, and track every submitter's progress on DocuSeal Cloud or your self-hosted instance. - [DocuSign Integration](https://flowrunner.ai/integrations/docusign.md): Automate the complete document signing lifecycle. Agents send envelopes from templates, track signing status, handle declines, and trigger downstream workflows the moment a document is fully executed. - [DocuWare Integration](https://flowrunner.ai/integrations/docuware.md): Archive and retrieve documents in DocuWare, Ricoh's enterprise document management platform. Agents upload files with index-field metadata, search cabinets with DocuWare Query Language, and download stored documents into FlowRunner storage for downstream steps. - [Dokan Integration](https://flowrunner.ai/integrations/dokan.md): Dokan is the multivendor marketplace plugin for WooCommerce. Agents onboard and approve vendors, publish and maintain marketplace products, track orders, withdrawals and commission, manage coupons and reviews, and pull store reports. - [Domotron Control Integration](https://flowrunner.ai/integrations/domotron-control.md): Connect AI agents to Domotron, a Slovak smart home control system. Agents read environment sensors and control lighting, climate, shading, ventilation, and scenes, and read the stream URLs behind each camera. - [DonnaJames Integration](https://flowrunner.ai/integrations/donnajames.md): Build and operate chatbots with in-chat forms through DonnaJames, managing agents and growing the knowledge base. Agents keep bot answers grounded in current source material and collect structured responses mid-conversation. - [Donorbox Integration](https://flowrunner.ai/integrations/donorbox.md): Bring Donorbox fundraising data into your workflows. Agents list donations, donors, campaigns, and recurring plans, and log offline gifts so donation data flows into CRMs, spreadsheets, and accounting tools. - [Dotdigital Integration](https://flowrunner.ai/integrations/dotdigital.md): Connect AI agents to Dotdigital marketing automation. Agents create and update contacts with custom data fields, add them to address books, and browse and send email campaigns scoped to your account's hosting region. - [DotSimple Integration](https://flowrunner.ai/integrations/dotsimple.md): Connect AI agents to DotSimple, a social media content planning and publishing tool. Agents upload media, schedule posts across connected accounts, and organize them with tags. - [Doubao Integration](https://flowrunner.ai/integrations/doubao.md): Call ByteDance's Doubao model family through the Volcano Engine Ark data-plane API. Agents run Chinese-language generation and embeddings on a platform that already meets local data requirements. - [DPD Germany Integration](https://flowrunner.ai/integrations/dpd-germany.md): Connect AI agents to DPD Germany, the parcel carrier's public shipping web services. Agents obtain an auth token, create parcel shipments with printable labels, and retrieve tracking data so German parcel dispatch runs straight from a flow. - [DPD Romania Integration](https://flowrunner.ai/integrations/dpd-romania.md): Connect AI agents to DPD Romania, the Romanian arm of the DPD parcel network. Agents calculate prices, create shipments, print labels, track parcels, and request courier pickups so Romanian shipping runs end to end without the carrier portal. - [DQE Integration](https://flowrunner.ai/integrations/dqe.md): Validate and correct email addresses, landline and mobile numbers, and international postal addresses in real time with DQE. Agents fix contact data at the point of capture rather than after a campaign fails. - [Dr.Tracker Integration](https://flowrunner.ai/integrations/dr-tracker.md): Dr.Tracker and the Dr. Mailer platform behind it combine mailing, list hygiene, and lead routing. Agents run Autopilot mailings, clean lists through Leads Surgeon, and route and report on tracked leads. - [Dribbble Integration](https://flowrunner.ai/integrations/dribbble.md): Connect AI agents to Dribbble, the designer portfolio and community platform. Agents publish and schedule shots, organize them into projects, attach assets, and look up a job listing by id. - [Drift Integration](https://flowrunner.ai/integrations/drift.md): Connect AI agents to Drift conversational marketing. Agents create and update contacts, reply to live chats, post internal notes, triage conversation status, and sync ABM accounts. - [Drip Integration](https://flowrunner.ai/integrations/drip.md): Connect AI agents to Drip, the ecommerce CRM and email marketing platform. Agents upsert subscribers by email, apply and remove tags, record ecommerce events, enroll people in email series campaigns, and read campaign and workflow lists for reporting. - [Dropbox Integration](https://flowrunner.ai/integrations/dropbox-service.md): Full Dropbox integration for AI workflows. Agents list, upload, download, move, copy, and search files. Manage shared links and folder members. Polling triggers react to new files, modified files, and new folders in watched directories. - [Dropcontact Integration](https://flowrunner.ai/integrations/dropcontact.md): Enrich contacts with verified professional emails and company data through a GDPR-compliant, EU-focused service that never buys or resells personal data. Agents clean lists in batches and append French SIREN, NAF, and VAT data. - [DropFunnels Integration](https://flowrunner.ai/integrations/dropfunnels.md): Pull leads, orders, and course activity from DropFunnels, covering sites, funnels, every order type, and course membership. Agents route new buyers and students into fulfilment and onboarding the moment they convert. - [Drupal Integration](https://flowrunner.ai/integrations/drupal.md): Read and write content on a Drupal site over its core JSON:API module. Agents create, update, and delete nodes, taxonomy terms, and users, and reach any entity type through typed CRUD actions plus a raw JSON:API request. - [Dub Integration](https://flowrunner.ai/integrations/dub.md): Create and manage branded short links in Dub, the open source link management platform. Agents create, update, and upsert links, organize them with tags and domains, and pull click, lead, and sale attribution analytics for a workspace or a single link. - [Dumpling AI Integration](https://flowrunner.ai/integrations/dumplingai.md): Reach the tools an AI agent needs behind one Dumpling AI key: web scraping and crawling, Google search, news and places, document and image extraction, and video work. Agents research and extract without a connector per task. - [Dust Integration](https://flowrunner.ai/integrations/dust.md): Connect AI agents to Dust, the enterprise AI agent platform. Agents discover and mention Dust agents, run conversations and poll for replies, keep data source knowledge current by upserting documents and table rows, search it semantically, and execute custom Dust apps. - [Dux-Soup Integration](https://flowrunner.ai/integrations/dux-soup.md): Queue LinkedIn outreach activity through the Dux-Soup Remote Control API, manage the action queue, and read prospect conversations and tags. Agents pace outreach against real limits instead of tripping them. - [Dynalist Integration](https://flowrunner.ai/integrations/dynalist.md): Dynalist is the outliner for structured notes and lists. Agents read and edit documents and folders, insert and update items inside a document, capture to the inbox, upload files, and read account preferences. - [Dynamic Mockups Integration](https://flowrunner.ai/integrations/dynamic-mockups.md): Dynamic Mockups renders product mockups by placing artwork into PSD smart objects. Agents render single mockups, batches, or whole collections, export print files, generate AI mockup scenes, and animate a still into a short motion video. - [Microsoft Dynamics 365 Integration](https://flowrunner.ai/integrations/dynamics-365.md): Read, create, update, and delete records in any Dataverse table behind Dynamics 365, and run FetchXML queries for aggregations and complex joins. - [Amazon DynamoDB Integration](https://flowrunner.ai/integrations/dynamodb-service.md): Read, write, query, and scan Amazon DynamoDB tables from automated flows. Agents persist workflow data, run PartiQL statements, and batch-load items with values returned as plain JSON. - [Dynosend Integration](https://flowrunner.ai/integrations/dynosend.md): Dynosend is an email marketing and transactional messaging platform. Agents manage audiences, contacts, and events, start and resume broadcast campaigns, send transactional messages, and maintain the account blocklist. - [e-conomic Integration](https://flowrunner.ai/integrations/e-conomic.md): e-conomic is the Danish accounting platform from Visma. Agents manage customers, suppliers, and products with currency specific pricing, issue and book invoices, and read the ledger and reports the business runs on. - [E-goi Integration](https://flowrunner.ai/integrations/e-goi.md): Run multichannel campaigns on E-goi, the Portuguese marketing platform, through the official Marketing API V3. Agents manage contacts and lists, send email and SMS, and read campaign results back into the flow. - [E2B Integration](https://flowrunner.ai/integrations/e2b.md): Give AI agents disposable cloud machines with E2B sandboxes. Agents spawn secure micro-VMs from templates, pause and resume them to preserve state between turns, snapshot known-good checkpoints, and monitor metrics and logs across the fleet before killing sandboxes when work completes. - [Easelly Integration](https://flowrunner.ai/integrations/easelly.md): Easelly turns a design built in its editor into a personalized infographic. Agents swap tagged text, colors, and images per recipient, then export the result as PNG, PDF, SVG, or JPG for reports, one-pagers, and campaign visuals. - [easiware Integration](https://flowrunner.ai/integrations/easiware.md): Connect AI agents to easiware, a French omnichannel customer service and CRM platform. Agents open tickets, post messages, route by competence group, manage webhook subscriptions, and read the response templates and event history behind each case. - [EasyCargo Integration](https://flowrunner.ai/integrations/easy-cargo.md): Connect AI agents to EasyCargo, a 3D truck and container load planning tool. Agents create shipments and cargo spaces, retrieve calculated load plans and item placements, and share public links so load planning results feed dispatch decisions automatically. - [Easy Project Integration](https://flowrunner.ai/integrations/easy-project.md): Easy Project by Easy Software is the Redmine-based suite for project, CRM and resource management. Agents manage projects, issues, time entries, versions and people, work the CRM's contacts, cases and leads, and start flows from Easy Project web hooks. - [easybill Integration](https://flowrunner.ai/integrations/easybill.md): easybill is German online invoicing and bookkeeping. Agents manage customers and the article catalog with stock, and create the single document object that serves as invoice, offer, credit note, or order, then send it and track payments. - [EasyCSV Integration](https://flowrunner.ai/integrations/easycsv.md): Queue CSV, XLSX, and Google Sheets imports with EasyCSV by posting a public file URL to an import page. Agents turn a spreadsheet that landed in a mailbox into rows in the destination system. - [Easypay Integration](https://flowrunner.ai/integrations/easypay.md): Easypay is a Portuguese payment gateway covering Multibanco, MB WAY, cards, direct debit, and wallets. Agents create single and frequent payments, run subscriptions and Pay by Link checkouts, capture, void, and refund, and reconcile settlements and out payments. - [EasyPost Integration](https://flowrunner.ai/integrations/easypost.md): Buy shipping labels across USPS, UPS, FedEx, and DHL, verify addresses, prepare customs declarations, and process batch shipments, pickups, insurance, and refunds. - [Easyship Integration](https://flowrunner.ai/integrations/easyship.md): Connect to 250+ couriers through Easyship to compare rates, create shipments, generate labels, schedule pickups, produce manifests, and track parcels worldwide. - [EasyWeek Integration](https://flowrunner.ai/integrations/easyweek.md): Connect AI agents to EasyWeek, appointment scheduling software for service businesses. Agents check real-time availability, create and reschedule bookings, and look up staff and services so schedules stay full without manual coordination. - [Ecofleet Integration](https://flowrunner.ai/integrations/ecofleet-cz.md): Connect AI agents to Ecofleet, a GPS fleet tracking and telematics platform. Agents read live vehicle data and trip history, assign drivers to trips, lock or unlock vehicles, and pull reports so fleet operations stay visible without the telematics console. - [Ecologi Integration](https://flowrunner.ai/integrations/ecologi.md): Connect AI agents to Ecologi, a UK climate action service funding tree planting and carbon removal. Agents purchase climate impact as a step inside a flow and pull the reporting that evidences it. - [Ecomail Integration](https://flowrunner.ai/integrations/ecomail-cz.md): Ecomail is a Czech email marketing platform. Agents manage lists and subscribers, run campaigns and automations, send transactional email, feed the behavioral tracker and e-commerce transactions, and manage coupons and sending domains. - [Eden AI Integration](https://flowrunner.ai/integrations/edenai.md): Reach hundreds of AI models from dozens of providers through one Eden AI key. Agents run OCR, moderation, translation, and entity extraction through the Universal AI endpoint, plus chat completions, embeddings, and image generation through an OpenAI-compatible surface with automatic fallbacks. - [Tencent EdgeOne Image Render Integration](https://flowrunner.ai/integrations/edgeone-image-render.md): Tencent EdgeOne Image Render turns an HTML and CSS template into a finished image at the edge. Workflows list templates, render images with dynamic data, and build signed image URLs safe to embed in email or dashboards. - [eDock Integration](https://flowrunner.ai/integrations/edock.md): eDock, sold today as MarketRock, is an Italian multichannel commerce platform that publishes one catalog to Amazon, eBay, Zalando and dozens of other marketplaces. Agents keep prices and stock in step everywhere at once, pull marketplace orders into a back office, and book the couriers that ship them. - [Edusign Integration](https://flowrunner.ai/integrations/edusign.md): Connect AI agents to Edusign, a digital attendance and signature platform for schools and training organizations. Agents manage students, courses, and groups, send signature requests, and mark attendance states so attendance records complete themselves without paper sheets. - [EenvoudigFactureren Integration](https://flowrunner.ai/integrations/eenvoudigfactureren.md): EenvoudigFactureren is Belgian small business invoicing software. Agents manage clients, contacts, and stock items, create and send documents by email or post, record payments and reminder costs, and pull document files and events. - [Revver Integration](https://flowrunner.ai/integrations/efilecabinet.md): Revver, formerly eFileCabinet, is enterprise document management organized as a tree of folders and files. Agents file documents into the right node, share them through links and collections, issue document requests, and pull signed contracts out for the next step. - [Egnyte Integration](https://flowrunner.ai/integrations/egnyte.md): Manage files and folders in Egnyte, the enterprise content collaboration platform. Agents browse folders, read metadata, upload and download files, move, copy, and delete content, create shared links with access controls, and look up domain users. - [8x8 Integration](https://flowrunner.ai/integrations/eight-x-eight.md): Send SMS and rich channel messages through 8x8 CPaaS. Agents deliver transactional SMS, reach customers on WhatsApp, Viber, Zalo, LINE, and KakaoTalk, schedule messages for future delivery, and poll delivery status to confirm receipt or catch failures. - [Elastic Email Integration](https://flowrunner.ai/integrations/elastic-email.md): Send transactional and bulk email through Elastic Email. Agents deliver receipts and password resets, send marketing email with HTML and plain text content, manage contacts and lists, and read account level statistics on delivery, opens, clicks, and bounces. - [Elasticsearch Integration](https://flowrunner.ai/integrations/elasticsearch.md): Connect AI agents to Elasticsearch and Elastic Cloud. Agents index and search documents with full Query DSL, run bulk NDJSON operations, update or delete by query, and manage indices. - [EleAPIs Integration](https://flowrunner.ai/integrations/eleapis.md): Connect AI agents to EleAPIs (Express-Chat), the omnichannel chatbot and WhatsApp Business platform. Agents search and segment inbox contacts across website, WhatsApp, and social channels and reply where the customer started. - [Eledo Integration](https://flowrunner.ai/integrations/eledo.md): Eledo renders a designed template plus your data into a PDF. Agents generate invoices, contracts, and certificates on demand and either receive the file or hand out a download link. - [Element451 Integration](https://flowrunner.ai/integrations/element451.md): Connect AI agents to Element451, a higher education enrollment CRM. Agents update person profiles, create tasks and notes, register event attendees, and enroll people in journeys so admissions teams keep every prospect moving. - [Elementor Integration](https://flowrunner.ai/integrations/elementor.md): Run flows the moment an Elementor Pro form is submitted on your WordPress site. Paste the FlowRunner callback URL into the form's webhook action and each submission arrives parsed, with field values keyed by ID and an optional form name filter to target a single form. - [Elements.cloud Integration](https://flowrunner.ai/integrations/elements-cloud.md): Elements.cloud is the change-intelligence platform for Salesforce teams. Workflows create, update, and list Stories and Requirements in the backlog, with statuses and custom fields, so change requests stay in sync with the systems that raise them. - [ElevenLabs Integration](https://flowrunner.ai/integrations/elevenlabs.md): Add voice synthesis to FlowRunner workflows with ElevenLabs. Agents convert text to natural speech, generate sound effects, clone and design voices, and store the audio for downstream use. - [elopage Integration](https://flowrunner.ai/integrations/elopage.md): elopage, now ablefy, is the German platform for selling digital products, courses, memberships and downloads. Agents grant access with free orders, cancel subscriptions, read payments and affiliate publishers, and manage funnels, upsells and webhook endpoints. - [EmailListVerify Integration](https://flowrunner.ai/integrations/email-list-verify.md): EmailListVerify checks email addresses in bulk and in real time. Agents verify single addresses, run asynchronous verification jobs, upload and download bulk lists, check disposable domains and blacklists, find business emails, and test inbox placement. - [EmailOctopus Integration](https://flowrunner.ai/integrations/email-octopus.md): Keep EmailOctopus audiences in sync with the rest of your stack. Agents create and update contacts, manage lists, look up subscribers by email address, and read campaign reports on opens, clicks, bounces, and unsubscribes to act on engagement automatically. - [Emailable Integration](https://flowrunner.ai/integrations/emailable.md): Verify email addresses before they enter your lists. Agents check deliverability in real time at signup, run batch verification jobs on lists up to 50,000 addresses, flag disposable, role, and accept-all addresses, and monitor remaining credits so flows pause before running out. - [Emailkampane Integration](https://flowrunner.ai/integrations/emailkampane.md): Emailkampane.cz, sold internationally as goemail.eu, is a Czech email marketing platform. Agents manage contacts, lists, and custom attributes, build and send campaigns from templates, and read the full range of campaign statistics. - [Emailvalidation.io Integration](https://flowrunner.ai/integrations/emailvalidation.md): emailvalidation.io by everapi validates email addresses in real time. Agents check a single address for deliverability, syntax, disposable and catch-all status, and read the remaining quota before a batch. - [Emelia Integration](https://flowrunner.ai/integrations/emelia.md): Run cold email and LinkedIn outreach through Emelia, covering multichannel campaigns, contact lists, and the built-in email and phone finder and verifier. Agents build a verified list and launch the sequence in one motion. - [Emercury Integration](https://flowrunner.ai/integrations/emercury.md): Emercury is an email marketing platform with audiences, automations, and autoresponders. Agents import subscribers, manage segments, custom fields, and suppression lists, schedule and send campaigns, and pull reporting. - [Emma Integration](https://flowrunner.ai/integrations/emma.md): Manage your Emma by Marigold email audience from a flow. Agents add and look up members by email, segment them into groups ahead of a send, opt contacts out to honor unsubscribes, and read mailings and custom field definitions for reporting and data mapping. - [Emporix Commerce Integration](https://flowrunner.ai/integrations/emporix-commerce.md): Emporix Commerce Engine is an API-first headless commerce platform. Agents sync products, prices and inventory from a PIM or ERP, manage carts, checkout, orders and returns, work customer and company accounts, and react to commerce webhooks in real time. - [Emporix Orchestration Engine Integration](https://flowrunner.ai/integrations/emporix-oe.md): Emporix Orchestration Engine runs value streams inside the Emporix Commerce Orchestration Platform, and its external surface is a CloudEvents receiver. Agents send signed events that start a new value stream instance or wake one paused on a human decision or a slow third party. - [Encharge Integration](https://flowrunner.ai/integrations/encharge.md): Drive Encharge marketing automation from your flows. Agents upsert people by email, apply and remove tags for segmentation, track product events on a contact's timeline, and enroll contacts into onboarding, nurture, or re-engagement flows. - [EnforcedFlow Integration](https://flowrunner.ai/integrations/enforcedflow.md): Route work to the right person with EnforcedFlow's round-robin engine, picking the next available agent from a group while honoring availability rules. Agents assign ownership without a manager arbitrating every handoff. - [Engage Integration](https://flowrunner.ai/integrations/engage.md): Sync customers and send lifecycle messages with Engage, covering customers, accounts, events, and campaigns. Agents keep lifecycle messaging driven by product behavior rather than a periodic list upload. - [EngageBay Integration](https://flowrunner.ai/integrations/engagebay.md): Manage EngageBay CRM records end to end. Agents create, update, and delete contacts, deals, tasks, and companies, tag, note, and score contacts to prioritize outreach, and move deals across pipeline tracks. - [Enginemailer Integration](https://flowrunner.ai/integrations/enginemailer.md): Enginemailer is a Malaysian email marketing and transactional email platform. Agents submit transactional email, manage subscribers and custom fields in batches, and build, assign, and send campaigns. - [Microsoft Entra ID Integration](https://flowrunner.ai/integrations/entra-id.md): Connect AI agents to Microsoft Entra ID (formerly Azure Active Directory). Agents create and manage users and groups, control membership and ownership, invite external guests, and audit directory roles, applications, and service principals. - [ERPLY Books Integration](https://flowrunner.ai/integrations/erply-books.md): ERPLY Books is Estonian double entry accounting software. Agents manage accounts, contacts, and documents, confirm and email invoices, send e-invoices, match payments, and import bank statements. - [ERPNext Integration](https://flowrunner.ai/integrations/erpnext.md): Run generic CRUD, querying, and remote-method calls across every ERPNext (Frappe) DocType, including Customer, Sales Order, Item, Sales Invoice, and any custom DocType. - [eSignatures.com Integration](https://flowrunner.ai/integrations/esignatures-io.md): eSignatures.com turns a Markdown template into a contract, fills its placeholders, and collects signatures in a single call. Agents create and send contracts, save drafts for human approval, manage signers and templates, and embed signing in your own product. - [eSIM Access Integration](https://flowrunner.ai/integrations/esim.md): Connect AI agents to eSIM Access, a global eSIM reseller API. Agents browse the data plan catalog, place orders, provision eSIMs to travelers, and set the webhook URL that reports activation and usage. - [EspoCRM Integration](https://flowrunner.ai/integrations/espo-crm.md): Read and write records in a self-hosted EspoCRM install. Agents run CRUD on any entity type including custom ones, link and unlink related records, and use typed actions for contacts, accounts, leads, opportunities, and activities, with a raw request escape hatch for anything else. - [ESPY Integration](https://flowrunner.ai/integrations/espy.md): Connect AI agents to the ESPY IRBIS OSINT API for people, data, and compliance lookups. Agents run people searches, check breach and deep web exposure, analyze images, and screen subjects against compliance lists. - [Etsy Integration](https://flowrunner.ai/integrations/etsy.md): Run an Etsy shop from your flows. Agents create and publish listings with images and inventory, sync orders into fulfillment and accounting systems, push tracking numbers back to buyers, and monitor reviews and the payment ledger through the Etsy Open API v3. - [Eventbrite Integration](https://flowrunner.ai/integrations/eventbrite.md): Create and publish Eventbrite events, manage ticket classes and venues, and sync attendees and orders into your other systems through the Eventbrite API. - [EventSquare Integration](https://flowrunner.ai/integrations/eventsquare.md): Connect AI agents to EventSquare, a Belgian event ticketing platform. Agents read store inventory, build and adjust carts, and read completed orders, handing checkout back to EventSquare's own hosted flow. - [Eventzilla Integration](https://flowrunner.ai/integrations/eventzilla.md): Connect AI agents to Eventzilla, an event registration and ticketing platform. Agents pull attendees, orders, and transactions, manage free ticket types, check attendees in, confirm or cancel orders, and open or close ticket sales for an event. - [Everhour Integration](https://flowrunner.ai/integrations/everhour.md): Track time in Everhour from your flows. Agents create tasks in the right project, log time when work completes elsewhere, start and stop live timers, and pull team wide time records for a date range to feed reporting, billing, or payroll. - [Evernote Integration](https://flowrunner.ai/integrations/evernote.md): Work with Evernote notes and notebooks through a personal developer token. Agents list notebooks, search notes with Evernote's search grammar, read a note's full content, and create notebooks and notes from a flow. - [Xodo Sign Integration](https://flowrunner.ai/integrations/eversign.md): Xodo Sign, formerly eversign, is Apryse's electronic signature platform. Agents send documents for signature, reuse templates, drive bulk sends from a CSV, download completed PDFs, and read the audit trail behind every document. - [EverWebinar Integration](https://flowrunner.ai/integrations/everwebinar.md): Register people for evergreen webinars in EverWebinar. Agents list automated webinars, look up a webinar's schedules and session slots, and register attendees to a session, capturing each person's unique join link for follow-up. - [eWay-CRM Integration](https://flowrunner.ai/integrations/eway-crm.md): Connect AI agents to eWay-CRM, a Czech CRM that runs inside Outlook. Agents manage companies, contacts, deals, projects, and tasks, and write journal entries against any of them. - [Exa Integration](https://flowrunner.ai/integrations/exa-ai.md): Give AI agents meaning-based web search through Exa. Agents search by neural embeddings or keywords, extract full page contents and summaries, find pages similar to a known URL, get cited answers to natural language questions, and run autonomous multi-source research tasks. - [Exact Online Integration](https://flowrunner.ai/integrations/exact-online.md): Connect AI agents to Exact Online, the Dutch cloud ERP and accounting platform. Agents manage CRM accounts and contacts, keep logistics items in step with external catalogs, raise draft sales invoices and orders, and pull general ledger data for reporting and reconciliation. - [Exact Spotter Integration](https://flowrunner.ai/integrations/exact-spotter.md): Connect AI agents to Exact Spotter, a lead qualification and pre-sales engagement platform from Exact Sales. Agents create and route leads, manage the pre-sales funnel, and read the meetings booked between pre-sales and sales. - [Extra Mobile Integration](https://flowrunner.ai/integrations/exm.md): Search calls, pull AI call analysis and recordings, and manage bookings on Extra Mobile Israeli telephony. Agents attach call evidence to the record that the conversation was actually about. - [Expedy Integration](https://flowrunner.ai/integrations/expedy-by-makemarket.io.md): Drive Expedy cloud printing and digital signage, pushing order tickets straight to a thermal printer in a kitchen, bar, or warehouse. Agents put the ticket where the work happens without a person relaying it. - [Expensify Integration](https://flowrunner.ai/integrations/expensify.md): Automate expense operations in Expensify. Agents create reports and pre-populate them with expense line items, push transactions in from other systems, export report data as JSON, CSV, or PDF for reconciliation, list policies, and bulk update employees and approval managers. - [Experian Aperture Integration](https://flowrunner.ai/integrations/experian-aperture.md): Experian Aperture is Experian's data quality API for address, email and phone validation. Agents validate and format addresses interactively, verify emails and phone numbers in real time, and run bulk cleansing batches on whole lists. - [Explorium AgentSource Integration](https://flowrunner.ai/integrations/explorium-agentsource.md): Explorium AgentSource is a B2B data platform built for AI agents and go-to-market teams. Agents match and enrich companies and prospects across twenty datasets, search by firmographic filters, track business signal events and run AI research jobs with a defined output schema. - [ezeep Blue Integration](https://flowrunner.ai/integrations/ezeep-blue-printing.md): Send documents from a flow straight to any cloud-connected printer with ezeep Blue, choosing copies, color, duplex, and paper format. Agents print picking lists, labels, and signed paperwork where a physical copy is still required. - [EZOfficeInventory Integration](https://flowrunner.ai/integrations/ezofficeinventory.md): Connect AI agents to EZOfficeInventory, an asset tracking and maintenance platform. Agents check assets in and out, reserve them ahead of need, move stock between locations, and raise work orders when maintenance is due. - [EZRentOut Integration](https://flowrunner.ai/integrations/ezrentout.md): Connect AI agents to EZRentOut, an equipment rental management platform. Agents build rental orders, price them, move them through the rental lifecycle, and keep customer and business contact records current. - [Facebook Integration](https://flowrunner.ai/integrations/facebook.md): Publish and manage Facebook Page posts and photos, read and moderate comments, like content, and pull Page and post insights via the Facebook Graph API. - [Facebook Catalogs Integration](https://flowrunner.ai/integrations/facebook-catalogs.md): Manage the Meta product catalogs behind Facebook and Instagram Shops and dynamic ads. Agents sync external product feeds with batch operations, add and update individual products with price and availability, and browse catalogs and product sets through the Graph API. - [Facebook Conversions API Integration](https://flowrunner.ai/integrations/facebook-conversions-api.md): Send server-side conversion events to the Meta Conversions API. Agents track purchases, leads, and any standard or custom funnel event, batch historical conversions, and deduplicate against browser Pixel events, with customer data automatically SHA-256 hashed to Meta's rules. - [Facebook Custom Audiences Integration](https://flowrunner.ai/integrations/facebook-custom-audiences.md): Manage Meta ad targeting from your workflows. Agents create and maintain Facebook Custom Audiences, add or remove hashed user lists, and keep ad account audiences synced with your CRM data. - [Facebook Lead Ads Integration](https://flowrunner.ai/integrations/facebook-lead-ads.md): Build and manage Meta ad campaigns and lead forms. Agents create campaigns, ad sets, ads, and lead forms. Retrieve performance insights across the full campaign structure. - [Facebook Messenger Integration](https://flowrunner.ai/integrations/facebook-messenger.md): Connect AI agents to Facebook Messenger through your Facebook Pages. Agents send text, media, button and generic templates, quick replies, and typing indicators, read conversations, and configure the page's greeting, Get Started button, and persistent menu. - [FaceUp Integration](https://flowrunner.ai/integrations/faceup.md): Connect AI agents to FaceUp, the anonymous whistleblowing and reporting platform. Agents pull aggregate report statistics and run custom GraphQL queries so compliance teams can monitor reporting activity from their workflows. - [Factorial Integration](https://flowrunner.ai/integrations/factorial.md): Connect AI agents to Factorial HR. Agents onboard employees, create and track leave requests, review attendance shifts, and pull team and leave-type data to keep HR workflows moving. - [Fakturoid Integration](https://flowrunner.ai/integrations/fakturoid.md): Fakturoid is Czech online invoicing. Agents manage subjects, issue invoices across six document kinds, log expenses, track inventory with stock moves, and run invoice templates on recurring schedules. - [fal.ai Integration](https://flowrunner.ai/integrations/fal-ai.md): Run generative media models on fal.ai directly from your workflows. Agents generate and transform images, create video from text or images, upload files, and manage queued jobs across hosted models with sync and async execution. - [FAPI Integration](https://flowrunner.ai/integrations/fapi.md): FAPI is a Czech invoicing and sales-automation platform for course sellers, membership sites, and small businesses. Agents create and send invoices, record payments, manage clients, order forms, and discount codes, apply vouchers, and pull sales statistics. - [FAPI Member Integration](https://flowrunner.ai/integrations/fapi-member.md): FAPI Member is the WordPress plugin that turns FAPI orders into member sections, drip-fed courses, and paid membership areas. Agents grant and update memberships, manage sections and unlock rules, create users, and read churn and activity statistics. - [FareHarbor Integration](https://flowrunner.ai/integrations/fareharbor.md): Connect AI agents to FareHarbor, a booking platform for tours and activities. Agents read the catalog and live availability, create bookings, and check guests in on the day. - [FastBill Integration](https://flowrunner.ai/integrations/fastbill.md): FastBill is German invoicing and bookkeeping. Agents manage customers, contacts, and articles, take invoices through their full lifecycle, send estimates, log project work time, and track revenues, expenses, and recurring invoices. - [FastField Integration](https://flowrunner.ai/integrations/fastfield.md): FastField is a mobile forms, inspection and dispatch platform from Merge Mobile. Agents publish form templates, take completed submissions apart field by field, dispatch work to crews and recall it, sync offline data tables and choice lists, and manage users and groups. - [Fathom Integration](https://flowrunner.ai/integrations/fathom.md): Turn Fathom AI notetaker recordings into workflow fuel. Agents list meetings, pull transcripts and AI summaries, browse teams and members, and register webhooks so every recorded call can trigger downstream follow-up. - [Fatture in Cloud Integration](https://flowrunner.ai/integrations/fatture-in-cloud.md): Fatture in Cloud is the leading Italian invoicing platform from TeamSystem, built around electronic invoicing. Agents file documents with the Sistema di Interscambio, pick up supplier e-invoices as they arrive, and manage clients, products, and receipts in between. - [Favro Integration](https://flowrunner.ai/integrations/favro.md): Favro is collaborative planning where collections hold boards and backlogs, and columns hold cards. Agents create and move cards, set custom field values on cards, manage collections and widgets, and react to card changes as they land. - [Firebase Cloud Messaging Integration](https://flowrunner.ai/integrations/fcm.md): Send push notifications through Firebase Cloud Messaging. Agents deliver messages to individual device tokens, topics, or condition expressions, and manage topic subscriptions, authenticating with a Google service account. - [FeedHive Integration](https://flowrunner.ai/integrations/feedhive.md): Connect AI agents to FeedHive, an AI-assisted social media scheduling platform. Agents draft and schedule posts across connected socials, organize them with labels, and read the analytics behind each one. - [Feedier Integration](https://flowrunner.ai/integrations/feedier.md): Feedier is a French customer feedback and experience management platform that pulls in surveys, reviews, email, SMS, chat and call center feedback and enriches it with AI topics and sentiment. Agents query feedback with FQL segments, read surveys, manage users and invitations, add recipients to push campaigns, and react to new feedback. - [Feedly Integration](https://flowrunner.ai/integrations/feedly.md): Give agents a monitored view of the web through Feedly. They subscribe to feeds, pull stream contents and entries, tag and save articles, and mark items read, turning your Feedly boards into a research pipeline. - [Feishu Drive Integration](https://flowrunner.ai/integrations/feishdrive.md): Feishu Drive is the cloud storage behind Feishu, ByteDance's workplace suite. Agents upload and organize files, export live Feishu documents to PDF or Word, import Office files as native documents, manage collaborators and share links, and read versions and comments. - [Feishu Group Robot Integration](https://flowrunner.ai/integrations/feishu-group-robot.md): Post messages into a Feishu group chat through the custom bot webhook. Agents deliver alerts and approvals to the group that owns the decision, with optional signature verification on every send. - [Feishu Base Integration](https://flowrunner.ai/integrations/feishubase.md): Feishu Base is the Bitable database inside Feishu. Agents create bases and tables, manage fields and views, search, create, update and delete records in bulk, and administer roles on bases with advanced permissions. - [Feishu Docs Integration](https://flowrunner.ai/integrations/feishudocument.md): Feishu Docs is the block-based document editor in Feishu, ByteDance's workplace suite. Agents create documents, read them as plain text or blocks, write generated reports with headings and lists intact, and convert Markdown or HTML into native Feishu documents. - [FetchFox Integration](https://flowrunner.ai/integrations/fetchfox.md): Scrape the web with FetchFox by describing the fields you want in plain English. FetchFox crawls, reads, and returns structured records, so agents collect data from sites that publish no API. - [Fibery Integration](https://flowrunner.ai/integrations/fibery.md): Work with your Fibery workspace at the schema level. Agents discover databases, query entities with filters, create, update, and delete records, and run raw Fibery commands for anything the dedicated operations do not cover. - [Fidoo Integration](https://flowrunner.ai/integrations/fidoo-expense-management.md): Fidoo is a Czech corporate expense-management platform with prepaid company cards, expense approval, and travel reports. Agents load and unload cards, manage users, approve expenses, read travel requests and reports, and read the accounting dimensions each item is coded against. - [Figma Integration](https://flowrunner.ai/integrations/figma.md): Read Figma files, export images, manage comments and reactions, browse teams and projects, and inspect components and styles from FlowRunner agents through the Figma REST API. - [FileMaker Integration](https://flowrunner.ai/integrations/filemaker.md): Connect AI agents to a FileMaker Server or FileMaker Cloud database through the Data API. Agents read, create, edit, find, and delete records, run scripts, and inspect layout metadata. - [Files.com Integration](https://flowrunner.ai/integrations/files-com.md): Automate file operations on your Files.com site. Agents upload, download, move, copy, and delete files, manage folders, create password-protected share bundles, and list users for governed file exchange. - [Filestack Integration](https://flowrunner.ai/integrations/filestack.md): Handle file ingestion and processing through Filestack. Agents store files from URLs, read metadata, build transformation URLs for resizing and conversion, run tag detection, and delete assets when they expire. - [Filestage Integration](https://flowrunner.ai/integrations/filestage.md): Filestage is the review and approval platform for creative content. Agents create projects and sections, upload files and versions, invite reviewer groups, collect decisions and threaded comments with markers, pull review reports, and react to review events in real time. - [FillFaster Integration](https://flowrunner.ai/integrations/fillfaster.md): FillFaster turns a PDF into a fillable, signable web form. Agents create prefilled links per recipient, track what happens to them, fetch the signed PDF, fill a PDF outright with no signing step, and react to form events. - [Fillout Integration](https://flowrunner.ai/integrations/fillout.md): Work with Fillout forms and their data. Agents list forms, retrieve and create submissions, and delete records, making form responses available to downstream workflow steps the moment you need them. - [finaX Integration](https://flowrunner.ai/integrations/finax.md): finaX is an e-invoicing API for XRechnung, ZUGFeRD, Factur-X, and EN 16931 documents. Agents convert between JSON invoice models, CII or UBL XML, and ZUGFeRD PDFs in every direction, and validate the result against EN 16931 before it goes out. - [Findymail Integration](https://flowrunner.ai/integrations/findymail.md): Find and verify B2B work emails and look up direct phone numbers with Findymail. Agents complete a contact record and confirm it is deliverable before the sequence spends a send on it. - [Fingertip Integration](https://flowrunner.ai/integrations/fingertip.md): Connect AI agents to Fingertip, the all-in-one platform for business sites with bookings, a store, forms, and contacts. Agents read sites and blog posts, create contacts, accept, decline, or reschedule bookings, and list orders so site operations run without manual clicks. - [finlight Integration](https://flowrunner.ai/integrations/finlight.md): finlight is a financial news API with sentiment analysis and resolved company entities. Agents search articles by company, ticker, or keyword, look up a story by URL, and browse the catalog of news sources to feed research and alerting flows. - [Fio banka Integration](https://flowrunner.ai/integrations/fio.md): Fio banka is a Czech bank with a token-authenticated API for account movements and payment orders. Agents pull new transactions and statements, track card transactions, and stage domestic, euro, and foreign payments or whole batches, which a person then authorizes in internet banking. - [Firecrawl Integration](https://flowrunner.ai/integrations/firecrawl.md): Give agents production-grade web data through Firecrawl. They scrape single URLs into clean Markdown, crawl entire sites, map site structure, search the web, and extract structured JSON with LLM-powered extraction, with credit and token usage tracking built in. - [Fireflies.ai Integration](https://flowrunner.ai/integrations/fireflies-service.md): Bring Fireflies.ai meeting intelligence into FlowRunner. Agents retrieve transcripts and AI summaries, extract action items, search past meetings, and send the notetaker bot to live calls. - [FireText Integration](https://flowrunner.ai/integrations/firetext.md): Connect AI agents to FireText, a UK SMS marketing and transactional messaging gateway. Agents send and schedule SMS, manage contacts, groups, and keywords, and pull delivery and click reports so UK text campaigns run and report automatically. - [FirstPromoter Integration](https://flowrunner.ai/integrations/firstpromoter.md): Run affiliate operations in FirstPromoter without manual bookkeeping. Agents create and update promoters, track referral signups and sales, and review referrals and commissions so payouts stay accurate. - [FIT CTU KOS Integration](https://flowrunner.ai/integrations/fit-cvut-kos.md): Connect AI agents to KOSapi, the study information system of the Czech Technical University in Prague. Agents list courses, parallels, exam terms, semesters, and programmes and look up rooms and teachers so academic catalog data flows into schedules and reports. - [Flaree Integration](https://flowrunner.ai/integrations/flaree.md): Connect AI agents to Flaree, the employee recognition platform where colleagues send each other point-bearing cards. Agents trigger recognition off real milestones instead of a quarterly reminder. - [Flashy Integration](https://flowrunner.ai/integrations/flashyapp.md): Run omnichannel marketing automation on Flashy, the Israeli platform, through its official API. Agents manage contacts and segments, trigger campaigns, and read engagement back into the customer record. - [Flatfox Integration](https://flowrunner.ai/integrations/flatfox.md): Connect AI agents to Flatfox, the Swiss rental and sale marketplace and property management platform. Agents read public listings and similar-property matches, pull tenant applications and inventory, and update the status of a property ticket. - [Fleep Integration](https://flowrunner.ai/integrations/fleep.md): Send messages, manage conversations and hooks, and search chat history with Fleep, a team messenger with built-in task tracking. Agents post updates into the thread that owns the work. - [ABRA Flexi Integration](https://flowrunner.ai/integrations/flexibee.md): ABRA Flexi, formerly FlexiBee, is a Czech accounting and ERP system built on generic registers. Agents read and write any register, issue invoices and orders, maintain price lists, track bank, cash, and stock movements, and react to changes in real time. - [Flexie CRM Integration](https://flowrunner.ai/integrations/flexie-crm.md): Connect AI agents to Flexie CRM, a CRM and workflow automation platform with fully custom entities. Agents create contacts, leads, deals, and cases, create records in custom entities, add records to lists, and read the workflow definitions that act on them. - [Flexmail Integration](https://flowrunner.ai/integrations/flexmail.md): Flexmail is a Belgian email marketing and transactional email platform. Agents manage contacts, interests, preferences, and opt-ins, run bulk imports and segments, send transactional messages, and track message events and webhooks across both APIs. - [Flickr Integration](https://flowrunner.ai/integrations/flickr.md): Search and read Flickr's photo library from your workflows. Agents search photos, fetch photo info and sizes, look up users and their public photos, browse interestingness, and call any other read-level Flickr REST method. - [Fliki Integration](https://flowrunner.ai/integrations/fliki.md): Produce AI voiceovers with Fliki from inside a flow. Agents browse languages, dialects, voices, and styles, generate audio synchronously or as background tasks, poll generation status, and check account usage. - [Flixcheck Integration](https://flowrunner.ai/integrations/flixcheck.md): Flixcheck is a German platform for digital customer interaction: a check is a personalized form sent by SMS, email or WhatsApp and returned as structured data, files and signatures. Agents create and send checks, read the answers and documents that come back, download the PDF, and manage templates, teams and users. - [Float Integration](https://flowrunner.ai/integrations/float.md): Keep resource plans current in Float. Agents manage people, projects, tasks, and clients, and read time-off records so scheduling decisions reflect real capacity. - [Flock Integration](https://flowrunner.ai/integrations/flock.md): Post messages, manage channels, and look up people with Flock. Agents deliver workflow notifications and route questions to the person who can answer them. - [Flodesk Integration](https://flowrunner.ai/integrations/flodesk.md): Manage your Flodesk email list from your workflows. Agents create or update subscribers, add them to segments, remove them when they churn, and read segment membership to drive targeted sends. - [Flow Blockchain Integration](https://flowrunner.ai/integrations/flow-blockchain.md): Flow is a public blockchain with an open HTTP Access API. Agents read blocks, collections, transactions and their results, accounts, keys, and events, execute read-only Cadence scripts, and submit signed transactions to mainnet or testnet. - [Flow XO Integration](https://flowrunner.ai/integrations/flow-xo.md): Automate Flow XO chatbots, messaging end users on any channel and managing contacts, segments, transcripts, knowledge bases, and translations. Agents keep bot content current and act on what conversations reveal. - [FLOWii Integration](https://flowrunner.ai/integrations/flowii.md): Connect AI agents to FLOWii, a Slovak CRM and business management suite for small businesses. Agents create partners and business cases, log communications, assign tasks, and read the orders and goods catalog behind each deal. - [Flowlu Integration](https://flowrunner.ai/integrations/flowlu.md): Run business operations in Flowlu end to end. Agents manage CRM accounts and leads, create projects and tasks, and read invoices, connecting sales activity to delivery work in one system. - [FluentCRM Integration](https://flowrunner.ai/integrations/fluentcrm.md): Manage FluentCRM subscribers inside your WordPress install. Agents create or update contacts, apply and detach tags and lists, manage companies, and read campaign data across 20 actions. - [Fluents AI Integration](https://flowrunner.ai/integrations/fluents-ai.md): Place and manage AI voice-agent phone calls with the Fluents Hosted API, starting outbound calls, ending live calls, and pulling transcripts. Agents run phone work and keep the transcript with the record. - [FogBugz Integration](https://flowrunner.ai/integrations/fogbugz.md): FogBugz, now sold as Manuscript, is issue tracking and case management with wikis and time tracking attached. Agents open, assign, edit and resolve cases, manage projects, areas and milestones, and keep the case queue current with what happens in other systems. - [Folderit Integration](https://flowrunner.ai/integrations/folderit-dms.md): Folderit is a cloud document management system with folders, versions, metadata, tags, sharing, and a full audit trail. Agents file incoming documents, build client folder structures, search by content, tag and describe files, and read who did what. - [folk Integration](https://flowrunner.ai/integrations/folk.md): Keep your folk CRM in sync automatically. Agents create, update, and delete people and companies, manage group membership, read custom fields, and look up workspace users across 17 actions. - [Follow Up Boss Integration](https://flowrunner.ai/integrations/follow-up-boss.md): Work leads in Follow Up Boss the moment they arrive. Agents create and update people, log events and notes, manage deals, and read pipelines, stages, and users to route real estate follow-up. - [Fomo Integration](https://flowrunner.ai/integrations/fomo.md): Push purchase, signup, and review events into a live Fomo social proof feed and manage the templates that render them. Agents surface real activity on the site as it happens rather than on a scheduled refresh. - [ForceManager Integration](https://flowrunner.ai/integrations/force-manager.md): Connect AI agents to ForceManager (now Sage Sales Management), a mobile-first field sales CRM. Agents create accounts and contacts, open opportunities, log field activities, and raise sales orders from the road. - [Force24 Integration](https://flowrunner.ai/integrations/force24.md): Run marketing automation on Force24, the UK platform, through its official API. Agents manage contacts and lists, trigger journeys, and read engagement data back for scoring and routing. - [Form.taxi Integration](https://flowrunner.ai/integrations/form-taxi.md): Form.taxi is a GDPR-compliant form backend for static websites. Agents submit entries, read stored submissions, create endpoints, and react to each new submission through a verified webhook. - [Formaloo Integration](https://flowrunner.ai/integrations/formaloo.md): Formaloo is a no-code form builder and customer data platform. Agents build forms and fields, read and write submissions, organize boards and folders, run email and WhatsApp campaigns, generate PDFs, and react the moment a form is submitted, edited or paid. - [Formbricks Integration](https://flowrunner.ai/integrations/formbricks.md): Automate survey operations in Formbricks. Agents create and delete surveys, list and inspect responses, submit new responses, and read contacts to close the loop between feedback and action. - [FormCan Integration](https://flowrunner.ai/integrations/formcan.md): FormCan, formerly PlatoForms, is an online and PDF form builder. Agents read forms and fields, search and update submissions, create pre-filled invitations, download generated PDFs, attachments and signature certificates, keep dropdown options in sync with an external system, and react on every submission. - [Formcrafts Integration](https://flowrunner.ai/integrations/formcrafts.md): Formcrafts is a drag and drop form, survey and quiz builder from Dusseldorf, Germany. Agents read workspaces and forms, pull live responses, resolve them against the form definition, download respondent uploads, and react when a response arrives. - [Formidable Forms Integration](https://flowrunner.ai/integrations/formidable-forms.md): Work with Formidable Forms data on your WordPress site. Agents list forms, and create, update, and delete entries through the REST API using WordPress Application Password auth. - [Formify Integration](https://flowrunner.ai/integrations/formify.md): Formify is a Swedish electronic signature and digital contract platform. Agents upload or merge PDFs, send them for signature with digital ink, BankID or ID verification, pre-fill and harvest form field values, chase unsigned parties, revoke documents, download the sealed PDF, and react to every document event. - [Form.io Integration](https://flowrunner.ai/integrations/formio.md): Manage Form.io forms, their field definitions, and the submissions collected against them, pushing external records in and reading submission data out to other systems. - [Formlets Integration](https://flowrunner.ai/integrations/formlets.md): Formlets by Oxopia is an online form and survey builder. Agents list and inspect forms, read field designs, report on form reach and state, build publish links, and react when a form is created or taken offline. - [Formsite Integration](https://flowrunner.ai/integrations/formsite.md): Read Formsite form data into your workflows. Agents list forms, inspect form items, and retrieve results individually or in bulk for processing downstream. - [Formspree Integration](https://flowrunner.ai/integrations/formspree.md): Work with Formspree form submissions. Agents submit entries to a form endpoint and read stored submissions and forms through the management API. - [Formstack Integration](https://flowrunner.ai/integrations/formstack.md): Manage Formstack forms, submissions, fields, folders, and webhooks, and sync new submissions into a CRM, spreadsheet, or database as they arrive. - [ForSign Integration](https://flowrunner.ai/integrations/forsign.md): ForSign is a Brazilian electronic signature platform. Agents upload a PDF, create a signing operation with members and signing order, request identity documents or a selfie before signature, collect form answers, and download the signed files. - [Fortnox Integration](https://flowrunner.ai/integrations/fortnox.md): Connect AI agents to Fortnox, the Swedish cloud accounting platform. Agents create and bookkeep invoices, manage customers, articles, orders, offers, and supplier invoices, and work with vouchers, accounts, and financial years across 39 actions. - [4leads Integration](https://flowrunner.ai/integrations/four-leads.md): Connect AI agents to 4leads, a German CRM and marketing automation platform. Agents sync contacts, apply tags and custom fields, enroll them in campaigns, and record opt-ins for consent tracking. - [Foursquare Integration](https://flowrunner.ai/integrations/foursquare.md): Give AI agents location intelligence from the Foursquare Places API. Agents search places by text or coordinates, pull rich venue details like hours, ratings, and price tier, run type-ahead autocomplete, and surface user photos and tips for any venue. - [Foxentry Integration](https://flowrunner.ai/integrations/foxentry.md): Validate, correct, and enrich customer data in real time across all five Foxentry modules: email, phone, name, address, and company. Agents clean a record at capture so downstream systems never inherit the mess. - [Frame.io Integration](https://flowrunner.ai/integrations/frame-io.md): Connect AI agents to a Frame.io video review workspace. Agents navigate teams, projects, and asset hierarchies, create folders and file placeholders, and read or post review comments, including notes pinned to a specific timestamp in a video. - [FreeFinance Integration](https://flowrunner.ai/integrations/free-finance.md): FreeFinance, also sold as FinanzFenster, is Austrian online accounting and invoicing software. Agents manage customers, suppliers, and items, create and finalize invoices, record payments, and pull invoice PDFs and XML for each bookkeeping client. - [FreeAgent Integration](https://flowrunner.ai/integrations/freeagent.md): Connect AI agents to FreeAgent, the UK cloud accounting platform. Agents create and update contacts, invoices, estimates, bills, and expenses, explain bank transactions for reconciliation, and pull company data for reporting across 45 actions. - [Freedcamp Integration](https://flowrunner.ai/integrations/freedcamp.md): Freedcamp is a project management suite covering tasks, milestones, issues, time tracking, calendars and files. Agents create and update tasks and issues, log time, manage projects and groups, and keep the calendar and files in step with outside systems. - [Freelo Integration](https://flowrunner.ai/integrations/freelo.md): Freelo is Czech project management built around projects, tasklists and tasks, with time tracking and invoice records attached to the same work. Agents create tasks and subtasks, assign and comment, log time, and mark billed work as invoiced. - [Freepik Integration](https://flowrunner.ai/integrations/freepik.md): Let AI agents source and generate visuals through the Freepik API. Agents search Freepik's stock library of photos, vectors, and icons, pull download and license details for a chosen resource, and generate images from text prompts with Text To Image or high resolution Freepik Mystic. - [FreeScout Integration](https://flowrunner.ai/integrations/freescout.md): Connect AI agents to FreeScout, the open source shared inbox help desk. Agents open and reply to conversations, manage threads across mailboxes, tag them, and keep customer records attached. - [freispace Integration](https://flowrunner.ai/integrations/freispace.md): Connect AI agents to freispace, the post-production scheduling platform. Agents create and update bookings, projects, tasks, resources, staff, and absences so studio schedules stay accurate without manual entry. - [FreJun Integration](https://flowrunner.ai/integrations/frejun.md): Automate calls, users, virtual numbers, and calling lists on FreJun. Agents supply the day's call list from live pipeline data and log outcomes against the right record. - [FreshBooks Integration](https://flowrunner.ai/integrations/freshbooks.md): Connect AI agents to FreshBooks for invoicing, payments, expenses, and project-based time tracking. Agents manage clients, invoices, estimates, expenses, vendors, bills, projects, and pull reports via OAuth2. - [Freshchat Integration](https://flowrunner.ai/integrations/freshchat.md): Connect AI agents to Freshchat, Freshworks' customer messaging platform. Agents open conversations, send messages and private notes, resolve or reassign chats to the right agent or group, and keep users, channels, and agent rosters in sync. - [Freshdesk Integration](https://flowrunner.ai/integrations/freshdesk.md): Connect AI agents to Freshdesk, the Freshworks customer support and helpdesk platform. Agents create and triage tickets, post replies and internal notes, search tickets and contacts, and sync customers as contacts and companies through the Freshdesk REST API v2. - [Freshservice Integration](https://flowrunner.ai/integrations/freshservice.md): Connect AI agents to Freshservice, Freshworks' IT service management platform. Agents automate the full ITSM lifecycle across tickets, changes, problems, releases, CMDB assets, agents, and requesters through the Freshservice API v2. - [Freshworks CRM Integration](https://flowrunner.ai/integrations/freshworks-crm.md): Connect AI agents to Freshworks CRM (formerly Freshsales). Agents manage contacts, sales accounts, and deals, log tasks and appointments, and upsert records without duplicates. - [Front Integration](https://flowrunner.ai/integrations/front-service.md): Connect FlowRunner agents to the Front shared inbox. Agents triage and reply to conversations across email, SMS, and chat, post internal comments, and keep the contact directory in sync. - [Frosty AI Integration](https://flowrunner.ai/integrations/frosty-ai.md): Send prompts through a Frosty AI router for multi-model routing, fallback, and observability. Agents keep working when one provider degrades and keep a record of which model answered. - [Fulcrum Integration](https://flowrunner.ai/integrations/fulcrum.md): Connect AI agents to Fulcrum, the mobile field data collection platform. Agents list forms and their field schemas, create, update, and manage geolocated records, and pull photo attachments captured in the field into downstream systems. - [FullEnrich Integration](https://flowrunner.ai/integrations/fullenrich.md): Let AI agents enrich B2B contacts through FullEnrich's waterfall enrichment API. Agents start bulk enrichment jobs from a name plus company or LinkedIn URL, retrieve verified work emails and phone numbers, and check remaining credits before launching large batches. - [FunnelCockpit Integration](https://flowrunner.ai/integrations/funnelcockpit.md): Connect AI agents to FunnelCockpit, the German marketing funnel builder suite, through its official API. Agents read funnel leads and orders and push buyers into fulfilment and follow-up. - [Fusioo Integration](https://flowrunner.ai/integrations/fusioo.md): Fusioo is a collaborative online database organized as apps containing records, with discussions, files and users alongside the data. Agents search and write records, count and filter them, post discussions and comments, manage files and react to record changes through webhooks. - [1001fx Integration](https://flowrunner.ai/integrations/fx1001.md): Use a curated set of 1001fx utility functions covering PDF and Office document work, conversion, and data handling. Agents close the small gaps in a workflow without standing up a service for each one. - [GakuNin RDM Integration](https://flowrunner.ai/integrations/gakunin-rdm.md): Connect AI agents to GakuNin RDM, the Japanese academic research data management platform. Agents create and update projects, manage contributors, upload and download files, and read registrations so research data stays organized across teams. - [Gamfi Integration](https://flowrunner.ai/integrations/gamfi.md): Connect AI agents to Gamfi, the employee onboarding and lifecycle platform. Agents add employees, start and manage workflow processes, complete actions, and download reports so onboarding runs on schedule with a full paper trail. - [Gamma Integration](https://flowrunner.ai/integrations/gamma-app.md): Let AI agents turn a prompt or source content into polished Gamma decks, documents, webpages, and social posts. Agents create generations, poll them to completion, and return a shareable Gamma link plus optional PPTX, PDF, or PNG exports. - [Ganttic Integration](https://flowrunner.ai/integrations/ganttic.md): Ganttic is resource planning that books people, rooms and machines against projects on a shared timeline. Agents create and reschedule bookings, manage resources and projects, and read utilization so the plan matches real demand. - [GanttPRO Integration](https://flowrunner.ai/integrations/ganttpro.md): GanttPRO is online Gantt chart software where work is a tree of tasks with durations, dependencies and resources laid over a working calendar. Agents open projects from approved plans, add tasks and dependencies, assign resources, log time and post comments. - [GatewayAPI Integration](https://flowrunner.ai/integrations/gatewayapi.md): Connect AI agents to GatewayAPI, a Danish SMS gateway delivering messages worldwide. Agents send and schedule SMS, look up delivery status and phone numbers, manage two-way keywords, and check balance and pricing so high-volume texting stays observable. - [GatherContent Integration](https://flowrunner.ai/integrations/gathercontent.md): Connect AI agents to GatherContent, now Bynder Content Workflow, the content operations platform. Agents create and update content items, apply templates, organize folders, and manage webhooks so structured content moves from plan to CMS without manual handoffs. - [Geckoboard Integration](https://flowrunner.ai/integrations/geckoboard.md): Push live business metrics into Geckoboard KPI dashboards from your flows. Agents define dataset schemas, then keep dashboards current by replacing or appending rows as sales, revenue, or conversion numbers change in source systems. - [Gemini AI Integration](https://flowrunner.ai/integrations/gemini-ai.md): Add Google's multimodal AI to your agent workflows. Agents generate content, analyze documents and images, extract structured data, and process audio and video files through Gemini's Files API. - [Gender API Integration](https://flowrunner.ai/integrations/gender-api.md): Infer the gender most commonly associated with a personal name through the Gender API Unified API. Agents split a full name, look up a first name, and personalize salutations with a stated confidence level. - [Genderize.io Integration](https://flowrunner.ai/integrations/genderize.md): Predict the gender most commonly associated with a first or full name using the Genderize.io frequency database. Every lookup returns a probability, so agents can route low-confidence cases to a person. - [GenerateBanners Integration](https://flowrunner.ai/integrations/generatebanners.md): Render marketing banners, social images, and PDFs from GenerateBanners templates by substituting headlines, colors, and logos. Agents produce on-brand assets per record without a design request. - [geoCapture Integration](https://flowrunner.ai/integrations/geocapture.md): geoCapture is the German field workforce platform for time tracking, GPS tracking, disposition, and orders. Agents manage employees, book time stamps and absences, create orders, tours, and routes, read tracker and GPS records, run approvals, and geocode addresses. - [GetAccept Integration](https://flowrunner.ai/integrations/getaccept.md): Connect AI agents to GetAccept, the digital sales room and e-signature platform. Agents create documents from files or templates, attach recipients, send them for signing, track status through the deal lifecycle, and keep contacts in sync with your CRM. - [GetEmail.io Integration](https://flowrunner.ai/integrations/getemail-io.md): Find and verify professional email addresses with GetEmail.io, using asynchronous lookups from a name plus domain. Agents resolve a contact before the outreach step spends a send on a guess. - [Getform Integration](https://flowrunner.ai/integrations/getform.md): Getform is a form backend for developers: a website form posts to a Getform endpoint and Getform stores, emails and forwards it. Agents submit entries, read stored submissions back out, verify webhook signatures, and react on every new submission. - [Guru Integration](https://flowrunner.ai/integrations/getguru.md): Connect AI agents to Guru, the company knowledge base. Agents search and read verified cards before drafting answers, capture knowledge from other tools as new cards in the right collection, update runbooks programmatically, and archive stale content. - [GetMyInvoices Integration](https://flowrunner.ai/integrations/getmyinvoices.md): GetMyInvoices collects invoices and receipts automatically from supplier portals, email inboxes, and bank transactions. Agents pull the documents it has captured, read their extracted data, tag transactions, and hand the files to the next system in the flow. - [GetProspect Integration](https://flowrunner.ai/integrations/getprospect.md): Find and verify work emails and pull B2B prospect data through GetProspect. Agents enrich a partial lead into a workable record before it enters the pipeline. - [GetResponse Integration](https://flowrunner.ai/integrations/getresponse.md): Connect AI agents to GetResponse. Agents add and update subscribers, search contacts, manage tags and custom fields, and create newsletters and campaigns across your email marketing lists. - [GetTranscribe Integration](https://flowrunner.ai/integrations/gettranscribe.md): Transcribe social video from Instagram, TikTok, YouTube, Facebook, X, Pinterest, and Google Drive into text, and resolve direct media download links. Agents turn published video into searchable, quotable copy. - [Ghost Integration](https://flowrunner.ai/integrations/ghost.md): Connect AI agents to Ghost. Agents create, publish, and manage posts and pages, curate tags, onboard members and tiers, upload images, and read published content through the Admin and Content APIs. - [GIPHY Integration](https://flowrunner.ai/integrations/giphy.md): Let AI agents drop the right GIF into any message. Agents search GIPHY's library of GIFs and stickers, surface trending content, translate a phrase into a matching GIF, and pull full rendition URLs and metadata for a known GIF ID. - [GIRITON Integration](https://flowrunner.ai/integrations/giriton.md): Connect AI agents to GIRITON, a Czech attendance and time tracking HR system. Agents update employees, record attendance events, add vacations, and process attendance requests so time data reaches payroll without manual chasing. - [Gist Integration](https://flowrunner.ai/integrations/gist.md): Manage contacts and lifecycle marketing with Gist, covering events, tags, segments, forms, and drip campaigns. Agents keep segmentation driven by product behavior rather than a stale import. - [GitHub Integration](https://flowrunner.ai/integrations/github.md): Connect AI agents to GitHub via OAuth2. Agents manage repos, issues, pull requests, files, releases, organizations, teams, secrets, variables, and trigger and monitor Actions workflows. - [GitLab Integration](https://flowrunner.ai/integrations/gitlab.md): Manage GitLab projects, issues, merge requests, repository content, CI/CD pipelines, and releases from your flows over the GitLab REST API v4. Works with GitLab SaaS and self-managed instances. - [Glide Integration](https://flowrunner.ai/integrations/glide.md): Read and write the data behind your Glide apps through the Glide Tables API. Agents add, update, and fetch rows in Big Tables, inspect and create table schemas, and sync records from CRMs, stores, and forms into a no-code app. - [Global Exchange Rates Integration](https://flowrunner.ai/integrations/global-exchange-rates.md): Global Exchange Rates publishes currency rates from many named providers instead of one blended feed. Agents list currencies and providers, fetch latest and historical rates, and convert amounts with a source you can audit. - [Gmail Integration](https://flowrunner.ai/integrations/gmail-service.md): Full Gmail integration for FlowRunner agents. Agents send and draft email, manage labels, save attachments to storage, and trigger flows on new messages, threads, and attachments. - [Gmelius Integration](https://flowrunner.ai/integrations/gmelius.md): Manage shared inboxes, kanban boards, and email sequences with Gmelius. Agents triage a shared mailbox, move work onto a board, and keep sequences aligned with what the thread actually says. - [Go4Clients Integration](https://flowrunner.ai/integrations/go4clients.md): Connect AI agents to Go4Clients, a multichannel campaign platform for SMS, voice, and email. Agents send messages and calls, run drip and verification campaigns, manage contacts and blacklists, and pull channel analytics so multichannel outreach runs from a single flow. - [GoAffPro Integration](https://flowrunner.ai/integrations/goaffpro.md): Run a Shopify affiliate programme with the GoAffPro Admin API, onboarding and approving affiliates and recording referred orders. Agents pay and police the programme against real order data. - [GoCanvas Integration](https://flowrunner.ai/integrations/gocanvas.md): Pull completed field submissions out of GoCanvas, the mobile forms platform. Agents list forms and reference datasets, retrieve submission records with their captured field data, and route inspections, work orders, and checklists into downstream systems. - [GoCardless Integration](https://flowrunner.ai/integrations/gocardless.md): Connect AI agents to GoCardless bank-to-bank payments. Agents set up Direct Debit mandates, take one-off, recurring, and instalment payments, issue refunds, reconcile payouts, and run hosted or own-page authorisation flows, with six event triggers that react to payment, mandate, subscription, refund, payout, and billing-request activity. - [GoDaddy Integration](https://flowrunner.ai/integrations/godaddy.md): Manage GoDaddy domains end to end from your flows. Agents check availability, register and renew domains, suggest brandable names, add or replace DNS records, and keep registrant contacts in sync through the GoDaddy Domains API. - [GoHighLevel Integration](https://flowrunner.ai/integrations/gohighlevel.md): Manage GoHighLevel contacts, opportunities, conversations, calendar, tasks, notes, tags, invoices, and products, automating lead management, deal tracking, and multi-channel messaging. - [Golemio Integration](https://flowrunner.ai/integrations/golemio.md): Golemio is the City of Prague's open data platform, operated by Operator ICT. Agents read air quality, traffic and pedestrian counts, public transport departures, parking, waste stations, city districts and municipal facilities as structured data. - [Gong Integration](https://flowrunner.ai/integrations/gong.md): Bring Gong revenue intelligence into your flows. Agents read calls and transcripts, pull rep-activity stats, run GDPR lookups and erasures, sync CRM objects, and automate Engage flows. - [Good Grants Integration](https://flowrunner.ai/integrations/good-grants.md): Connect AI agents to Good Grants, the grantmaking platform from Creative Force. Agents receive and update applications, assign reviewers, record allocations and payment records, create funds, and read the categories, chapters, and statuses a program is configured with. - [GoodBarber Integration](https://flowrunner.ai/integrations/goodbarber.md): GoodBarber Commerce is the ecommerce side of the no-code mobile app builder. Workflows manage products, variants, images, and collections, read and update orders and shipping, and reach customers with push and marketing tools. - [Google Ads Integration](https://flowrunner.ai/integrations/google-ads.md): Connect AI agents to Google Ads. Agents run GAQL reports, list accessible accounts and campaigns, pull campaign metrics, and pause or resume campaigns on spend or performance thresholds. - [Google Analytics Integration](https://flowrunner.ai/integrations/google-analytics.md): Run standard and realtime Google Analytics 4 reports, discover available dimensions and metrics, and list accounts and properties through the GA4 Data and Admin APIs with read-only OAuth access. - [Google Books Integration](https://flowrunner.ai/integrations/google-books.md): Search and browse books and magazines through the Google Books catalog: look up volume metadata by ISBN, title, or author, run keyword searches, and read public users' bookshelves. - [Google Business Profile Integration](https://flowrunner.ai/integrations/google-business-profile.md): Manage Google Business Profile accounts, business locations, and customer reviews. Monitor and reply to reviews, keep location details in sync, and aggregate ratings across locations. - [Google Calendar Integration](https://flowrunner.ai/integrations/google-calendar.md): Automate scheduling and calendar management. Agents create, update, retrieve, and delete Google Calendar events as part of operational workflows. - [Google Chat Integration](https://flowrunner.ai/integrations/google-chat.md): Post and manage messages, send interactive card messages, create and configure spaces, and manage space membership in Google Chat as the connected user over the Google Chat API. - [Dialogflow Integration](https://flowrunner.ai/integrations/google-cloud-dialogflow.md): Connect AI agents to Google Cloud Dialogflow ES. Agents detect intent from natural language messages, fire events to start scripted conversations, manage intents, entity types, and session contexts, and retrain the agent programmatically. - [Google Cloud Pub/Sub Integration](https://flowrunner.ai/integrations/google-cloud-pubsub.md): Bridge your flows into Google Cloud Pub/Sub. Agents publish messages with attributes and ordering keys, manage topics and subscriptions, and pull and acknowledge messages from pull subscriptions, with payloads decoded to plain text. - [Google Cloud Speech-to-Text Integration](https://flowrunner.ai/integrations/google-cloud-speech.md): Transcribe audio with Google Cloud Speech-to-Text across 125+ languages. Agents recognize short clips synchronously and submit long recordings as asynchronous jobs polled to completion, with word timestamps, speaker diarization, and phrase hints. - [Google Cloud Storage Integration](https://flowrunner.ai/integrations/google-cloud-storage.md): Manage Google Cloud Storage buckets and objects from FlowRunner agents: list, upload, download, copy, and delete, plus V4 signed URLs for time-limited sharing and direct uploads. - [Google Cloud Text-to-Speech Integration](https://flowrunner.ai/integrations/google-cloud-tts.md): Turn text and SSML into natural sounding speech with Google Cloud Text-to-Speech. Agents synthesize audio saved to FlowRunner file storage as a ready URL, render long form content as asynchronous jobs, and browse the voice catalog for localization. - [Google Contacts Integration](https://flowrunner.ai/integrations/google-contacts.md): Connect AI agents to Google Contacts through the People API. Agents create, update, search, and delete contacts and manage contact groups using the connected Google account. - [Looker Studio Integration](https://flowrunner.ai/integrations/google-data-studio.md): Manage Looker Studio assets and their sharing permissions through the Looker Studio API. Agents provision report access as people join or leave a team, so dashboard permissions follow the org chart instead of drifting from it. - [Google Docs Integration](https://flowrunner.ai/integrations/google-docs.md): Create, read, edit, export, and generate Google Docs documents from templates through the Google Docs and Drive APIs. Agents fill placeholder tokens with workflow data to produce contracts, invoices, and reports, then export to PDF. - [Google Drive Integration](https://flowrunner.ai/integrations/google-drive-api.md): Connect AI agents to Google Drive for file management, content processing, and collaborative workflows. Agents upload, organize, share, copy, export, and read files and react to file changes via OAuth2. - [Google Firestore Integration](https://flowrunner.ai/integrations/google-firestore.md): Connect AI agents to Cloud Firestore. Agents create, read, update, delete, list, query, and aggregate documents with plain JSON in and plain JSON out. - [Google Forms Integration](https://flowrunner.ai/integrations/google-forms.md): Process Google Forms responses and manage form structure. Agents create forms, retrieve responses by ID or list, update field configuration, and manage form lifecycle. - [Google Keep Integration](https://flowrunner.ai/integrations/google-keep.md): Google Keep is the note taking app in Google Workspace, and its API serves Workspace accounts only. Agents list, read, create, and delete notes, share them with people and groups, and download attachments. - [Google Maps Integration](https://flowrunner.ai/integrations/google-maps.md): Give AI agents the Google Maps Platform. Agents geocode and validate addresses, search and autocomplete places, pull place details and photos, compute routes and distance matrices, look up elevation and time zones, and generate static map images. - [Google Meet Integration](https://flowrunner.ai/integrations/google-meet.md): Connect AI agents to Google Meet through the Meet REST API. Agents create meeting spaces with auto recording and transcription enabled, pull conference records, participants, and recordings, and assemble full plain text transcripts ready for AI summarization. - [Google Cloud Natural Language Integration](https://flowrunner.ai/integrations/google-natural-language.md): Extract meaning and structure from unstructured text with Google Cloud Natural Language: entities, sentiment, entity sentiment, syntax, and content classification. - [Google Photos Integration](https://flowrunner.ai/integrations/google-photos.md): Manage app-created Google Photos content from your flows. Agents create and populate albums, register uploaded media, and search app-created items by content category, media type, or date range through the Photos Library API. - [Google Custom Search Integration](https://flowrunner.ai/integrations/google-search.md): Give AI agents live web lookup through Google's Custom Search JSON API. Agents run web, image, and site restricted searches against your Programmable Search Engine and return ranked titles, links, and snippets to ground downstream steps. - [Google Search Console Integration](https://flowrunner.ai/integrations/google-search-console.md): Connect AI agents to Google Search Console. Agents query search performance by page, query, country, and device, submit and audit sitemaps, inspect URL index status and canonicals, and manage properties across accounts. - [Google Sheets Integration](https://flowrunner.ai/integrations/google-sheets.md): Read, write, and manage Google Sheets data with FlowRunner agents. Agents append and update rows, export sheets to files, import CSV, and trigger flows on new or updated rows. - [Google Merchant Center Integration](https://flowrunner.ai/integrations/google-shopping.md): Manage your Google Shopping product feed from AI agent workflows. Agents insert, list, and delete Merchant Center products, audit per-product approval status and data quality issues, and review data feeds and orders through the Content API for Shopping. - [Google Slides Integration](https://flowrunner.ai/integrations/google-slides.md): Create and manage Google Slides presentations, fill slides, run template-based deck generation, and export to PDF, PPTX, or thumbnail images through the Google Slides and Drive APIs. Agents generate customer-ready decks from workflow data. - [Google Tasks Integration](https://flowrunner.ai/integrations/google-tasks.md): Manage Google Tasks lists and tasks from your flows: create, complete, reopen, move, and clear completed items. Agents capture actionable email and events as tasks and confirm before sweeping a shared list. - [Google Translate Integration](https://flowrunner.ai/integrations/google-translate.md): Translate text between 130-plus languages, detect the source language of any text, and list supported languages using the Google Cloud Translation API. - [Google Vertex AI Integration](https://flowrunner.ai/integrations/google-vertex-ai.md): Generate text and multimodal content with Gemini on Google Cloud's enterprise Vertex AI platform, create embeddings, generate images, and browse Model Garden. - [Google Workspace Admin Integration](https://flowrunner.ai/integrations/google-workspace-admin.md): Connect AI agents to the Google Workspace Admin SDK Directory API. Agents create and manage users, groups, and organizational units, control membership, manage admin roles and aliases, and audit devices and domains. - [Google Cloud Vision Integration](https://flowrunner.ai/integrations/googlecloudvision.md): Give AI agents production-grade image understanding through Google Cloud Vision. Agents run OCR on scans and multi-page PDFs, auto-tag images with labels, objects, and logos, moderate uploads with SafeSearch, and match photos against an indexed product catalog across 17 actions. - [GoPay Integration](https://flowrunner.ai/integrations/gopay.md): GoPay is a Czech payment gateway for cards, bank transfers, and wallets. Agents create standard, pre-authorized, and recurring payments, capture and refund, charge recurrences on demand, pull account statements and EET receipts, and check payment status. - [GoSMS Integration](https://flowrunner.ai/integrations/gosms.md): Connect AI agents to GoSMS, a Czech SMS gateway operated by ZooControl. Agents send transactional and marketing SMS to numbers, contacts, or groups, preview message pricing, and track delivery and replies so Czech SMS traffic stays predictable and tracked. - [Gotify Integration](https://flowrunner.ai/integrations/gotify.md): Send and manage push notifications through a self-hosted Gotify server. FlowRunner agents deliver prioritized alerts and maintain Gotify applications and clients as part of a workflow. - [gotoHuman Integration](https://flowrunner.ai/integrations/gotohuman.md): Send AI-generated content or a critical action to gotoHuman for review and resume once a person approves. Agents use it as an external review surface when the approval already lives in gotoHuman. - [GoTo Webinar Integration](https://flowrunner.ai/integrations/gotowebinar.md): Schedule and update GoTo Webinar webinars, register attendees and hand them their personal join URLs, and pull attendance, poll, question, and survey data for reporting. - [GPS Tools Integration](https://flowrunner.ai/integrations/gps-tools.md): Run coordinate mathematics entirely inside FlowRunner with no API key, vendor account, or outbound call. Agents measure distances and bearings and project waypoints as part of routing and dispatch logic. - [GPT Maker Integration](https://flowrunner.ai/integrations/gpt-maker.md): Build and operate AI customer service agents with GPT Maker, managing agents, training them, and conversing over the API. Agents keep support answers grounded in current documentation. - [GPT-trainer Integration](https://flowrunner.ai/integrations/gpt-trainer.md): Build, train, and operate GPT-trainer chatbots and their multi-agent configuration from FlowRunner. Agents keep the knowledge base current as source files and pages change. - [Grafana Integration](https://flowrunner.ai/integrations/grafana.md): Manage Grafana dashboards, folders, and data sources, record annotations, query metrics, and inspect alert rules from your flows over the Grafana HTTP API. Works with Grafana Cloud and self-hosted. - [Gravitec Integration](https://flowrunner.ai/integrations/gravitec.md): Send and track web push notifications with the Gravitec Server API, reaching everyone or a precise audience. Agents push time-sensitive alerts and read campaign performance back. - [Gravity Forms Integration](https://flowrunner.ai/integrations/gravity-forms.md): Process WordPress form submissions through operational workflows. Agents create entries, retrieve submissions, send notifications, validate data, and manage form lifecycle. - [GREEN-API Integration](https://flowrunner.ai/integrations/green-api.md): Send and receive WhatsApp messages through your own GREEN-API instance. Agents deliver texts and files by URL, verify numbers are on WhatsApp, read chat history, and poll the notification queue for inbound messages without configuring webhooks. - [Greenhouse Integration](https://flowrunner.ai/integrations/greenhouse.md): Connect AI agents to Greenhouse Recruiting via the Harvest API. Agents create and update candidates, attach resumes and notes, advance, move, reject, or hire applications, and read scorecards, interviews, and offers across 33 actions covering the full hiring pipeline. - [Greip Integration](https://flowrunner.ai/integrations/greip.md): Connect AI agents to Greip, a fraud prevention and IP intelligence API. Agents look up IP reputation and geolocation, validate submitted data, and score a transaction for fraud risk before it proceeds. - [Grimp Integration](https://flowrunner.ai/integrations/grimp.md): Connect AI agents to Grimp, an education partner platform for school career services. Agents manage student records, publish job posts, confirm recruitment outcomes, and tag the contacts and enterprises behind each employer relationship. - [Grist Integration](https://flowrunner.ai/integrations/grist.md): Connect AI agents to Grist, the relational spreadsheet. Agents run record CRUD including upsert, manage tables and columns, discover documents, and run read-only SQL queries. - [Groove Integration](https://flowrunner.ai/integrations/groove.md): Work the Groove shared inbox from AI agent workflows. Agents list and triage support conversations, reply to customers, add internal notes, look up contacts, and run raw GraphQL queries against Groove's v2 API for anything the dedicated actions don't cover. - [Groq Integration](https://flowrunner.ai/integrations/groq.md): Run ultra-fast inference on open models like Llama, Qwen, and Whisper through the Groq API. Chat, vision, audio transcription, batches, and agentic Compound systems in one connector. - [GroupMe Integration](https://flowrunner.ai/integrations/groupme.md): Read and post group chat messages and manage bots, members, and blocks with GroupMe. Agents deliver updates into the group that already coordinates the work. - [GrowSurf Integration](https://flowrunner.ai/integrations/growsurf.md): Run referral and affiliate programs with GrowSurf, adding participants and crediting referrals. Agents award credit off verified conversions rather than a self-reported claim. - [Gumlet Integration](https://flowrunner.ai/integrations/gumlet.md): Gumlet is a video and image delivery platform with VOD transcoding, live streaming, and an image optimization CDN. Agents upload and manage video assets, run live streams, organize playlists and channels, purge caches, pull viewer analytics, and react to Gumlet webhook events. - [Gumroad Integration](https://flowrunner.ai/integrations/gumroad.md): Manage Gumroad products, sales, subscribers, license keys, and discount offer codes from automated flows. Agents track new sales, verify and revoke software licenses, run discount campaigns, and register webhook subscriptions for store events. - [Habitica Integration](https://flowrunner.ai/integrations/habitica.md): Habitica is the open source habit tracker built as a role playing game. Agents create and score habits, dailies, and to-dos, manage tags and checklists, read character stats and inventory, work with parties, guilds, and challenges, and react to webhook events. - [Hacker News Integration](https://flowrunner.ai/integrations/hackernews.md): Read-only access to Hacker News through its Firebase and Algolia APIs: fetch items and users, pull story-list feeds, hydrate front-page stories, and run full-text search. No credentials required. - [HacknPlan Integration](https://flowrunner.ai/integrations/hacknplan.md): HacknPlan is project management built for game development teams, with milestones, kanban boards, work items and a game design model. Agents turn bug reports into work items, move items across board stages, log work and keep the design model linked to the tasks that implement it. - [HaloPSA Integration](https://flowrunner.ai/integrations/halopsa.md): Connect AI agents to HaloPSA, the all-in-one MSP and IT service management (PSA) platform. Agents manage tickets, actions, clients, sites, end users, assets, agents, and invoices across the Halo API. - [Handlebars Integration](https://flowrunner.ai/integrations/handlebars.md): Render Handlebars templates against flow data using the official handlebars package, entirely inside FlowRunner. Agents build message bodies, documents, and payloads from a template without an outbound call. - [Handwrytten Integration](https://flowrunner.ai/integrations/handwrytten.md): Write and post real handwritten cards with robots through Handwrytten, covering single sends, bulk baskets, and gift cards. Agents send a physical note when a renewal or a save deserves more than an email. - [Happierleads Integration](https://flowrunner.ai/integrations/happierleads.md): Identify B2B website visitors with Happierleads, reading identified leads, enriching domains, and managing tracking. Agents route a warm account to sales while the visit is still recent. - [Happyforms Integration](https://flowrunner.ai/integrations/happyforms.md): Happyforms is a self-hosted WordPress form builder from The Theme Foundry. Agents submit responses to a form the way a browser does, read a form's field structure off the rendered page, find which pages embed which forms, and read and delete stored responses through the WordPress REST API. - [HappyFox Chat Integration](https://flowrunner.ai/integrations/happyfox-chat.md): Read live chat data from HappyFox Chat, covering transcripts, offline messages, visitors with full browsing attribution, agents, and canned responses. Agents pull the conversation history a follow-up actually needs. - [HappyFox Integration](https://flowrunner.ai/integrations/happyfox-help-desk.md): Run HappyFox ticketing from AI agent workflows. Agents create tickets from inbound emails, forms, or chat, post staff and user replies, register contacts, and look up categories and staff to route and attribute tickets automatically. - [HARPA AI Integration](https://flowrunner.ai/integrations/harpa-ai.md): Dispatch browser work to your own Chrome, Edge, Brave, or Opera through the HARPA AI GRID API. Agents reach pages behind a real logged-in session that a headless scraper cannot. - [Harvest Integration](https://flowrunner.ai/integrations/harvest.md): Run Harvest time tracking, projects, and invoicing from your flows: log time, sync clients and projects, and draft invoices from tracked hours. Agents assemble billing and route each invoice to a person before it is sent. - [Heartbeat Integration](https://flowrunner.ai/integrations/heartbeat.md): Manage members, channels, threads, chat, events, groups, invitations, and courses in a Heartbeat community. Agents onboard members and keep community access aligned with billing and program status. - [HelloAsso Integration](https://flowrunner.ai/integrations/helloasso.md): HelloAsso is the French association payments and fundraising platform, with Plus Billetterie for ticketing. Agents create checkout intents, manage forms, orders, and payments for an organization, run ticketing events, and react to notifications in real time. - [helloHQ Integration](https://flowrunner.ai/integrations/hellohq.md): helloHQ is agency management software for German-speaking agencies: projects and tasks, time tracking, a light CRM, a service catalog and cost and revenue planning. Agents open projects from won deals, book time to the right task and keep companies and contacts in step with the CRM. - [Dropbox Sign Integration](https://flowrunner.ai/integrations/hellosign.md): Send documents for legally binding e-signature with Dropbox Sign. Agents kick off signature requests from file URLs or reusable templates, track signing progress, remind slow signers, cancel in-flight requests, and retrieve the fully signed PDF. - [Help Scout Integration](https://flowrunner.ai/integrations/help-scout.md): Connect AI agents to Help Scout. Agents create and triage conversations across shared mailboxes, add replies and notes, assign and tag threads, and sync customer records into support queues. - [Helpwise Integration](https://flowrunner.ai/integrations/helpwise.md): Connect AI agents to Helpwise, a shared team inbox from SaaS Labs. Agents read and reply to conversations across inboxes, tag them, keep contact and custom field data current, and pull team reports. - [HERE Integration](https://flowrunner.ai/integrations/here.md): Add HERE location intelligence to agent workflows. Agents geocode addresses into coordinates and back, search nearby places, power address autocomplete, calculate driving, walking, cycling, or truck routes, and enrich place IDs with full details. - [Bullhorn Automation Integration](https://flowrunner.ai/integrations/herefish.md): Connect AI agents to Bullhorn Automation, formerly Herefish, the recruiting automation platform for staffing firms. Agents add or update contacts, manage list membership, sync placements, and match candidates to jobs so recruiting outreach stays current automatically. - [HERO Software Integration](https://flowrunner.ai/integrations/hero-software.md): Connect AI agents to HERO Software, a German business platform for craft and trade businesses. Agents convert leads into projects and save documents against a job, then read the calendar, materials, and contact records the crew works from. - [Hexomatic Integration](https://flowrunner.ai/integrations/hexomatic.md): Control Hexomatic no-code web scraping and data extraction workflows from FlowRunner. Agents launch a scrape, wait for it to finish, and pull the tabular results into the next step. - [Hexowatch Integration](https://flowrunner.ai/integrations/hexowatch.md): Create and manage Hexowatch monitors that watch any web page and report when it changes. Agents react to a competitor price change or a policy edit the moment it is detected. - [HeyGen Integration](https://flowrunner.ai/integrations/heygen.md): Produce AI avatar video at scale with HeyGen. Agents turn scripts into avatar-narrated videos, translate and localize existing footage, clone voices, generate lipsync, auto-clip long-form content, and manage avatars, assets, and webhook endpoints across 98 actions on the v3 API. - [HeyReach Integration](https://flowrunner.ai/integrations/heyreach.md): Run LinkedIn outreach at agency scale with HeyReach, driving campaigns, lead lists, sender accounts, and the unified inbox. Agents route replies to the right owner and keep sender health in view. - [TimeCamp Planner Integration](https://flowrunner.ai/integrations/heyspace.md): TimeCamp Planner, formerly HeySpace, pairs kanban boards with team chat in one workspace. Agents create and move cards, manage lists, checklists, estimates and tags, post chat messages and handle people and groups across spaces. - [HeySummit Integration](https://flowrunner.ai/integrations/heysummit.md): Connect AI agents to HeySummit, a platform for virtual and hybrid summits. Agents register attendees, and read the events, talks, and tickets behind a summit as it fills. - [Heyy Integration](https://flowrunner.ai/integrations/heyy.md): Automate customer messaging through a Heyy workspace, sending messages and WhatsApp templates across WhatsApp, Instagram, and other connected channels. Agents keep the conversation in one place regardless of where it started. - [HiBob Integration](https://flowrunner.ai/integrations/hibob.md): Connect AI agents to HiBob, the HRIS for mid-size companies. Agents search and retrieve employee profiles, onboard new hires by creating people records, sync time-off changes incrementally, submit leave requests, and read company named lists. - [Hireflix Integration](https://flowrunner.ai/integrations/hireflix.md): Connect AI agents to Hireflix, a one-way asynchronous video interview platform. Agents invite candidates, retrieve interview results, and create shareable playback links so hiring teams review screens without scheduling live calls. - [Hirely Integration](https://flowrunner.ai/integrations/hirely.md): Connect AI agents to Hirely, a German applicant tracking system for recruiting teams and agencies. Agents publish job postings, capture and progress applications, and hire or reject applicants so every candidate moves through the pipeline on time. - [HITL Integration](https://flowrunner.ai/integrations/hitl.md): Create review loops on the HITL.sh platform, send content to human reviewers, and resume once a person responds. Agents use it as an external review surface when the reviewer already works in HITL.sh. - [HitPay Integration](https://flowrunner.ai/integrations/hitpay.md): HitPay is a Singapore-based payment gateway for SMEs. Agents create payment requests and invoices, manage customers, products, and subscription plans, refund charges, send transfers, run card readers and QR codes, and react to payment events as they arrive. - [Hive Integration](https://flowrunner.ai/integrations/hive.md): Manage Hive projects from AI agent workflows. Agents create projects and actions (Hive's tasks), update statuses, assignees, and due dates, and look up workspace members to route work across a Hive workspace. - [Holded Integration](https://flowrunner.ai/integrations/holded.md): Holded is Spanish business management software that puts invoicing, accounting, inventory, CRM, projects, and payroll in one account. Agents create and send documents, manage contacts and products, ship orders, record payments, and react to changes as they happen. - [Home Assistant Integration](https://flowrunner.ai/integrations/home-assistant.md): Read entity states, control devices through Home Assistant services, fire and inspect events, render templates, and query history and calendars through the Home Assistant REST API. - [HomeRunner Integration](https://flowrunner.ai/integrations/homerunner.md): Connect AI agents to HomeRunner, a Nordic multi-carrier shipping platform. Agents look up carrier products, find nearby servicepoints, create shipments with labels, and track parcels so Nordic and international deliveries get booked from one flow. - [Hootsuite Integration](https://flowrunner.ai/integrations/hootsuite.md): Schedule and manage social publishing through Hootsuite. Agents queue posts to connected social profiles, review and cancel scheduled messages, register media uploads, and read account and organization details via the Hootsuite REST API. - [Hotmart Integration](https://flowrunner.ai/integrations/hotmart.md): Connect AI agents to Hotmart, the digital product sales platform. Agents pull sales history, summaries, participants, and commissions, monitor and cancel subscriptions, and list products to reconcile creator revenue. - [HrFlow.ai Integration](https://flowrunner.ai/integrations/hrflow-ai.md): Connect AI agents to HrFlow.ai, an AI talent data platform for parsing and searching HR data. Agents parse resumes, index profiles and jobs, and search across sources so candidate and job data becomes structured and searchable. - [HTML 2 PDF Integration](https://flowrunner.ai/integrations/html-2-pdf.md): Convert web pages and raw HTML into PDF documents with html2pdf.app, rendered in headless Chromium with full modern CSS support. Agents produce invoices, statements, and reports that look the way the browser shows them. - [HTML/CSS to Image Integration](https://flowrunner.ai/integrations/html-css-to-image.md): Turn HTML and CSS into pixel-perfect PNG, JPG, WebP, or PDF files with the HTML/CSS to Image API, rendering exactly like Chrome. Agents generate social cards and report visuals per record. - [HTML to IMG API Integration](https://flowrunner.ai/integrations/html-to-img-api.md): Convert live web pages or raw HTML and CSS into images and PDFs, render charts from data, and run PDF utilities. Agents produce visual output from markup without running a browser themselves. - [HttpStatus Integration](https://flowrunner.ai/integrations/httpstatus.md): Check HTTP status codes, redirect chains, and response headers for any URL with the HttpStatus API. Agents verify a link resolves before it goes out in a campaign or a published page. - [Hubhus Integration](https://flowrunner.ai/integrations/hubhus.md): Connect AI agents to Hubhus, a Danish sales and field service platform that runs lead to order. Agents capture and update leads into the right campaign, book and reschedule calendar work, manage users, and read fleet tracking, time, and consent records. - [HubSpot Integration](https://flowrunner.ai/integrations/hubspot.md): Keep your CRM current without manual data entry. AI agents create contacts, update deals, and associate records automatically. Your sales team works from clean data. Always. - [Hubstaff Integration](https://flowrunner.ai/integrations/hubstaff.md): Pull time-tracking and workforce analytics from Hubstaff. Agents read organizations, projects, members, and tasks, create projects, and retrieve activity, screenshot, and worked-time reports for payroll, billing, and utilization automations. - [Hugging Face Integration](https://flowrunner.ai/integrations/huggingface.md): Reach thousands of open models through Hugging Face Inference Providers and the Hub API: chat, image and audio generation, embeddings, text analysis, and model discovery. - [Humanitix Integration](https://flowrunner.ai/integrations/humanitix.md): Read Humanitix event ticketing data in agent workflows. Agents list events, pull orders per event for reconciliation and CRM enrichment, build attendee lists from issued tickets, and run incremental syncs with a modified-since filter. - [Hunter.io Integration](https://flowrunner.ai/integrations/hunter.md): Find, verify, and enrich professional email addresses. Agents discover every email at a domain, guess a named person's address, verify deliverability before sending, enrich contacts, and manage leads and leads lists. - [Hyros Integration](https://flowrunner.ai/integrations/hyros.md): Feed Hyros ad attribution with clean conversion data. Agents push leads, sales, and sales-call events into Hyros so revenue is attributed to the right ad and campaign, and look up leads and order history to verify tracking was recorded. - [SafetyCulture Integration](https://flowrunner.ai/integrations/iauditor.md): Automate SafetyCulture (formerly iAuditor) inspection workflows. Agents search and retrieve inspections, export them to PDF, browse templates, raise and track corrective actions when audits fail, and look up users. - [IC Project Integration](https://flowrunner.ai/integrations/ic-project.md): IC Project is a Polish project management platform organized as projects, stages, boards and tasks, with time tracking, a light CRM and invoicing on top. Agents open projects with their stages and boards, create tasks on the right column, close work and log time against it. - [iClosed Integration](https://flowrunner.ai/integrations/iclosed.md): Connect AI agents to iClosed, a sales-focused appointment scheduling platform. Agents book and reschedule event calls, manage contacts and deals, record call outcomes, and sync transactions so the sales pipeline reflects every booked call. - [iCount Integration](https://flowrunner.ai/integrations/icount.md): iCount is the Israeli cloud accounting and invoicing system. Agents manage clients and their balances, create, cancel, and email documents, send pay now links, log and scan expenses, and react to new activity in real time. - [Icypeas Integration](https://flowrunner.ai/integrations/icypeas.md): Find and verify emails and pull B2B lead data through Icypeas, with async single and bulk searches and a searchable people database. Agents build a verified list without a manual export cycle. - [IdentityCheck Integration](https://flowrunner.ai/integrations/identitycheck.md): Connect AI agents to IdentityCheck by StackGo, a KYC identity verification service. Agents create verification requests that invite a contact into a hosted capture session and read the media it captured, while the decision itself arrives at a webhook endpoint you configure in the vendor dashboard. - [Ideogram Integration](https://flowrunner.ai/integrations/ideogram.md): Generate and edit images with Ideogram v3, known for accurate in-image text rendering. Agents create visuals with real typography, inpaint with masks, remix and reframe images, replace backgrounds, upscale, and caption images, with results saved to durable FlowRunner file URLs. - [Idiligo Integration](https://flowrunner.ai/integrations/idiligo.md): Run scripted online meetings through Idiligo, creating meetings from a script, connecting participants, and reading back prefilled content. Agents make sure every customer meeting covers the same required ground. - [iDoklad Integration](https://flowrunner.ai/integrations/idoklad.md): iDoklad is Czech and Slovak online invoicing software from Solitea. Agents manage contacts and documents, send invoices and payment reminders, record payments, pair bank statements, and maintain price list items. - [IFTTT Integration](https://flowrunner.ai/integrations/ifttt.md): Fire a named IFTTT event from any flow through the Webhooks service, triggering every applet listening for it. Agents reach consumer devices and services that expose no direct business API. - [ilert Integration](https://flowrunner.ai/integrations/ilert.md): ilert is an incident management, on-call alerting, and status page platform. Agents post events and alerts, accept or resolve them, look up who is on call, manage escalation policies, schedules, and maintenance windows, and update status pages. - [iLovePDF Integration](https://flowrunner.ai/integrations/ilovepdf.md): Process PDFs end to end with iLovePDF. Agents compress, merge, and split documents, convert Office files to PDF, turn pages into JPGs, password-protect or unlock files, and stamp watermarks, with every result stored as a FlowRunner file URL. - [Image-Charts Integration](https://flowrunner.ai/integrations/image-charts.md): Render chart images on demand from a single URL with Image-Charts, turning numbers, Chart.js configurations, DOT graphs, or plain text into PNG, GIF, or SVG. Agents embed live visuals in email and chat where JavaScript cannot run. - [Img2Go Integration](https://flowrunner.ai/integrations/img2go.md): Convert files with Img2Go on QaamGo's shared API2Convert REST API, covering image, document, and media formats. Agents normalize whatever arrives into the format the next system will accept. - [ImgBB Integration](https://flowrunner.ai/integrations/imgbb.md): Host images on ImgBB and get instant public links, accepting a FlowRunner file, a public URL, or base64 content. Agents publish generated or processed images so downstream systems can embed them. - [Imgur Integration](https://flowrunner.ai/integrations/imgur-com.md): Imgur is an image hosting community with albums, a public gallery, tags, and comments. Agents upload images, build albums, search and share to the gallery, follow tags, and post or moderate comments on the connected account. - [Infinity Integration](https://flowrunner.ai/integrations/infinity.md): Infinity is flexible work management where boards hold folders and items and every column is an attribute you define. Agents create and update items, manage boards and folders, set custom attributes and react to item changes in real time. - [Infobip Integration](https://flowrunner.ai/integrations/infobip.md): Reach customers worldwide over Infobip's omnichannel CPaaS. Agents send SMS, free-form and template WhatsApp messages, and transactional email, then track delivery reports and read inbound messages for two-way conversations. - [Infor M3 Integration](https://flowrunner.ai/integrations/infor-m3.md): Connect AI agents to Infor M3, a tenant-hosted enterprise ERP reached through the Infor ION API gateway. Agents look up customers, items, warehouses, and company data through named read actions, and run any M3 program through a generic MI transaction when a write is needed. - [InforUMobile Integration](https://flowrunner.ai/integrations/inforumobile.md): Connect AI agents to InforUMobile, an Israeli multichannel messaging platform. Agents send SMS, WhatsApp, email, voice, and push messages, manage contacts, groups, and unsubscribes, and pull delivery receipts and incoming messages so multichannel campaigns run under one roof. - [InfraNodus Integration](https://flowrunner.ai/integrations/infranodus.md): Turn text into a knowledge graph with topic clustering and graph-grounded AI insight through InfraNodus. Agents find the gaps and blind spots in a body of feedback or research rather than just summarizing it. - [Initiative CRM Integration](https://flowrunner.ai/integrations/initiative-crm.md): Connect AI agents to Initiative CRM, a French B2B CRM built on the Vtiger engine. Agents create and update contacts and records, raise tickets, attach documents, and inspect module metadata for custom fields. - [inMobile Integration](https://flowrunner.ai/integrations/inmobile-sms-gateway.md): Connect AI agents to inMobile, a Danish gateway for SMS, RCS, and transactional email. Agents send template-based messages, manage recipient lists and blacklists, and retrieve status reports and incoming messages so messaging compliance and delivery stay visible. - [Innform Integration](https://flowrunner.ai/integrations/innform.md): Connect AI agents to Innform, an employee training LMS. Agents invite and update learners, organize groups and locations, assign training, and browse courses and learning paths so onboarding and compliance training run on autopilot. - [Inoreader Integration](https://flowrunner.ai/integrations/inoreader.md): Turn Inoreader feeds into an automation-ready content source. Agents pull the newest unread articles, subscribe to and unsubscribe from RSS feeds, browse folders and tags, star noteworthy items, and mark articles as read. - [InOut Personal WhatsApp Integration](https://flowrunner.ai/integrations/inout-personal-whatsapp.md): Send messages to your own WhatsApp number through InOut.bot, a personal notification API built on the official Meta stack. Agents deliver alerts to the phone you actually check. - [Inro Integration](https://flowrunner.ai/integrations/inro.md): Connect AI agents to Inro, an Instagram DM automation platform. Agents manage Instagram contacts, run messaging scenarios and campaigns, and organize conversations into folders. - [Insightly Integration](https://flowrunner.ai/integrations/insightly.md): Connect AI agents to Insightly CRM. Agents manage contacts, organizations, opportunities, projects, leads, tasks, calendar events, and notes across 39 actions, with pipeline and stage reference data for advancing deals. - [Instagram for Business Integration](https://flowrunner.ai/integrations/instagram-business.md): Publish and manage Instagram Business content through the Meta Graph API. Agents publish photos, reels, stories, and carousels with automatic status polling, moderate comments with replies and hides, pull account and media insights, and search public hashtags. - [Instamojo Integration](https://flowrunner.ai/integrations/instamojo.md): Instamojo is an Indian payments platform for collecting money through payment requests. Agents create payment requests, raise orders from them, read payment details, and issue refunds through the resolution center. - [Instant DM Integration](https://flowrunner.ai/integrations/instant-dm.md): Connect AI agents to Instant DM, an Instagram and Facebook automation platform. Agents publish and read Instagram content, send and receive DMs, reply to ad and page comments, and read ad account performance. - [Instantly Integration](https://flowrunner.ai/integrations/instantly.md): Orchestrate cold email campaigns at scale. 17 real-time activity triggers fire on every email event. Agents manage campaigns, leads, and email accounts. Full control over outreach workflows. - [Instapage Integration](https://flowrunner.ai/integrations/instapage.md): Sync Instapage landing-page leads into the rest of your stack. Agents list subaccounts and pages, pull captured form submissions filtered by page or date window, and read publish status and URL details for campaign audits. - [Instasent Integration](https://flowrunner.ai/integrations/instasent.md): Connect AI agents to Instasent, a Spanish messaging platform combining an SMS gateway with a customer data platform. Agents send transactional SMS, run number lookups, push contacts and events into the audience, and query segments and campaigns so messaging and customer data stay in sync. - [IntegrityNext Integration](https://flowrunner.ai/integrations/integritynext.md): Connect AI agents to IntegrityNext, a supply chain sustainability and compliance platform. Agents create carbon assessments, raise EUDR orders and update their products, and read supplier compliance data and due diligence statements. - [Intellivy Integration](https://flowrunner.ai/integrations/intellivy.md): Intellivy is a consumer research platform that polls a real panel to test Amazon listings, images, prices and product concepts. Agents pull finished polls and their aggregated results into a flow and hand them to an AI model for a listing recommendation. - [Intercom Integration](https://flowrunner.ai/integrations/intercom.md): Run your Intercom support and CRM workspace from a flow. Manage contacts and companies, triage conversations and tickets, send proactive messages, tag and segment people, submit events, and author Help Center content. - [Interhyp Gruppe Integration](https://flowrunner.ai/integrations/interhyp-gruppe.md): Connect AI agents to the Interhyp Submission API from German mortgage broker Interhyp AG. Agents submit financing applications, send submission commands, read pre-scoring results, and download the documents a submission produced. - [Microsoft Intune Integration](https://flowrunner.ai/integrations/intunes.md): Manage your device fleet in Microsoft Intune with AI agents. List and inspect managed devices, trigger sync, retire, or wipe actions, and review device configurations, compliance policies, and app inventories through Microsoft Graph. - [Invoice Ninja Integration](https://flowrunner.ai/integrations/invoice-ninja.md): Connect AI agents to Invoice Ninja. Agents create clients and issue invoices when a deal closes, record and apply payments, email or mark invoices sent and paid, generate and approve quotes, and sync the product catalog through the Invoice Ninja v5 API on both hosted and self-hosted installs. - [Invoiced Integration](https://flowrunner.ai/integrations/invoiced.md): Run accounts receivable on autopilot with Invoiced. Agents create customers and invoices, record payments, manage estimates, subscriptions, credit notes, plans, and coupons, and read balances to drive dunning and collections across 51 actions. - [Invoicing Plus Integration](https://flowrunner.ai/integrations/invoicing-plus.md): Invoicing.plus is a French electronic invoicing platform. Agents issue invoices, estimates, and recurring invoices with their lines and PDFs, manage customers, contacts, and products, and keep company data in sync. - [involve.me Integration](https://flowrunner.ai/integrations/involve-me.md): involve.me is a funnel builder for quizzes, calculators, surveys and lead forms. Agents build personalized funnel links and embed snippets from flow data, list funnels, and start a flow on every funnel submission. - [IONOS Integration](https://flowrunner.ai/integrations/ionos.md): Manage IONOS DNS from agent workflows. Agents list zones, create, update, and delete records in bulk, and enable Dynamic DNS so routers and scripts keep A and AAAA records current through the IONOS Developer DNS API. - [IP2Location.io IP Geolocation Integration](https://flowrunner.ai/integrations/ip2location-io-ip-geolocation.md): Resolve IPv4 and IPv6 addresses to geolocation and network intelligence with IP2Location.io, returning country, region, city, ZIP, and carrier data. Agents localize experiences and flag traffic that does not fit the account. - [IP2Proxy Integration](https://flowrunner.ai/integrations/ip2proxy.md): Connect AI agents to IP2Proxy, a proxy and VPN detection service from IP2Location. Agents check whether an IP belongs to a proxy, VPN, or Tor exit before a flow trusts the request behind it. - [IP2WHOIS Integration](https://flowrunner.ai/integrations/ip2whois.md): Look up complete WHOIS records for any registered domain across 1221 TLDs through IP2WHOIS. Agents check domain age and registrant details before trusting an inbound signup. - [IPEX Integration](https://flowrunner.ai/integrations/ipex.md): Work with the IPEX Czech VoIP and PBX platform, covering call records, live call control, recordings, transcripts, and contact centre reporting. Agents attach call evidence to the record it belongs to. - [IPGeolocation.io Integration](https://flowrunner.ai/integrations/ipgeolocation.md): Look up geolocation, currency, ASN, company, and network data for any IP address with ipgeolocation.io. Agents localize content and score risk from a single lookup. - [IPLocate Integration](https://flowrunner.ai/integrations/iplocate.md): Turn any IPv4 or IPv6 address into geolocation, network, and threat intelligence with IPLocate. Agents localize an experience and flag proxy or hosting traffic before it reaches a signup flow. - [IPQualityScore Integration](https://flowrunner.ai/integrations/ipqualityscore.md): Screen risk before it enters your funnel. Agents score IP addresses for proxy, VPN, Tor, and bot activity, validate email deliverability and reputation, verify phone numbers, and scan URLs for malware and phishing through IPQualityScore's fraud scoring API. - [IQM Reports Integration](https://flowrunner.ai/integrations/iqmreports.md): Connect AI agents to IQM, a programmatic advertising demand-side platform. Agents build and run campaign reports, schedule them to recur, and download the results for downstream analysis. - [ISO Codes Integration](https://flowrunner.ai/integrations/iso-codes.md): Convert and look up standard ISO country, currency, and language codes entirely offline, with no API key or external call. Agents normalize codes between systems that each picked a different representation. - [IssueBadge Integration](https://flowrunner.ai/integrations/issuebadge.md): Connect AI agents to IssueBadge, the digital badge and certificate platform. Agents create badge templates and issue badges with public verification URLs so recipients get credentials the moment they earn them. - [Iterable Integration](https://flowrunner.ai/integrations/iterable.md): Connect AI agents to Iterable. Agents manage user profiles, track events, work with lists and campaigns, send triggered email and push, and pull campaign metrics across channels. - [iUcto Integration](https://flowrunner.ai/integrations/iucto.md): iÚčto is Czech online accounting and invoicing software. Agents manage parties and documents, post invoices and payments, upload incoming documents, import bank transactions, and track stock cards and warehouses. - [IVR Solutions Integration](https://flowrunner.ai/integrations/ivr-solutions.md): Run voice automation through an IVR Solutions account, bridging agents and customers with click-to-call and managing the flows behind it. Agents connect the right person to the right caller without a manual transfer. - [Jasper Integration](https://flowrunner.ai/integrations/jasper-ai.md): Connect AI agents to Jasper, the AI marketing content platform. Agents run commands and templates to generate on-brand copy, execute Jasper agent tasks, manage brand voices, style guides, audiences, and knowledge bases, edit marketing imagery, and file results into projects and documents. - [Jenkins Integration](https://flowrunner.ai/integrations/jenkins.md): Manage Jenkins jobs, trigger and monitor builds, inspect the build queue, and query system information from your flows over the Jenkins remote access API. - [Jepto Integration](https://flowrunner.ai/integrations/jepto.md): Connect AI agents to Jepto, a marketing intelligence platform that centralizes channel data. Agents manage clients, read the insights that anomaly detection, budget tracking, and KPI forecasting produce, and acknowledge or resolve each one. - [Jestor Integration](https://flowrunner.ai/integrations/jestor.md): Jestor is a Brazilian no-code platform for internal tools built on relational tables. Agents list, fetch, create, update and delete records, read table schemas and field options, manage users and tasks, and upload files. - [JetAPI Integration](https://flowrunner.ai/integrations/jetapi.md): Send messages through JetAPI, a gateway reaching WhatsApp, Telegram and SMS behind one API. Agents pick the channel per recipient without a connector per network. - [Jibble Integration](https://flowrunner.ai/integrations/jibble.md): Pull time tracking data from Jibble. Agents list people, groups, activities, and locations, read individual time entries, and build timesheet summaries for payroll prep and attendance reporting. - [Jina AI Integration](https://flowrunner.ai/integrations/jina.md): Build search and RAG workflows with Jina AI: embeddings, reranking, web-page reading, web search, and zero-shot classification through one search-foundation connector. - [Jira Issues Integration](https://flowrunner.ai/integrations/jira-issues.md): Connect operational workflows to your engineering and project tracking. Agents create, update, transition, assign, and search Jira issues. Full JQL query support. - [Jitbit Helpdesk Integration](https://flowrunner.ai/integrations/jitbit.md): Connect AI agents to Jitbit Helpdesk, a ticketing helpdesk available as SaaS or self-hosted. Agents open tickets, post comments, route by category, and keep user and custom field data current. - [JOBka Integration](https://flowrunner.ai/integrations/jobka.md): Connect AI agents to JOBka, a Czech employee communication and HR app for frontline teams. Agents manage employees and groups, publish and notify posts, and move files in and out of storage so company updates reach every worker automatically. - [JobNimbus Integration](https://flowrunner.ai/integrations/jobnimbus.md): Connect AI agents to JobNimbus, the CRM and job management platform for roofing and contracting. Agents create contacts and jobs, build estimates from the price list, raise invoices, log activities and tasks, and upload photos from the site. - [Jodoo Integration](https://flowrunner.ai/integrations/jodoo.md): Jodoo, known as Jiandaoyun in China, is a no-code application platform built on forms and records. Agents read and write form records with file uploads, query aggregate tables, drive workflow instances and their approval tasks, and manage members, departments, and roles. - [JoggAI Integration](https://flowrunner.ai/integrations/joggai.md): Create talking-avatar and template videos, lip-sync clips, and ad scripts with JoggAI, dubbing into more than 40 languages. Agents produce localized video variants without a shoot per market. - [JOIN Integration](https://flowrunner.ai/integrations/join.md): Connect AI agents to JOIN, the European job multiposting and applicant tracking platform. Agents create jobs, manage applications, tag candidates, and add notes so hiring pipelines stay organized across every job board. - [Joomla Integration](https://flowrunner.ai/integrations/joomla.md): Run a Joomla site from your workflows. Agents create, update, and publish articles, manage categories, tags, and user accounts, and audit menus, modules, banners, template styles, and global configuration through the Joomla Web Services API. - [Jotform Integration](https://flowrunner.ai/integrations/jotform.md): Create forms, retrieve submissions, and monitor usage. Agents build forms programmatically, pull submission data into operational workflows, and manage form questions. - [journy.io Integration](https://flowrunner.ai/integrations/journy-io.md): Connect AI agents to journy.io, a B2B customer data platform. Agents sync users and accounts, track product events, attach metadata, and trigger in-app messages from a flow. - [JSON2Video Integration](https://flowrunner.ai/integrations/json2video.md): JSON2Video renders finished MP4 videos from a Movie JSON document of scenes, text, voice, subtitles, HTML, and audiograms. Agents submit a movie, wait for the render, manage templates and media folders, and react when a render finishes. - [JsonCut Integration](https://flowrunner.ai/integrations/jsoncut.md): JsonCut renders images and videos from a JSON configuration of layers, clips, transitions, text, and HTML. Agents validate a job config, submit and poll render jobs, upload source files, and publish public links to the output. - [Judge.me Integration](https://flowrunner.ai/integrations/judge-me.md): Work product reviews in Judge.me from your flows. Agents list, create, update, and moderate Shopify store reviews, look up products, read total and per-product review counts with average ratings, and register webhooks for review events. - [JustCall Integration](https://flowrunner.ai/integrations/justcall.md): Give agents your JustCall cloud phone system. They log call notes and dispositions, send and reply to SMS and MMS, poll for inbound text replies, sync contacts, and report on calls and texts by agent, number, or date range. - [JW Player Integration](https://flowrunner.ai/integrations/jw-player.md): Manage a JW Player video library from your workflows. Agents ingest videos from external URLs with metadata set at creation, update titles, descriptions, and tags, check encoding status before publishing downstream, and list media and playlists. - [Kajabi Integration](https://flowrunner.ai/integrations/kajabi.md): Connect AI agents to Kajabi. Agents manage contacts and tags, grant or revoke offers, and read products, courses, purchases, orders, and forms to automate enrollment and follow-up across your knowledge commerce business. - [Kaleyra Integration](https://flowrunner.ai/integrations/kaleyra.md): Send SMS and WhatsApp messages worldwide through Kaleyra, the Tata Communications CPaaS. Agents deliver transactional, OTP, and promotional messages from approved sender IDs, run OTP verification flows, and check delivery status for auditing. - [Kanban Tool Integration](https://flowrunner.ai/integrations/kanban-tool.md): Kanban Tool is kanban boards with swimlanes, card types, subtasks and built-in time tracking from a Polish vendor. Agents create and move cards, manage subtasks and comments, log time and read board state to keep work visible. - [KanbanFlow Integration](https://flowrunner.ai/integrations/kanbanflow.md): KanbanFlow pairs a WIP-limited kanban board with a built-in Pomodoro timer and stopwatch. Agents create and move tasks, manage subtasks, labels and dates, read time entries and start flows when board events fire. - [Kanban Zone Integration](https://flowrunner.ai/integrations/kanbanzone.md): Kanban Zone is a Kanban board and flow-metrics tool built on lean principles. Agents create, move and search cards, manage checklists, comments and labels, read the eight flow reports and start flows from card events. - [Kantree Integration](https://flowrunner.ai/integrations/kantree.md): Kantree is a flexible work management platform from the French cooperative Digicoop, built on a model where everything is a card. Agents create and update cards, manage projects, groups and attributes, post comments and react to card changes as they happen. - [Kartra Integration](https://flowrunner.ai/integrations/kartra.md): Keep Kartra in sync with the rest of your stack. Agents create and update leads, assign and remove tags to drive Kartra automations, subscribe contacts to mailing lists, and grant membership site access after a purchase. - [KashFlow Integration](https://flowrunner.ai/integrations/kashflow.md): IRIS KashFlow is British bookkeeping software for small businesses. Agents manage customers and suppliers, create and email invoices, convert quotes, record receipts and bank transactions, and apply credit notes. - [Katana Integration](https://flowrunner.ai/integrations/katana-mrp.md): Connect AI agents to Katana cloud manufacturing. Agents sync products and sales orders from your store, raise manufacturing orders when producible stock runs low, report live inventory and committed stock, and look up materials and customers. - [Keap Integration](https://flowrunner.ai/integrations/keap.md): Connect AI agents to Keap (formerly Infusionsoft). Agents manage contacts, companies, opportunities, orders, tags, notes, and tasks, and react in real time to Keap events. - [Keboola Integration](https://flowrunner.ai/integrations/keboola.md): Operate a Keboola data project from your flows. Agents manage Storage buckets and tables, preview and import CSV data with full or incremental loads, create component configurations, and trigger jobs and orchestrations while polling their status. - [Keywords Everywhere Integration](https://flowrunner.ai/integrations/keywords-everywhere-api.md): Pull search volume, CPC, keyword expansion, and backlink data from the Keywords Everywhere API. Agents research and prioritize keywords as a step in a content workflow rather than in a spreadsheet. - [Kickbox Integration](https://flowrunner.ai/integrations/kickbox.md): Protect sender reputation before a campaign goes out. Agents verify email addresses in real time with Sendex quality scores, run batch verification over large lists, download results as CSV, catch typos with suggestions, and watch remaining credits. - [Kimi Integration](https://flowrunner.ai/integrations/kimi.md): Bring Kimi by Moonshot AI into your workflows. Agents run chat completions with very long context windows, native tool calling, and structured JSON output, extract text from PDFs and scanned documents server-side, and estimate token counts before sending long prompts. - [Kinderpedia Integration](https://flowrunner.ai/integrations/kinderpedia.md): Connect AI agents to Kinderpedia, a school and kindergarten management platform. Agents create students and parents and update student records, manage groups, families, and custom fields, and react to new enrollments so student records stay in sync with the rest of your stack. - [Kintone Integration](https://flowrunner.ai/integrations/kintone.md): Treat Kintone apps as a live database for your agents. They query records with conditions and pagination, create or update up to 100 records at once, post comments with mentions to record timelines, and read app schemas to build valid payloads dynamically. - [KIS Integration](https://flowrunner.ai/integrations/kis.md): Connect AI agents to KIS, a no-code platform for internal micro-apps built around datatables. Agents read and write the datatables a team already runs its process on. - [Kissflow Integration](https://flowrunner.ai/integrations/kissflow.md): Drive Kissflow processes from outside Kissflow. Agents launch process instances like purchase orders and leave requests, read process items for reporting, keep datasets in sync with an external source of truth, and read board cards to track work in flight. - [Kit Integration](https://flowrunner.ai/integrations/kit.md): Connect AI agents to Kit (formerly ConvertKit). Agents create and tag subscribers, add them to forms and sequences, manage custom fields, and create and measure broadcasts on the creator email platform. - [Kixie Integration](https://flowrunner.ai/integrations/kixie.md): Put Kixie's sales dialer to work in your flows. Agents trigger click-to-call from a rep's line the moment a lead arrives, queue calls into PowerLists, and send personal or shared-number SMS follow-ups after forms, purchases, or bookings. - [Kizeo Forms Integration](https://flowrunner.ai/integrations/kizeo-forms.md): Kizeo Forms is a French mobile data collection platform that replaces paper forms in the field. Agents read entries with crash-safe read markers, push forms to users, manage external lists, export entries to file storage, and react to new field submissions. - [Klaviyo Integration](https://flowrunner.ai/integrations/klaviyo.md): Run Klaviyo email and SMS marketing from your workflows. Agents upsert profiles and manage consent, lists, and segments, track custom events, query time-bucketed metric aggregates for revenue reporting, send campaigns, and handle suppression and data privacy requests. - [Klenty Integration](https://flowrunner.ai/integrations/klenty.md): Feed Klenty sales outreach from your other systems. Agents create and enrich prospects, look them up by email, enroll them in multi-step cadences, and pull opens, clicks, replies, and bounces into reporting or CRM records. - [KlickTipp Integration](https://flowrunner.ai/integrations/klicktipp.md): KlickTipp is a German email marketing and marketing automation platform. Agents upsert and tag contacts, manage data fields and opt-in processes, start automations by tagging, and use the public sign-up routes of the Listbuilding API. - [Klippa Integration](https://flowrunner.ai/integrations/klippa.md): Turn receipts and invoices into structured data with Klippa OCR. Agents extract totals, VAT lines, merchant names, and line items, classify documents by type before routing, and anonymize sensitive fields ahead of sharing or archiving. - [Klippa DocHorizon Integration](https://flowrunner.ai/integrations/klippa-dochorizon.md): Klippa DocHorizon, now presented as Doxis AI.dp, extracts structured data from invoices, receipts, identity documents, and bank statements. Agents capture documents with pretrained or custom models, prompt for their own fields, manipulate files, and route results to a human reviewer before release. - [Knack Integration](https://flowrunner.ai/integrations/knack.md): Read and write Knack no-code databases from your flows. Agents discover objects and field keys, filter records to drive downstream automation, and create, update, and delete records via the Knack REST API. - [KnowBe4 Integration](https://flowrunner.ai/integrations/knowbe4.md): Watch security awareness posture from your workflows. Agents pull KnowBe4 users with Phish-Prone Percentages, report training campaign completion, flag risky clicks in phishing security tests, and surface account-level risk scores through the read-only Reporting API. - [KoBoToolbox Integration](https://flowrunner.ai/integrations/kobotoolbox.md): Connect AI agents to KoBoToolbox, a field data collection and survey platform. Agents manage forms (assets), read and filter collected submissions, delete records, and generate CSV or XLS exports via an account API token. - [Kollie AI Integration](https://flowrunner.ai/integrations/kollie.md): Run AI voice agents on the phone with Kollie AI, covering agents and skills, outbound calls, campaigns, contacts and lists, and knowledge bases. Agents place calls grounded in current source material. - [Komercni banka Integration](https://flowrunner.ai/integrations/komercni-banka.md): Komerční banka is a Czech bank with an Account Direct Access API for business accounts. Agents list accounts, read balances and transaction history, download PDF statements, and manage event subscriptions that fire when new transactions arrive. - [Koncile Integration](https://flowrunner.ai/integrations/koncile-advanced-ocr.md): Turn invoices, delivery notes, and other documents into structured data with Koncile, an AI OCR platform. Define folders, templates, and fields once, and agents extract typed data from every document that follows. - [konfipay Integration](https://flowrunner.ai/integrations/konfipay.md): konfipay by windata is a German corporate banking gateway that reaches bank accounts over EBICS. Agents manage bank accounts and balances, trigger EBICS jobs, create and submit payment files, and track their status through the bank. - [Kraken.io Integration](https://flowrunner.ai/integrations/kraken-io.md): Kraken.io is an image optimization API with lossy and lossless compression, resizing, and format conversion. Agents optimize images by URL or upload, generate responsive image sets, deliver output straight to your own cloud storage, and check remaining quota. - [Krisp Integration](https://flowrunner.ai/integrations/krisp.md): Krisp is an AI meeting assistant that records, transcribes and summarizes calls. FlowRunner agents read the meetings a key holder can see, pull transcripts, AI notes and action items, import outside recordings for transcription, and react the moment Krisp finishes a note. - [KrispCall Integration](https://flowrunner.ai/integrations/krispcall.md): Work with virtual numbers, calls, SMS, and contacts on KrispCall. Agents provision numbers per campaign or region and log every conversation against the right record. - [Ksaar Integration](https://flowrunner.ai/integrations/ksaar.md): Ksaar is the French no-code application platform. Agents discover applications and their workflows, create, search, and update workflow records with Ksaar's JSON search, manage application users, and read the files attached to either. - [CNB Exchange Rates Integration](https://flowrunner.ai/integrations/kurzy-cnb.md): The Czech National Bank publishes koruna exchange rates, PRIBOR and CZEONIA money market rates, and FX forward points through an open API. Agents pull daily, monthly, and average rate series with no account or key required. - [Kutt Integration](https://flowrunner.ai/integrations/kutt.md): Shorten target URLs into branded short links with Kutt, an open-source and self-hostable shortener, and manage links and custom domains. Agents mint tracked links without handing traffic data to a third party. - [kwtSMS Integration](https://flowrunner.ai/integrations/kwtsms.md): Connect AI agents to kwtSMS, a Kuwaiti SMS gateway covering all four Kuwait carriers. Agents send single, bulk, and personalized SMS, validate numbers before sending, and check delivery reports and credit balance so regional messaging spends credits only on reachable numbers. - [La Growth Machine Integration](https://flowrunner.ai/integrations/lagrowthmachine.md): Run multichannel sales prospecting with La Growth Machine, covering campaigns and stats, audiences, leads and enrichment, and the LinkedIn and email inbox. Agents build the audience and work the replies in one system. - [AWS Lambda Integration](https://flowrunner.ai/integrations/lambda-service.md): Invoke AWS Lambda functions as a step in any flow. Agents run functions synchronously to compute or transform data, fire-and-forget for background work, or dry-run to validate parameters before going live. - [Landbot Integration](https://flowrunner.ai/integrations/landbot.md): Connect AI agents to Landbot chatbots on WhatsApp and web. Agents enrich customer profiles with CRM data, send proactive WhatsApp templates, route conversations to bots or human agents, and sync captured leads into downstream systems. - [Landingi Integration](https://flowrunner.ai/integrations/landingi.md): Connect AI agents to Landingi, a landing page builder with a programmatic page generation API. Agents start programmatic processes, list and inspect runs, and retrieve the generated landing pages so hundreds of personalized page variants ship without manual page building. - [Langdock Integration](https://flowrunner.ai/integrations/langdock.md): Run workspace agents and generate completions and embeddings on Langdock, the GDPR-compliant enterprise AI platform. Agents work under a data residency posture legal has already signed off on. - [Language Weaver Integration](https://flowrunner.ai/integrations/language-weaver.md): Run enterprise machine translation through the RWS Language Weaver Cloud API, with terminology dictionaries that hold your approved wording. Agents translate at volume without drifting off approved terms. - [Lark Base Integration](https://flowrunner.ai/integrations/larksuitebase.md): Use Lark Base tables as agent-writable databases. Agents list and create tables, discover fields, and create, update, delete, and search records with structured filters, ideal for syncing form submissions and app data into ByteDance's Lark suite. - [Lark Docs Integration](https://flowrunner.ai/integrations/larksuitedocument.md): Lark Docs is the block-based document editor in Lark, ByteDance's workplace suite. Agents create documents, read them as plain text or blocks, write generated reports with headings and lists intact, and convert Markdown or HTML into native Lark documents. - [Lark Drive Integration](https://flowrunner.ai/integrations/larksuitedrive.md): Lark Drive is the cloud storage behind Lark, ByteDance's workplace suite. Agents upload and organize files, export live Lark documents to PDF or Word, import Office files as native documents, manage collaborators and share links, and read versions and comments. - [Lark Group Bot Integration](https://flowrunner.ai/integrations/larksuitegrouprobot.md): Send text, rich posts, images, group share cards, and interactive cards to a Lark group chat through the custom bot webhook. Agents put an alert or an approval card in front of the group that owns it. - [LastPass Integration](https://flowrunner.ai/integrations/lastpass.md): Automate LastPass Enterprise user lifecycle. Agents bulk-provision employees from an HR source, disable departing users without deleting their vaults, pull login and admin activity reports for compliance, and reach any provisioning command through a raw escape hatch. - [Layerise Integration](https://flowrunner.ai/integrations/layerise.md): Layerise is a product registration and post-sale customer experience platform. Agents query its GraphQL Organisation API to read registered products and customers and to push data into the post-purchase journey. - [lc.cx Integration](https://flowrunner.ai/integrations/lccx.md): Create trackable short URLs on your own domains with lc.cx, organize them with tags, and read click statistics. Agents mint a link per send and see which placements produced the traffic. - [LDAP Integration](https://flowrunner.ai/integrations/ldap.md): Connect AI agents to LDAP directory servers, including OpenLDAP, Active Directory, and any RFC 4511-compliant server. Agents search and read entries, provision and de-provision accounts, update attributes, and validate user credentials. - [Lead Agent Integration](https://flowrunner.ai/integrations/leadagent.md): Connect AI agents to Lead Agent, a Czech omnichannel CRM and marketing platform. Agents import leads in bulk, move them between groups with a destination status, create funnels and contacts, and send mobile verification codes before outreach. - [LeadSquared Integration](https://flowrunner.ai/integrations/leadsquared.md): Run the lead lifecycle in LeadSquared from your flows. Agents capture and upsert leads, search by keyword or structured criteria, log custom activities to timelines, create tasks and appointments for reps, and advance opportunities as deals progress. - [Leady Integration](https://flowrunner.ai/integrations/leady.md): Identify website visitors in Czechia and Slovakia with Leady, reading identified company and person sessions and reconstructing their path. Agents route a researching account to sales before the interest cools. - [Leafy Plant Integration](https://flowrunner.ai/integrations/leafy-plant.md): Identify plants from a photo and look up horticultural care data with the Leafy Plant API. Search 40,000-plus species and fetch watering, light, soil, and temperature guidance. - [LearnDash Integration](https://flowrunner.ai/integrations/learndash.md): Connect AI agents to LearnDash, the WordPress LMS plugin. Agents create courses, lessons, topics, and groups, enroll or remove learners, and read course progress so curriculum changes and enrollments never wait on manual admin. - [LearningSuite Integration](https://flowrunner.ai/integrations/learningsuite.md): Connect AI agents to LearningSuite, a German online course platform. Agents create and update members, assign courses and bundles, manage groups, and read, set, or reset course progress so member access always matches what customers bought. - [LearnWorlds Integration](https://flowrunner.ai/integrations/learnworlds.md): Connect AI agents to your LearnWorlds school. Agents create and update learners, enroll and unenroll them in courses and bundles, track course progress, browse the catalog, and reconcile payment transactions through the Admin REST API v2. - [LEAV Engine Integration](https://flowrunner.ai/integrations/leav-engine.md): LEAV Engine is an open-source, self-hosted no-code data platform with a GraphQL API. Agents manage libraries, attributes, trees and views, create and deactivate records, save values in bulk and run GraphQL queries against your own deployment. - [legacy-use Integration](https://flowrunner.ai/integrations/legacy-use.md): Turn a legacy desktop application into a REST API with legacy-use, which drives it through an AI computer-use agent over RDP, VNC, or TeamViewer. Agents automate systems that were never given an interface. - [Legito Integration](https://flowrunner.ai/integrations/legito.md): Connect AI agents to Legito, a no-code document automation and contract lifecycle platform. Agents generate documents from templates, manage records through their lifecycle, control sharing, and react to workspace events. - [Legnext.ai Integration](https://flowrunner.ai/integrations/legnext.md): Drive Midjourney image and video generation over REST through Legnext.ai, with no Discord account required. Agents generate and iterate on visual assets inside the workflow that needs them. - [Lemlist Integration](https://flowrunner.ai/integrations/lemlist.md): Connect AI agents to Lemlist. Agents add and update leads in outreach campaigns, track engagement activity, mark leads interested or not interested, and maintain the team-wide unsubscribe list. - [Lemon Squeezy Integration](https://flowrunner.ai/integrations/lemon-squeezy.md): Connect AI agents to Lemon Squeezy, the merchant of record for digital products. Agents read stores, products, orders, subscriptions, and customers, create customer records, cancel subscriptions, and generate hosted checkout URLs for buyers. - [Leonardo.Ai Integration](https://flowrunner.ai/integrations/leonardo-ai.md): Generate images and motion videos with Leonardo.Ai directly from your flows. Agents run text-to-image and image-to-video generations, upscale and remove backgrounds, train custom models and Elements from datasets, and save every result to durable FlowRunner file storage. - [Leroy Merlin Integration](https://flowrunner.ai/integrations/leroy-merlin.md): Leroy Merlin's marketplace runs on Mirakl, and this connector covers the whole seller workflow. Agents accept orders as they arrive, push shipment confirmations and tracking from a warehouse system, handle returns and refunds, manage offers, prices and stock, and reconcile settlements. - [Lettermint Integration](https://flowrunner.ai/integrations/lettermint.md): Lettermint is an EU hosted transactional and broadcast email platform. Agents send single and batched messages with idempotency keys, follow message activity and events, manage suppressions, and administer domains, projects, routes, and webhooks. - [LetterXpress Integration](https://flowrunner.ai/integrations/letterxpress.md): Submit a PDF to LetterXpress and it prints, folds, franks, and posts the physical letter, with electronic delivery and billing also covered. Agents send legally required paper mail without a mailroom. - [lexoffice Integration](https://flowrunner.ai/integrations/lexoffice.md): Connect AI agents to lexoffice, the German SMB accounting platform. Agents manage contacts, create invoices, quotations, credit notes, and dunnings, render document PDFs, record bookkeeping vouchers with file attachments, and register webhook event subscriptions. - [LIFX Integration](https://flowrunner.ai/integrations/lifx.md): Connect AI agents to LIFX smart lights, controlled through the LIFX cloud HTTP API. Agents set colors and brightness, activate scenes, and run effects so a flow can signal state in the physical room. - [Lightr Integration](https://flowrunner.ai/integrations/lightr.md): Automate the full order flow on Lightr, the Dutch direct-mail platform that produces genuinely handwritten cards and posts them. Agents trigger a physical card off a milestone the CRM already recorded. - [Lightspeed eCom Integration](https://flowrunner.ai/integrations/lightspeed-ecom.md): Read products, variants, orders, customers, and categories straight from your Lightspeed eCom store. Agents sync catalog and inventory data, export recent orders for reporting, and look up customer records to enrich fulfillment and support workflows. - [Limble CMMS Integration](https://flowrunner.ai/integrations/limble-cmms.md): Connect AI agents to Limble CMMS, maintenance management software. Agents create and update maintenance tasks, track assets and parts across locations, and assign work to the right people. - [LINE Integration](https://flowrunner.ai/integrations/line.md): Send and manage messages through the LINE Messaging API for a LINE Official Account. Push and reply to chats, multicast, broadcast, and narrowcast, download media users send you, and read profile and quota data. - [Linear Integration](https://flowrunner.ai/integrations/linear.md): Manage issues, projects, comments, teams, users, workflow states, and labels over Linear's GraphQL API, and react in real time to Linear events. Turn tickets, alerts, and form submissions into tracked work. - [LingvaNex Integration](https://flowrunner.ai/integrations/lingvanex.md): Translate text or HTML between hundreds of languages with the LingvaNex B2B Cloud API, list supported languages, and auto-detect the source language of any input. - [Linkbreakers Integration](https://flowrunner.ai/integrations/linkbreakers.md): Create and manage branded short links, QR codes, and vCard contact cards in Linkbreakers, organized by directories and tags. Agents mint trackable assets per campaign and follow the visitor events they produce. - [Linked API Integration](https://flowrunner.ai/integrations/linked-api.md): Linked API drives a real LinkedIn account through a dedicated cloud browser. Agents send messages and connection requests, search people and companies, sync the inbox and network, and react to new messages and accepted connections while staying inside the account's limits. - [LinkedIn Integration](https://flowrunner.ai/integrations/linkedin.md): Publish and manage LinkedIn content as a member or company page. Create text, article-link, and image posts, read and delete posts, comment on and like posts, and list administered organizations. - [LinkedIn Ads Campaign Management Integration](https://flowrunner.ai/integrations/linkedin-ads-campaign-mgmt.md): Manage LinkedIn advertising without opening Campaign Manager. Agents search ad accounts, organize spend into campaign groups, create and update campaigns, pause or activate them, adjust budgets and bids, and list creatives through LinkedIn's versioned Marketing API. - [LinkedIn Ads Reports Integration](https://flowrunner.ai/integrations/linkedin-ads-reports.md): Pull LinkedIn advertising performance and revenue-attribution analytics through the LinkedIn Marketing Reporting API. Agents read campaign results back into the systems where the revenue actually landed, closing the loop on what the spend produced. - [LinkedIn Events Integration](https://flowrunner.ai/integrations/linkedin-events.md): Connect AI agents to the LinkedIn Event Management API. Agents create and update LinkedIn events and read their details as part of a broader campaign flow. - [LinkedIn Lead Gen Forms Integration](https://flowrunner.ai/integrations/linkedin-lead-forms.md): Pull LinkedIn Lead Gen Form submissions into your workflows. Agents list the forms a sponsored ad account owns, inspect form definitions, and retrieve lead responses with each member's name, email, and company answers for CRM sync and follow-up. - [LinkedIn Matched Audiences Integration](https://flowrunner.ai/integrations/linkedin-matched-audiences.md): Upload first-party audiences to LinkedIn Ads through the official DMP Segments API, creating segments and streaming hashed identifiers. Agents keep ad audiences synchronized with the CRM rather than with last quarter's export. - [LinkedIn Offline Conversions Integration](https://flowrunner.ai/integrations/linkedin-offline-conversions.md): Stream server-side conversion events to LinkedIn Ads through the official Conversions API, creating rules and associating campaigns. Agents report closed revenue back to the ad platform that sourced it. - [Linkly Integration](https://flowrunner.ai/integrations/linkly.md): Create and track branded short links with Linkly, covering link management, custom domains, and click data. Agents mint a tracked link per campaign and read attribution back into reporting. - [Linkuma Integration](https://flowrunner.ai/integrations/linkuma.md): Buy and track backlinks on Linkuma, a French netlinking marketplace, through its REST API. Agents order placements and monitor whether the links actually went live. - [Linkup Integration](https://flowrunner.ai/integrations/linkup.md): Search the web with Linkup, an API built for grounding language models, returning ranked pages with full content or written answers with citations. Agents answer from current sources instead of stale model memory. - [LinkupAPI Integration](https://flowrunner.ai/integrations/linkupapi.md): LinkupAPI connects an agent to LinkedIn and WhatsApp with B2B enrichment on top. Workflows search people, companies, and Sales Navigator leads, send connection requests and messages, post and engage with content, enrich emails, and react to replies and accepted invitations. - [Linx Commerce Integration](https://flowrunner.ai/integrations/linx-commerce.md): Linx Commerce is the Brazilian ecommerce platform from Linx S.A., part of StoneCo. Agents poll orders into an ERP, push stock and price feeds from a warehouse system, manage SKUs, promotions and marketplace offers, and serve LGPD privacy requests. - [LionWheel Integration](https://flowrunner.ai/integrations/lionwheel-delivery.md): Connect AI agents to LionWheel, a local delivery management and route planning platform. Agents create and update deliveries, optimize daily driver routes, and manage customer companies so last-mile dispatch runs without manual coordination. - [LiquidPlanner Integration](https://flowrunner.ai/integrations/liquidplanner.md): LiquidPlanner, now sold as Tempo Portfolio Manager, is predictive project management built on ranged estimates and a scheduling engine. Agents create and update projects, tasks and assignments, adjust estimates and read the schedule to keep forecasts honest. - [LiveChatAI Integration](https://flowrunner.ai/integrations/live-chat-ai.md): Connect AI agents to LiveChatAI, an AI chatbot platform for customer support. Agents drive conversations, post operator replies and close threads, manage the AI agents and knowledge base behind the bot, and read conversation analytics. - [LiveAgent Integration](https://flowrunner.ai/integrations/liveagent.md): Connect AI agents to LiveAgent, the multichannel help desk. Agents open, update, and route support tickets, post replies and internal notes into conversations, sync customer contacts, and read agents and departments for rules-based assignment. - [LiveChat Integration](https://flowrunner.ai/integrations/livechat.md): Manage LiveChat conversations from your flows. Agents read active chats, post messages and internal notes, start proactive outbound chats, close resolved conversations, and create or list agents on the account. - [Livespace Integration](https://flowrunner.ai/integrations/livespace-crm.md): Connect AI agents to Livespace, a Polish sales CRM with a structured sales process. Agents create people and companies, move deals through defined stages, assign tasks, and search across the full record set. - [Livestorm Integration](https://flowrunner.ai/integrations/livestorm.md): Connect AI agents to Livestorm webinars. Agents create and list events, browse scheduled sessions, register people by email, and pull registrants and attendees for CRM follow-up and attendance reporting. - [LiveWebinar Integration](https://flowrunner.ai/integrations/livewebinar.md): Connect AI agents to LiveWebinar, a webinar, meeting, and virtual event platform. Agents schedule webinars, register attendees, read participation data, and manage and monitor the recorders behind a session. - [LivingMetrics Integration](https://flowrunner.ai/integrations/livingmetrics.md): Connect AI agents to LivingMetrics, an SMB CRM and marketing data platform. Agents submit leads, create contacts and deals, move and transfer products between holders, assign tasks, and read the history of form submissions, visits, and email opens behind a record. - [Meta Llama Integration](https://flowrunner.ai/integrations/llama.md): Bring Meta's first-party Llama API into your flows. Agents run chat completions with tool calling and structured JSON output, ask single-prompt questions, screen content with Llama Guard moderation, and discover which models a key can use. - [Lnk.Bio Integration](https://flowrunner.ai/integrations/lnk-bio.md): Connect AI agents to Lnk.Bio, a link-in-bio landing page service. Agents update links, reorder pages and blocks, and restyle the profile so the bio page tracks whatever a campaign is promoting. - [Lob Integration](https://flowrunner.ai/integrations/lob.md): Send physical mail programmatically with Lob. Agents verify and standardize US addresses, maintain reusable address records, create postcards, letters, and checks from HTML or templates, and audit sent mail with status and proof URLs. - [Lobstr.io Integration](https://flowrunner.ai/integrations/lobstr.md): Lobstr.io runs a catalog of ready-made crawlers for sites like Google Maps, LinkedIn, and Amazon. Workflows configure squids, add tasks, start and schedule runs, collect results, and react when a run finishes or fails. - [Lokalise Integration](https://flowrunner.ai/integrations/lokalise.md): Connect AI agents to Lokalise to run the full localization lifecycle. Agents create projects, define translation keys, add languages, read and update translations, upload source files, export translation bundles, and monitor localization tasks. - [Loopify Integration](https://flowrunner.ai/integrations/loopify.md): Run marketing automation on Loopify, the Norwegian platform, through its official Web API. Agents manage contacts and segments, trigger campaigns, and read engagement back into the record. - [Loops Integration](https://flowrunner.ai/integrations/loops.md): Connect AI agents to Loops, the email platform built for SaaS. Agents create and update contacts, manage mailing list subscriptions, send transactional emails, and fire custom events that trigger automated email loops. - [Loopy Loyalty Integration](https://flowrunner.ai/integrations/loopy-loyalty.md): Run digital stamp-card loyalty programmes with Loopy Loyalty, enrolling customers into Apple and Google Wallet passes. Agents award stamps off verified transactions and push the updated card to the phone. - [Loqate Integration](https://flowrunner.ai/integrations/loqate.md): Verify and standardize postal addresses in 245+ countries with Loqate. Agents autocomplete address capture, cleanse addresses against reference data, validate phone numbers and email addresses, and geocode locations to coordinates. - [Loyverse Integration](https://flowrunner.ai/integrations/loyverse.md): Connect AI agents to Loyverse POS. Agents manage catalog items and categories, register loyalty customers, record and export sales and refund receipts, monitor inventory stock levels, and read stores, payment types, and employees. - [Luma AI Integration](https://flowrunner.ai/integrations/luma-ai.md): Generate video and images with Luma AI's Dream Machine API. Agents turn prompts and keyframes into video clips with the Ray models, create images with Photon, extend, upscale, reframe, or add audio to results, and persist ephemeral assets to durable FlowRunner storage. - [Lumin Integration](https://flowrunner.ai/integrations/lumin.md): Lumin is a New Zealand document workflow platform combining PDF editing, Lumin Sign eSignatures, and AgreementGen template generation. Agents send documents for signature in order, chase signers, require verified identity, render templates into PDFs, and file documents into a workspace. - [Lusha Integration](https://flowrunner.ai/integrations/lusha.md): Enrich leads and accounts with Lusha's verified B2B data. Agents append work emails, direct dials, titles, and firmographics to people and companies, bulk-enrich lists, and search the prospecting database to build targeted contact and company lists. - [Luxafor Integration](https://flowrunner.ai/integrations/luxafor.md): Connect AI agents to Luxafor, a USB busy light controlled through a cloud webhook. Agents set colors, blink patterns, and built-in sequences so a flow can signal status on a physical desk. - [Magento 2 Integration](https://flowrunner.ai/integrations/magento.md): Manage Magento 2 and Adobe Commerce products, orders, customers, categories, and sales documents from automated flows. Agents sync catalog and stock, invoice and ship orders, keep customer records in step, and report on sales activity through the Magento REST API. - [Magentrix Integration](https://flowrunner.ai/integrations/magentrix.md): Connect AI agents to Magentrix, a partner relationship management and customer community platform. Agents read and write records across any Magentrix object, resolve schema metadata, and manage the files attached to them. - [Magic Eden Integration](https://flowrunner.ai/integrations/magic-eden.md): Connect AI agents to Magic Eden, a multi-chain NFT marketplace. Agents read collection and token data, inspect wallet holdings, and query AMM pool state across supported chains. - [Zee Integration](https://flowrunner.ai/integrations/magic-zee.md): Connect AI agents to Zee (ZEETECH), a WhatsApp conversational agent platform, to send and schedule WhatsApp messages. Agents keep customer conversations running on the channel they already use. - [Magileads Integration](https://flowrunner.ai/integrations/magileads.md): Run B2B multichannel prospecting with Magileads, covering contact lists, enrichment, workflows, statistics, suppression, and tags. Agents build the audience and keep suppression honored on every send. - [Mail Komplet Integration](https://flowrunner.ai/integrations/mail-komplet.md): Mail Komplet is a Czech email marketing and transactional platform. Agents manage mailing lists, recipients, and custom columns, run campaigns and distributions, send transactional email and SMS, and handle unsubscribes and sale coupons. - [mail2many Integration](https://flowrunner.ai/integrations/mail2many.md): mail2many is a German email marketing platform by Atrivio. Agents manage subscribers, groups, and customer fields, set up and send trigger mailings, publish articles and channels, and read tracking data. - [MailBluster Integration](https://flowrunner.ai/integrations/mailbluster.md): MailBluster is a bulk email platform that sends through your own Amazon SES account. Agents create and update leads by email, manage custom fields and tags, and sync products and orders for targeted campaigns. - [Mailbox (SMTP/IMAP) Integration](https://flowrunner.ai/integrations/mailbox.md): Read inbound email and send outbound from any mailbox. Agents monitor inboxes via IMAP, send messages via SMTP, and manage read status. The universal email integration for any mail provider. - [MailboxValidator Integration](https://flowrunner.ai/integrations/mailboxvalidator.md): MailboxValidator validates email addresses and cleans lists. Agents run a full single address validation, check whether an address belongs to a free provider, and flag disposable addresses before they enter a list. - [Mailcheck Integration](https://flowrunner.ai/integrations/mailcheck.md): Mailcheck verifies email addresses and cleans lists. Agents check a single address for validity, a disposable flag, catch-all and mailbox existence, and a trust rate before sending to it or storing it. - [Mailchimp Marketing Integration](https://flowrunner.ai/integrations/mailchimp-marketing.md): Manage audiences, campaigns, and subscriber engagement. Agents add and remove list members, tag contacts, send campaigns, and retrieve reports. - [Mailchimp Transactional Integration](https://flowrunner.ai/integrations/mailchimp-transactional.md): High-deliverability transactional email with template management and analytics. Agents send individual and templated emails, manage rejection lists, and retrieve message tracking data. - [Maileon Integration](https://flowrunner.ai/integrations/maileon.md): Maileon is the email marketing and automation platform from XQueue. Agents manage contacts, fields, filters, and target groups, build and send mailings, schedule dispatchings and quality checks, maintain blocklists, and push transactions and data extensions. - [MailerSend Integration](https://flowrunner.ai/integrations/mailer-send.md): Send transactional emails and react to delivery events. 10 real-time email triggers let agents respond to opens, clicks, bounces, and unsubscribes. - [MailerCheck Integration](https://flowrunner.ai/integrations/mailercheck.md): MailerCheck is MailerLite's email verification and list cleaning tool. Agents verify addresses in real time or asynchronously, build verification lists from arrays or files, run them, read the results, and watch credits. - [Mailercloud Integration](https://flowrunner.ai/integrations/mailercloud.md): Mailercloud puts email marketing, a transactional Email API, and an email verifier behind one connection. Agents manage campaigns, contacts, lists, segments, and automations, send transactional and mail merge email, test inbox placement, and verify addresses. - [MailerLite Integration](https://flowrunner.ai/integrations/mailerlite.md): Connect AI agents to MailerLite. Agents upsert subscribers, organize groups, inspect fields, and build, schedule, and send email campaigns, with a real-time trigger for subscriber events. - [Maileroo Integration](https://flowrunner.ai/integrations/maileroo.md): Maileroo is an SMTP relay and email delivery platform. Agents send single, templated, bulk, and scheduled email, read logs and suppressions, manage templates, domains, sending keys, and webhooks, and pull delivery statistics. - [Mailgun Integration](https://flowrunner.ai/integrations/mailgun.md): Connect AI agents to Mailgun. Agents send plain, HTML, and templated email with attachments, validate addresses, manage mailing lists and members, maintain suppression lists, and pull sending statistics. - [Mailjet Integration](https://flowrunner.ai/integrations/mailjet.md): Connect AI agents to Mailjet. Agents send transactional and bulk email, manage contacts and lists with custom properties, work with templates, and monitor deliverability through message events and statistics. - [Mailkit Integration](https://flowrunner.ai/integrations/mailkit.md): Mailkit is a Czech email and SMS platform. Agents manage mailing lists and recipients, schedule campaigns, run transactional and mass delivery and SMS, drive journeys, read reports, and work with the account profile and file manager. - [Mailmodo Integration](https://flowrunner.ai/integrations/mailmodo.md): Connect AI agents to Mailmodo, the interactive AMP email platform. Agents upsert contacts, manage list membership, trigger campaigns with personalization data, enroll contacts into journeys, and pull campaign performance reports. - [Mailparser Integration](https://flowrunner.ai/integrations/mailparser-io.md): Consume the structured data Mailparser extracts from inbound emails. Agents list mailboxes and their inbound addresses, retrieve parsed results over a date range, and look up individual parsed emails to route extracted fields downstream. - [Mailrelay Integration](https://flowrunner.ai/integrations/mailrelay.md): Mailrelay is a Spanish email marketing platform with SMS, WhatsApp, and e-commerce tracking. Agents manage subscribers, groups, tags, and custom fields, run campaigns, RSS campaigns, and A/B tests, send transactional email, message over SMS and WhatsApp, and sync stores, products, and carts. - [Mails.so Integration](https://flowrunner.ai/integrations/mails-so.md): Mails.so is an email validation API for keeping lists clean before you send. Agents validate a single address in real time or submit a batch job and collect the results when it finishes. - [Mailshake Integration](https://flowrunner.ai/integrations/mailshake.md): Run Mailshake cold outreach from your flows. Agents add recipients to campaigns, track leads flagged for follow-up, monitor opens, clicks, and replies, and list campaigns and senders to drive campaign-aware automations. - [Mailtrap Integration](https://flowrunner.ai/integrations/mailtrap.md): Send and test email through Mailtrap. Agents deliver transactional emails from verified sending domains, route test sends to sandbox inboxes, and inspect captured messages down to their rendered HTML and plain-text bodies. - [Mailvio Integration](https://flowrunner.ai/integrations/mailvio.md): Mailvio is an email marketing and deliverability platform whose own control panel runs on its API. Agents manage subscribers, groups, tags and custom fields, build and send campaigns, run deliverability tests, send transactional mail, and react to webhook events. - [MaintainX Integration](https://flowrunner.ai/integrations/maintainx.md): Connect AI agents to MaintainX, the work order and maintenance management platform. Agents open work orders when equipment faults are detected, update and assign them to technicians, manage assets and locations, and read parts inventory and team members. - [MakePlans Integration](https://flowrunner.ai/integrations/makeplans.md): Connect AI agents to MakePlans, an online appointment, class, and event booking platform. Agents create and confirm bookings, check available slots, and manage services, people, and resources so schedules update without manual entry. - [Mallabe Integration](https://flowrunner.ai/integrations/mallabe.md): Reach Mallabe's utility API suite behind a single account, grouped into categories of everyday automation helpers. Agents handle small conversion and lookup chores without a service per task. - [Mamo Integration](https://flowrunner.ai/integrations/mamo.md): Mamo Business is a UAE payments platform for links, invoices, and payouts. Agents create payment links and invoices, initiate, capture, and refund payments, run subscriptions, send local and international payouts, and act on webhook events. - [ManageEngine ADManager Plus Integration](https://flowrunner.ai/integrations/manageengine-admanager.md): Connect AI agents to a self-hosted ManageEngine ADManager Plus instance. Agents read and update Active Directory users, groups, and OUs, create computers and contacts, disable computers, and run orchestration templates for the full account lifecycle. - [Mangopay Integration](https://flowrunner.ai/integrations/mangopay.md): Run marketplace payment operations on Mangopay. Agents onboard natural users, create and inspect e-money wallets, move funds between wallets with configurable platform fees, and confirm card or bank-transfer pay-ins have settled before releasing goods. - [MantisBT Integration](https://flowrunner.ai/integrations/mantisbt.md): MantisBT is the open-source, self-hosted bug tracker. Agents create, update, and move issues, add notes, attachments, tags, and relationships, and manage projects, versions, and users over the built-in REST API. - [ManyChat Integration](https://flowrunner.ai/integrations/manychat.md): Connect AI agents to ManyChat's chat marketing across Messenger, Instagram, WhatsApp, Telegram, and SMS. Agents manage subscribers, tags, and custom fields, send content or trigger existing flows on any connected channel, and look up subscriber profiles for routing. - [ManyContacts Integration](https://flowrunner.ai/integrations/manycontacts.md): Connect AI agents to ManyContacts, a WhatsApp team inbox and CRM. Agents send and read WhatsApp messages, apply templates, tag contacts, and route conversations to the right team. - [ManyReach Integration](https://flowrunner.ai/integrations/manyreach.md): Run cold email outreach with ManyReach, covering campaigns and sequences, the prospect CRM, mailing lists, sending mailboxes, tags, and blacklists. Agents keep sender health and suppression lists intact as volume grows. - [Mapy.cz Integration](https://flowrunner.ai/integrations/mapy-cz.md): Use mapping, geocoding, and routing from the Mapy.com REST API by Seznam. Agents convert addresses to coordinates, autocomplete places, and plan routes for dispatch and field work. - [Daylite Integration](https://flowrunner.ai/integrations/marketcircle-daylite.md): Connect AI agents to Daylite by Marketcircle, a CRM and productivity app for Apple-centric small businesses. Agents manage contacts, companies, opportunities, projects, tasks, and appointments, and maintain webhook subscriptions against those records. - [Salesforce Marketing Cloud Integration](https://flowrunner.ai/integrations/marketing-cloud.md): Connect AI agents to Salesforce Marketing Cloud. Agents upsert and read Data Extension rows, fire Journey Builder entry events, send triggered transactional emails, and look up contacts to power segmentation and lifecycle campaigns. - [Marketo Integration](https://flowrunner.ai/integrations/marketo.md): Connect AI agents to Adobe Marketo Engage. Agents sync leads, manage list membership, run and schedule smart campaigns, work with assets and CRM objects, and run bulk import/export jobs. - [Marketstack Integration](https://flowrunner.ai/integrations/marketstack.md): Access real-time, intraday, and historical stock market data from the Marketstack v2 API: end-of-day and intraday prices, stock splits and dividends, and reference data for tickers, exchanges, currencies, and timezones. - [Mastodon Integration](https://flowrunner.ai/integrations/mastodon.md): Publish and manage statuses on Mastodon from AI agents: post toots with controlled visibility, boost and favourite, read home and public timelines, search the fediverse, and pull notifications for triage. - [Matrix Integration](https://flowrunner.ai/integrations/matrix.md): Connect to a Matrix homeserver over the Client-Server API. Send messages and events, provision and administer rooms, manage memberships, read profiles, and move media in and out of the homeserver media repository. - [MatrixFlows Integration](https://flowrunner.ai/integrations/matrixflows.md): MatrixFlows is the AI knowledge enablement and collaboration platform. Agents read table schemas, create and update records with comments, attachments, and images, work the inbox, and react to record events in real time. - [Mattermost Integration](https://flowrunner.ai/integrations/mattermost.md): Integrate a self-hosted or Cloud Mattermost instance. Create and manage posts, channels, teams, users, files, and reactions over the Mattermost REST API using a personal access token or bot token. - [Mautic Integration](https://flowrunner.ai/integrations/mautic.md): Connect AI agents to Mautic, the open-source marketing automation platform. Agents manage contacts, segments, campaigns, companies, and emails, and move contacts through funnel stages. - [Mavenlink (Kantata) Integration](https://flowrunner.ai/integrations/mavenlink.md): Connect AI agents to Mavenlink (Kantata OX) professional services automation. Agents create and track client workspaces, manage tasks and stories, log time entries for billing and utilization, and sync project and user data with other systems. - [Medium Integration](https://flowrunner.ai/integrations/medium.md): Connect AI agents to Medium. Agents read the current user, create posts on a profile or under a publication, list publications and contributors, and upload images. The Medium public API is deprecated, so this serves accounts that already hold an integration token. - [Meet.bot Integration](https://flowrunner.ai/integrations/meetbot.md): Connect AI agents to Meet.bot, an AI-first meeting scheduling platform. Agents check availability, book meetings, manage connected calendar events, and maintain scheduling pages so meeting coordination happens without email ping-pong. - [MeetGeek Integration](https://flowrunner.ai/integrations/meetgeekai.md): Put MeetGeek meeting intelligence into agent workflows. Agents dispatch a recording bot to live calls, submit recordings for AI analysis, and pull transcripts, summaries, highlights, and per-meeting KPIs into downstream tools. - [Meet Hour Integration](https://flowrunner.ai/integrations/meethour.md): Schedule and run video meetings on Meet Hour, covering one-off and recurring meetings, address book contacts and groups, join tokens, and recordings. Agents book the meeting and file the recording where the team will find it. - [Meetime Integration](https://flowrunner.ai/integrations/meetime.md): Connect AI agents to Meetime, a Brazilian sales engagement platform for prospecting cadences. Agents edit leads and add them to a cadence, and read the calls, demos, and prospecting activity that moved a deal forward. - [MEETOVO Integration](https://flowrunner.ai/integrations/meetovo.md): Connect AI agents to MEETOVO, the German funnel builder for recruiting and lead generation, through its official API. Agents pull funnel submissions and route candidates and leads to the right owner. - [Meetup Integration](https://flowrunner.ai/integrations/meetup.md): Read Meetup data inside agent workflows: look up your member profile and groups, list a group's upcoming events, retrieve full event details, and reach any other field via raw GraphQL queries. - [MegaAPI Integration](https://flowrunner.ai/integrations/megaapi-start.md): Connect AI agents to MegaAPI, a Brazilian REST API that links a WhatsApp account to your automations. Agents send text, media, and interactive messages, manage chats and groups, check numbers on WhatsApp, and receive inbound messages through a webhook trigger so WhatsApp conversations flow into and out of your workflows. - [MeisterTask Integration](https://flowrunner.ai/integrations/meistertask.md): Connect AI agents to MeisterTask kanban boards. Agents create projects and tasks, move work between sections, update status and checklist items, and read project members to route assignments. - [Melo Integration](https://flowrunner.ai/integrations/melo.md): Connect AI agents to Melo, an aggregated and deduplicated French real estate listings API. Agents search properties across sources, read market insights by location, and maintain saved searches whose match webhook can start a flow in realtime. - [Mem Integration](https://flowrunner.ai/integrations/mem.md): Mem is the AI note taking app. Agents create notes and read their versions, organize collections, search across notes and attachments, capture content through Mem It, work the assistant's tasks, projects, and follow-ups, read the calendar, and poll the account's event feed. - [Mem0 Integration](https://flowrunner.ai/integrations/mem0.md): Give AI agents durable long-term memory with Mem0. Agents add, search, and update memories, set expirations so stale context drops out, audit memory history, manage entities, and submit relevance feedback through the Mem0 Platform API. - [Memberful Integration](https://flowrunner.ai/integrations/memberful.md): Manage your Memberful membership base from AI agents: list and look up members with their subscriptions and renewal state, create and update member records with custom metadata, discover plans, and run custom GraphQL for anything else. - [MemberKit Integration](https://flowrunner.ai/integrations/memberkit.md): Connect AI agents to MemberKit, a Brazilian membership and online course platform. Agents create or update members, generate passwordless magic links, and read courses, classrooms, and subscriber lists so member access follows every sale. - [MemberPress Integration](https://flowrunner.ai/integrations/memberpress.md): Connect AI agents to MemberPress on your WordPress site. Agents onboard members, create transactions that grant plan access, and read membership, billing, and subscription data through the MemberPress REST API. - [Memberspot Integration](https://flowrunner.ai/integrations/memberspot.md): Connect AI agents to Memberspot, a German online course and membership platform. Agents manage users, grant or revoke offer and chapter access, set order states, and read course progress and exam results so access provisioning stays aligned with payments. - [Memberstack Integration](https://flowrunner.ai/integrations/memberstack.md): Manage Memberstack members and plans from AI agents: sync new sign-ups as members, update custom fields and metadata, attach or remove free plans to control access, and look up members to drive downstream automation. - [MemberVault Integration](https://flowrunner.ai/integrations/membervault.md): Connect AI agents to MemberVault, a course and membership platform. Agents fetch products, add or remove users from products, and delete users so enrollment stays aligned with purchases and refunds. - [MemoMeister Integration](https://flowrunner.ai/integrations/memomeister.md): Connect AI agents to MemoMeister, a construction and trades documentation platform from Freiraum GmbH. Agents file field memos into project folders, label them, control folder permissions, and search across active memos with autocomplete. - [Mentortools Integration](https://flowrunner.ai/integrations/mentortools.md): Connect AI agents to Mentortools, a German online course and membership platform. Agents grant and revoke member course access on each scheduled run so purchases unlock content and refunds close it without manual work. - [Cisco Meraki Integration](https://flowrunner.ai/integrations/meraki.md): Connect AI agents to your Cisco Meraki cloud-managed network. Agents enumerate organizations, networks, and devices, monitor device availability, audit connected clients and data usage, and update device metadata across the deployment. - [Merci Facteur Integration](https://flowrunner.ai/integrations/merci-facteur.md): Connect AI agents to Merci Facteur, a French postal mail API that prints and sends letters through La Poste. Agents send standard, tracked, or registered letters, estimate postage, track and cancel mailings, and manage addresses so physical correspondence runs from the same workflows as digital channels. - [Merk Integration](https://flowrunner.ai/integrations/merk.md): Resolve companies against the complete Czech and Slovak business registers with Merk, matching on name, ICO, VAT number, email, or bank account. Agents verify a counterparty against the official record before onboarding or paying. - [MessageBird Integration](https://flowrunner.ai/integrations/messagebird.md): Send SMS and text-to-speech voice messages, run one-time-password verification, validate phone numbers, and manage contacts through the MessageBird (Bird) platform. - [MessengerOS Integration](https://flowrunner.ai/integrations/messenger-os.md): Send email and SMS campaigns through a MessengerOS account, using template-based or custom content. Agents deliver campaign messages and handle the subscribe and unsubscribe lifecycle. - [Meta Ads Integration](https://flowrunner.ai/integrations/meta-ads.md): Run the full Meta advertising stack from AI agents: create and manage campaigns, ad sets, ads, and creatives, build customer-file custom audiences with automatic hashing, and pull Insights performance metrics for reporting, across 28 actions. - [Metabase Integration](https://flowrunner.ai/integrations/metabase.md): Run saved questions and ad-hoc queries, manage cards, collections, dashboards, and databases, and export results as JSON or CSV from your flows over the Metabase API. - [Metricool Integration](https://flowrunner.ai/integrations/metricool.md): Schedule and measure social content through Metricool. Agents discover connected brands, create and audit scheduled posts across networks like Instagram, LinkedIn, and TikTok, and pull follower and engagement analytics for dashboards. - [MFR (Mobile Field Report) Integration](https://flowrunner.ai/integrations/mfr-field-service-management.md): Connect AI agents to MFR (Mobile Field Report), a field service platform from simPlias GmbH. Agents open service requests, schedule appointments, track service objects and parts, and read the reports technicians file. - [Microsoft Advertising Campaign Management Integration](https://flowrunner.ai/integrations/microsoft-ad-campaign-mgmt.md): Manage Microsoft Advertising campaigns from AI agents through the Campaign Management v13 SOAP service. Agents create and update campaigns in bulk, list ad groups, and reach any other v13 operation via a raw SOAP escape hatch. - [Microsoft Advertising Reports Integration](https://flowrunner.ai/integrations/microsoft-advertising-reports.md): Automate Microsoft Advertising performance reporting: agents submit report requests for campaign, keyword, and audience metrics, poll until generation completes, and retrieve the download URL for the finished file. - [Dynamics 365 Business Central Integration](https://flowrunner.ai/integrations/microsoft-d365-bc.md): Connect AI agents to Dynamics 365 Business Central. Agents manage customers, vendors, and items, create and post sales invoices and orders, record general journal lines, and sync purchase documents across 45 accounting actions. - [Microsoft Excel 365 Integration](https://flowrunner.ai/integrations/microsoft-excel.md): Read and write Excel 365 workbooks in OneDrive from your flows: append table rows, read ranges as objects, and manage worksheets and tables. Agents keep spreadsheets current and confirm before overwriting shared regions. - [Microsoft OneDrive Integration](https://flowrunner.ai/integrations/microsoft-onedrive.md): Browse, search, upload, and download files in the connected user's OneDrive from FlowRunner agents through the Microsoft Graph API, plus folder management and sharing links. - [Microsoft OneNote Integration](https://flowrunner.ai/integrations/microsoft-onenote.md): Turn OneNote into an agent-writable knowledge surface. Agents create pages from HTML, append structured updates to existing pages, scaffold notebooks and sections, read page content as XHTML, and run async copy operations with built-in polling. - [Microsoft Planner Integration](https://flowrunner.ai/integrations/microsoft-planner.md): Connect AI agents to Microsoft Planner boards. Agents create plans in Microsoft 365 groups, organize buckets, create and assign tasks with dates, priority, and checklists, and update progress with etag handling managed automatically. - [Microsoft Power Automate Integration](https://flowrunner.ai/integrations/microsoft-power-automate.md): Microsoft Power Automate is Microsoft's cloud workflow service. Agents list, turn on, and turn off cloud and desktop flows through the Dataverse Web API, read and write any Dataverse table in the environment, export and import solutions, and call a flow's HTTP request trigger. - [Microsoft Power BI Integration](https://flowrunner.ai/integrations/microsoft-power-bi.md): Operate Power BI from AI agents: refresh datasets and verify outcomes, run DAX queries for KPIs, stream rows into push datasets, and export reports to PDF, PowerPoint, or PNG into FlowRunner file storage across 36 actions. - [Microsoft Teams Integration](https://flowrunner.ai/integrations/microsoft-teams.md): Work with Microsoft Teams over the Microsoft Graph API. List teams and channels, send and read channel messages, reply to threads, create and delete channels, run one-on-one and group chats, and inspect membership as the connected user. - [Microsoft To Do Integration](https://flowrunner.ai/integrations/microsoft-todo.md): Manage Microsoft To Do task lists and tasks from your flows: create tasks with due dates and reminders, complete or reopen them, and manage checklist steps. Agents capture work and confirm before escalating a task to a person. - [Milvus Integration](https://flowrunner.ai/integrations/milvus.md): Open-source vector database (and Zilliz Cloud) for similarity search and agent memory. Manage collections, write and delete entities, run vector search, and administer indexes and partitions. - [MIME Tools Integration](https://flowrunner.ai/integrations/mime-tools.md): Map file extensions to MIME types and back, entirely offline with no API key or external call. Agents set the correct content type before a file moves between systems that care about it. - [Mindee Integration](https://flowrunner.ai/integrations/mindee.md): Extract structured data from invoices, receipts, IDs, and custom document types with the Mindee V2 API. Runs OCR and parsing jobs and returns clean fields ready for downstream systems. - [MindStudio Integration](https://flowrunner.ai/integrations/mindstudio-ai.md): Invoke your published MindStudio agents from FlowRunner flows, synchronously or via async callback, passing launch variables in and returning structured output for downstream steps. - [Mirakl Integration](https://flowrunner.ai/integrations/mirakl.md): Run marketplace operations on Mirakl from AI agents: list and accept incoming orders, attach tracking and validate shipments, bulk-import offers, look up catalog products, and route seller message threads to your team. - [Miro Integration](https://flowrunner.ai/integrations/miro.md): Connect AI agents to Miro boards. Agents spin up boards for projects and workshops, turn incoming ideas and tasks into sticky notes and cards, browse board items for syncing, and manage sharing and member access. - [MISP Integration](https://flowrunner.ai/integrations/misp.md): Connect AI agents to MISP, the open-source threat intelligence platform. Agents create and manage threat events, attach and search indicators of compromise, apply TLP tags, publish to communities, and record sightings. - [Missive Integration](https://flowrunner.ai/integrations/missive.md): Connect AI agents to Missive shared inboxes. Agents read conversations, send email drafts, post internal comments and notifications, manage contacts and contact books, apply shared labels, and route work to the right team, across 20 actions. - [Mistral AI Integration](https://flowrunner.ai/integrations/mistral-ai.md): Run the full Mistral AI platform in your flows: chat and vision, Document AI OCR, embeddings, code completion, moderation, audio, agents, and batch processing. - [Mitto Integration](https://flowrunner.ai/integrations/mitto-sms.md): Connect AI agents to Mitto, a Swiss enterprise CPaaS for A2P SMS, one-time passwords, and number lookup. Agents send single and bulk SMS, deliver and verify OTP codes, and look up numbers via HLR so customer notifications and verification steps run inside governed workflows. - [Mixmax Integration](https://flowrunner.ai/integrations/mixmax.md): Drive Mixmax sales engagement from AI agents: enroll leads into sequences, keep contacts in sync with your system of record, send tracked emails on behalf of the authenticated user, and look up sequences for routing. - [Mixpanel Integration](https://flowrunner.ai/integrations/mixpanel.md): Connect AI agents to Mixpanel product analytics across US, EU, and India residencies. Agents ingest events and profiles, stitch identities, run segmentation, funnel, and retention queries, export raw events, and sync Lexicon schemas across 36 actions. - [Mobivate SMS Integration](https://flowrunner.ai/integrations/mobivate-sms.md): Connect AI agents to Mobivate SMS, a UK bulk messaging gateway with campaigns, contacts, and wallet management. Agents send single and batch SMS, run template campaigns, manage contacts, groups, and opt-outs, and check wallet balances so outbound messaging stays organized and compliant. - [Mocean Integration](https://flowrunner.ai/integrations/mocean.md): Send SMS, run two-factor verification flows, look up carrier and portability data, and monitor account balance and pricing through the MoceanAPI network. - [Mockingbyrd Integration](https://flowrunner.ai/integrations/mockingbyrd.md): Read and act on SMS conversations in Mockingbyrd, an AI SMS platform for collecting and following up with leads. Agents work the routine replies and hand the thread to a person when it turns into a real negotiation. - [MOCO Integration](https://flowrunner.ai/integrations/moco.md): MOCO is the German agency ERP: projects, time tracking, resource planning, invoicing, offers, deals and purchasing in one system. Agents create projects and activities, log time, issue offers and invoices, manage deals and start flows from MOCO web hooks. - [Mollie Integration](https://flowrunner.ai/integrations/mollie.md): Connect AI agents to Mollie to run the full payment lifecycle: create hosted checkout payments and payment links, issue refunds, track chargebacks, and manage customers, mandates, and subscriptions for recurring billing. Agents also reconcile payouts through balances, settlements, and Mollie fee invoices. - [Monday.com Integration](https://flowrunner.ai/integrations/monday.md): Automate Monday.com boards, items, groups, and workspaces through the GraphQL API: create and update items, change column values, post updates, and watch boards for changes. Agents run routine updates and gate bulk column changes behind a person. - [Moneybird Integration](https://flowrunner.ai/integrations/moneybird.md): Moneybird is the Dutch bookkeeping platform. Agents manage contacts, sales and purchase invoices, estimates, recurring invoices, and subscriptions, book receipts and journal entries, reconcile banking, and log project time entries. - [MongoDB Integration](https://flowrunner.ai/integrations/mongodb.md): Connect AI agents to MongoDB and MongoDB Atlas. Agents run the full document lifecycle, aggregation pipelines, collection and index management, and Atlas vector search. - [Monica Integration](https://flowrunner.ai/integrations/monica.md): Connect AI agents to Monica, the personal relationship management app. Agents manage contacts, notes, activities, tasks, reminders, calls, tags, and journal entries. - [Moodle Integration](https://flowrunner.ai/integrations/moodle.md): Connect AI agents to your Moodle LMS to create user accounts, provision courses, enrol learners, and pull rosters, assignments, and grades through Moodle Web Services. Useful for onboarding learners from signup systems and syncing grade data into reporting tools. - [Moon Invoice Integration](https://flowrunner.ai/integrations/moon-invoice.md): Moon Invoice is an invoicing and billing platform for small businesses. Agents issue invoices, sales receipts, proforma invoices, estimates, delivery challans, and credit notes, and manage purchase orders, bills, and expenses on the other side. - [Moosend Integration](https://flowrunner.ai/integrations/moosend.md): Keep Moosend email marketing lists in sync from your flows: agents subscribe, look up, unsubscribe, and remove members with custom fields, discover lists with member counts, and review campaign send history for reporting. - [MoreApp Integration](https://flowrunner.ai/integrations/moreapp-forms.md): MoreApp is a Dutch digital forms and work order platform for field teams. Agents read submissions and their PDF reports, create and dispatch tasks, manage forms, folders, users and groups, keep data sources in sync, and react on every new submission. - [Morning Integration](https://flowrunner.ai/integrations/morning.md): Morning, formerly Green Invoice, is the Israeli invoicing and expense platform. Agents create documents, manage clients, suppliers, and catalog items, log expenses, generate hosted payment pages, and charge cards a customer already saved by paying through one. - [Moskit CRM Integration](https://flowrunner.ai/integrations/moskitcrm.md): Connect AI agents to Moskit CRM, a sales CRM for Brazilian small and medium businesses. Agents create companies, contacts, and deals, log activities and notes against them, and add products to the catalog those deals draw on. - [Motion Integration](https://flowrunner.ai/integrations/motion.md): Create tasks that Motion's AI calendar schedules automatically: agents add work with priority, due date, and duration, manage projects and workspaces, and read users and schedules to drive assignment logic. - [Motionlab Integration](https://flowrunner.ai/integrations/motionlab.md): Motionlab renders one personalized video per recipient, composing their name, photo, location, or offer into the footage. Agents generate videos from a campaign, poll render status, read campaign parameters, and run test mode before spending credits on a real send. - [Microsoft Advertising Offline Conversions Integration](https://flowrunner.ai/integrations/ms-advertising-conversions.md): Report offline conversions back to Microsoft Advertising so bidding and reporting reflect real revenue: agents apply conversions and adjustments in batches of up to 1,000 against the Microsoft Click ID, and can invoke any Campaign Management v13 SOAP operation through the escape hatch. - [Microsoft Graph Security Integration](https://flowrunner.ai/integrations/ms-graph-security.md): Connect AI agents to the Microsoft Graph Security API across Microsoft Defender and Sentinel. Agents triage alerts and incidents, track Microsoft Secure Score posture, and manage threat intelligence indicators. - [MSG91 Integration](https://flowrunner.ai/integrations/msg91.md): Connect AI agents to MSG91. Agents send and verify one-time passwords, send transactional SMS and WhatsApp messages through approved templates, send transactional email, and check credit balances. - [MuleSoft Anypoint Integration](https://flowrunner.ai/integrations/mulesoft.md): MuleSoft Anypoint Platform is Salesforce's integration platform. Workflows search Exchange for assets, list environments, start, stop, and update CloudHub applications, read their logs and deployments, and inspect managed APIs. - [Mumara Integration](https://flowrunner.ai/integrations/mumara.md): Mumara Campaigns is a self-hosted email marketing platform you run on your own server. Agents manage contact lists and contacts, schedule broadcasts, pull delivery statistics, and handle suppression, bounces and feedback loops on your installation. - [Mumble Integration](https://flowrunner.ai/integrations/mumble.md): Connect AI agents to Mumble, a WhatsApp business messaging platform for customer conversations. Agents manage customers and labels, send session and template messages, route conversations to agents, teams, or chatbots, and create message templates so WhatsApp outreach and support run without manual triage. - [Mural Integration](https://flowrunner.ai/integrations/mural.md): Put agent output onto Mural whiteboards: navigate the workspace, room, and mural hierarchy, create new murals, read a board's widgets, and post sticky notes so AI brainstorms land on a shared canvas. - [Murf AI Integration](https://flowrunner.ai/integrations/murf-ai.md): Generate studio-quality AI voice overs with Murf AI: agents turn text into natural speech across a catalog of voices, re-voice existing recordings, translate narration scripts, and dub audio or video into new languages, with output stored durably in FlowRunner. - [MusicGPT Integration](https://flowrunner.ai/integrations/musicgpt.md): Generate songs, sound effects, lyrics, and speech, and reshape existing recordings, with the MusicGPT public API. Agents produce audio assets inside the workflow that needs them. - [MyCarTracks Integration](https://flowrunner.ai/integrations/mycartracks.md): Connect AI agents to MyCarTracks, a vehicle GPS tracking and mileage logging service. Agents read vehicles, live positions, recorded trips, geofence alerts, and driver statistics so fleet mileage and location data feed reports and workflows automatically. - [Myphoner Integration](https://flowrunner.ai/integrations/myphoner.md): Connect AI agents to Myphoner, a cold calling and lead management CRM for outbound teams. Agents load lists and leads, distribute them to callers, and record dispositions and call outcomes. - [MyPreferences Integration](https://flowrunner.ai/integrations/mypreferences.md): Connect AI agents to MyPreferences by PossibleNOW, an enterprise consent and preference management platform. Agents add, deactivate, and permanently merge a customer's communication preferences, and read the programs, segments, and distribution lists those preferences belong to. - [MySQL Integration](https://flowrunner.ai/integrations/mysql.md): Connect AI agents to MySQL and MySQL-compatible servers such as MariaDB, PlanetScale, and Aurora. Agents run parameterized SQL, do CRUD without SQL, inspect schemas, and bulk-load rows. - [Namsor Integration](https://flowrunner.ai/integrations/namsor.md): Classify personal names with Namsor to infer likely gender, country of origin, country of residence, and diaspora. Every result carries a confidence score, so agents can route uncertain cases to a person. - [Nanonets Integration](https://flowrunner.ai/integrations/nanonets.md): Extract structured data from invoices, receipts, and other documents with trained Nanonets OCR models. Agents run synchronous or asynchronous extraction on files and URLs, poll prediction results, and upload labeled training data to improve custom OCR and classification models. - [NASA Integration](https://flowrunner.ai/integrations/nasa.md): Access NASA's open APIs for astronomy imagery, Mars rover photos, near-Earth object tracking, full-disc Earth imagery, Landsat imagery, space-weather events, and the NASA Image and Video Library. - [natif.ai Integration](https://flowrunner.ai/integrations/natif-ai.md): Submit documents to natif.ai workflows for invoice extraction, receipt extraction, and OCR on the German intelligent document processing platform. Agents turn inbound paperwork into typed fields ready for approval. - [Navan Integration](https://flowrunner.ai/integrations/navan.md): Pull corporate travel and expense data out of Navan: agents list and retrieve expense and card transactions, fetch receipts for audit evidence, surface embedded travel booking details, and read custom fields to map spend into accounting systems. - [Neaktor Integration](https://flowrunner.ai/integrations/neaktor.md): Neaktor is business process and project management where task models define fields and status workflows and tasks are instances of a model. Agents create tasks against the right model, move them through their workflow, mirror comments and report spent time. - [NectarCRM Integration](https://flowrunner.ai/integrations/nectar-crm.md): Connect AI agents to NectarCRM, a Brazilian B2B sales CRM from Colmeia Solucoes. Agents create contacts and opportunities, run qualifications, schedule appointments, and log notes against each deal. - [Nedzo AI Integration](https://flowrunner.ai/integrations/nedzo.md): Place outbound AI voice calls and manage agents, contacts, and workspaces with the Nedzo REST API. Agents run phone conversations and read the outcome back into the pipeline. - [nele.ai Integration](https://flowrunner.ai/integrations/nele-ai.md): Reach chat completions across multiple models through nele.ai, the German company AI assistant gateway, with one API key. Agents run generation under a data posture German compliance teams already accept. - [NeoGate (SMSbrána) Integration](https://flowrunner.ai/integrations/neogate.md): Send SMS through the Czech SMSbrana gateway using its SMS Connect API. Agents deliver transactional and alert messages to Czech recipients and read delivery status back. - [Neon CRM Integration](https://flowrunner.ai/integrations/neoncrm.md): Run nonprofit operations against Neon CRM: agents create and search constituent accounts, record donations against the right campaign and fund, enroll memberships, log activities, and pull pledges, grants, and event registrations for reporting. - [Nero AI Integration](https://flowrunner.ai/integrations/nero-ai.md): Upscale, restore, edit backgrounds, and animate images with Nero AI. Agents prepare visual assets for listings and campaigns without a manual editing pass. - [NetHunt CRM Integration](https://flowrunner.ai/integrations/nethunt.md): Work records in NetHunt, the Gmail-native CRM: agents discover folders and their field schemas, find, create, update, and delete records, and log comments and calls to keep interaction history complete. - [Netlify Integration](https://flowrunner.ai/integrations/netlify.md): Manage Netlify sites, deploys, environment variables, forms, and DNS from your flows over the Netlify API. Trigger builds, roll back to a known-good deploy, and sync configuration. - [Neto Integration](https://flowrunner.ai/integrations/neto.md): Neto, now sold as Maropost Commerce Cloud, is an Australian ecommerce, POS and inventory suite. Agents keep stock in step with warehouses and suppliers, hand new orders to a 3PL or accounting system, manage products, customers and vouchers, and quote shipping. - [NetSuite Integration](https://flowrunner.ai/integrations/netsuite.md): Bridge your ERP and your AI agents. NetSuite integration gives agents direct access to customers, vendors, invoices, sales orders, payments, and custom SuiteQL queries. - [Netwo Integration](https://flowrunner.ai/integrations/netwo.md): Connect AI agents to Netwo, a French wholesale telecom marketplace for fiber, mobile, and connectivity. Agents check eligibility, place and track orders, handle number portability, update mobile subscriptions, and read consumption per SIM and customer. - [NeverBounce Integration](https://flowrunner.ai/integrations/neverbounce.md): Verify email addresses before they hurt deliverability: agents check a single address in real time, run bulk NeverBounce verification jobs over whole lists, poll job progress, and pull per-email results with typo corrections and remaining credit balances. - [Super News Integration](https://flowrunner.ai/integrations/news.md): Super News is the TopHub Data API for trending content rankings across the Chinese internet, from Weibo and Zhihu to Douyin, Baidu, and GitHub. Agents list hot boards, read board history, search trending content, and react when a new item enters a ranking. - [Newsman Integration](https://flowrunner.ai/integrations/newsman.md): Newsman is a Romanian email service provider that puts marketing email, transactional SMTP, SMS, automations, coupons and ecommerce remarketing behind one API. Agents manage lists and subscribers, send newsletters and transactional messages, and sync store data for remarketing. - [Nextcloud Integration](https://flowrunner.ai/integrations/nextcloud.md): Manage files, folders, shares, and users on your own Nextcloud server from FlowRunner agents through the WebDAV and OCS APIs, authenticated with a scoped app password. - [Nimba SMS Integration](https://flowrunner.ai/integrations/nimbasms.md): Connect AI agents to Nimba SMS, an SMS gateway serving Guinea and the wider West African region. Agents send SMS and WhatsApp template messages, verify phone numbers, manage contacts, and check delivery status and balances so regional customer outreach runs without manual dispatch. - [Nimble Integration](https://flowrunner.ai/integrations/nimble.md): Connect AI agents to Nimble, the social relationship CRM. Agents look up contacts by email before creating duplicates, enrich people and companies with tags and notes, log activities, schedule follow-up tasks, and move deals through their pipelines. - [Ninja Forms Integration](https://flowrunner.ai/integrations/ninja-forms.md): Pull Ninja Forms data out of WordPress: agents list a site's forms, read stored submissions, and retrieve individual entries to route captured leads into CRMs, spreadsheets, and notification tools. - [Ninox Integration](https://flowrunner.ai/integrations/ninox.md): Read and write your Ninox low-code databases: agents browse teams, databases, and tables, list and filter records, create or update them in bulk, and run Ninox script queries to aggregate data across tables. - [NIP24 Integration](https://flowrunner.ai/integrations/nip24-pl.md): Look up verified Polish company data and tax compliance status from official government registries with NIP24, searching by NIP, REGON, or KRS number. Agents confirm a counterparty is real and in good standing before invoicing. - [NocoDB Integration](https://flowrunner.ai/integrations/nocodb.md): Connect AI agents to NocoDB, the open-source no-code database. Agents read and write table records, manage linked records, provision tables, and inspect base and view metadata. - [NoCodeTable Integration](https://flowrunner.ai/integrations/nocodetable.md): NoCodeTable is a hosted spreadsheet-database with documents, tables, fields, views and records. Agents create and query records, manage tables, fields and views, validate formula fields and set document, table and record configuration. - [noCRM.io Integration](https://flowrunner.ai/integrations/nocrm-io.md): Drive a noCRM.io sales pipeline end to end: agents create leads, change status and pipeline step, assign reps, log comments, feed cold prospects into prospecting lists and convert them into leads, and read users, pipelines, and tags for routing. - [Noloco Integration](https://flowrunner.ai/integrations/noloco.md): Noloco is a no-code internal app builder, and every app gets its own GraphQL API. Workflows list, create, update, and delete records, describe tables, and run raw queries against the app's data. - [Notificacoes Inteligentes Integration](https://flowrunner.ai/integrations/notificacoes-inteligentes.md): Connect AI agents to Notificacoes Inteligentes, a Brazilian platform that sends WhatsApp notifications for ecommerce stores. Agents dispatch order, shipping, and cart events, manage contacts, tags, and lists, and maintain custom variables so every store milestone reaches customers on WhatsApp automatically. - [Notify Integration](https://flowrunner.ai/integrations/notify.md): Send SMS, email, push, Slack, and webhook notifications from a Notify account, which bundles many providers behind one connection. Agents pick the channel per recipient without a contract per provider. - [Notion Integration](https://flowrunner.ai/integrations/notion.md): Turn Notion into an active workspace. Agents create and update pages, manage databases, build content, and collaborate on records. 22 actions for complete Notion management. - [Nozbe Integration](https://flowrunner.ai/integrations/nozbe-teams.md): Nozbe, formerly Nozbe Teams, is the team task and project manager. Agents create projects and sections, add tasks with comments, attachments, reminders, and recurrences, manage tags and project access, and poll for new or changed tasks. - [npm Registry Integration](https://flowrunner.ai/integrations/npm.md): Read the public npm registry to fetch package metadata and versions, resolve dist-tags, search packages, and pull download statistics. Public reads need no token; an optional token reads private packages. - [Nuelink Integration](https://flowrunner.ai/integrations/nuelink.md): Connect AI agents to Nuelink, a social media scheduling and publishing platform. Agents upload media and schedule posts into a collection, which distributes them to every channel in it. - [Numverify Integration](https://flowrunner.ai/integrations/numverify.md): Validate and look up phone numbers across 232 countries and territories with Numverify by APILayer. Agents confirm a number is real and identify its carrier and line type before a message is sent. - [Nusii Integration](https://flowrunner.ai/integrations/nusii.md): Nusii is proposal software for agencies and freelancers. Agents raise proposals from CRM deals using templates, create clients, build sections and priced line items, follow the activity trail, and react when a client views or accepts. - [Nutshell Integration](https://flowrunner.ai/integrations/nutshell.md): Manage sales in Nutshell CRM through its JSON-RPC API: agents search people and companies by email before creating duplicates, open and advance leads through pipelines, log calls and meetings as activities, and read team, tag, and pipeline configuration. - [NVIDIA NIM Integration](https://flowrunner.ai/integrations/nvidia.md): Run hundreds of GPU-hosted open models through NVIDIA NIM's OpenAI-compatible endpoint: agents call Llama, Nemotron, and DeepSeek chat models with tool calling, generate embeddings and rerank passages with NeMo Retriever for RAG, and discover models from the live catalog. - [Nvoip Integration](https://flowrunner.ai/integrations/nvoip.md): Use the Nvoip Brazilian VoIP and SMS provider for outbound calls, SMS, voice blasts, one-time passcodes, virtual numbers, and sub-accounts. Agents reach Brazilian customers on the channel each situation calls for. - [OANDA Exchange Rates Integration](https://flowrunner.ai/integrations/oanda-exchange-rates.md): OANDA Exchange Rates is a currency data API backed by OANDA and central bank sources. Agents fetch spot rates and candles, pull historical ranges, convert amounts, list currencies, and check remaining plan quota. - [Oblio Integration](https://flowrunner.ai/integrations/oblio.eu-facturare.md): Oblio is Romanian invoicing and stock software. Agents issue proformas, delivery notices, and invoices, record collections, file e-Factura to the national SPV system, and look up the nomenclatures every document has to reference. - [Octopus Energy Integration](https://flowrunner.ai/integrations/octopus-energy.md): Connect AI agents to Octopus Energy, the UK energy supplier. Agents read public tariffs and product data, pull half-hourly meter consumption, and check account and industry records. - [Odoo Integration](https://flowrunner.ai/integrations/odoo.md): A generic ORM client for any Odoo model. Search, read, create, update, and delete records, discover model fields, and call arbitrary methods over Odoo’s External API. - [Odyssee Field Service Integration](https://flowrunner.ai/integrations/odyssee-field-service.md): Connect AI agents to Odyssee Field Service (also branded Wello), field service management software. Agents open and update work orders, dispatch tasks to technicians, manage companies, contacts, and projects, book meetings, and read the article and purchase order records behind a job. - [OfficeGuy Integration](https://flowrunner.ai/integrations/officeguy.md): OfficeGuy, also known as SUMIT, is the Israeli business management and billing platform. Agents create accounting documents, manage customers, income items, and expenses, run payments and recurring billing, and work the CRM entity store and stock underneath. - [Officely Integration](https://flowrunner.ai/integrations/officely.md): Officely AI runs conversational AI agents across WhatsApp, Zendesk, Intercom, and web chat. Agents send a chat flow or a single message to a customer, hand a conversation between the bot and a human, schedule calls, and run an AI agent team. - [OfficeRnD Integration](https://flowrunner.ai/integrations/officernd.md): Connect AI agents to OfficeRnD Flex, the coworking and flexible workspace management platform. Agents onboard members and companies, check them in and out, book space, open memberships and tickets, and read the invoices billing produces. - [OkaySend Integration](https://flowrunner.ai/integrations/okaysend.md): OkaySend collects documents from clients through a checklist delivered as a secure portal, then chases them until every item is in. Agents send onboarding requests the moment a deal closes, hold them as drafts for review, track progress, and manage the client records behind them. - [Okta Integration](https://flowrunner.ai/integrations/okta.md): Connect AI agents to your Okta org. Agents manage users, groups, and group rules, assign applications and admin roles, run MFA and password help-desk flows, read the System Log, rotate credentials, and revoke sessions and OAuth grants. - [Olark Integration](https://flowrunner.ai/integrations/olark.md): Keep Olark live chat staffed correctly: agents provision and update chat operators, remove departing ones, and manage group membership so incoming conversations route to the right team. - [Ollama Integration](https://flowrunner.ai/integrations/ollama.md): Run open-source models like Llama, Qwen, DeepSeek, and Gemma on your own hardware and call them directly from flows. Generation, embeddings, and model management with no data leaving your infrastructure. - [Olostep Integration](https://flowrunner.ai/integrations/olostep.md): Olostep is a scraping, crawling, and search API built for AI pipelines. Agents scrape single pages with structured extraction, map and crawl whole sites, batch up to 100,000 URLs, search the web, get retrieval-backed answers, and monitor pages for changes. - [OLX Integration](https://flowrunner.ai/integrations/olx.md): Publish and manage classifieds on the OLX marketplace: agents create, update, and change the status of adverts, look up leaf categories, required attributes, and city IDs before posting, and read and reply to buyer message threads. - [Opn Payments Integration](https://flowrunner.ai/integrations/omise.md): Opn Payments, formerly Omise, is a Thai payment gateway for cards, sources, and recurring charges. Agents create and capture charges, tokenize cards, issue refunds, manage customers and schedules, and track disputes with their evidence deadlines. - [Omnichat Integration](https://flowrunner.ai/integrations/omnichat.md): Automate the Omnichat chat commerce platform across LINE, WhatsApp, Facebook, Instagram, WeChat, and web. Agents run conversational selling in Taiwan and Hong Kong markets where chat is the storefront. - [Omnisend Integration](https://flowrunner.ai/integrations/omnisend.md): Feed your Omnisend ecommerce marketing from any workflow: agents sync contacts with tags and subscription status, trigger custom events that launch automations, add products to the catalog, and read campaigns and abandoned carts. - [Omnivery Integration](https://flowrunner.ai/integrations/omnivery.md): Omnivery is a transactional email delivery platform focused on deliverability. Agents send mail from a verified domain, parse and validate addresses, manage suppressions, detect bot traffic, and configure domains and webhooks. - [Onboard Integration](https://flowrunner.ai/integrations/onboard.md): Connect AI agents to Onboard.io, a customer onboarding project management platform. Agents create customer roadmaps and tasks with an assignee, update the customer-facing plan, and read the team and customer members behind each launch. - [One Simple API Integration](https://flowrunner.ai/integrations/one-simple-api.md): Connect over a single API token to a utility toolkit that bundles screenshots, PDFs, QR codes, currency conversion, email validation, domain and SSL expiry checks, and URL and metadata tools. - [123FormBuilder Integration](https://flowrunner.ai/integrations/one-two-three-form-builder.md): Bring 123FormBuilder responses into your workflows: agents list account forms, inspect a form's field structure, and retrieve submissions to sync entries into CRMs, spreadsheets, and notification flows. - [OneClick for Pipedrive Integration](https://flowrunner.ai/integrations/oneclick-for-pipedrive.md): Complete the round trip for OneClick for Pipedrive by Zimple, the add-on that puts custom buttons on Pipedrive deals. Agents receive the button press and run the work the salesperson just asked for. - [1CRM Integration](https://flowrunner.ai/integrations/onecrm.md): Connect AI agents to a 1CRM instance, a small business CRM and order management suite. Agents create accounts, contacts, quotes, and invoices, read and write any module through generic record operations, link related records, and upload files. - [OneDeck Integration](https://flowrunner.ai/integrations/onedeck.md): OneDeck is a no-code business management platform built on boards holding records, plus generated documents such as work orders and quotes. Agents read and write records on any board and pull the documents produced from them. - [OnePageCRM Integration](https://flowrunner.ai/integrations/onepagecrm.md): Connect AI agents to OnePageCRM, an action-focused sales CRM that turns every contact into a next action. Agents create contacts and deals, set the next action, log calls and meetings, and post notes against a record. - [OneSignal Integration](https://flowrunner.ai/integrations/onesignal.md): Send push notifications, email, and SMS through OneSignal and manage the audience behind them: agents create and update users, aliases, and channel subscriptions, build segments, send with templates, schedule or cancel messages, and pull delivery and outcome analytics. - [Onfleet Integration](https://flowrunner.ai/integrations/onfleet.md): Last-mile delivery management for FlowRunner. Create and dispatch delivery tasks, manage drivers and teams, define reusable destinations and recipients, and auto-dispatch routes over the Onfleet API. - [OnlineCheckWriter Integration](https://flowrunner.ai/integrations/onlinecheckwriter.md): OnlineCheckWriter, powered by Zil Money, prints, mails, and electronically sends payments from your own bank accounts. Agents create and mail checks, send ACH, wire, RTP, and virtual card payouts, manage payees and wallets, and collect money through payment links. - [Onoff Business Integration](https://flowrunner.ai/integrations/onoff-business.md): Manage members, numbers, departments, contacts, and call and SMS logs on Onoff Business. Agents provision business lines and keep call activity attached to the right record. - [Onsefy Integration](https://flowrunner.ai/integrations/onsefy.md): Connect AI agents to Onsefy, an API-first fraud detection and user validation platform. Agents validate a user at signup, and apply the same check again at any high-value action a flow reaches. - [Ontraport Integration](https://flowrunner.ai/integrations/ontraport.md): Connect AI agents to Ontraport, the marketing automation and CRM platform. Agents create and update contacts, apply and remove tags for segmentation, work with any object type including deals and custom objects, and look up email messages. - [OOPSpam Integration](https://flowrunner.ai/integrations/oopspam-anti-spam.md): Connect AI agents to OOPSpam, a spam detection API for forms, comments, and signups. Agents score submissions for spam, check a domain against DNS blocklists, and report misdetections back for retraining. - [OpenPhone Integration](https://flowrunner.ai/integrations/open-phone.md): Work your OpenPhone business phone system from any flow: agents send SMS and MMS from shared team numbers, pull call history with recordings, AI summaries, and transcripts, sync contacts with custom fields, and register webhooks for real-time message and call events. - [OpenAI Integration](https://flowrunner.ai/integrations/openai-ai.md): Integrate OpenAI's full platform: the Responses and Chat Completions APIs with structured output and web search, embeddings, DALL-E and gpt-image-1, Whisper and TTS, files, batches, vector stores, and video. - [OpenCart Integration](https://flowrunner.ai/integrations/opencart.md): Connect AI agents to your OpenCart store through a REST API extension. Agents create orders, look up products, categories, orders, and customers, and sync store data into fulfillment, CRM, and reporting workflows. - [OpenGraph.io Integration](https://flowrunner.ai/integrations/opengraph-io.md): Turn any public URL into structured data with OpenGraph.io, unfurling link metadata, scraping raw HTML, and extracting selected elements. Agents preview and validate links before they are published. - [OpenMic Integration](https://flowrunner.ai/integrations/openmic.md): Place outbound phone calls from a registered number with OpenMic AI voice agents and manage the agents that run them. Agents handle the call and pass the transcript and outcome downstream. - [OpenRouter Integration](https://flowrunner.ai/integrations/openrouter.md): Reach hundreds of models from OpenAI, Anthropic, Google, Meta, Mistral, and DeepSeek through one unified OpenRouter API, with automatic routing, fallbacks, and cost insights. - [openrouteservice Integration](https://flowrunner.ai/integrations/openrouteservice.md): Plan routes for cars, heavy vehicles, bikes, walking, hiking, and wheelchair users with openrouteservice, built on OpenStreetMap. Agents compute realistic travel times and service areas for dispatch and scheduling. - [OpenThesaurus Integration](https://flowrunner.ai/integrations/openthesaurus.md): Look up German-language synonyms using the free, public OpenThesaurus thesaurus. Results group into synonym sets, with optional phonetically similar words, substring and prefix matches, hyponyms, hypernyms, and base forms. No authentication required. - [OpenWeatherMap Integration](https://flowrunner.ai/integrations/openweathermap.md): Access current weather, multi-day forecasts, air quality data, and geocoding from OpenWeatherMap plus the One Call 3.0 API. Locations can be given by city name, coordinates, ZIP code, or city ID. - [Atlassian Opsgenie Integration](https://flowrunner.ai/integrations/opsgenie.md): Atlassian Opsgenie is an incident alerting and on-call platform. Agents create, acknowledge, close, and escalate alerts, open incidents with timelines and postmortems, look up who is on call, and manage schedules, teams, and escalations. - [Optimizely Campaign Integration](https://flowrunner.ai/integrations/optimizely-campaign.md): Manage recipients and mailings through the Optimizely Campaign REST API, covering recipient lists and Smart Campaigns. Agents keep the audience current and trigger sends off events in the systems of record. - [Oracle Database Integration](https://flowrunner.ai/integrations/oracle-database.md): Connect AI agents to Oracle Database in Thin mode (no Instant Client). Agents run SQL and PL/SQL, do CRUD without SQL, and inspect schemas as governed steps in a flow. - [Oracle Eloqua Integration](https://flowrunner.ai/integrations/oracle-eloqua.md): Connect AI agents to Oracle Eloqua, Oracle's B2B marketing automation platform. Agents create and update contacts keyed on email address and browse campaigns, emails, custom objects, and shared contact lists through the Eloqua REST API. - [Oracle Fusion Cloud ERP Integration](https://flowrunner.ai/integrations/oracle-fusion-cloud-erp.md): Connect AI agents to Oracle Fusion Cloud ERP financials and procurement. Agents create AP invoices, suppliers, payments, receivables invoices, and GL journal entries, track purchase orders, requisitions, and expense reports, and reach any other Fusion REST resource through generic advanced actions. - [Oracle Fusion Cloud HCM Integration](https://flowrunner.ai/integrations/oracle-fusion-cloud-hcm.md): Connect AI agents to Oracle Fusion Cloud HCM. Agents onboard workers, read worker assignments, sync active employees, search the person directory, and reach any other HCM REST resource through generic advanced actions. - [Oracle Fusion Cloud Sales Integration](https://flowrunner.ai/integrations/oracle-fusion-cloud-sales.md): Connect AI agents to Oracle Fusion Cloud CX Sales, the CRM inside Oracle Fusion Cloud Customer Experience. Agents create and update accounts, contacts, leads, and opportunities, and log the activities against each. - [Order Desk Integration](https://flowrunner.ai/integrations/order-desk.md): Connect AI agents to Order Desk to centralize and route ecommerce orders. Agents create, update, and delete orders, attach shipment tracking, keep inventory items in sync, and read store settings to discover fulfillment folders. - [Orderry Integration](https://flowrunner.ai/integrations/orderry.md): Connect AI agents to Orderry, a business management platform for repair shops and field service. Agents open orders and estimates, take bookings and inquiries, raise invoices and transactions, and read the product and service catalog a job draws on. - [Origami Integration](https://flowrunner.ai/integrations/origami.md): Origami is the Israeli no-code business platform that runs CRM, HR, and ERP on one schemaless data model. Agents discover entities, query records with Origami's filter language, update fields and repeatable groups, attach files, read history, send email and SMS, and work the invoice system. - [Orsay AI Integration](https://flowrunner.ai/integrations/orsay.md): Connect AI agents to Orsay AI, an AI lead capture platform that runs outreach and replies across social channels. Agents create and activate sequences, read the AI profiles and agents behind them, and read the conversations those agents held with prospects. - [Orshot Integration](https://flowrunner.ai/integrations/orshot.md): Render Orshot Studio designs or prebuilt library templates into PNG, JPEG, WebP, or AVIF images. Agents generate branded visuals per record without a design request. - [Ortto Integration](https://flowrunner.ai/integrations/ortto.md): Connect AI agents to Ortto. Agents merge or create people by email, record custom activities to drive journeys, look people up, and discover typed custom field ids for downstream steps. - [Oura Integration](https://flowrunner.ai/integrations/oura.md): Connect over a Personal Access Token to read sleep, readiness, activity, biometric, and account data from an Oura Ring through the Oura API v2. - [Outline Integration](https://flowrunner.ai/integrations/outline.md): Connect AI agents to Outline, your team knowledge base. Agents create, update, and full-text search wiki documents, append meeting notes and runbooks automatically, browse collections, and look up workspace users. - [Outlook Integration](https://flowrunner.ai/integrations/outlook.md): Read, send, and organize email. Manage calendar events and contacts. Agents draft, send, forward, and reply to emails and create calendar events across Microsoft 365 accounts. - [Outreach Integration](https://flowrunner.ai/integrations/outreach.md): Connect AI agents to Outreach for sales engagement. Agents manage prospects, accounts, and opportunities, enroll prospects in sequences, queue tasks, log calls, and react to real-time events. - [Outseta Integration](https://flowrunner.ai/integrations/outseta.md): Connect AI agents to Outseta, an all-in-one membership platform combining CRM, billing, and support. Agents create people, accounts, and deals, add payments to invoices, open and reply to support cases, and read the subscriptions behind each account. - [Quant Overledger Integration](https://flowrunner.ai/integrations/overledger.md): Quant Overledger puts many blockchains behind one REST gateway. Workflows read network and contract data, send JSON-RPC calls, record deposits, swaps, and withdrawals, and react to address activity and contract events. - [OxaPay Integration](https://flowrunner.ai/integrations/oxapay-crypto-pay-gtw.md): OxaPay is a cryptocurrency payment gateway with hosted invoices and static receiving addresses. Agents generate invoices and white label payments, track payment history, send crypto payouts, run currency swaps, and read balances and price feeds. - [Oxylabs Integration](https://flowrunner.ai/integrations/oxylabs.md): Give AI agents web data collection at scale through the Oxylabs Scraper API. Agents scrape any URL or dedicated targets like Google and Amazon with JavaScript rendering and structured parsing, and run large crawls as asynchronous batch jobs they poll to completion. - [Paced Email Integration](https://flowrunner.ai/integrations/paced-email.md): Paced Email creates inbox aliases that batch incoming mail into daily, weekly or monthly digests. Agents list accounts and custom domains, create and delete alias inboxes, and check how much mail is waiting for the next digest. - [Paddle Integration](https://flowrunner.ai/integrations/paddle.md): Connect AI agents to Paddle, the merchant-of-record billing platform for SaaS and digital products. Agents manage products and prices, look up customers and their subscriptions, run the pause, resume, and cancel lifecycle, create transactions and fetch invoice PDFs, manage discount codes, and issue refunds or credits through the Paddle Billing API. - [Pagar.me Integration](https://flowrunner.ai/integrations/pagar-me.md): Pagar.me is a Brazilian payment gateway from Stone covering cards, Pix, and boleto. Agents create customers, orders, and charges, tokenize cards, run plans and subscriptions, manage marketplace recipients and transfers, and reconcile receivables and settlements. - [PagePixels Screenshots Integration](https://flowrunner.ai/integrations/pagepixels-screenshots.md): Capture, schedule, and monitor website screenshots with the PagePixels Screenshot API, including rendering raw HTML into images. Agents keep visual evidence of what a page looked like when it mattered. - [PagerDuty Integration](https://flowrunner.ai/integrations/pagerduty.md): Automate on-call incident response. Create and run the full incident lifecycle, fire alerts and change events through the Events API v2, and administer services, escalation policies, schedules, users, teams, and maintenance windows. - [Pagerly Integration](https://flowrunner.ai/integrations/pagerly.md): Pagerly manages on-call rotations for Slack and Microsoft Teams. Workflows list teams, look up who is on call right now in Slack or Jira-ready shape, and subscribe a callback URL to rotation changes. - [PandaDoc Integration](https://flowrunner.ai/integrations/pandadoc.md): Connect AI agents to PandaDoc. Agents generate documents from templates or uploaded files, send them for eSignature, create embedded signing links, download completed PDFs, and manage templates, contacts, folders, and webhook subscriptions. - [Paperform Integration](https://flowrunner.ai/integrations/paperform.md): Connect AI agents to Paperform. Agents list forms, retrieve and clean up submissions, and create percentage or fixed-amount discount coupons on payment forms to automate promo campaigns. - [Pappers Integration](https://flowrunner.ai/integrations/pappers.md): Pappers is the French company register and legal data API over the RCS, INSEE Sirene, INPI and BODACC registers. Agents look up companies and associations by SIREN, search directors and beneficial owners, download official extracts and filings, and maintain watchlists. - [Papyrs Integration](https://flowrunner.ai/integrations/papyrs.md): Papyrs is the drag-and-drop intranet and wiki. Agents create and update pages and the widgets inside them, read and write form records, search the site, look up the people directory, and follow the activity stream. - [ParseHub Integration](https://flowrunner.ai/integrations/parsehub.md): ParseHub is a visual web scraping tool. Workflows run the projects you built in the ParseHub client, pass parameters into them, and collect the extracted data when a run is ready. - [Parseur Integration](https://flowrunner.ai/integrations/parseur.md): Turn unstructured documents into structured agent inputs. Parseur extracts data from invoices, emails, and PDFs. Two trigger modes let agents act instantly or poll on a schedule. - [Parsio Integration](https://flowrunner.ai/integrations/parsio.md): Import PDFs, scanned documents, and emails into Parsio, an AI-powered parser, and pull structured fields back. Agents turn inbound paperwork into typed data the moment it arrives. - [Partnero Integration](https://flowrunner.ai/integrations/partnero.md): Run affiliate and referral programs with Partnero, onboarding partners, attributing referred customers, and recording commissions. Agents credit partners off verified conversions rather than a claim. - [PassKit Integration](https://flowrunner.ai/integrations/passkit.md): Create and manage Apple Wallet and Google Wallet passes with PassKit, covering membership and loyalty programs, tiers, members, points, and single-use passes. Agents issue and update passes straight to the customer's phone. - [Pastebin Integration](https://flowrunner.ai/integrations/pastebin.md): Pastebin is the text and code snippet sharing service. Workflows create pastes with expiry and privacy settings, list and read a member's pastes, and pull recent public pastes through the PRO scraping API. - [Patreon Integration](https://flowrunner.ai/integrations/patreon.md): Connect AI agents to your Patreon creator account. Agents pull campaigns, patron pledge and charge details, and published posts to power supporter rosters, thank-you workflows, and revenue dashboards. - [PayFunnels Integration](https://flowrunner.ai/integrations/payfunnels.md): PayFunnels is a payment-link and subscription platform. Agents create one-time, recurring, plan, and pay-what-you-want links, list and refund payments, cancel, pause, and resume subscriptions, and read setup and processing fees. - [Paylocity Integration](https://flowrunner.ai/integrations/paylocity.md): Connect AI agents to Paylocity payroll and HCM. Agents add and update employee records, manage earnings lines, audit local tax setup, and look up department codes to keep payroll in sync with your HR source of record. - [Trolley Integration](https://flowrunner.ai/integrations/payment-rails.md): Trolley, formerly Payment Rails, is a global payouts platform. Agents onboard recipients and their payout accounts, build and submit payment batches, quote currency, manage invoices and invoice payments, and react to payment and recipient events. - [Paymo Integration](https://flowrunner.ai/integrations/paymo.md): Paymo is project management, time tracking, resource scheduling and invoicing for small teams. Agents create projects and tasks, log time, schedule bookings, issue invoices and estimates, and start flows when Paymo records change. - [PayPal Integration](https://flowrunner.ai/integrations/paypal.md): Connect AI agents to PayPal. Agents create and capture Checkout orders, refund or void payments, draft, send, and track invoices, manage subscriptions across the activate, suspend, and cancel lifecycle, and send batch payouts to many recipients at once across the Checkout, Payments, Invoicing, Billing, and Payouts APIs. - [Payrexx Integration](https://flowrunner.ai/integrations/payrexx.md): Payrexx is a Swiss online payment gateway with paylinks, gateways, and card terminals. Agents create gateways and paylinks, charge, capture, and refund transactions, tokenize cards, run subscriptions, issue merchant invoices, and drive ECR terminals. - [Payrexx Platforms Integration](https://flowrunner.ai/integrations/payrexx-platforms.md): Payrexx Platforms is the white-label marketplace side of Payrexx. Agents onboard and manage merchant instances, read and charge their transactions, and run the point-of-sale terminal estate from models and stock to orders and returns. - [Paystack Integration](https://flowrunner.ai/integrations/paystack.md): Accept and reconcile payments across Africa with Paystack. Agents initialize and verify hosted checkout transactions, manage customers and subscription plans, and pay out to bank accounts and mobile-money wallets via transfers. - [pCloud Integration](https://flowrunner.ai/integrations/pcloud.md): Manage a pCloud drive from your flows. Agents upload and download files, create and organize folders, generate public share links, and check account storage, with content moving through FlowRunner file storage for use in other steps. - [PDF-app.net Integration](https://flowrunner.ai/integrations/pdf-app-net.md): Use PDF-app.net for document automation covering PDF generation, conversion between PDF, Word, HTML, and image formats, and document security. Agents produce and protect finished documents inside a workflow. - [PDF Export by PDFCrowd Integration](https://flowrunner.ai/integrations/pdf-export-by-pdfcrowd.md): Convert documents between formats with the PDFCrowd HTTP API, rendering web pages and HTML into PDFs or images and extracting text from existing files. Agents generate and read documents in the same connector. - [PDF Generator API Integration](https://flowrunner.ai/integrations/pdf-generator-api.md): Generate transactional PDF and HTML documents by merging JSON data into browser-editable templates with PDF Generator API, covering the full template lifecycle. Agents produce per-record documents that non-developers can restyle. - [PDF Munk Integration](https://flowrunner.ai/integrations/pdf-munk.md): Convert HTML, CSS, and live web pages into pixel-accurate PDFs and images with PDF Munk, plus a full set of PDF utilities. Agents produce finished documents without maintaining a headless browser. - [PDF Vector Integration](https://flowrunner.ai/integrations/pdf-vector.md): Parse documents, answer questions about them, and extract structured data with PDF Vector, converting PDFs, Office files, and images into clean Markdown. Agents read a document and act on what it says. - [PDF4me Integration](https://flowrunner.ai/integrations/pdf4me.md): Run document automation on the PDF4me v2 REST API, converting, merging, splitting, compressing, securing, watermarking, and OCRing PDF and Office files. Agents handle whatever shape a document arrives in. - [PDF.co Integration](https://flowrunner.ai/integrations/pdfco.md): Give agents full PDF processing power. Convert, merge, split, extract, classify, compress, secure, and parse invoices with AI. 17 actions for every document operation agents need. - [pdfFiller Integration](https://flowrunner.ai/integrations/pdffiller.md): pdfFiller, from airSlate, turns a PDF into something people can complete and sign online. Agents publish fillable links, read submissions back as data, send signature requests, pull the signed copy with its certificate, and organize documents into folders. - [Pdfless Integration](https://flowrunner.ai/integrations/pdfless.md): Generate PDF documents from HTML and CSS templates designed in Pdfless, merging JSON data into template variables with optional password encryption. Agents produce per-record documents that stay on brand. - [PDFMonkey Integration](https://flowrunner.ai/integrations/pdfmonkey.md): Generate polished PDFs from reusable PDFMonkey templates by sending a JSON payload and tracking the asynchronous render. Agents produce invoices, contracts, and reports without a design pass per document. - [pdf noodle (pdforge) Integration](https://flowrunner.ai/integrations/pdforge.md): Generate pixel-perfect PDFs from reusable pdforge templates or raw HTML, and post-process existing documents. Agents merge record data into Handlebars variables and produce a finished file. - [PDF Snake Integration](https://flowrunner.ai/integrations/pdfsnake.md): Impose PDF, JPEG, and PNG documents for print production with the PDF Snake Web API, driven by a steps document you design visually. Agents prepare print-ready files without a prepress operator in the loop. - [Peaka Integration](https://flowrunner.ai/integrations/peaka.md): Peaka is a zero-ETL data platform that connects databases and SaaS APIs and exposes them as queryable catalogs. Agents manage projects and connections, browse catalogs and schemas, run SQL across sources, build internal tables and ask Peaka's AI agent questions about the data. - [Peedief Integration](https://flowrunner.ai/integrations/peedief.md): Turn HTML or a saved dashboard template into a finished PDF through Peedief's managed rendering service. Agents produce documents without operating a Chromium instance inside the flow. - [Peekalink Integration](https://flowrunner.ai/integrations/peekalink.md): Connect over an API key to turn any URL into structured link-preview metadata, including titles, descriptions, images, favicons, and platform-specific details for recognized services. - [Peliqan Integration](https://flowrunner.ai/integrations/peliqan.md): Peliqan is a data platform combining one-click ETL from 250+ sources, a built-in or bring-your-own data warehouse, SQL and low-code Python transformations and published APIs. Agents sync applications, manage warehouse tables and fields, run pipelines and data apps and publish API endpoints. - [Pennylane Integration](https://flowrunner.ai/integrations/pennylane-v2.md): Pennylane is the French accounting and invoicing platform. Agents manage customer and supplier invoices, quotes, customers, suppliers, and products, read the general ledger and bank transactions, and tag analytical categories. - [People Data Labs Integration](https://flowrunner.ai/integrations/people-data-labs.md): Enrich people and companies with the People Data Labs dataset. Agents turn an email or domain into a full profile with job, employer, and firmographics, resolve ambiguous identities into ranked matches, and build lead lists with Elasticsearch or SQL search. - [People HR Integration](https://flowrunner.ai/integrations/people-hr.md): Connect AI agents to People HR, the Access PeopleHR management platform for UK businesses. Agents create and update employees, book holidays, and pull absence and salary data so HR records stay current across systems. - [peoplefone Integration](https://flowrunner.ai/integrations/peoplefone.md): Manage the peoplefone Swiss business VoIP PBX and send SMS. Agents provision extensions and numbers and keep call activity attached to the right record. - [Peppol E-Invoicing Integration](https://flowrunner.ai/integrations/peppol-e-invoicing.md): Send and receive compliant e-invoices over the Peppol network through your certified Access Point provider. Agents transmit invoices and credit notes, track delivery status, poll the inbox for incoming documents, and validate trading partners' Peppol IDs before sending. - [Perplexity Integration](https://flowrunner.ai/integrations/perplexity.md): Get web-grounded AI answers with Perplexity Sonar, run the standalone Search API, use the Agent API with built-in research tools, and create embeddings, all with live citations. - [Persat Integration](https://flowrunner.ai/integrations/persat.md): Connect AI agents to Persat, a field service management platform used across Latin America. Agents dispatch work orders, track technicians and devices, create and track deliveries, and read the digital forms a technician submitted, moving them through their workflow states. - [Personio Integration](https://flowrunner.ai/integrations/personio.md): Connect AI agents to the Personio HR platform. Agents manage people, employments, time off, time tracking, documents, recruiting, reports, and react to HR changes via API credentials. - [Perspective Integration](https://flowrunner.ai/integrations/perspective.md): Score text for toxicity and related attributes with Google's Perspective API, and submit suggested scores as feedback. Built for moderation and community-health workflows. - [PetOffice Integration](https://flowrunner.ai/integrations/petoffice.md): Connect AI agents to PetOffice, animal shelter and rescue management software used in Germany and beyond. Agents read the animals a shelter has published for adoption and generate ready to embed list widgets for a website. - [PGVector Integration](https://flowrunner.ai/integrations/pgvector.md): Open-source vector-similarity extension for PostgreSQL. Enable the extension, create vector tables and ANN indexes, insert and upsert embeddings, and run nearest-neighbor search to build a RAG store on Postgres. - [Phantombuster Integration](https://flowrunner.ai/integrations/phantombuster.md): Launch and manage Phantombuster automation agents (phantoms) and retrieve their scraped results. Agents start a phantom, poll its run to completion, pull back the collected data, and monitor organization quotas. - [Phaxio Integration](https://flowrunner.ai/integrations/phaxio.md): Send faxes, poll their status, download received documents, and manage fax numbers and blocking rules with Phaxio. Agents reach the healthcare, legal, and government counterparties that still require fax. - [Philips Hue Integration](https://flowrunner.ai/integrations/philips-hue.md): Control and read Philips Hue smart lighting through the Hue Bridge CLIP v2 API. Control lights, grouped lights, rooms, zones, and scenes, and read devices and sensors. Authenticates with a hue-application-key header sent to the bridge local IP. - [PhotoRoom Integration](https://flowrunner.ai/integrations/photoroom.md): Give AI agents production-grade image editing with PhotoRoom. Agents remove backgrounds, generate AI lifestyle backgrounds, add shadows, upscale, expand, and relight product photos, and strip unwanted text, saving every result to FlowRunner file storage. - [PHP Point Of Sale Integration](https://flowrunner.ai/integrations/php-point-of-sale.md): PHP Point Of Sale is a retail and restaurant POS available hosted or self-hosted. Agents keep the register catalog in step with an online store, pull daily sales and closeout figures for bookkeeping, manage customers, suppliers and receivings, and run any of its canned reports. - [PiAPI Integration](https://flowrunner.ai/integrations/piapi.md): Reach dozens of generative AI models behind one PiAPI key and one unified task API, covering the major image, video, and audio families. Agents generate media without an account per model provider. - [PicDefense Integration](https://flowrunner.ai/integrations/picdefense.md): Connect AI agents to PicDefense, an image copyright and theft risk scanner. Agents check an image for copyright exposure before a flow publishes it, and read the analysis behind the score. - [Pickaform Integration](https://flowrunner.ai/integrations/pickaform.md): Pickaform, now AirProcess, is a French no-code platform for building business applications out of forms, data models and workflows. Agents browse workspaces, applications, models and views, then create, update, search and delete the records behind them. - [Picnie Integration](https://flowrunner.ai/integrations/picnie.md): Generate branded images and PDFs from Picnie templates, then compress, resize, crop, convert, watermark, and inspect them. Agents produce and finish visual assets in a single connector. - [Picqer Integration](https://flowrunner.ai/integrations/picqer-debesis.md): Connect AI agents to Picqer, a warehouse management system for e-commerce fulfillment. Agents create and process orders, manage products, customers, and picklists, and read warehouse and supplier data so order fulfillment keeps moving without manual data entry. - [Picsart Integration](https://flowrunner.ai/integrations/picsart.md): Edit images programmatically with the Picsart Image API. Agents remove backgrounds, upscale low-resolution images, restore faces, and apply photo effects from a public image URL, no file upload needed. - [Pictory Integration](https://flowrunner.ai/integrations/pictory.md): Turn scripts and articles into finished videos with Pictory. Agents build storyboards from text, render them into captioned, narrated MP4s, and poll job status until the downloadable video URL is ready. - [Pigment Integration](https://flowrunner.ai/integrations/pigment.md): Connect AI agents to Pigment, the enterprise business planning and FP&A platform. Agents move data in and out of planning models and read audit logs, so a forecast reflects what the source systems say rather than a stale export. - [PIMMS Integration](https://flowrunner.ai/integrations/pimms.md): Create smart deep links with PIMMS and attribute conversions back to them, tracking leads and sales rather than clicks alone. Agents connect revenue to the specific link that produced it. - [Pinboard Integration](https://flowrunner.ai/integrations/pinboard.md): Pinboard is the no-nonsense bookmarking service. Agents add, update, and delete bookmarks, manage tags, keep notes, and read recent or dated bookmark sets for research and archiving flows. - [Pinecone Integration](https://flowrunner.ai/integrations/pinecone.md): Managed vector database for agent memory. Manage serverless indexes, upsert and query vectors, work with integrated-embedding text records, and call hosted models for embeddings and reranking. - [PingBell Integration](https://flowrunner.ai/integrations/pingbell.md): Ring a PingBell virtual doorbell to increment its live counter on every paired screen, from phones to Apple TV and Fire TV. Agents raise a visible, physical-feeling alert on the floor when something needs attention now. - [Pingdom Integration](https://flowrunner.ai/integrations/pingdom.md): Connect AI agents to Pingdom uptime monitoring. Agents create, update, and pause checks as infrastructure changes, pull uptime, outage, and performance summaries for SLA reporting, inspect per-probe raw results, and audit alerting actions. - [Pingmee Integration](https://flowrunner.ai/integrations/pingmee.md): List and inspect conversations and send messages through Pingmee, a business messaging platform built around customer conversations. Agents handle routine threads and escalate the ones that need a person. - [Pinterest Integration](https://flowrunner.ai/integrations/pinterest.md): Connect AI agents to Pinterest. Agents publish image, carousel, and async video pins, organize boards and sections, repin curated content, search owned pins and boards, and pull account and per-pin analytics into reporting workflows. - [Pipedrive Integration](https://flowrunner.ai/integrations/pipedrive.md): Connect AI agents to Pipedrive CRM via OAuth2. Agents manage deals, leads, persons, organizations, activities, projects, products, and account configuration end-to-end. - [Pipedrive Resellers Integration](https://flowrunner.ai/integrations/pipedrive-resellers.md): Connect AI agents to the Pipedrive Resellers partner API. Agents provision client Pipedrive accounts, manage subscriptions, and add or remove users on behalf of the customer. - [Pipefy Integration](https://flowrunner.ai/integrations/pipefy.md): Connect AI agents to Pipefy business processes. Agents create cards from forms or tickets, move them between phases, update field values, post comments, and run arbitrary GraphQL against anything the dedicated actions do not cover. - [Pipeliner CRM Integration](https://flowrunner.ai/integrations/pipelinercrm.md): Connect AI agents to Pipeliner CRM. Agents get full create, read, update, and delete access to accounts, contacts, leads, opportunities, and tasks, plus a generic entity request that reaches appointments, notes, and custom records. - [Pixfizz Integration](https://flowrunner.ai/integrations/pixfizz.md): Pixfizz is a web-to-print ecommerce platform for personalized products. Agents bring orders in from an outside storefront and start print fulfillment, render personalized product images and production PDFs, and manage galleries, promo codes and gift vouchers. - [pixx.io Integration](https://flowrunner.ai/integrations/pixx-io.md): pixx.io is a German digital asset management platform for images, videos, and documents. Agents push approved assets into the library with keywords and folders set, pull resized or converted versions, retag whole selections, and hand out shares and upload links. - [Placetel Integration](https://flowrunner.ai/integrations/placetel.md): Manage numbers, extensions, contacts, and call data on the Placetel cloud phone system by Cisco. Agents keep telephony configuration in step with the team and log calls against the right record. - [Placid Integration](https://flowrunner.ai/integrations/placid-app.md): Generate on-brand creatives from Placid templates. Agents fill named template layers with dynamic text and images to render social images, multi-page PDFs, and videos, polling asynchronous renders until the output URL is ready. - [PlanOK GCI Integration](https://flowrunner.ai/integrations/planok-gci.md): Connect AI agents to PlanOK GCI, the real estate commercial management platform for Latin American developers. Agents create leads, register natural-person clients, and log tracking entries, then read the projects, units, reservations, promises, and quotations a deal moves through. - [Planyo Integration](https://flowrunner.ai/integrations/planyo.md): Connect AI agents to Planyo, an online reservation and booking system. Agents make and modify reservations, check availability and pricing, and confirm, cancel, or check in bookings so reservation workflows run end to end. - [PlatoForms Integration](https://flowrunner.ai/integrations/platoforms.md): PlatoForms turns PDFs into fillable online forms. Agents submit forms to generate filled PDFs, pull documents, attachments and answers back out, manage fields, sharing links, invitations, workflows and dropdown lists, and react on every submission. - [PlayHT Integration](https://flowrunner.ai/integrations/playht.md): Give AI agents an ultra-realistic voice with PlayHT text-to-speech. Agents generate lifelike speech in 36 languages from hundreds of stock voices, clone a brand voice from an audio sample, and store finished audio in FlowRunner file storage with ready download URLs. - [Plentific Integration](https://flowrunner.ai/integrations/plentific.md): Connect AI agents to Plentific, a property operations and maintenance platform. Agents raise work orders against a property, dispatch and track tasks, run inspections, and process contractor invoices. - [PlentyONE Integration](https://flowrunner.ai/integrations/plentymarkets.md): PlentyONE, until recently branded plentymarkets, is the multichannel commerce ERP from plentysystems in Germany. Agents sync orders out to warehouse, accounting or reporting systems, push stock levels to marketplaces, and maintain the catalog and accounting documents from one back end. - [Plex Integration](https://flowrunner.ai/integrations/plex.md): Plex Media Server hosts your own library of movies, shows, music, and photos. Agents browse libraries and metadata, search, build playlists, watch active sessions, set watched state and playback progress, and react when new media is added. - [Plivo Integration](https://flowrunner.ai/integrations/plivo.md): Send SMS and MMS, place outbound voice calls, and manage rented phone numbers and applications through the Plivo cloud communications platform. - [Tamio (Plug&Paid) Integration](https://flowrunner.ai/integrations/plug-paid.md): Tamio, formerly plug&paid, is a hosted commerce platform with products, invoices, subscriptions and payment pages. Agents keep the catalog and stock in step with an ERP or supplier feed, create and issue invoices with hosted payment links, and manage customers, orders, discounts and affiliates. - [Plumsail Documents Integration](https://flowrunner.ai/integrations/plumsail-documents.md): Generate finished documents with Plumsail Documents. Agents merge CRM or spreadsheet data into a prebuilt Process template to produce DOCX or PDF output, running synchronously for fast templates or start-then-fetch for long jobs. - [Plumsail Forms Integration](https://flowrunner.ai/integrations/plumsail-forms.md): Plumsail Forms is a drag and drop builder for public web forms shared by link or embedded in a page. Agents create, update and duplicate forms, read and clear submissions and drafts, manage uploaded attachments, and run a flow every time a chosen form is submitted. - [PlusVibe Integration](https://flowrunner.ai/integrations/plusvibe.md): PlusVibe.ai, formerly pipl.ai, is a cold email outreach platform. Agents run campaigns and subsequences, add and manage leads, work the Unibox, manage sending accounts and warmup, run inbox placement tests, and react to campaign events as they happen. - [Plutio Integration](https://flowrunner.ai/integrations/plutio.md): Plutio is the all-in-one business app for freelancers and small agencies: projects, tasks, time, clients, invoices and files in one workspace. Agents create and update projects and tasks, log time, manage clients and issue invoices from a single connection. - [Pneumatic Integration](https://flowrunner.ai/integrations/pneumatic.md): Pneumatic is workflow and SOP automation built around turning a written procedure into live, assigned work. Agents launch workflows from templates, complete and reassign tasks, fill form fields and start flows when a Pneumatic workflow moves. - [PocketAlert Integration](https://flowrunner.ai/integrations/pocketalert.md): Send push notifications to iOS and Android devices through a PocketAlert account, reaching all devices or a single one. Agents deliver time-sensitive alerts straight to the phone. - [Podio Integration](https://flowrunner.ai/integrations/podio.md): Connect AI agents to Podio, the low-code work management platform. Agents navigate the organization, workspace, and app hierarchy, filter and update items, create new records from flow data, assign tasks, and post comments. - [Pointagram Integration](https://flowrunner.ai/integrations/pointagram.md): Connect AI agents to Pointagram, a gamification platform for sales and support teams. Agents create players and teams, award points to score series, and read competitions and leaderboards so performance incentives update automatically as work gets done. - [Pointerpro Integration](https://flowrunner.ai/integrations/pointerpro-au.md): Pointerpro, formerly Survey Anyplace, is an assessment and questionnaire platform that delivers personalized PDF reports. Agents manage questionnaires, contacts and lists, read and upload responses, send report emails, run invitation campaigns, and start a flow on every new response. - [Poper Integration](https://flowrunner.ai/integrations/poper.md): Read projects, domains, campaigns, captured leads, and performance analytics from the Poper AI popup builder. Agents pull captured leads into the systems that will actually work them. - [Porsline Integration](https://flowrunner.ai/integrations/porsline.md): Porsline is an online survey builder for forms, questionnaires, quizzes and lead capture. Agents build surveys across every question type, manage folders, settings and access codes, generate personalized response links, read results and exports, and react on every new response. - [Post My Link Integration](https://flowrunner.ai/integrations/post-my-link.md): Turn long URLs into branded, trackable short links and dynamic QR codes with Post My Link, covering the full link lifecycle. Agents repoint a live code without reprinting anything and read scan data back. - [Postalytics Integration](https://flowrunner.ai/integrations/postalytics.md): Put physical mail into automated workflows through Postalytics. Agents create and list recipient contacts, review direct mail campaigns, and fire triggered drops that mail a postcard or letter to a single address in real time. - [PostFast Integration](https://flowrunner.ai/integrations/postfast.md): Connect AI agents to PostFast, a social media scheduling platform. Agents upload media, schedule posts across connected accounts, and work the social inbox, replying publicly or privately and moderating items on the source platform. - [PostgreSQL Integration](https://flowrunner.ai/integrations/postgresql.md): Connect AI agents to PostgreSQL. Agents run parameterized SQL, read and write rows without SQL, inspect schemas, and bulk-load records as governed steps in a flow. - [PostGrid Integration](https://flowrunner.ai/integrations/postgrid.md): Send real-world mail from FlowRunner through PostGrid. Agents verify and standardize mailing addresses, maintain a reusable contact address book, and send letters and postcards programmatically, with test mode for development. - [PostHog Integration](https://flowrunner.ai/integrations/posthog.md): Capture product events, manage persons and feature flags, run HogQL queries, and create annotations from your flows over the PostHog API. Works with US Cloud, EU Cloud, and self-hosted. - [Postiz Integration](https://flowrunner.ai/integrations/postiz.md): Connect AI agents to Postiz, an open source social media scheduling tool. Agents upload media and schedule posts across every connected channel from a single flow. - [Postmark Integration](https://flowrunner.ai/integrations/postmark.md): Connect AI agents to Postmark. Agents send single, templated, and batch transactional email, search outbound messages and bounces, manage message-stream suppressions, and read delivery statistics. - [Powerlink (Fireberry) Integration](https://flowrunner.ai/integrations/powerlink.md): Connect AI agents to Powerlink (now branded Fireberry), an Israeli CRM with a generic record model. Agents create accounts, contacts, and opportunities, read and write any object through generic record operations by system name, run queries, and inspect metadata. - [PowerOffice Go Integration](https://flowrunner.ai/integrations/poweroffice.md): PowerOffice Go is the Norwegian cloud accounting platform. Agents manage customers, suppliers, and products, invoice sales orders, post vouchers to the general ledger, and read the customer and supplier sub ledgers, projects, and departments. - [PracticePanther Integration](https://flowrunner.ai/integrations/practicepanther.md): Connect AI agents to PracticePanther legal practice management. Agents intake clients by creating contacts and matters, log billable time entries, create follow-up tasks, and surface open matters and outstanding invoices for reporting. - [Prapii Integration](https://flowrunner.ai/integrations/prapii-app.md): Connect AI agents to Prapii, the platform for building, securing, and exposing AI-powered APIs. Agents call an organization's own governed AI endpoints as a step in a flow. - [PreCallAI Integration](https://flowrunner.ai/integrations/precallai.md): Place outbound AI sales calls with PreCallAI and manage the assistants, dialers, contact segments, and campaigns behind them. Agents qualify at volume and pass live interest to a person. - [PredictLeads Integration](https://flowrunner.ai/integrations/predict-leads-app.md): PredictLeads is company intelligence sourced directly from company websites, covering 120M+ companies. Agents discover companies, job openings, technology detections, news, financing events, SEC filings, products and connections, and follow companies for ongoing tracking. - [Prefinery Integration](https://flowrunner.ai/integrations/prefinery.md): Run viral waitlists and referral launches with Prefinery, adding people to a waitlist and returning their share link and position. Agents move people up the list off verified referrals. - [Press Advantage Integration](https://flowrunner.ai/integrations/press-advantage.md): Submit, review, and distribute press releases with Press Advantage, including agency-style management across client accounts. Agents prepare the release and hold distribution until a person signs off. - [PrestaShop Integration](https://flowrunner.ai/integrations/prestashop.md): Run a self-hosted PrestaShop store from FlowRunner. Agents manage products, categories, stock levels, customers, orders, and addresses, advance order statuses to drive fulfillment, and reach any other webservice resource through a generic escape hatch. - [Print.one Integration](https://flowrunner.ai/integrations/print-one.md): Send physical postcards, letters, and greeting cards programmatically with Print.one, where artwork is HTML with merge fields. Agents trigger real mail off an event the CRM already recorded. - [Printavo Integration](https://flowrunner.ai/integrations/printavo.md): Connect AI agents to Printavo, shop management software for screen printers and decorators. Agents create quotes and line items, register customers and inquiries, request and record payments, and email customers about status. - [Printful Integration](https://flowrunner.ai/integrations/printful.md): Automate print-on-demand fulfillment with Printful. Agents browse the catalog, manage sync products and variants, create draft orders, estimate costs, confirm orders for fulfillment, calculate shipping and tax rates, and manage files and webhooks. - [Printify Integration](https://flowrunner.ai/integrations/printify.md): Run a Printify print-on-demand shop from FlowRunner. Agents build products from catalog blueprints and print providers, publish them to connected sales channels like Shopify and Etsy, upload artwork, and submit orders to production. - [PrintNode Integration](https://flowrunner.ai/integrations/printnode.md): Send documents from workflows to physical printers through PrintNode cloud printing. Agents discover connected computers and printers, check account credits, submit PDF or raw print jobs by URL or Base64, and track job status for auditing. - [Pro Crew Schedule Integration](https://flowrunner.ai/integrations/pro-crew-schedule.md): Connect AI agents to Pro Crew Schedule, construction crew scheduling and resource management. Agents build crew schedules, assign tasks to employees, track time and time off, and manage inventory against each job. - [ProAbono Integration](https://flowrunner.ai/integrations/proabono.md): ProAbono is a French subscription-billing platform with a Live API and a BackOffice API. Agents create customers and subscriptions, meter feature usage, manage offers and billing addresses, suspend or invalidate accounts, and react to billing events in real time. - [Process Street Integration](https://flowrunner.ai/integrations/process-street.md): Drive checklist-based processes in Process Street. Agents launch workflow runs from templates, complete or reopen tasks as work progresses, update run statuses and due dates, and read Data Sets to keep recurring SOPs moving. - [Procountor Integration](https://flowrunner.ai/integrations/procountor.md): Procountor is the Finnish cloud accounting platform. Agents create sales and purchase invoices and drive their approval chain, manage business partners and products, record payments and ledger receipts, and pull dimensions and accounting reports. - [ProdPad Integration](https://flowrunner.ai/integrations/prodpad.md): Feed your ProdPad backlog from every channel. Agents capture ideas and customer feedback automatically, tag feedback to contacts, post comments recording decisions on ideas, and read products, personas, and roadmaps for reporting. - [Product Hunt Integration](https://flowrunner.ai/integrations/product-hunt.md): Track Product Hunt launches from FlowRunner. Agents fetch daily featured posts ordered by votes or rank, read launch comments, browse curated collections and topics, and run custom GraphQL queries against the v2 API for anything else. - [Productboard Integration](https://flowrunner.ai/integrations/productboard.md): Connect AI agents to Productboard. Agents capture customer feedback as notes, create and update features, browse the product hierarchy of products and components, link feedback to companies, and track release groups across product lines. - [Productify.ai Integration](https://flowrunner.ai/integrations/productify-ai.md): Productify.ai generates ecommerce product content from a name and an image. Agents fill in missing marketing copy, headlines, features and SEO keywords across a catalog, digitize product images into tables and back-of-pack data, and translate listings into 47 languages. - [ProfitWell Integration](https://flowrunner.ai/integrations/profitwell.md): Connect AI agents to ProfitWell Metrics by Paddle. Agents read month-over-month and day-by-day MRR, active customers, ARPU, and churn, break a single metric down over time by plan, and push subscription creations, upgrades, migrations, and churn events into manual, API-based ProfitWell accounts. - [Projectworks Integration](https://flowrunner.ai/integrations/projectworks.md): Projectworks is professional services automation for consultancies and agencies: projects, budgets, timesheets, leave, expenses, forecasting, resourcing and invoicing. Agents create projects and assignments, submit and unsubmit timesheets, record expenses and pull invoicing and forecast data. - [PromptLayer Integration](https://flowrunner.ai/integrations/promptlayer.md): Manage prompts, observe model calls, and run evaluations with PromptLayer, fetching versioned prompts by release label. Agents run the prompt version that was approved rather than one pasted into code. - [Pronnel Integration](https://flowrunner.ai/integrations/pronnel.md): Pronnel is a board-and-item work platform that has grown a CRM and an omni-channel inbox around its kanban core. Agents create and move items, update fields, manage boards and contacts, and work the inbox from a single connection. - [ProofHub Integration](https://flowrunner.ai/integrations/proofhub.md): Manage ProofHub projects from FlowRunner. Agents create tasks from inbound events, update dates, assignees, and completion status, organize work into task lists, review discussion topics, and resolve account members for assignment. - [Proofly Integration](https://flowrunner.ai/integrations/proofly.md): Manage social proof notification widgets with Proofly, reading usage, listing campaigns, switching them on or off, and pulling captured leads. Agents surface genuine recent activity on the site. - [ProSMS Integration](https://flowrunner.ai/integrations/prosms-se.md): Connect AI agents to ProSMS, a Scandinavian SMS gateway operated by Compaya with EU-resident data storage. Agents send and schedule SMS to individuals or contact groups, read delivery logs, and manage contacts and sender names so Nordic customer messaging stays auditable and on time. - [ProvenExpert Integration](https://flowrunner.ai/integrations/proven-expert.md): Read profiles and aggregated review summaries on ProvenExpert, the German review and rating platform, and work the inbound customer requests and leads they generate. Agents route new reviews and review-sourced leads to an owner. - [Pulseem Integration](https://flowrunner.ai/integrations/pulseem.md): Pulseem is an Israeli multi-channel messaging platform for transactional email, SMS and WhatsApp Business. Agents send messages across all three channels, manage suppression lists, and administer reseller sub-accounts. - [Referly (Push Lap Growth) Integration](https://flowrunner.ai/integrations/push-lap-growth.md): Run a startup affiliate program with the Referly API, onboarding affiliates and attributing referrals. Agents credit and pay partners off verified conversion data. - [Pushbullet Integration](https://flowrunner.ai/integrations/pushbullet.md): Push notes, links, and files to your devices, share pushes with other people, and send SMS through a connected phone using Pushbullet. - [Pushcut Integration](https://flowrunner.ai/integrations/pushcut.md): Send rich, actionable mobile notifications and run Shortcuts, HomeKit scenes, and automations on Apple devices through the Pushcut Web API. - [PushEngage Integration](https://flowrunner.ai/integrations/pushengage.md): Send web push campaigns through PushEngage. Agents fire notifications to all subscribers or a targeted segment, pull subscriber counts, enrich profiles with custom attributes, remove subscribers, and inspect segments and drip campaigns. - [Pushinator Integration](https://flowrunner.ai/integrations/pushinator.md): Send push notifications to iOS and Android devices through a Pushinator account, delivering to channel subscribers. Agents reach a subscribed audience without building a mobile backend. - [Pushover Integration](https://flowrunner.ai/integrations/pushover.md): Send push notifications to phones, tablets, and desktops through Pushover, with emergency-priority alerts that repeat until a recipient acknowledges them. - [Pushwoosh Integration](https://flowrunner.ai/integrations/pushwoosh.md): Send mobile push and omnichannel messages through Pushwoosh. Agents create or schedule notifications, cancel scheduled messages, check delivery status, register and unregister devices, and set tags to drive audience segmentation. - [Put.io Integration](https://flowrunner.ai/integrations/put-io.md): Put.io is cloud storage that downloads for you from an HTTP URL or a magnet link. Agents start transfers, watch RSS feeds and fetch matching items, unpack archives in place, bundle files into ZIPs, and share or pull the results down. - [Qapla Integration](https://flowrunner.ai/integrations/qapla.md): Connect AI agents to Qapla', an Italian multi-carrier shipment tracking platform. Agents push shipments and orders, pull tracking updates, detect couriers, and create labels so post-shipping visibility stays in sync across carriers. - [Qdrant Integration](https://flowrunner.ai/integrations/qdrant.md): Open-source vector database and similarity search engine for agent memory. Manage collections, upsert and retrieve points with vectors and payloads, run similarity searches, and maintain point metadata. - [Qonto Integration](https://flowrunner.ai/integrations/qonto.md): Automate finance work on your Qonto business accounts. Agents read organizations and account balances, list and inspect transactions for reconciliation, pull invoice and receipt attachments, and look up beneficiaries, labels, and team members. - [Quable PIM Integration](https://flowrunner.ai/integrations/quable-pim.md): Quable is a PIM and DAM for product content. Agents sync a storefront catalog from the PIM incrementally, enrich documents, variants and assets, run import and export profiles, and react to workflow steps and asset changes in real time. - [Quaderno Integration](https://flowrunner.ai/integrations/quaderno.md): Quaderno automates sales tax and invoicing for businesses selling across borders. Agents issue invoices, receipts, credit notes, and proformas, calculate tax and validate tax IDs, manage contacts and products, and record the location evidence EU digital VAT requires. - [Qualiobee Integration](https://flowrunner.ai/integrations/qualiobee.md): Connect AI agents to Qualiobee, administrative software for French training organizations built around Qualiopi compliance. Agents manage learners, customers, training programs, sessions, trainers, and modules so enrollment and compliance records stay current without manual upkeep. - [Qualtrics Integration](https://flowrunner.ai/integrations/qualtrics.md): Work Qualtrics surveys end to end. Agents list surveys and pull full authoring definitions, export responses asynchronously with filters, record new responses, manage XM Directory contacts and mailing lists, and schedule email distributions. - [Quanda Integration](https://flowrunner.ai/integrations/quanda.md): Quanda is a Czech B2B marketing automation platform that combines surveys, quizzes, web forms and email marketing. Agents manage contacts and tags, build and send email campaigns, send transactional email and survey invitations, read survey structures and responses, and react through webhooks. - [Quentn Integration](https://flowrunner.ai/integrations/quentn.md): Connect AI agents to Quentn, the German email marketing and automation platform, through its official API. Agents manage contacts and tags and trigger campaigns off events in the systems of record. - [QuestDB Integration](https://flowrunner.ai/integrations/questdb.md): Connect AI agents to QuestDB, the high-performance time-series SQL database. Agents run queries and DDL/DML over the REST API, export result sets as CSV, and validate connectivity. - [Quick Base Integration](https://flowrunner.ai/integrations/quickbase.md): Connect AI agents to Quick Base through the JSON REST API. Agents query, insert, update, and delete records, manage tables and fields, run reports, and download file attachments. - [QuickBooks Online Integration](https://flowrunner.ai/integrations/quickbooks-online.md): Automate your accounting layer. Agents create invoices, record bills, manage vendors and customers, run P&L reports, and sync financial data across systems. - [ArcGIS QuickCapture Integration](https://flowrunner.ai/integrations/quickcapture.md): ArcGIS QuickCapture is Esri's single tap field data capture app where a button press lands a located record in a hosted feature layer. Agents manage projects, buttons, data sources and resources, read captured records, configure webhooks, and react when a crew captures data. - [QuickChart Integration](https://flowrunner.ai/integrations/quickchart.md): Generate chart images, QR codes, barcodes, and word clouds with QuickChart. Charts use the standard Chart.js configuration model and work keyless on the free tier. - [QuickEmailVerification Integration](https://flowrunner.ai/integrations/quickemailverification.md): QuickEmailVerification is an email verification and list cleaning API. Agents verify a single address in real time, upload a list for bulk verification, poll job status, download the cleaned results, and delete finished jobs. - [QuickFile Integration](https://flowrunner.ai/integrations/quickfile.md): QuickFile is the UK cloud accounting platform. Agents raise invoices and estimates, manage clients, suppliers, and purchases, post journals, banking, and payments, and read the reports built on them. - [Quill Booking Integration](https://flowrunner.ai/integrations/quill-booking.md): Connect AI agents to Quill Booking, the WordPress appointment scheduling plugin. Agents list events and calendars, cancel or reschedule bookings, and react when bookings are created or changed on your WordPress site so schedule changes propagate instantly. - [Quill Forms Integration](https://flowrunner.ai/integrations/quill-forms.md): Quill Forms is an open source conversational form builder that runs as a WordPress plugin. Agents read and write forms and their block-based questions, manage submissions, submit entries, manage themes and provider integrations, generate a form from a plain language brief, and react on every new entry. - [QuintaDB Integration](https://flowrunner.ai/integrations/quintadb.md): QuintaDB is an online database and form builder where databases hold forms, fields and records, with saved reports as filtered views. Agents create and update databases and forms, search and write records, update cells in bulk, upload files and run configured actions. - [Quipu Integration](https://flowrunner.ai/integrations/quipu.md): Quipu is the Spanish invoicing and accounting platform. Agents issue invoices, simplified invoices, and additional incomes, manage contacts, paysheets, and numbering series, attach files, and read the book entries underneath. - [QuizCube Integration](https://flowrunner.ai/integrations/quizcube.md): QuizCube is an AI powered quiz and assessment builder that generates questions from a document, text or URL. Agents create quizzes, questions and collections, drive the AI assistant, manage participants and magic links, read analytics and exports, and react on quiz completion. - [Qwen Integration](https://flowrunner.ai/integrations/qwen-ai.md): Bring Alibaba's Qwen models into your flows via Model Studio. Agents generate text and run function-calling loops, analyze images with Qwen-VL, create embeddings, produce Wan images and video, and convert speech both ways with TTS and ASR. - [Qwilr Integration](https://flowrunner.ai/integrations/qwilr.md): Automate proposals and quotes in Qwilr. Agents create branded pages from templates with merge-field substitutions, publish and update page settings, and read acceptance and analytics data to trigger downstream steps when a buyer views or accepts. - [Ragic Integration](https://flowrunner.ai/integrations/ragic.md): Ragic is a no-code database builder where every sheet's own URL is its API endpoint. Agents list, create, update and delete records on any sheet, upload files, add comments, lock and approve records in bulk, run action buttons and download record documents. - [Raindrop.io Integration](https://flowrunner.ai/integrations/raindrop.md): Connect over an access token to manage collections, bookmarks, tags, and text highlights through the Raindrop.io REST API v1, with automatic page metadata parsing on new bookmarks. - [Raklet Integration](https://flowrunner.ai/integrations/raklet.md): Connect AI agents to Raklet, a membership and community management platform. Agents create and enrich contacts, manage tags and subscriptions, record donations and debts, publish posts, and read the events the community runs. - [Ramp Integration](https://flowrunner.ai/integrations/ramp-service.md): Connect AI agents to the Ramp corporate spend platform. Agents monitor transactions, manage cards and users, sync vendors and bills, and route reimbursement approvals to a human. - [RANDOM.ORG Integration](https://flowrunner.ai/integrations/random.org.md): Generate true random numbers from atmospheric noise through the RANDOM.ORG API, rather than the pseudo-random generators built into code. Agents run fair draws, sampling, and audit selection that will survive scrutiny. - [RankYak Integration](https://flowrunner.ai/integrations/rankyak.md): Pull AI-written SEO articles out of RankYak and push them into your own publishing pipeline. Agents route drafts to an editor before anything goes live under your byline. - [Rapid Indexer Integration](https://flowrunner.ai/integrations/rapid-indexer.md): Submit URLs and backlinks to Rapid Indexer for search engine indexing and check whether they made it in. Agents confirm new pages were actually indexed rather than assuming it. - [RapidReg Integration](https://flowrunner.ai/integrations/rapidreg.md): Connect AI agents to RapidReg, a UK contactless registration platform. Agents read registrations as they are captured, along with the brands, items, and account records behind each event. The connector is read-only. - [Rav Messer Integration](https://flowrunner.ai/integrations/rav-messer.md): Rav Messer, sold in English as Responder, is an Israeli email and SMS marketing platform. Agents manage lists and subscribers, personal fields and tags, maintain both suppression lists, and register webhooks for subscriber events. - [RAYNET CRM Integration](https://flowrunner.ai/integrations/raynet-crm-v2.md): Connect AI agents to RAYNET CRM, a Czech cloud CRM for sales teams. Agents create accounts, contacts, and leads, convert leads into deals, assign tasks as the activity trail, and read the offers and product catalog behind each deal. - [Razorpay Integration](https://flowrunner.ai/integrations/razorpay.md): Run payments in India end to end with Razorpay. Agents create orders and capture payments, issue refunds, send payment links and UPI QR codes, generate invoices, manage plans and subscriptions, and reconcile settlements across 70 actions. - [RD Station Integration](https://flowrunner.ai/integrations/rd-station.md): Sync leads into RD Station Marketing from any workflow. Agents upsert contacts by email, register conversion events, send custom events, enrich leads with tags and custom fields, and flag qualified leads as sales opportunities. - [ReachInbox Integration](https://flowrunner.ai/integrations/reachinbox.md): ReachInbox is a cold outreach platform with built-in deliverability tooling. Agents manage sending accounts and warmup, create campaigns and load leads, work replies in the unified Onebox, pull analytics, maintain the blocklist, and run inbox placement tests. - [Read AI Integration](https://flowrunner.ai/integrations/read.md): Turn Read AI meeting reports into downstream automation. Agents list and search past meetings, retrieve full reports, and pull out summaries, transcripts, action items, key questions, topics, and engagement analytics field by field. - [Readwise Integration](https://flowrunner.ai/integrations/readwise.md): Readwise is the highlights library and Readwise Reader is its read-later app. Agents save and tag highlights and books, pull the daily review, and create, list, update, and tag Reader documents from one token. - [RealMail Integration](https://flowrunner.ai/integrations/realmail.md): RealMail is a single-endpoint email verification API built for SaaS signup flows. Agents check an address at signup and act on the verdict before a fake or disposable email gets into your user base. - [Re:amaze Integration](https://flowrunner.ai/integrations/reamaze.md): Handle customer conversations in Re:amaze from FlowRunner. Agents create conversations from inbound events, post staff or customer replies across channels, keep contacts in sync, and surface knowledge base articles for self-service deflection. - [Rebill Integration](https://flowrunner.ai/integrations/rebill.md): Rebill is a Latin American payments and subscriptions platform. Agents create customers, cards, and hosted payment links, run card and alternative-method checkouts, manage plans, coupons, and subscriptions, refund payments, and act on webhook events. - [Rebrandly Integration](https://flowrunner.ai/integrations/rebrandly.md): Automate branded short links with Rebrandly. Agents shorten URLs onto your connected domains, repoint destinations without changing published links, track click counts, organize links with tags, and list domains for auditing. - [ReceitaWS Integration](https://flowrunner.ai/integrations/receitaws.md): ReceitaWS returns Brazilian company registration data for a CNPJ straight from the Receita Federal. Agents validate a CNPJ and pull the registered name, address, activities, partners and status, with an optional freshness limit on cached results. - [Recharge Integration](https://flowrunner.ai/integrations/recharge.md): Automate recurring revenue operations in Recharge. Agents create, update, and cancel customer subscriptions, look up upcoming and historical charges for reconciliation and retention workflows, and browse subscription-enabled products. - [Reclaim.ai Integration](https://flowrunner.ai/integrations/reclaim-ai.md): Connect AI agents to Reclaim.ai, the AI scheduling assistant for Google Calendar. Agents create and update tasks, mark them done, and read habits and calendars so Reclaim finds the time on your team's schedule automatically. - [Recommand Integration](https://flowrunner.ai/integrations/recommand.md): Recommand is a Belgian Peppol access point. Agents put invoices, credit notes, self billing documents, and raw UBL XML onto the Peppol network, read what arrives, and act on message level responses. - [Recruitee Integration](https://flowrunner.ai/integrations/recruitee.md): Connect AI agents to Recruitee, an applicant tracking system. Agents manage candidates, jobs, pipelines, interviews, notes, tasks, requisitions, and candidate communication via API token. - [RecruSpace Integration](https://flowrunner.ai/integrations/recruspace.md): Connect AI agents to RecruSpace, an AI recruitment sourcing and screening platform. Agents submit applications, create and tag candidates, and organize talent pools so sourcing pipelines fill up and stay organized. - [Recurly Integration](https://flowrunner.ai/integrations/recurly.md): Connect AI agents to Recurly subscription billing. Agents create accounts and subscriptions, cancel plans, and pull invoices and plan catalogs to keep billing operations and revenue reporting in sync. - [Reddit Integration](https://flowrunner.ai/integrations/reddit.md): Connect AI agents to Reddit on behalf of a connected user. Agents submit and edit posts and comments, vote, save and hide content, browse and search subreddits, manage subscriptions, and send private messages. - [Redis Integration](https://flowrunner.ai/integrations/redis.md): Connect AI agents to Redis. Agents cache values with TTL, maintain atomic counters and locks, work with hashes, lists, sets and sorted sets, publish messages, and read server stats. - [Redmine Integration](https://flowrunner.ai/integrations/redmine.md): Connect AI agents to your Redmine instance. Agents file and update issues, log time entries, and read projects, trackers, statuses, and users through the REST API using your instance URL and API key. - [Referral Factory Integration](https://flowrunner.ai/integrations/referral-factory.md): Build and run referral programs with Referral Factory, adding referrers and capturing referred leads. Agents issue rewards off verified referral events. - [ReferralHero Integration](https://flowrunner.ai/integrations/referralhero.md): Run referral and waitlist campaigns with ReferralHero, adding subscribers and attributing referrals. Agents keep reward eligibility tied to confirmed conversions. - [Refiner Integration](https://flowrunner.ai/integrations/refiner.md): Refiner runs in-app microsurveys for NPS, CSAT, CES and product-market fit at the moment a user is in your product. Agents push user traits and events in, pull responses and aggregated scores out, manage surveys and segments, and react when new feedback arrives. - [RegFox Integration](https://flowrunner.ai/integrations/regfox.md): Connect AI agents to RegFox, the event registration product from Webconnex. Agents pull registration records and read the forms, inventory, and coupons behind them, across a fully read-only surface. - [Rejstriky.info Integration](https://flowrunner.ai/integrations/rejstriky-info.md): Rejstriky.info is a Czech business register lookup covering the insolvency register (ISIR) and the register of enforcement proceedings. Agents check a subject for insolvency and enforcement records, pull case details and events, and manage a watch list of monitored subjects. - [Relatel Integration](https://flowrunner.ai/integrations/relatel.md): Manage employees, calls, contacts, SMS, and switchboards on Relatel, the Danish phone system formerly known as Firmafon. Agents keep phone routing aligned with who is actually working. - [RelationCity Integration](https://flowrunner.ai/integrations/relationcity.md): Connect AI agents to RelationCity, a Danish SMS marketing and contact management platform. Agents manage contacts, tags, and custom fields, send transactional SMS and template messages, run tag-targeted SMS campaigns, and fire trigger-based emails so customer outreach stays coordinated across channels. - [Release0 Integration](https://flowrunner.ai/integrations/release0.md): Use the Release0 builder API for its no-code conversational agent and form builder, covering agents, submissions, analytics, and workspaces. Agents pull submissions into the systems that act on them. - [Relevance AI Integration](https://flowrunner.ai/integrations/relevance.md): Orchestrate Relevance AI from FlowRunner. Agents trigger Relevance agents and wait for results, run tools synchronously or async, and manage knowledge sets and documents to keep retrieval data fresh. - [remberg Integration](https://flowrunner.ai/integrations/remberg-de.md): Connect AI agents to remberg, a German asset and service management XRM for equipment manufacturers and operators. Agents create assets and parts, open tickets and work orders, approve work requests, and read the forms a technician completed. - [Remote Retrieval Integration](https://flowrunner.ai/integrations/remote-retrieval.md): Connect AI agents to Remote Retrieval, a device return service for remote employee offboarding. Agents create device return orders, look up device pricing and company details, and track order status so laptops come back on schedule when employees leave. - [remove.bg Integration](https://flowrunner.ai/integrations/removebg.md): Give agents one-call background removal through remove.bg. Strip backgrounds from product shots, portraits, and car photos, returning transparent images ready for listings, ads, and design pipelines. - [Rendi Integration](https://flowrunner.ai/integrations/rendi.md): Rendi is FFmpeg as a service: send a command string and a map of file aliases, and Rendi runs it on its own infrastructure. Agents transcode, trim, concatenate, and probe media, chain commands, store and manage files, and react when a command finishes. - [Rentman Integration](https://flowrunner.ai/integrations/rentman.md): Connect AI agents to Rentman, the rental and event production platform used by AV and staging companies. Agents create projects from requests, maintain the equipment catalog, assign tasks to crew, state crew availability, and read the equipment and crew planned on each project along with their shortages. - [Reoon Email Verifier Integration](https://flowrunner.ai/integrations/reoon-email-verifier.md): Reoon Email Verifier checks whether email addresses are real and safe to send to. Agents verify a single address in quick or power mode, create a bulk verification task, and retrieve the results when it completes. - [RepairShopr Integration](https://flowrunner.ai/integrations/repairshopr.md): Connect AI agents to RepairShopr, the CRM, ticketing, and point of sale platform for repair shops. Agents open tickets against a customer, quote and invoice the work, take payments, and track leads through to the repair. - [RepliQ Integration](https://flowrunner.ai/integrations/repliq.md): Generate personalized videos, images, and outreach copy with RepliQ, driven by a template you control. Agents produce per-prospect creative without a manual pass per contact. - [Reply.io Integration](https://flowrunner.ai/integrations/reply-io.md): Connect AI agents to Reply.io sales engagement. Agents add contacts and push them into campaigns, mark replies and finished sequences, and manage people and email accounts so outreach reacts to what prospects actually do. - [Reportei Integration](https://flowrunner.ai/integrations/reportei.md): Connect AI agents to Reportei, a marketing report and dashboard builder for agencies. Agents generate client reports from templates, pull metrics across connected integrations, and post timeline events that explain a change. - [RescueTime Integration](https://flowrunner.ai/integrations/rescuetime.md): Pull RescueTime focus data into workflows. Agents fetch analytic reports, daily summaries, and alerts, and post highlights, turning attention data into standups, reviews, and productivity nudges. - [Resend Integration](https://flowrunner.ai/integrations/resend.md): Send transactional email through Resend from any workflow. Agents send single or batch emails, schedule and cancel sends, verify domains, and manage audiences, contacts, and broadcasts across 29 actions. - [Reservanto Integration](https://flowrunner.ai/integrations/reservanto.md): Connect AI agents to Reservanto, a Czech online booking system for appointments, classes, and courses. Agents check available start times, create and confirm bookings, and manage customers so reservations stay accurate across locations. - [Resource Guru Integration](https://flowrunner.ai/integrations/resource-guru.md): Connect AI agents to Resource Guru scheduling. Agents create bookings against resources and projects, check leave, and keep clients and project assignments current so capacity plans reflect reality. - [Respond.io Integration](https://flowrunner.ai/integrations/respond-io.md): Connect AI agents to Respond.io conversations. Agents send messages, open and close conversations, tag and comment on contacts, and update custom fields across every channel your customers write in on. - [ResponseSuite Integration](https://flowrunner.ai/integrations/responsesuite.md): ResponseSuite is a survey marketing platform whose API exists to read surveys and submissions back out. Agents list and look up surveys, read completed submissions from a point in time, page a whole submission history into one array, and react when a new submission or survey appears. - [REST Countries Integration](https://flowrunner.ai/integrations/rest-countries.md): Pull reference data for more than 250 countries and territories from REST Countries, with more than 90 normalized fields per record. Agents normalize country, currency, and language data between systems that disagree. - [restdb.io Integration](https://flowrunner.ai/integrations/restdb-io.md): restdb.io is a hosted NoSQL database whose REST API is generated from your schema, with MongoDB-style querying. Agents list, create, replace, update and delete documents, read and add child documents, query and manage media, read metadata and send email from the database. - [Restyaboard Integration](https://flowrunner.ai/integrations/restya-core.md): Restyaboard is the open source, self hosted, Trello style kanban platform from Restya. Agents create boards, lists and cards, move and assign work, add comments, checklists and attachments, and manage users on your own installation. - [Retable Integration](https://flowrunner.ai/integrations/retable.md): Retable is an online spreadsheet-database organized as workspaces containing projects containing tables. Agents get, insert, update, delete and search rows, create workspaces, projects and tables, add and remove columns and upload files to a project. - [Retell AI Integration](https://flowrunner.ai/integrations/retell-ai.md): Put Retell AI voice agents to work inside FlowRunner. Agents place phone and web calls, fetch transcripts and recordings, and manage Retell agents, LLM configurations, voices, and phone numbers end to end. - [Retently Integration](https://flowrunner.ai/integrations/retently.md): Connect AI agents to Retently customer feedback. Agents send NPS surveys, read feedback and scores, and create or update customers so follow-up runs the moment a detractor responds. - [Reverse Contact Integration](https://flowrunner.ai/integrations/reversecontact.md): Connect AI agents to Reverse Contact, a real-time email and contact enrichment API. Agents turn an email address or a name and company into a resolved profile before a flow decides how to route the lead. - [ReviewStudio Integration](https://flowrunner.ai/integrations/reviewstudio.md): ReviewStudio is the online proofing and creative approval platform. Agents create clients, projects, and reviews, add files and remote versions, assign reviewers, read approval status and notes, apply labels and workflows, and read the webhook delivery log. - [Revolut Business Integration](https://flowrunner.ai/integrations/revolut-business.md): Connect AI agents to the Revolut Business banking API. Agents look up accounts and multi-currency balances, manage saved recipients, read and cancel transactions, transfer between accounts, make payments, prepare payment drafts, exchange currencies, create payout links, and verify UK recipients, with four triggers that react to transactions and payout-link status. - [Revox Integration](https://flowrunner.ai/integrations/revox.md): Connect AI agents to Revox, the outbound AI voice calling platform, building reusable assistants with their own prompt, voice, and retry policy. Agents place calls and read outcomes back into the flow. - [RevSend Integration](https://flowrunner.ai/integrations/revsend.md): Send sales gifts and rewards with RevSend, covering digital gift cards by email or link and a physical gift catalogue. Agents trigger a gift off a milestone and hold spend behind an approval. - [Reward Sciences Integration](https://flowrunner.ai/integrations/rewardsciences.md): Track customer activity for loyalty and rewards with the Reward Sciences API, recording the actions that earn credit. Agents log qualifying events from the systems where they actually happen. - [Rhombus Integration](https://flowrunner.ai/integrations/rhombus.md): Connect AI agents to Rhombus, a cloud-managed physical security platform for cameras and sensors. Agents read camera and sensor state, pull security events, and resolve the location behind each one. - [RingCentral Integration](https://flowrunner.ai/integrations/ringcentral.md): Connect AI agents to RingCentral. Agents send SMS and fax, place RingOut calls, read call logs and recordings, manage contacts and extensions, and post to team chats across 30 actions. - [Ringover Integration](https://flowrunner.ai/integrations/ringover.md): Connect AI agents to your Ringover phone system. Agents trigger click-to-call, send SMS, read call logs, voicemails, and conversations, and manage contacts, numbers, and users. - [LogicGate Risk Cloud Integration](https://flowrunner.ai/integrations/riskcloud.md): Connect AI agents to LogicGate Risk Cloud, a governance, risk, and compliance platform. Agents read records and their linked context, build and maintain the program structure of applications, workflows, steps, and routing paths, and audit who had access. - [Robly Integration](https://flowrunner.ai/integrations/robly.md): Robly is an email marketing platform for small organizations and nonprofits. Agents manage mailing lists, contacts and their fields, apply tags and list memberships, run background CSV imports, and read campaign statistics. - [Robocorp Integration](https://flowrunner.ai/integrations/robocorp.md): Robocorp Control Room (now Sema4.ai) runs and supervises Python and Robot Framework automations. Workflows start process runs, manage work items and their files, read step run output and artifacts, keep assets and secrets current, and react when a run completes or needs attention. - [Robolytix Integration](https://flowrunner.ai/integrations/robolytix.md): Connect AI agents to Robolytix, a real-time process monitoring and analytics platform. Agents send Sonar checkpoint messages and report each run's outcome so automation health is measured rather than assumed. - [Rocketbot Integration](https://flowrunner.ai/integrations/rocketbot.md): Rocketbot Orchestrator runs RPA robots and the Xperience form queues that carry data into them. Workflows start robots, deploy new robot files, add and update queue items, and read assets. - [Rocket.Chat Integration](https://flowrunner.ai/integrations/rocketchat.md): Post and manage messages, work with channels, private groups, and direct messages, administer users, and upload files against your self-hosted or cloud Rocket.Chat server over its REST API. - [RocketReach Integration](https://flowrunner.ai/integrations/rocketreach.md): Enrich people and companies with RocketReach. Agents look up verified emails and phone numbers, search people and companies by criteria, and check lookup status and credit balances mid-workflow. - [Roezan Integration](https://flowrunner.ai/integrations/roezan.md): Connect AI agents to Roezan, an SMS marketing platform built for course creators and webinar funnels. Agents send SMS and MMS, manage contacts, lists, and tags, and draft, preview, and schedule broadcasts so launch sequences and reminders go out on schedule. - [Romulus Integration](https://flowrunner.ai/integrations/romulus.md): Connect AI agents to Romulus, which qualifies and answers inbound leads across voice calls, WhatsApp, SMS, and web. Agents pick up the qualified lead and run the work that follows. - [Clientary Integration](https://flowrunner.ai/integrations/ronin.md): Clientary, formerly Ronin, is the invoicing, time tracking, and project management platform. Agents manage clients and contacts, log time against projects, issue invoices and estimates, and record payments as work gets billed. - [Rossum Integration](https://flowrunner.ai/integrations/rossum-elis.md): Connect AI agents to Rossum intelligent document processing. Agents upload documents, wait for extraction, confirm or reject annotations, export queues, and manage schemas, workspaces, and hooks across 41 actions. - [Rows Integration](https://flowrunner.ai/integrations/rows.md): Read and write Rows spreadsheets from workflows. Agents read cell ranges, write values, and append rows, turning a Rows spreadsheet into a live input or output for any automation. - [RudderStack Integration](https://flowrunner.ai/integrations/rudderstack.md): Stream events to RudderStack from any workflow. Agents send track, identify, page, screen, group, and batch calls to the data plane and manage sources, destinations, and connections on the control plane. - [RUIAN API Integration](https://flowrunner.ai/integrations/ruian-api.md): Look up and validate Czech addresses against RUIAN, the official state register of territorial identification, addresses, and real estate. Agents confirm an address exists before it reaches a contract or a delivery. - [Rundeck Integration](https://flowrunner.ai/integrations/rundeck.md): Run and control Rundeck jobs, adhoc commands, and scripts, and inspect executions from your flows over the Rundeck API. Runbook automation and job scheduling for operations. - [Runo Call Management CRM Integration](https://flowrunner.ai/integrations/runo-call-management-crm.md): Connect AI agents to Runo, a SIM-based mobile call management and telecalling CRM used by teams in India. Agents allocate leads to callers, log interactions that upsert the customer record by phone, and pull call logs from previous days. - [Runware Integration](https://flowrunner.ai/integrations/runware-ai.md): Generate images and video, upscale, and remove backgrounds across hundreds of models through Runware's fast, low-cost inference. Agents produce visual assets at volume inside the workflow that needs them. - [Runway Integration](https://flowrunner.ai/integrations/runway.md): Generate media with Runway inside your workflows. Agents create images and video from text or reference media, dub and isolate audio, build avatar and product ad videos, and run Runway workflows, then wait on tasks and save outputs to files across 51 actions. - [Ryver Integration](https://flowrunner.ai/integrations/ryver.md): Post chat messages, run topics and tasks, and manage teams with the Ryver REST API. Agents deliver updates into the workspace where the team already tracks the work. - [Amazon S3 / Cloud Storage Integration](https://flowrunner.ai/integrations/s3.md): Store, retrieve, and manage files across 13+ cloud storage providers. Agents upload processed documents, generate secure links, and manage buckets across Amazon S3, Cloudflare R2, and more. - [Sage Accounting Integration](https://flowrunner.ai/integrations/sage-accounting.md): Connect AI agents to Sage Accounting. Agents create and update contacts, sales and purchase invoices, credit notes, quotes, products, and journals, and read ledger and bank accounts to keep the books current across 49 actions. - [Sage Intacct Integration](https://flowrunner.ai/integrations/sage-intacct.md): Connect AI agents to Sage Intacct, the cloud financial management and ERP platform. Agents run generic create, read, update, delete, and list operations over 400+ object types, plus dedicated Accounts Payable and Accounts Receivable actions, Cash Management reconciliation, Contracts and Revenue Management lifecycles, and Purchasing approvals through the Sage Intacct REST API. - [Sakari Integration](https://flowrunner.ai/integrations/sakari-sms.md): Send and track SMS through Sakari. Agents message contacts, review delivery history, and manage contacts and groups for two-way business texting campaigns. - [SalesTrigger Integration](https://flowrunner.ai/integrations/sales-trigger.md): Build multi-step LinkedIn connection and message sequences with the SalesTrigger OutReach API and enrol leads in them. Agents pace outreach and route replies to the right owner. - [SalesBlink Integration](https://flowrunner.ai/integrations/salesblink.md): Run AI cold email outreach with SalesBlink, covering prospect lists and leads, multi-step sequences, the unified inbox, and send activity. Agents keep sequences supplied and replies routed. - [Salesflare Integration](https://flowrunner.ai/integrations/salesflare.md): Connect AI agents to Salesflare CRM. Agents manage contacts, accounts, opportunities, tasks, and tags, log interactions, and read pipelines and stages to keep deal data moving without manual entry. - [Salesforce Essentials Integration](https://flowrunner.ai/integrations/salesforce-essentials.md): Core Salesforce CRM automation for contacts, leads, and campaigns. 7 focused actions for essential CRM operations: create records, find entries, update data, manage campaign membership. - [Salesforce Pardot Integration](https://flowrunner.ai/integrations/salesforce-pardot.md): Connect AI agents to Salesforce Pardot, now Marketing Cloud Account Engagement. Agents create and update prospects, add them to lists, and read campaigns to keep B2B marketing data synchronized. - [Salesforce Pro Integration](https://flowrunner.ai/integrations/salesforce-pro.md): Full Salesforce CRM automation for complex sales operations. Agents create leads, convert contacts, query with SOQL, send emails, manage campaigns, and handle file attachments. 25 actions. - [Salesforge Integration](https://flowrunner.ai/integrations/salesforge.md): Run AI cold email and multichannel sending with Salesforge, covering workspaces, contacts and the do-not-contact list, sequences, and contact validation. Agents keep suppression honored on every send. - [Salesloft Integration](https://flowrunner.ai/integrations/salesloft.md): Connect AI agents to Salesloft. Agents manage people and accounts, add prospects to cadences, create and complete tasks, log calls and notes, and read emails and meetings across 41 actions. - [SALESmanago Integration](https://flowrunner.ai/integrations/salesmanago.md): Connect AI agents to SALESmanago marketing automation. Agents upsert contacts, manage tags, and record external events like purchases and visits so campaigns respond to real customer behavior. - [Salesmate Integration](https://flowrunner.ai/integrations/salesmate.md): Connect AI agents to Salesmate CRM. Agents create and update contacts, companies, and deals, advance deals through pipeline stages, and log activities from your flows. - [Salesmsg Integration](https://flowrunner.ai/integrations/salesmsg.md): Send SMS and MMS from your Salesmsg business lines directly from agent workflows. Agents create and update contacts, browse message and conversation history, and pick the right inbox (phone number) before dispatching a text. - [Salla Integration](https://flowrunner.ai/integrations/salla.md): Salla is the Saudi ecommerce platform, and this connector covers its Merchant API. Agents sync orders into an ERP or fulfillment system as they are placed, update stock in bulk, manage products, customers, coupons and shipments, and react to store webhooks. - [SambaNova Integration](https://flowrunner.ai/integrations/sambanova.md): Run chat completions, single-prompt generation, and embeddings on SambaNova Cloud's RDU hardware through an OpenAI-compatible API. Agents call Llama, DeepSeek, and Qwen models with tool calling and structured JSON output, and transcribe audio with Whisper. - [SamCart Integration](https://flowrunner.ai/integrations/samcart.md): Pull products, orders, customers, subscriptions, and charges from your SamCart checkout platform. Agents reconcile revenue over date ranges, monitor subscription status for churn signals, and sync buyers into CRMs and reporting systems. - [Sameday Integration](https://flowrunner.ai/integrations/sameday-curier-ro.md): Connect AI agents to Sameday, a Romanian courier with an EasyBox locker network. Agents estimate shipping costs, create and track AWBs, look up lockers and cities, and download printable labels so Romanian parcel dispatch runs hands free. - [SAP Business One Integration](https://flowrunner.ai/integrations/sap-business-one.md): Automate SAP Business One through the OData Service Layer. Agents manage business partners and items, run the full sales and purchasing document lifecycle, apply payments, and post journal entries. - [SAP S/4HANA Integration](https://flowrunner.ai/integrations/sap-s4hana.md): Connect AI agents to SAP S/4HANA through its OData v2 API services. Agents read business partners, sales orders, products, and purchase orders, create sales orders with automatic CSRF token handling, and run arbitrary OData queries against any exposed API service. - [SAP SuccessFactors Integration](https://flowrunner.ai/integrations/sap-successfactors.md): Read core HR data from SAP SuccessFactors through the OData v2 API. Agents look up users, employment records, effective-dated job information, biographical data, and legal entities to feed provisioning, payroll, and org reporting workflows. - [SAP SuccessFactors Recruiting Integration](https://flowrunner.ai/integrations/sap-successfactors-recruiting.md): Connect AI agents to SAP SuccessFactors Recruiting, the applicant tracking module of the SAP SuccessFactors HXM suite. Agents read job requisitions, applications, status history, and offers so recruiting pipeline data feeds dashboards and downstream automations. - [SatisMeter Integration](https://flowrunner.ai/integrations/satismeter.md): SatisMeter runs NPS, CES, CSAT, PMF and custom surveys inside web apps, mobile apps and email. Agents export responses and statistics, identify and update users, insert responses, manage unsubscribes, and react to each new response. - [Scalelist Integration](https://flowrunner.ai/integrations/scalelist.md): Connect AI agents to Scalelist, a B2B lead generation and data enrichment platform. Agents find verified contact details for a person or domain, and run bulk enrichment across a batch of prospects as an asynchronous job. - [ScanOrders Integration](https://flowrunner.ai/integrations/scanorders.md): ScanOrders is barcode pick and fulfillment scanning for Shopify and WooCommerce sellers. Agents read orders and their picked state, add order lines, and poll for newly picked fulfillments to trigger the next step in the warehouse. - [Scarf Integration](https://flowrunner.ai/integrations/scarf-club.md): Scarf is a Brazilian all-in-one creator platform with courses, community spaces, memberships and checkout under one club. Agents sell courses and memberships from outside Scarf, grant access, run renewal and dunning pipelines on invoices, and manage members and spaces through the GraphQL API. - [OnceHub (ScheduleOnce) Integration](https://flowrunner.ai/integrations/scheduleonce.md): Read bookings and their customer details from OnceHub (formerly ScheduleOnce). Agents browse booking pages and master pages, manage contacts, and look up team members to route follow-ups when meetings are scheduled, rescheduled, or canceled. - [Schogini Integration](https://flowrunner.ai/integrations/schogini.md): Use the Schogini utility API for image editing and inspection, QR codes, text analysis, encoding, and link shortening. Agents close small gaps in a workflow behind one credential. - [SchoolMaker Integration](https://flowrunner.ai/integrations/schoolmaker.md): Connect AI agents to SchoolMaker, a French online course and community platform. Agents manage members and offer access, create coaching appointments, send community messages, and react to school events so student operations run without manual touchpoints. - [HasData Integration](https://flowrunner.ai/integrations/scrape-it-cloud.md): HasData (formerly Scrape-It.Cloud) is a web scraping platform with JavaScript rendering and AI extraction. Agents scrape any page, query dozens of source-specific scrapers for Google, Maps, Amazon, Zillow, Indeed, and more, and run asynchronous crawler jobs. - [ScrapeGraphAI Integration](https://flowrunner.ai/integrations/scrapegraphai.md): Extract structured data from any web page with a plain-English prompt through ScrapeGraphAI's LLM-powered scraping API. Agents enforce JSON output schemas, run search-and-scrape queries across the web, convert pages to LLM-ready Markdown, and poll long-running jobs by request ID. - [Scrapeless Integration](https://flowrunner.ai/integrations/scrapeless.md): Scrapeless is a web scraping and unblocking platform. Agents run site-specific scrapers, unlock and render protected pages, crawl whole sites, and drive a scraping browser with sessions, profiles, and stored credentials. - [ScrapeNinja Integration](https://flowrunner.ai/integrations/scrapeninja.md): ScrapeNinja is a web scraping API with a fast HTTP engine, a real browser engine, and a server-side Cheerio extractor. Workflows fetch pages, extract structured data with JavaScript extractors, capture screenshots, and grab AJAX responses. - [Scrapfly Integration](https://flowrunner.ai/integrations/scrapfly.md): Scrapfly is a managed web scraping and browser automation platform. Workflows scrape with anti-bot protection and JavaScript rendering, batch and schedule scrapes, take screenshots, extract structured data, and crawl whole sites. - [ScrAPI Integration](https://flowrunner.ai/integrations/scrapi.md): ScrAPI is a privacy-first web scraping API with browser automation, CAPTCHA solving, and proxy rotation. Agents scrape pages as HTML, Markdown, or JSON, capture screenshots and PDFs, and run hosted ecommerce price and stock monitors. - [ScrapingBee Integration](https://flowrunner.ai/integrations/scrapingbee.md): Scrape JavaScript-heavy pages through ScrapingBee with premium proxies, country geolocation, screenshots, and structured extraction. Agents pull structured Google SERP data and check remaining credits before launching large scraping jobs. - [Scraptio Integration](https://flowrunner.ai/integrations/scraptio.md): Scraptio extracts the readable text of a web page, optionally narrowed to chosen elements. Workflows pull article bodies without navigation and footer noise, validate the key, and check usage. - [screenshotbase Integration](https://flowrunner.ai/integrations/screenshotbase.md): screenshotbase is a website screenshot API. Workflows capture full-page or viewport screenshots with ad, cookie-banner, and chat blocking, save them to FlowRunner files or a hosted URL, and check the account quota. - [Scrive Integration](https://flowrunner.ai/integrations/scrive.md): Run the full e-signature lifecycle in Scrive. Agents create documents from PDFs or reusable templates, start the signing process, remind signatories who have not signed, cancel or trash documents, and download the signed PDF with its audit trail. - [SCRNIFY Integration](https://flowrunner.ai/integrations/scrnify.md): SCRNIFY is a pay-as-you-go screenshot and video capture API. Workflows render any public URL into PNG, JPEG, or WebP images or MP4, WebM, and GIF recordings, and generate signed capture URLs safe to publish. - [SE Ranking Integration](https://flowrunner.ai/integrations/se-ranking.md): Pull keyword rankings, ranking history, and tracked-keyword audits from your SE Ranking projects. Agents run keyword research and competitor domain overviews, and check the API credit balance before credit-heavy research operations. - [SeaTable Integration](https://flowrunner.ai/integrations/seatable.md): Connect AI agents to a SeaTable base. Agents read, append, and update rows in bulk, run read-only SQL over base data, maintain row links, and discover table and column names. - [SectorFlow Integration](https://flowrunner.ai/integrations/sectorflow.md): Manage SectorFlow multi-model workspaces and chat against them from FlowRunner. Agents compare model output across providers without a connection per vendor. - [SecurityScorecard Integration](https://flowrunner.ai/integrations/securityscorecard.md): Connect AI agents to SecurityScorecard. Agents assess any company's cyber posture by domain, pull grades, factor scores, and issue findings, benchmark against industry peers, manage portfolios, and generate reports. - [SeekTable Integration](https://flowrunner.ai/integrations/seektable.md): Connect AI agents to SeekTable, a pivot table BI and reporting tool. Agents run and export saved reports against a cube, import CSV data into a cube, and share a report by email. - [SEEN Integration](https://flowrunner.ai/integrations/seen.md): SEEN renders one personalized film per recipient and delivers it by email, SMS, push, or landing page. Agents create data items in a workspace to trigger a render, list and retrieve items, and react as new recipients are added. - [Segment Integration](https://flowrunner.ai/integrations/segment.md): Connect AI agents to Segment's Public and Tracking APIs. Agents provision sources and destinations, build and activate Engage audiences, govern data and IAM, monitor delivery and usage, and send live customer events. - [SegMetrics Integration](https://flowrunner.ai/integrations/segmetrics.md): Connect AI agents to SegMetrics, marketing attribution for funnels, courses, and memberships. Agents sync contacts and orders, tag them, and read the attribution reports that show which ads produced revenue. - [Segmind Integration](https://flowrunner.ai/integrations/segmind.md): Run serverless GPU inference across more than 200 generative AI models covering image, video, audio, and text, billed per use. Agents generate media without provisioning or paying for idle hardware. - [Seliom Integration](https://flowrunner.ai/integrations/seliom.md): Seliom is no-code business process automation from Spain: design a process template once, then run cases through it. Agents start cases, read and advance them, and pull case data so a process that lives in Seliom can be driven from anywhere. - [Seller Assistant Integration](https://flowrunner.ai/integrations/seller-assistant.md): Seller Assistant is an Amazon product research and sourcing tool with Keepa data and profit calculation. Agents turn a supplier price list into go or no-go decisions, watch ASINs for buy box, price and rank movement, run the repricer, and manage FBA inbound shipments. - [Sellercloud Integration](https://flowrunner.ai/integrations/sellercloud.md): Manage omnichannel e-commerce operations in Sellercloud. Agents read and create catalog products, orders, and customers, check inventory levels by SKU, and reconcile purchase orders and vendors across all sales channels from a single connection. - [SellIntegro CloudPrint Integration](https://flowrunner.ai/integrations/sellintegro-cloud-print.md): SellIntegro CloudPrint sends documents from a flow to a physical printer. Agents print a shipping label on a warehouse thermal printer the moment an order is ready, or an invoice or packing slip when it is paid, from a PDF or raw ZPL and EPL commands. - [Sellsy Integration](https://flowrunner.ai/integrations/sellsy.md): Connect AI agents to Sellsy, the French CRM, invoicing, and prospecting suite. Agents manage companies, contacts, and individuals, open opportunities, and turn them into estimates and invoices. - [Sempico Solutions Integration](https://flowrunner.ai/integrations/sempico-solutions-sms.md): Connect AI agents to Sempico Solutions, an A2P SMS gateway also sold under the Gatum brand. Agents send single, batch, and group campaigns, run HLR and MNP lookups, search sent traffic, and manage contact groups and blacklists so high-volume messaging stays deliverable and accountable. - [Semrush Integration](https://flowrunner.ai/integrations/semrush.md): Pull live SEO intelligence from Semrush: domain overviews, organic keyword rankings, keyword metrics, related-keyword ideas, and backlink summaries by regional Google database. Agents get raw and parsed results and can check remaining API units before large exports. - [SendApp Integration](https://flowrunner.ai/integrations/sendapp-cloud.md): Connect AI agents to SendApp, an Italian WhatsApp marketing and conversational AI platform. Agents send WhatsApp and SMS messages, manage Meta templates and contacts, launch bulk campaigns, and query the built-in AI assistant so customer outreach runs across channels from one place. - [Sendbot Integration](https://flowrunner.ai/integrations/sendbot.md): Automate the Sendbot AI chatbot platform for WhatsApp and web, managing workspaces and bots and driving chat sessions end to end. Agents keep conversational flows current as the underlying content changes. - [Sendcloud Integration](https://flowrunner.ai/integrations/sendcloud.md): Automate shipping through Sendcloud. Agents create and cancel parcels, buy carrier labels, look up shipping methods and sender addresses, and monitor return shipments to keep e-commerce fulfilment moving across European carriers. - [Sender Integration](https://flowrunner.ai/integrations/sender.md): Manage your Sender email and SMS marketing audience. Agents create and update subscribers, organize them into groups for segmentation, delete lapsed contacts, and review email and SMS campaign activity for reporting. - [SendFox Integration](https://flowrunner.ai/integrations/sendfox.md): Keep your SendFox creator audience in sync. Agents create contacts, build and organize lists, unsubscribe opt-outs by email, and page through subscribers and sent campaigns for reporting. - [SendGrid Integration](https://flowrunner.ai/integrations/sendgrid.md): Connect AI agents to SendGrid. Agents send transactional and templated email, upsert and search contacts, manage lists, maintain unsubscribe and bounce suppressions, validate addresses, and read email stats. - [Sendlane Integration](https://flowrunner.ai/integrations/sendlane.md): Manage your Sendlane e-commerce marketing audience end to end. Agents upsert contacts by email, subscribe them to lists, apply and remove tags for segmentation, and track behavioral events that fire Sendlane automations. - [Sendle Integration](https://flowrunner.ai/integrations/sendle.md): Book carbon-neutral shipping through Sendle. Agents create pickup orders, quote routes before checkout, list available service levels and ETAs, track parcels through transit, and cancel bookings while they are still cancellable. - [Sendme Integration](https://flowrunner.ai/integrations/sendme123.md): Manage contacts and send SMS and email campaigns with Sendme, maintaining a tagged address book with custom fields. Agents keep segmentation driven by live data rather than a periodic upload. - [SendOwl Integration](https://flowrunner.ai/integrations/sendowl.md): SendOwl sells and delivers digital goods: files, software licenses, bundles, subscriptions and drip content. Agents fulfill orders taken on another checkout, refund and revoke access, run discount campaigns, and validate software license keys from your own application. - [SendPulse Integration](https://flowrunner.ai/integrations/sendpulse.md): Manage SendPulse address books and their contacts, send transactional email through the SMTP service, and report on campaign statistics. Agents sync leads into mailing lists with merge-field variables and look up a contact's subscription state by email. - [sendSMS Integration](https://flowrunner.ai/integrations/sendsms.md): Connect AI agents to sendSMS, a Romanian CPaaS and SMS gateway with Viber support. Agents send messages, run batch campaigns, generate and check OTP codes, manage contacts and blocklists, and look up numbers via HLR and MNP so Romanian and international messaging runs end to end. - [SendX Integration](https://flowrunner.ai/integrations/sendx.md): SendX is an email marketing and automation platform. Agents manage contacts, lists, tags and custom fields, build and report on campaigns, send transactional email, record behavioral and revenue events, and register webhooks. - [Sendy Integration](https://flowrunner.ai/integrations/sendy.md): Connect AI agents to a self-hosted Sendy installation. Agents subscribe and unsubscribe list members, check subscription status and counts, retrieve brands, and draft or send email campaigns. - [SensiBot Integration](https://flowrunner.ai/integrations/sensibot.md): Automate the SensiBot AI WhatsApp platform, switching the assistant between AI and keyword replies and reading chat history. Agents hand a conversation to a human the moment it needs one. - [SensorPro Integration](https://flowrunner.ai/integrations/sensorpro.md): SensorPro is an email marketing and relay platform from Narragansett Technologies. Agents manage contacts and their consent state, build campaigns and segments, launch broadcasts, read engagement metrics, and send through the transactional relay. - [Sentry Integration](https://flowrunner.ai/integrations/sentry.md): Application error monitoring and performance tracking from your flows. Manage projects, triage issues, inspect events, and coordinate releases and deploys over the Sentry API using an auth token. - [Senuto Integration](https://flowrunner.ai/integrations/senuto.md): Pull SEO data from Senuto, the Polish visibility, keyword, and backlink platform, covering visibility analysis, keyword explorer, and link analysis. Agents track ranking movement as a workflow step rather than a monthly report. - [SerpApi Integration](https://flowrunner.ai/integrations/serpapi.md): Run live, structured searches across 15 engines through SerpApi, including Google Search, Maps, News, Scholar, Bing, YouTube, and Amazon. Agents get fully parsed JSON results, re-read archived searches without spending new credits, and monitor account usage. - [ServiceTitan Integration](https://flowrunner.ai/integrations/service-titan.md): Connect AI agents to ServiceTitan, the field service platform for the trades. Agents create and look up customers, read service locations, monitor jobs and appointments by status for dispatch workflows, and pull invoices for accounting sync. - [ServiceM8 Integration](https://flowrunner.ai/integrations/servicem8.md): Drive ServiceM8 field service work from agent workflows. Agents create jobs from inbound leads, advance them through the quote-to-completed lifecycle, sync clients, and read staff, materials, activities, and job contacts for scheduling and billing. - [ServiceNow Integration](https://flowrunner.ai/integrations/servicenow.md): Connect AI agents to ServiceNow's Now Platform. Agents run CRUD operations over the ITSM tables a business user works with, incidents, change requests, problems, requested items, and users, plus a generic Table Record escape hatch for any other table. - [SerwerSMS Integration](https://flowrunner.ai/integrations/serwersms.md): Connect AI agents to SerwerSMS, a Polish messaging gateway from Vercom covering SMS, MMS, voice, and RCS. Agents send and schedule messages, pull delivery reports and inbound replies, manage contacts, templates, and blacklists, and run HLR checks so Polish-market messaging stays reliable. - [Amazon SES Integration](https://flowrunner.ai/integrations/ses-service.md): Send transactional and templated email through Amazon SES v2. Agents deliver receipts and notifications, manage reusable Handlebars-style templates, and broadcast personalized email in a single batch. - [SetSmart Integration](https://flowrunner.ai/integrations/setsmart.md): Connect AI agents to SetSmart, an AI appointment setter that qualifies leads and books calls in Instagram, WhatsApp, and Messenger. Agents pick up the booked meeting and prepare whoever is taking it. - [sevdesk Integration](https://flowrunner.ai/integrations/sevdesk.md): sevdesk is the German accounting and invoicing platform. Agents manage contacts with their addresses and custom fields, issue invoices, credit notes, and orders, book vouchers, and resolve the reference data every write depends on. - [seven Integration](https://flowrunner.ai/integrations/seven.md): Send SMS and text-to-speech voice calls, look up and validate phone numbers, track delivery, and manage contacts through the seven.io gateway. - [7Loc Integration](https://flowrunner.ai/integrations/seven-loc.md): Connect AI agents to 7Loc, an AI-powered hospitality platform for guest mobile ordering. Agents start guest carts and scan receipts into loyalty checks, and read the store catalog, shipping options, loyalty configuration, and coupon terms behind each venue. - [Seven Senders Integration](https://flowrunner.ai/integrations/sevensenders.md): Connect AI agents to Seven Senders, a European parcel delivery orchestration and tracking platform. Agents create orders and shipments, push state events, generate carrier labels, and pull end to end tracking history so cross-border delivery data stays current in every system. - [SFTP Integration](https://flowrunner.ai/integrations/sftp.md): Move files on any SFTP server from agent workflows. Agents browse and stat directories, download files or read them inline as text, upload from FlowRunner storage, URLs, or raw text, and rename, delete, chmod, and manage remote folders. - [ShareAI Integration](https://flowrunner.ai/integrations/share-ai.md): Run open-source language models through the ShareAI decentralized GPU network. Agents reach open-weight models without provisioning hardware or signing a per-model contract. - [ShareFile Integration](https://flowrunner.ai/integrations/sharefile.md): Manage documents in ShareFile from agent workflows. Agents browse folders, upload and download files, create and organize items, and generate Send shares to distribute files or Request shares to collect uploads securely. - [SharePoint Integration](https://flowrunner.ai/integrations/sharepoint.md): Manage SharePoint sites, lists, list items, and document libraries from FlowRunner agents through the Microsoft Graph API, and react to new or updated list items and files. - [SharpSpring Integration](https://flowrunner.ai/integrations/sharpspring.md): Manage leads in SharpSpring, now Constant Contact Lead Gen and CRM. Agents bulk create and update leads, add them to active lists for campaign targeting, report on sales opportunities, and inspect lead field definitions before building automations. - [Ship24 Integration](https://flowrunner.ai/integrations/ship24.md): Track packages across hundreds of couriers worldwide through Ship24. Agents register tracking numbers, retrieve full checkpoint history and delivery status, track a parcel in a single call, and resolve courier codes before creating trackers. - [ShipBob Integration](https://flowrunner.ai/integrations/shipbob.md): Full 3PL automation with 5 real-time fulfillment triggers. Agents respond to shipment events, manage orders, track inventory, process returns, and handle warehouse receiving orders. - [shipcloud Integration](https://flowrunner.ai/integrations/shipcloud.md): Connect AI agents to shipcloud, a German multi-carrier shipping platform covering DHL, DPD, UPS, GLS, and more. Agents create shipments with labels, get shipment quotes, track parcels with trackers, and request pickups so multi-carrier shipping runs through one API. - [Shipday Integration](https://flowrunner.ai/integrations/shipday.md): Connect AI agents to Shipday for local delivery dispatch. Agents create delivery orders, advance them through pickup and delivery, assign orders to drivers, and keep the driver roster in sync with your POS or ordering systems. - [ShipHero Integration](https://flowrunner.ai/integrations/shiphero.md): Run warehouse and fulfillment operations from AI agents through ShipHero's GraphQL API. Agents list and create orders, browse the product catalog, check per-warehouse inventory before fulfilling, and reach any part of the schema with a raw GraphQL query. - [ShiPinHao Scraper Integration](https://flowrunner.ai/integrations/shipinhao.md): Connect AI agents to WeChat Channels (ShiPinHao), China's short video platform. Agents look up video accounts, read individual videos, and search public content for monitoring and research. - [Shippingbo Integration](https://flowrunner.ai/integrations/shippingbo.md): Connect AI agents to Shippingbo, a French order and warehouse management platform for ecommerce. Agents manage orders, products, and return orders, and look up configured carriers and warehouses so fulfillment data stays in sync across sales channels. - [ShippingEasy Integration](https://flowrunner.ai/integrations/shippingeasy.md): Connect AI agents to ShippingEasy, US shipping label software by Auctane. Agents search and retrieve orders account wide or per store, create orders, update order status, and cancel orders so fulfillment systems stay aligned with shipping. - [Shippo Integration](https://flowrunner.ai/integrations/shippo.md): Multi-carrier shipping across USPS, FedEx, UPS, and DHL. Create labels, compare live rates, track packages, handle customs, schedule pickups, and manage refunds and orders. - [ShipStation Integration](https://flowrunner.ai/integrations/shipstation.md): Automate e-commerce shipping with ShipStation. Manage orders, buy and void labels, compare live carrier rates, and keep customers, products, warehouses, stores, and webhooks in sync. - [Shop Apotheke Integration](https://flowrunner.ai/integrations/shop-apotheke.md): Shop Apotheke's marketplace runs on Mirakl, and this connector covers the seller surface. Agents pull new orders into an ERP or warehouse system and accept them, confirm shipments with tracking, handle returns and incidents, push stock and price imports, and read accounting documents. - [Shopify Integration](https://flowrunner.ai/integrations/shopify.md): Manage Shopify orders, products, collections, customers, multi-location inventory, and Shopify Payments from automated flows. Agents retrieve and refund orders, curate collections, sync stock across locations, and react in real time to new orders, customers, products, and disputes. - [Shoprocket Integration](https://flowrunner.ai/integrations/shoprocket.md): Shoprocket is an embeddable e-commerce layer for any website. Agents manage products, orders and customers, read shipping and tax settings, and react to store events as they happen. - [Short.io Integration](https://flowrunner.ai/integrations/short-io.md): Create and manage branded short links on your own domains with Short.io. Agents mint a tracked link per send and read click data back into reporting. - [Short Menu Integration](https://flowrunner.ai/integrations/short-menu.md): Create and manage branded short links through Short Menu, the shortener and link-analytics service from Appiculous. Agents attach a measurable link to every outbound message. - [ShortPen Integration](https://flowrunner.ai/integrations/shortpen.md): Manage branded links and click analytics through the ShortPen REST API. Agents mint links per campaign and see which placements produced real traffic. - [Shotstack Integration](https://flowrunner.ai/integrations/shotstack.md): Turn data into rendered video with Shotstack's cloud editing API. Agents submit JSON timeline edits, render reusable templates with merge fields for personalized videos at scale, and poll render status until the finished asset URL is ready. - [Shuffll Integration](https://flowrunner.ai/integrations/shuffll.md): Shuffll is an AI video creation platform: give it a topic, web page, Google Doc, or script and it writes the storyline, sources the media, applies your branding, and renders a video. Agents create and export projects, manage brand assets, pick from the template library, and react when a project is ready. - [Shuttle Integration](https://flowrunner.ai/integrations/shuttle-payment-links.md): Shuttle is a white-label payment orchestration platform that routes payments across gateways and processors. Agents create and capture payments, refund and void, run recurring contracts, generate hosted checkouts and payment links, and manage accounts, instances, and routing. - [Signable Integration](https://flowrunner.ai/integrations/signable.md): Signable is a UK electronic signature service. Agents send envelopes built from one or more templates, chase and track signers, add copy recipients, maintain the contact book and users, register webhooks, and manage partner sub accounts. - [SignalWire Integration](https://flowrunner.ai/integrations/signalwire.md): Send SMS and MMS, place and update phone calls, buy and manage phone numbers, and pull call recordings through SignalWire's Twilio-compatible LaML API, all from AI agents running against your SignalWire Space. - [Signaturit Integration](https://flowrunner.ai/integrations/signaturit.md): Signaturit is a Spanish electronic signature and certified delivery provider. Agents send a document for signature in a single call by email, SMS, or signing URL, require ID scans or SMS validation, download the evidence, manage templates and branding, and react to signature events. - [Signi Integration](https://flowrunner.ai/integrations/signi.md): Signi is a Czech electronic signature platform built around the contract. Agents upload a PDF with who signs, in what capacity and order, place signatures by coordinates or anchor text, hold drafts for a person to release, and read the audit trail. - [SIGNL4 Integration](https://flowrunner.ai/integrations/signl4.md): Send and resolve SIGNL4 alerts that reach the on-call team by push, SMS, and voice, tied together by a stable per-incident External ID. - [SignNow Integration](https://flowrunner.ai/integrations/signnow.md): Automate e-signature workflows with SignNow. Agents upload documents, auto-detect signature fields, send role-based or free-form signing invites, download the flattened signed PDF, and create documents from reusable templates. - [Signority Integration](https://flowrunner.ai/integrations/signority.md): Signority is a Canadian electronic signature platform, now part of FileCloud. Agents build documents as envelopes with roles for signers, viewers, and reviewers, hold them as drafts for release, send from templates, and organize them into folders. - [SigParser Integration](https://flowrunner.ai/integrations/sigparser.md): SigParser is a stateless email parsing API that extracts contact details from email signatures and strips signatures and reply chains from message bodies. Agents turn a raw or MIME email into clean contact records and a clean message body before pushing it to a CRM. - [SimpleCirc Integration](https://flowrunner.ai/integrations/simplecirc.md): Connect AI agents to SimpleCirc, subscriber management and circulation software for publishers. Agents create and renew subscriptions, update subscriber addresses, and keep circulation records accurate. - [Simpleen Integration](https://flowrunner.ai/integrations/simpleen-translation.md): Translate plain text, structured JSON and i18n locale objects, and complete localization files with Simpleen. Agents localize product strings without breaking the file structure around them. - [Simplero Integration](https://flowrunner.ai/integrations/simplero.md): Manage contacts, lists, courses, subscriptions, and revenue with the Simplero v2 API. Agents enrol buyers, grant course access, and keep billing and access in step. - [Simplesat Integration](https://flowrunner.ai/integrations/simplesat.md): Simplesat collects CSAT, NPS and CES feedback for service teams. Agents push feedback in, pull it back out, keep customers and team members in step with a helpdesk, send event based survey emails, and react to new feedback as it arrives. - [SimpleShop Integration](https://flowrunner.ai/integrations/simpleshop-cz2.md): SimpleShop is the Czech checkout and invoicing platform for selling online. Agents issue invoices, proformas and the other document types, email and settle them, maintain the address book, and export who bought what. - [SimpleTexting Integration](https://flowrunner.ai/integrations/simpletexting.md): Run US SMS marketing from AI agents through SimpleTexting. Agents send SMS and MMS messages, create and update contacts, and organize contact lists so campaigns stay in sync with your CRM. - [Simplified Integration](https://flowrunner.ai/integrations/simplified-webhooks.md): Publish and schedule social content, pull analytics, and generate AI images with the Simplified API. Agents queue posts across networks and hold anything sensitive for approval. - [SimplyBook.me Integration](https://flowrunner.ai/integrations/simplybook.md): Manage SimplyBook.me appointments from AI agents: create and cancel bookings, add clients, list services and providers, and call any admin method directly through the JSON-RPC escape hatch. - [SimplyMeet.me Integration](https://flowrunner.ai/integrations/simplymeet-me.md): Connect AI agents to SimplyMeet.me, the personal meeting scheduler from the SimplyBook.me group. Agents check availability, book and reschedule meetings, invite participants, and generate single-use booking links so meeting logistics run themselves. - [Simvoly Integration](https://flowrunner.ai/integrations/simvoly2.md): Connect AI agents to Simvoly, a website and funnel builder with a built-in store, memberships, and email lists. Agents manage contacts, members, products, and orders, read form submissions and bookings, and maintain subscriber lists so funnel data flows straight into the rest of the stack. - [Sinch Engage Integration](https://flowrunner.ai/integrations/sinch-engage.md): Message customers over WhatsApp, Facebook Messenger, Instagram, and Telegram through Sinch Engage. Agents send text, media, and approved template messages, segment users with tags, and react to inbound messages in real time. - [SingleCase Integration](https://flowrunner.ai/integrations/singlecase.md): Connect AI agents to SingleCase, a Czech legal practice management platform. Agents open cases against a client, file documents, track time and expenses, raise quick invoices and mark them sent to accounting, and read the retainers and case messages behind each matter. - [Site Search 360 Integration](https://flowrunner.ai/integrations/site-search-360.md): Connect AI agents to Site Search 360, a hosted site search service for websites. Agents run searches, fetch autocomplete suggestions, index or remove content, and read index status and query analytics so on-site search stays fresh and measurable. - [Site24x7 Integration](https://flowrunner.ai/integrations/site24x7.md): Keep uptime monitoring in step with your workflows through Site24x7, Zoho's monitoring platform. Agents create and manage monitors, organize monitor groups, read live up, down, or trouble status, and pull uptime and performance summary reports. - [Siteglide Integration](https://flowrunner.ai/integrations/siteglide.md): Connect AI agents to Siteglide, a digital experience platform agencies use to build sites, portals, and stores. Agents manage pages, module and WebApp items, users, categories, products, and form cases, and read orders so site content and CRM records stay in sync. - [Skai Integration](https://flowrunner.ai/integrations/skai.md): Manage paid media and pull performance data with the Skai API, covering campaigns, ad groups, ads, portfolios, tags, and reports. Agents adjust spend against results the business actually recorded. - [Sklonovani jmen Integration](https://flowrunner.ai/integrations/sklonovani-jmen.md): Inflect Czech and foreign personal names into any of the seven Czech grammatical cases and detect the gender behind a name. Agents address Czech recipients correctly instead of defaulting to the nominative. - [Skool Integration](https://flowrunner.ai/integrations/skool.md): Read Skool community data from AI agents: list members, look up a member's role, level, and points, and pull community metadata. Skool has no official public API, so an authenticated passthrough action reaches any Skool-compatible endpoint you have access to. - [SkootEco Integration](https://flowrunner.ai/integrations/skoot-eco.md): Connect AI agents to SkootEco, a climate action service for tree planting and ocean plastic recovery. Agents allocate climate impact as a step inside a flow and read the profile's all-time impact totals. - [Skyvern Integration](https://flowrunner.ai/integrations/skyvern-ai.md): Run AI agents inside cloud browsers with Skyvern to complete tasks on websites that have no API. Agents describe a goal in plain English or run a saved agent and collect the result. - [Slack Integration](https://flowrunner.ai/integrations/slack.md): Give AI agents a voice in your workspace. Slack triggers activate workflows on messages, mentions, and reactions. Agents send updates, route exceptions, and collect human decisions without leaving Slack. - [SleekFlow Integration](https://flowrunner.ai/integrations/sleekflow.md): Connect AI agents to SleekFlow's omnichannel inbox. Agents create, search, and update contacts, send messages over WhatsApp, SMS, and social channels, read conversation threads, and segment customers with labels for targeted broadcasts. - [SlickText Integration](https://flowrunner.ai/integrations/slicktext.md): Run keyword-based SMS marketing through SlickText's v1 API. Agents add and unsubscribe contacts, browse textwords and lists, and send and review SMS messages to keep subscriber engagement in sync with your other systems. - [SlickText (v2) Integration](https://flowrunner.ai/integrations/slicktext-v2.md): Send SMS and MMS and manage contacts through SlickText's modern brand-scoped v2 developer API. Agents discover brands, create and look up contacts, send messages, and browse lists across every SlickText account they can access. - [SlideSpeak Integration](https://flowrunner.ai/integrations/slidespeak.md): Generate, edit, translate, and narrate PowerPoint decks with the SlideSpeak API. Agents turn source material into a presentation and localize it without a design pass per market. - [SlimCRM Integration](https://flowrunner.ai/integrations/slimcrm.md): Connect AI agents to SlimCRM, a Vietnamese CRM for small and medium businesses. Agents create customers and leads, run projects through milestones, open support tickets, log expenses, and read the invoices behind each account. - [SlimEmail Integration](https://flowrunner.ai/integrations/slimemail.md): SlimEmail is a Vietnamese email marketing platform from SlimCRM. Agents subscribe, unsubscribe, check and delete subscribers, count a list, and create campaigns on your installation. - [Sling Integration](https://flowrunner.ai/integrations/sling.md): Connect AI agents to Sling by Toast, an employee shift scheduling and workforce management platform. Agents look up users and schedules, pull roster reports, and create and publish shifts so staffing gaps get filled without spreadsheet juggling. - [Slybroadcast Integration](https://flowrunner.ai/integrations/slybroadcast.md): Send ringless voicemail campaigns through Slybroadcast, delivering pre-recorded audio straight to voicemail boxes. Agents reach a list without ringing a phone, and hold the send behind an approval. - [Smaily Integration](https://flowrunner.ai/integrations/smaily.md): Smaily is an Estonian email marketing and automation platform. Agents manage subscribers and segments, launch campaigns and A/B tests, trigger automation workflows, send transactional messages, and read the event logs. - [SmartEmailing Integration](https://flowrunner.ai/integrations/smart-emailing.md): SmartEmailing is a Czech email marketing platform. Agents manage contacts and lists, send newsletters and transactional email, dispatch SMS, run automations, sync e-commerce orders, record GDPR purposes, and register webhooks. - [SmartBill Integration](https://flowrunner.ai/integrations/smartbill.md): SmartBill is the Romanian invoicing platform. Agents issue invoices, proformas, storno invoices, receipts, and fiscal till receipts, cancel or restore them, and pull the PDFs and payment status a workflow needs. - [Smartcar Integration](https://flowrunner.ai/integrations/smartcar.md): Connect AI agents to vehicles across 40+ car brands through Smartcar. Agents read odometer, GPS location, fuel, and battery telemetry, look up vehicle attributes, and send remote lock or unlock commands from a workflow. - [Smartcat Integration](https://flowrunner.ai/integrations/smartcat.md): Automate localization through Smartcat's Integration API. Agents create and cancel translation projects, track document status across workflow stages, and pull word-count statistics to estimate translation volume and cost. - [Smartlead Integration](https://flowrunner.ai/integrations/smartleadai.md): Run cold email outreach at scale through Smartlead. Agents create campaigns, add and pause leads, inspect email sequences, and list the sending mailboxes connected to your account. - [Smartlook Integration](https://flowrunner.ai/integrations/smartlook.md): Connect AI agents to Smartlook, a session replay and product analytics platform. Agents read sessions, events, and visitor records, pull funnel data, and manage webhook subscriptions for downstream systems. - [SmartPasses Integration](https://flowrunner.ai/integrations/smartpasses.md): Manage loyalty program members and reward offers with SmartPasses, enrolling customers, awarding points, and publishing offers. Agents keep wallet passes current as balances change. - [SmartReach.io Integration](https://flowrunner.ai/integrations/smartreach-io.md): SmartReach.io is a multichannel cold outreach platform. Agents add prospects, manage campaigns and read their stats, work tasks, maintain the do-not-contact list, and administer teams and users. - [Smartsheet Integration](https://flowrunner.ai/integrations/smartsheet.md): Give AI agents full control of Smartsheet: 49 actions covering sheets, rows, columns, attachments, discussions, workspaces, reports, search, and webhooks, including exporting sheets and downloading attachments straight to FlowRunner file storage. - [SmartSuite Integration](https://flowrunner.ai/integrations/smartsuite.md): Read and write SmartSuite work management data from AI agents. Agents browse solutions and tables, then run full record CRUD: listing, retrieving, creating, updating, and deleting rows in any table. - [SmartTask Integration](https://flowrunner.ai/integrations/smarttask-io.md): SmartTask is task and project management with a built-in CRM. Agents create and update tasks and projects, manage assignees and due dates, and start flows from SmartTask webhooks. - [Samsung SmartThings Integration](https://flowrunner.ai/integrations/smartthings.md): Control your smart home from AI agents through Samsung SmartThings. Agents list devices, read live device status, send control commands, browse locations and rooms, and execute saved scenes. - [Smith.ai Integration](https://flowrunner.ai/integrations/smith-ai.md): Request on-demand outbound calls handled by Smith.ai, the AI plus human virtual receptionist service. Agents escalate to a real receptionist when a call needs a person, not a script. - [Smoove Integration](https://flowrunner.ai/integrations/smoove.md): Connect AI agents to smoove, the Israeli marketing automation platform, through its official REST API. Agents manage contacts and lists and trigger campaigns off events in the systems of record. - [SMS Alert Integration](https://flowrunner.ai/integrations/sms-alert.md): Connect AI agents to SMS Alert, an Indian bulk and transactional SMS gateway by Cozy Vision. Agents send and schedule SMS, run OTP verification, manage contacts, groups, and templates, and pull delivery reports so transactional messaging in India stays traceable. - [SMS Masivos Integration](https://flowrunner.ai/integrations/sms-masivos.md): Connect AI agents to SMS Masivos, a Mexican bulk messaging platform covering SMS, WhatsApp, and voice. Agents send messages and verification codes, manage contact lists, pull delivery reports, and run the vendor's wallet and loyalty tooling so Mexican customer messaging and engagement run from one API. - [SMS Niaga Integration](https://flowrunner.ai/integrations/sms-niaga.md): Connect AI agents to SMS Niaga, a Malaysian bulk SMS platform from Web Impian. Agents send text messages to Malaysian numbers, check credit balances, and manage sender IDs and subscriber groups so local campaigns and alerts go out without manual handling. - [019 SMS Integration](https://flowrunner.ai/integrations/sms019.md): Connect AI agents to 019 SMS, the messaging platform of Israeli telecom operator 019 Mobile. Agents send SMS, voice, and WhatsApp messages, manage contact lists, blacklists, and one-time passwords, and read delivery reports and balances so Israeli-market messaging runs under full control. - [SMS8 Integration](https://flowrunner.ai/integrations/sms8-io.md): Connect AI agents to SMS8, an Android SMS gateway that sends messages through your own paired phones and SIM cards. Agents send SMS and MMS, run personalized batches, read inbox and delivery states, and manage contacts and OTP codes so messaging runs over your own SIMs without buying carrier routes. - [SMSAdvert Integration](https://flowrunner.ai/integrations/smsadvert-ro.md): Connect AI agents to SMSAdvert, a Romanian SMS platform from Synaptech Services. Agents send text messages immediately or schedule them into a future delivery window so customer notifications go out over the SMSAdvert network or your connected Android devices. - [SMSAlert Integration](https://flowrunner.ai/integrations/smsalert.md): Connect AI agents to SMSAlert, a Romanian messaging platform that unifies SMS gateways, GSM modems, Android phones, WhatsApp, Telegram, and push. Agents send, schedule, and bulk-send messages, check delivery status, and list sent and received traffic so one flow covers every attached channel. - [SMSAPI Integration](https://flowrunner.ai/integrations/smsapi.md): Connect AI agents to SMSAPI, a European messaging gateway from LINK Mobility with SMS, RCS, and OTP support. Agents send and schedule SMS to numbers or contact groups, verify one-time passcodes, run HLR lookups, and manage contacts, templates, and subaccounts so large-scale messaging stays organized. - [SMSEdge Integration](https://flowrunner.ai/integrations/smsedge.md): Connect AI agents to SMSEdge, a multi-channel campaign platform for SMS, cold email, and WhatsApp. Agents send single and bulk SMS, verify phone numbers, manage contact lists, run US 10DLC registry traffic, and pull sending statistics so high-volume campaigns stay compliant and measurable. - [SMS Everyone Integration](https://flowrunner.ai/integrations/smseveryone.md): Connect AI agents to SMS Everyone, an Australian SMS gateway for campaign-based messaging. Agents send and schedule campaigns, pause, resume, or cancel them, retrieve delivery receipts and replies, and manage recipient lists and opt-outs so Australian SMS programs stay controlled end to end. - [SMSFactor Integration](https://flowrunner.ai/integrations/smsfactor.md): Connect AI agents to SMSFactor, a French SMS gateway operated by Commify. Agents send messages and campaigns, react to replies, delivery reports, and opt-outs as they arrive, and manage contact lists, blacklists, and sub-accounts so French-market messaging runs as a closed loop. - [SMSGlobal Integration](https://flowrunner.ai/integrations/smsglobal.md): Connect AI agents to SMSGlobal, an Australian headquartered global SMS gateway. Agents send single or bulk SMS, review outgoing and inbound message history, manage sender IDs and dedicated numbers, and maintain opt-out lists and contact groups so customer notifications stay compliant and on schedule. - [SMSIndiaHub Integration](https://flowrunner.ai/integrations/smsindiahub.md): Connect AI agents to SMSIndiaHub, an Indian bulk, transactional, and OTP SMS gateway. Agents send single, scheduled, group, and bulk SMS, pull delivery reports, and check credit balances so high volume notifications go out reliably and spend stays visible. - [SMSKUB Integration](https://flowrunner.ai/integrations/smskub.md): Connect AI agents to SMSKUB, a Thai bulk SMS gateway with a managed OTP service. Agents send quick messages and scheduled campaigns, pull delivery reports, request and verify one time passwords, and check credit balances so Thai customer messaging and verification run without manual work. - [SMS Partner Integration](https://flowrunner.ai/integrations/smspartner.md): Connect AI agents to SMS Partner, a French SMS and RCS gateway. Agents send single and bulk SMS, deliver RCS rich cards and carousels, track delivery statuses, and manage contacts and stop lists so French and international campaigns stay deliverable and compliant. - [SMS Zasilam Integration](https://flowrunner.ai/integrations/smszasilam.md): Connect AI agents to SMS Zasilam, a Czech SMS gateway. Agents send single and bulk text messages, read inbound and outbound history, reconstruct two way conversations, and inspect or purge the outgoing queue so Czech customer messaging stays responsive and under control. - [SMTP2GO Integration](https://flowrunner.ai/integrations/smtp2go.md): Send transactional email through SMTP2GO from AI agents, with CC, BCC, reply-to, and custom headers. Agents audit delivery activity, pull deliverability stats, and manage the suppression list to protect sender reputation. - [Snack Prompt Integration](https://flowrunner.ai/integrations/snack-prompt.md): Store and share prompts, snippets, and documents through the Snack Prompt Integration API. Agents fetch the approved prompt version rather than one copied into a flow by hand. - [Snapchat Ads Integration](https://flowrunner.ai/integrations/snapchat-campaign-management.md): Manage Snapchat advertising from AI agents through the Marketing API. Agents walk the business hierarchy from organizations to ad accounts, ad squads, and ads, and list, create, and update campaigns, with new campaigns defaulting to paused so nothing spends before you activate it. - [Snapchat Conversions API Integration](https://flowrunner.ai/integrations/snapchat-conversions.md): Send server-side conversion events to Snapchat's Conversions API for a Snap Pixel. Agents post page views, add-to-cart, sign-up, and purchase events with customer PII automatically SHA-256 hashed before it leaves FlowRunner. - [Snappy Integration](https://flowrunner.ai/integrations/snappy.md): Send personalized gifts through the Snappy corporate gifting platform. Agents send gifts under campaigns when milestones hit, track redemption and shipping status, and browse recipients and gift collections to reconcile gifting spend. - [SnelStart Integration](https://flowrunner.ai/integrations/snelstart.md): SnelStart is the Dutch bookkeeping platform. Agents manage relations and articles, read price agreements, create sales orders and quotations, read sales invoices, and post to the sales, purchase, bank, cash, and journal ledgers. - [Snipcart Integration](https://flowrunner.ai/integrations/snipcart.md): Snipcart is a developer-first shopping cart that drops into any site. Agents manage products, inventory, orders and discounts, read and cancel subscriptions, and react to order events. - [Snov.io Integration](https://flowrunner.ai/integrations/snovio.md): Build and validate outreach audiences with Snov.io. Agents discover professional emails from names and company domains, verify deliverability, manage prospect lists, and monitor the account's credit balance. - [Snowflake Integration](https://flowrunner.ai/integrations/snowflake.md): Connect AI agents to Snowflake through the SQL API v2. Agents run parameterized SQL, load results into tables, poll long-running statements, and discover databases, schemas, and tables. - [Amazon SNS Integration](https://flowrunner.ai/integrations/sns-service.md): Publish messages to Amazon SNS topics or send SMS directly. Agents fan workflow events out to many subscribers, manage topics, and subscribe email, SMS, HTTP, SQS, or Lambda endpoints to a topic. - [SOAP Client Integration](https://flowrunner.ai/integrations/soap-client.md): Build correctly namespaced SOAP 1.1 or 1.2 envelopes, post them to any endpoint, and parse the reply. Agents reach the legacy enterprise services that never got a REST interface. - [SocialBee Integration](https://flowrunner.ai/integrations/socialbee.md): Schedule social content through SocialBee from AI agents. Agents browse a workspace's connected profiles and content categories, list existing posts, and file new posts into a category's queue for publishing across multiple profiles. - [Societe.com Integration](https://flowrunner.ai/integrations/societe-com.md): Societe.com is the French company-information provider keyed on SIREN, covering identity, directors, establishments, filed accounts, trademarks, official filings and a financial health score. Agents search and verify companies, pull directors and financials, download official documents and check a company's score. - [Softr Integration](https://flowrunner.ai/integrations/softr.md): Connect AI agents to Softr, the no-code app and client portal builder. Agents provision and invite app users, activate or deactivate accounts, generate magic-link sign-in URLs, and validate session tokens to authenticate requests from your Softr front end. - [Solana Integration](https://flowrunner.ai/integrations/solana.md): Query the Solana blockchain over JSON-RPC. Agents read wallet balances and SPL token holdings, look up transactions and signature history, fetch blockhashes and slots, broadcast pre-signed transactions, and call any other RPC method through an escape hatch. - [SOLAPI Integration](https://flowrunner.ai/integrations/solapi.md): Connect AI agents to SOLAPI, a South Korean messaging gateway covering SMS, KakaoTalk, RCS, fax, and voice. Agents send messages across channels, manage Kakao templates and sender numbers, track balances and statistics, and react to delivery reports and inbound faxes so Korean customer messaging runs end to end. - [SolarMarket Integration](https://flowrunner.ai/integrations/solarmarket.md): Connect AI agents to SolarMarket, a Brazilian CRM for solar energy sales integrators. Agents create and update clients and projects, set custom field values, and read the funnels and stages a project moves through. - [SolarWinds Service Desk Integration](https://flowrunner.ai/integrations/solarwinds.md): Connect AI agents to SolarWinds Service Desk, the cloud ITSM platform formerly known as Samanage. Agents create and update incidents, append comments, and look up users, hardware assets, and categories to triage and route tickets. - [Sonar Integration](https://flowrunner.ai/integrations/sonar.md): Connect AI agents to Sonar, a conversational SMS and MMS platform for business texting sold as Marchex Sonar. Agents send messages and campaigns, manage customers and follow-ups, track conversion events, and react to inbound messages and subscription changes in real time so text conversations get handled the moment they arrive. - [Sorry Integration](https://flowrunner.ai/integrations/sorry.md): Sorry (SorryApp) is a status page and incident communication platform. Agents open incident and maintenance notices, post updates, keep components accurate, manage subscribers and templates, and control whether each write publishes to customers. - [SOS Inventory Integration](https://flowrunner.ai/integrations/sos-inventory.md): SOS Inventory is inventory, order and manufacturing management built on QuickBooks Online. Agents manage items, sales orders, purchase orders and builds, read bills of materials, and keep stock, lots and serials in step with the books. - [SoundCloud Integration](https://flowrunner.ai/integrations/soundcloud.md): Connect AI agents to SoundCloud. Agents search the public track catalog, manage uploaded tracks and their visibility, build and populate playlists, and curate the connected account's liked tracks. - [SparkPost Integration](https://flowrunner.ai/integrations/sparkpost.md): Send transactional and template-based email through SparkPost. Agents deliver transmissions with per-recipient substitution data, manage reusable templates, maintain the suppression list to honor opt-outs, and audit sending domains and transmission history. - [Spectra Integration](https://flowrunner.ai/integrations/spectra-fm.md): Drive the Spectra asynchronous creative export pipeline, inspecting design projects, layouts, and templates. Agents render creative variants at volume without a designer per asset. - [Speechmatics Integration](https://flowrunner.ai/integrations/speechmatics.md): Transcribe pre-recorded audio and video with Speechmatics in 60+ languages. Agents submit batch jobs from uploads or URLs, add speaker diarization, translation, summarization, and sentiment, then export transcripts as JSON, text, or SRT subtitles into FlowRunner file storage. - [Splitwise Integration](https://flowrunner.ai/integrations/splitwise.md): Connect AI agents to Splitwise, the shared-expense tracker. Agents log expenses with equal splits, manage groups and friends, read running balances, and clean up mistaken entries as part of reconciliation flows. - [Splunk Integration](https://flowrunner.ai/integrations/splunk.md): Run SPL searches, manage saved searches, ingest events through HEC, and inspect indexes and server health from your flows over the Splunk REST API and HTTP Event Collector. - [Spotify Integration](https://flowrunner.ai/integrations/spotify.md): Connect over OAuth2 to search the Spotify catalog, read track, album, and artist details, manage playlists and the Liked Songs library, and control playback on a user device. - [Wayfront Integration](https://flowrunner.ai/integrations/spp.md): Wayfront, formerly Service Provider Pro (SPP.co), is a client portal and operations platform for productized service agencies: orders, tasks, tickets, invoicing, a CRM pipeline and a client-facing portal. Agents create and update orders, tasks and tickets, manage clients and invoices, and keep the portal current. - [SpreadsheetWeb Hub Integration](https://flowrunner.ai/integrations/spreadsheetweb-hub.md): SpreadsheetWeb Hub by Pagos turns Excel models into web applications and APIs. Agents run calculations against a published workbook, manage applications and their transactions, read and remove stored records, generate documents from print templates, and manage workspace users and share links. - [SproutVideo Integration](https://flowrunner.ai/integrations/sproutvideo.md): SproutVideo is business video hosting with live streaming, viewer logins, playlists, captions, and engagement analytics. Agents upload and organize videos, manage subtitles and calls to action, grant viewer access, run live streams, pull analytics, and react when a video is deployed. - [Microsoft SQL Server Integration](https://flowrunner.ai/integrations/sql-server.md): Connect AI agents to Microsoft SQL Server and Azure SQL Database. Agents run parameterized T-SQL, do CRUD without SQL, inspect schemas, and bulk-load rows as steps in a flow. - [SqlBak Integration](https://flowrunner.ai/integrations/sqlbak.md): SqlBak is a database backup and monitoring platform. Workflows check that scheduled backup and maintenance jobs actually ran, read run logs and upload results, and review servers, connections, and destinations. - [Amazon SQS Integration](https://flowrunner.ai/integrations/sqs-service.md): Send, receive, and delete messages on Amazon SQS standard and FIFO queues. Agents decouple workflow steps, batch-send up to 10 messages at once, long-poll for work, and inspect queue depth. - [Squad Integration](https://flowrunner.ai/integrations/squad.md): Squad by HabariPay is a Nigerian payment gateway. Agents initiate and verify checkout payments, charge cards, bank accounts, and USSD, run direct-debit mandates, issue refunds, manage disputes and virtual accounts, send payouts, and sell airtime and data. - [Square Integration](https://flowrunner.ai/integrations/square.md): Connect AI agents to Square. Agents charge cards and record payments, issue refunds, create and pay orders, publish invoices, sync catalog and multi-location inventory, manage customers and subscriptions, and reconcile payouts. - [Squarespace Integration](https://flowrunner.ai/integrations/squarespace.md): Connect AI agents to Squarespace Commerce. Agents list, create, update, and delete products, read and fulfill orders with shipment tracking, monitor stock levels and adjust quantities across variants, and react to new and fulfilled orders through two polling triggers. - [Stability AI Integration](https://flowrunner.ai/integrations/stability-ai.md): Generate and edit images with Stability AI's Stable Diffusion and Stable Image models. Agents create images from prompts, inpaint and outpaint, replace backgrounds, upscale to print resolution, animate stills into video, and produce 3D assets and audio, with every asset saved to FlowRunner file storage. - [Stackby Integration](https://flowrunner.ai/integrations/stackby.md): Connect AI agents to Stackby, the spreadsheet-database. Agents create, read, update, and delete rows in a Stack table as governed steps in a flow. - [Stack Exchange Integration](https://flowrunner.ai/integrations/stackexchange.md): Read the Stack Exchange network, including Stack Overflow, from your flows. Agents browse and search questions, fetch answers, look up tags and user profiles, and monitor topics to power triage, digests, and internal Q&A assistants. - [Stamped Integration](https://flowrunner.ai/integrations/stamped.md): Connect AI agents to Stamped, the ecommerce reviews and loyalty platform. Agents collect and moderate product reviews, browse the product catalog, and read or adjust customer loyalty points for rewards and make-good workflows. - [Stannp Integration](https://flowrunner.ai/integrations/stannp2.md): Print and post letters and postcards through the Stannp Direct Mail API, with UK address validation built in. Agents trigger real mail off a record and price the campaign before committing to it. - [Statuspage Integration](https://flowrunner.ai/integrations/statuspage.md): Manage Atlassian Statuspage from your flows. Agents open and advance incidents, publish scheduled maintenance windows, flip component statuses as health checks change, and add email, SMS, or webhook subscribers. - [Steady Integration](https://flowrunner.ai/integrations/steady.md): Steady, formerly Status Hero, runs asynchronous team check-ins, goals and activity. Agents read and post check-ins, track goals and pull team activity so standups and status roll-ups happen without a meeting. - [StealthSeminar Integration](https://flowrunner.ai/integrations/stealth-seminar.md): Connect AI agents to StealthSeminar, an evergreen and automated webinar platform. Agents read a webinar's type, timezone, and upcoming session times, then register attendees into a chosen session. There is no action to list or remove a registration. - [StealthGPT Integration](https://flowrunner.ai/integrations/stealthgpt.md): Use the StealthGPT API for AI writing, humanization, and AI-content detection. Agents draft and screen copy in the same workflow that will publish it. - [Steam Integration](https://flowrunner.ai/integrations/steam.md): Read public Steam player and game data through Valve's Steam Web API. Agents resolve vanity URLs to SteamIDs, pull player profiles and friend lists, report owned and recently played games, track achievement progress, and fetch game news. - [STEL Order Integration](https://flowrunner.ai/integrations/stel-order.md): STEL Order is the Spanish ERP and CRM for small businesses. Agents manage clients, prospects, suppliers, and contacts, maintain products, services, rates, and stock, and run the whole sales cycle from quote to invoice. - [Stencil Integration](https://flowrunner.ai/integrations/stencil.md): Stencil is an image and PDF generation API: design a template in its editor, then render it with your own text, images, QR codes, and barcodes. Agents create images and PDFs on demand, run collection batches, search past renders, and open editor sessions. - [Stood CRM Integration](https://flowrunner.ai/integrations/stood-crm.md): Connect AI agents to Stood CRM, a lightweight composable CRM from Hway Digital. Agents create accounts, contacts, and deals, log activities and posts, and organize records into folders and sub-collections. - [Storeman Integration](https://flowrunner.ai/integrations/storeman.md): Storeman from Cloud Sailor is a cloud inventory tool built on storage spaces that contain items, with barcode scanning and a mobile app. Agents read and update spaces and items, log stock movements, and keep the inventory in step with orders and purchases. - [Storyblok Integration](https://flowrunner.ai/integrations/storyblok.md): Connect AI agents to the Storyblok headless CMS. Agents read published or draft stories, datasource entries, links, and tags through the Content Delivery API, and create, update, publish, and delete stories through the Management API. - [Storydoc Integration](https://flowrunner.ai/integrations/storydoc-app.md): Storydoc makes interactive decks, proposals, and one-pagers that live at a URL. Agents create a personalized version of a story template per recipient, generate a first-draft deck from a prompt, and pull engagement analytics back out. - [Strapi Integration](https://flowrunner.ai/integrations/strapi.md): Connect AI agents to a self-hosted or Strapi Cloud instance through the v5 REST API. Agents create, read, update, and delete content-type entries, manage Draft and Publish workflows, and upload files to the Media Library. - [Strava Integration](https://flowrunner.ai/integrations/strava.md): Connect over OAuth to read athlete profiles and stats, create and update activities, explore segments, browse clubs, and look up gear and routes through the Strava API v3. - [Streak Integration](https://flowrunner.ai/integrations/streak.md): Connect AI agents to Streak, the CRM built into Gmail. Agents create and advance boxes through pipeline stages, set custom field values, log comments, read email threads, and search across records to keep deals moving. - [Streamtime Integration](https://flowrunner.ai/integrations/streamtime.md): Streamtime is project management for creative businesses, where jobs break into phases, items, assignments and checklists, and logged time drives quoting and invoicing. Agents open jobs from won deals, push timesheets in bulk, manage companies and contacts, and attach quotes and invoices to outgoing email. - [Stripe Integration](https://flowrunner.ai/integrations/stripe.md): Automate your full revenue cycle. Agents read payments, invoices, and disputes on a schedule, process charges, manage subscriptions, issue refunds, and escalate anomalies to your team before they become problems. - [Stylish Cost Calculator Integration](https://flowrunner.ai/integrations/stylish-cost-calculator.md): Stylish Cost Calculator is a WordPress quote and cost calculator plugin. Agents read the calculators defined on a site, pull the quote submissions they collect, and pick up each new quote in real time. - [SugarCRM Integration](https://flowrunner.ai/integrations/sugarcrm11.md): Connect AI agents to your SugarCRM instance via the v11 REST API. Agents create and update Accounts, Contacts, Leads, Opportunities, Cases, and Tasks, run advanced filtered searches, and traverse related records across any module, including custom ones. - [SuiteCRM Integration](https://flowrunner.ai/integrations/suitecrm7.md): Connect AI agents to SuiteCRM 7 through its V8 JSON:API. Agents create, update, and delete records in any module, link related records through named relationships, and work with Accounts, Contacts, Leads, Opportunities, Cases, and Tasks via typed operations. - [SuiteDash Integration](https://flowrunner.ai/integrations/suitedash.md): SuiteDash is the all-in-one client portal and business software with CRM, projects and marketing built in. Agents create and update contacts and companies, manage projects, subscribe audiences and read the attribute schemas that say what your own account accepts. - [Supabase Integration](https://flowrunner.ai/integrations/supabase.md): Real-time database operations and event triggers for Supabase. Agents react to record creation, updates, and deletion. Full CRUD with PostgREST filter support and dynamic table discovery. - [Supabase Management API Integration](https://flowrunner.ai/integrations/supabase-management.md): The Supabase Management API is the control plane for Supabase projects and organizations. Workflows create, pause, and restore projects, run SQL and migrations, manage Edge Functions, secrets, and auth settings, and read security and performance advisors. - [Supadata Integration](https://flowrunner.ai/integrations/supadata.md): Supadata is an API for video transcripts, social media metadata, AI video content extraction and web scraping across YouTube, TikTok, Instagram, X, Facebook and any public URL. Agents fetch and translate transcripts, list channel and playlist videos, scrape and crawl web pages and extract structured data from video. - [Super Manager Integration](https://flowrunner.ai/integrations/super-manager.md): Connect AI agents to the SuperHote property management platform through Super Manager, checking live availability and pricing for a rental property. Agents quote and confirm bookings against real inventory. - [Superchat Integration](https://flowrunner.ai/integrations/superchat.md): Run WhatsApp, SMS, email, Instagram, Facebook Messenger, Telegram, and live chat conversations from one Superchat workspace. Agents keep the thread in one place regardless of where the customer started it. - [SuperFaktura Integration](https://flowrunner.ai/integrations/superfaktura.md): SuperFaktura is the Slovak and Czech invoicing platform. Agents issue invoices of every kind with their items, record payments, send documents by email or post, generate PDFs, and manage clients and contact people. - [SuperHote Integration](https://flowrunner.ai/integrations/superhote.md): Connect AI agents to SuperHote, a French short-term and vacation rental management platform. Agents read availability across a host's properties and create reservations and bookings, with no update or cancel exposed, so each write is final. - [SuperOffice Integration](https://flowrunner.ai/integrations/superoffice.md): Connect AI agents to SuperOffice CRM. Agents manage companies, people, sales opportunities, projects, and appointments, search records with OData-style filters, and look up saved selections and users to drive downstream automations. - [SuperSaaS Integration](https://flowrunner.ai/integrations/supersaas.md): Connect AI agents to SuperSaaS, an online appointment scheduling service. Agents create and update appointments, check availability, maintain the user database, and react to new bookings so schedules stay current without manual upkeep. - [SupportBee Integration](https://flowrunner.ai/integrations/supportbee.md): Connect AI agents to SupportBee, an email-based help desk. Agents open tickets, reply to customers, post internal comments, and apply labels to route work, since the connector exposes no direct agent assignment. - [SureCart Integration](https://flowrunner.ai/integrations/surecart.md): Read your SureCart store from your flows. Agents list products, prices, orders, and subscriptions, look up order and customer details, monitor subscription status for retention workflows, and register new customers. - [Surense Integration](https://flowrunner.ai/integrations/surense.md): Connect AI agents to Surense, the Israeli pension and insurance licensee compliance registry. Agents create and update licensee records, retrieve their license documents, and validate that an agent is authorized before a flow proceeds. - [ArcGIS Survey123 Integration](https://flowrunner.ai/integrations/survey123.md): ArcGIS Survey123 is Esri's form centric field data collection product backed by hosted feature services. Agents find surveys, read, write, aggregate and attach to responses, generate feature reports, manage webhooks, and react to every new or edited submission. - [Surveybot Integration](https://flowrunner.ai/integrations/surveybot.md): Surveybot runs surveys as chatbot conversations inside Facebook Messenger and Workplace. Agents list surveys, pull responses with the full respondent profile, flatten answers to one row per question, resolve respondents, and react when a response arrives or a survey completes. - [SurveyMonkey Integration](https://flowrunner.ai/integrations/surveymonkey.md): Manage SurveyMonkey surveys, responses, collectors, pages, and questions, translate answer IDs into readable text, and export responses into your other systems for reporting. - [SurveySparrow Integration](https://flowrunner.ai/integrations/surveysparrow.md): Connect AI agents to SurveySparrow, the conversational survey platform. Agents list surveys, sync contacts, pull responses for analysis, and send survey invitations by email after purchases, signups, or support interactions. - [SurveyTale Integration](https://flowrunner.ai/integrations/surveytale.md): SurveyTale is an interactive, story-driven survey platform that runs surveys as hosted links, in-app widgets and in email. Agents read responses, manage contacts and attributes, build personalized survey links, manage webhooks and organization resources, and react to new responses. - [Survicate Integration](https://flowrunner.ai/integrations/survicate.md): Pull customer feedback out of Survicate through its Data Export API. Agents list surveys and questions, export NPS and CSAT responses for incremental syncs, look up respondents, and enrich them with custom attributes for segmentation. - [Swapcard Integration](https://flowrunner.ai/integrations/swapcard.md): Connect AI agents to Swapcard, the event engagement platform. Agents retrieve events and their details, list and look up attendees for CRM syncs, and run arbitrary GraphQL queries against the Event Admin API for anything beyond the dedicated actions. - [SwiftKanban Integration](https://flowrunner.ai/integrations/swiftkanban.md): SwiftKanban by Digité is the enterprise Kanban and Scrum board. Agents create, move and update cards, manage boards, lanes, sprints and releases, post comments and attachments, and act on behalf of other users where the activity log should show them. - [Swift Missive Integration](https://flowrunner.ai/integrations/swiftmissive.md): Swift Missive is an email marketing platform for creators and small businesses. Agents manage contacts and segments, maintain templates and sender personas, send email blasts, run campaigns and workflows, record events, and pull reporting. - [Swipe One Integration](https://flowrunner.ai/integrations/swipe-one.md): Connect AI agents to Swipe One, an AI CRM and email marketing platform for agencies and small businesses. Agents create contacts and set properties, apply tags and segments, assign tasks, and track events against each record. - [Swoogo Integration](https://flowrunner.ai/integrations/swoogo.md): Connect AI agents to Swoogo event management. Agents browse events, create and look up registrants, pull registrant lists for reports and follow-ups, and inspect session agendas for scheduling workflows. - [Swordfish AI Integration](https://flowrunner.ai/integrations/swordfish.md): Find work and mobile phone numbers, emails, and social profiles for a person with the Swordfish AI Search Person API. Agents complete a contact record before outreach spends a touch on a guess. - [SyncroMSP Integration](https://flowrunner.ai/integrations/syncromsp.md): Connect AI agents to SyncroMSP, the all-in-one PSA, RMM, and remote-access platform for managed service providers. Agents manage tickets, customers, contacts, assets, invoices, and RMM alerts from a single flow. - [Synthesia Integration](https://flowrunner.ai/integrations/synthesia.md): Generate AI avatar videos with Synthesia. Agents create videos from scripts or reusable templates, track rendering through webhooks, upload media assets, and save finished MP4s into FlowRunner file storage for durable URLs. - [Synthflow AI Integration](https://flowrunner.ai/integrations/synthflow-ai.md): Run AI voice phone agents with Synthflow AI. Agents create and configure assistants, attach phone numbers, place outbound calls, and pull back call status, transcripts, recordings, and collected variables. - [System Obsługi Najmu Integration](https://flowrunner.ai/integrations/system-obslugi-najmu.md): Connect AI agents to System Obslugi Najmu, a Polish rental property management platform. Agents create reservations, mark them paid as an irreversible income operation, and look up a tenant's bank account details. - [Systeme.io Integration](https://flowrunner.ai/integrations/systeme-io.md): Connect AI agents to Systeme.io, the all-in-one marketing platform. Agents sync contacts, assign and remove tags to trigger automations, enrich records through custom contact fields, and manage the contact list end to end. - [T2M URL Shortener Integration](https://flowrunner.ai/integrations/t2m-url-shortener.md): Create and manage branded short links with the T2M URL Shortener API, using custom slashtags and redirects on your own domain. Agents mint trackable links per campaign. - [TAAPI.IO Integration](https://flowrunner.ai/integrations/taapi-io.md): TAAPI.IO is a technical analysis API with 262 indicators across crypto and stock exchanges. Agents fetch RSI, MACD, moving averages, and any other indicator for a symbol, run bulk calculations across markets, compute indicators from your own candles, and pull raw OHLCV data. - [Tabidoo Integration](https://flowrunner.ai/integrations/tabidoo.md): Tabidoo is a Czech low-code database platform organized as applications containing tables containing records. Agents list applications and tables, search, count and summarize records, create, update, replace and delete records in bulk and check the current user. - [Tableau Integration](https://flowrunner.ai/integrations/tableau.md): Connect AI agents to Tableau dashboards and data. Agents render views as PNG images for scheduled reports, export summary data as CSV, download workbooks for backup, trigger extract refreshes, and audit users, groups, projects, and data sources across a site. - [Tadabase Integration](https://flowrunner.ai/integrations/tadabase.md): Tadabase is a no-code database and application builder with data tables, pages built from components, scheduled tasks, exports and PDF templates. Agents read and write records, attach files, recalculate equations, run scheduled tasks, submit form components and generate exports and PDFs. - [Taggun Integration](https://flowrunner.ai/integrations/taggun.md): Read receipts and invoices with Taggun OCR and get back structured totals, tax, dates, merchant details, and line items, each with a confidence score. Agents post the confident ones and route the rest to a person. - [Taiga Integration](https://flowrunner.ai/integrations/taiga.md): Manage Taiga user stories, tasks, issues, epics, and sprints from your flows, on hosted or self-hosted instances. Agents file and classify inbound work and route high-severity or ambiguous reports to a lead before they land. - [TalentHR Integration](https://flowrunner.ai/integrations/talenthr.md): Connect AI agents to TalentHR, the lightweight all-in-one HRIS by Epignosis. Agents hire, update, and terminate employees, process time off requests, and manage departments and locations so people data stays consistent from onboarding to offboarding. - [TalentLMS Integration](https://flowrunner.ai/integrations/talentlms.md): Connect AI agents to your TalentLMS training portal. Agents create users, enroll them in onboarding courses, generate one-time auto-login URLs, track completion and progress, and organize learners across branches, groups, and categories. - [Tally Integration](https://flowrunner.ai/integrations/tally.md): Connect AI agents to Tally forms. Agents create and update forms, list questions and submissions, manage webhooks, and react the moment a new submission arrives, routing respondent answers into the rest of your stack. - [Tallyfy Integration](https://flowrunner.ai/integrations/tallyfy.md): Tallyfy is the workflow platform where a template is authored once and launched as a process that hands tasks to people in turn. Agents launch processes, complete and reassign tasks, fill form fields, manage members and guests and read the audit trail. - [Tapfiliate Integration](https://flowrunner.ai/integrations/tapfiliate.md): Connect AI agents to Tapfiliate. Agents enroll and approve affiliates, add them to programs, record conversions and commissions, manage approval status, and attribute customers for referral tracking. - [TapHome Integration](https://flowrunner.ai/integrations/taphome.md): Connect AI agents to TapHome, a smart home automation system. Agents read device values and send control commands so a flow can act on and change the state of a building. - [Tars Integration](https://flowrunner.ai/integrations/tars.md): Send approved WhatsApp template notifications through a Tars webhook campaign. Agents deliver templated customer messages on the channel most likely to be read. - [Tarvent Integration](https://flowrunner.ai/integrations/tarvent.md): Tarvent is a marketing automation platform built around audiences and journeys. Agents manage contacts, groups and segments, run campaigns and journeys, send transactional email, work with forms, surveys and landing pages, and export data through the GraphQL API. - [Tavily Integration](https://flowrunner.ai/integrations/tavily.md): Give AI agents live web access through Tavily. Agents run ranked web searches, extract clean markdown from batches of URLs, crawl and map entire sites, launch deep-research tasks that return cited reports, and monitor API credit usage. - [tawk.to Integration](https://flowrunner.ai/integrations/tawkto.md): Connect AI agents to tawk.to live chat. Agents create support tickets from inbound emails and form submissions, pull chat history for reporting, and list agents and properties to drive assignment and routing logic. - [Teachable Integration](https://flowrunner.ai/integrations/teachable.md): Connect AI agents to your Teachable school. Agents look up courses and users, grant or revoke enrollments as customers sign up or churn elsewhere, and report on per-student course progress and completion. - [Team Password Manager Integration](https://flowrunner.ai/integrations/team-password-manager.md): Connect AI agents to a self-hosted Team Password Manager instance. Agents create and update stored passwords, generate replacements, archive or lock entries, organize them into projects, and manage the users and groups that can reach them. - [Teamgate Integration](https://flowrunner.ai/integrations/teamgate-crm.md): Connect AI agents to Teamgate, a sales CRM for small and medium businesses. Agents create leads and move them into people and companies, open deals, log activities, and read the product catalog a deal quotes from. - [Teamleader Integration](https://flowrunner.ai/integrations/teamleader.md): Connect AI agents to Teamleader Focus. Agents create and update contacts and companies, move deals through pipeline phases, manage tasks and projects, and pull invoices with temporary PDF or UBL download links for accounting workflows. - [Teamup Calendar Integration](https://flowrunner.ai/integrations/teamup.md): Connect AI agents to Teamup, a shared group calendar for teams. Agents create, update, and delete events, search the schedule, and list sub-calendars so group calendars stay accurate without manual editing. - [TeamViewer Integration](https://flowrunner.ai/integrations/teamviewer.md): Manage TeamViewer remote access from AI agents. Agents keep the Computers & Contacts list organized, provision and offboard company users, and open remote-control service cases with shareable supporter and customer links straight from a helpdesk ticket. - [Teamwork Integration](https://flowrunner.ai/integrations/teamwork.md): Connect AI agents to Teamwork.com. Agents spin up projects and task lists when a client is onboarded, turn tickets and form submissions into assigned tasks, log billable time entries, and keep tasks updated as work progresses. - [Teamwork Desk Integration](https://flowrunner.ai/integrations/teamwork-desk.md): Connect AI agents to Teamwork Desk, the shared inbox and ticketing product from Teamwork. Agents open and update tickets, keep customer records current, and read the desk configuration behind routing. - [Push by Techulus Integration](https://flowrunner.ai/integrations/techulus-push.md): Send instant push notifications to phones, tablets, and desktops from any flow with Push by Techulus. Agents deliver a title, message, and custom link straight to the device. - [Telegram Integration](https://flowrunner.ai/integrations/telegram.md): Use Telegram bots to deliver alerts, documents, and media as part of operational workflows. 1 message trigger and 15 actions including text, audio, documents, photos, and location messages. - [telli Integration](https://flowrunner.ai/integrations/telli.md): Place outbound phone calls and run an auto dialer with telli AI voice agents, managing contacts and their typed custom properties. Agents call from live pipeline data and log the outcome. - [Telnyx Integration](https://flowrunner.ai/integrations/telnyx.md): Send and track SMS and MMS through Telnyx. Agents deliver transactional messages, check delivery status and cost per recipient, validate numbers and detect line type before sending, and search available numbers by country or area code. - [TelTel Integration](https://flowrunner.ai/integrations/teltel-sms.md): Connect AI agents to TelTel, a Latvian messaging and call center platform. Agents send single, bulk, and two way SMS, run SMS and autodialer campaigns, manage contacts and phone numbers, and create click to call requests so outbound campaigns and customer follow up run automatically. - [Temi Integration](https://flowrunner.ai/integrations/temi.md): Submit audio or video to Temi by Rev for machine transcription and get back plain text or structured JSON with speaker labels and word timings. Agents make recorded conversation searchable and quotable. - [Templated Integration](https://flowrunner.ai/integrations/templated-io.md): Generate images, PDFs, videos, and HTML files from Templated designs by replacing text, image, color, and shape layers. Agents produce on-brand assets per record without a design request. - [TemplateDocs Integration](https://flowrunner.ai/integrations/templatedocs.md): TemplateDocs generates documents from Microsoft Word templates. Agents fill a DOCX template with data, receive a Word file or PDF, email it to the recipient in the same step, and version templates by uploading new files. - [Sign In Scheduling (10to8) Integration](https://flowrunner.ai/integrations/ten-to-eight.md): Connect AI agents to Sign In Scheduling, the online appointment booking platform formerly known as 10to8. Agents look up services, staff, and locations, find open slots, and book appointments for customers so scheduling requests get handled without staff intervention. - [Termene Integration](https://flowrunner.ai/integrations/termene-ro.md): Termene.ro is a Romanian company-data provider covering identity, financials, shareholders, insolvency, public procurement, court cases and tax-authority records. Agents look up a company by CUI and pull whichever data family a schema key selects. - [Termii Integration](https://flowrunner.ai/integrations/termii.md): Connect AI agents to Termii, a Nigerian messaging and verification platform. Agents send messages over SMS, WhatsApp, voice, and email, generate and verify one time passwords, manage phonebooks and campaigns, and check delivery reports so customer messaging and verification run at scale. - [HCP Terraform Integration](https://flowrunner.ai/integrations/terraform-cloud.md): Drive HCP Terraform from AI agents. Agents create and configure workspaces, queue plan and apply runs, gate downstream steps on run status, manage Terraform and environment variables including secrets, and read the current state version. - [Tess AI Integration](https://flowrunner.ai/integrations/tess-ai.md): Run agents and generate completions and embeddings on Tess AI by Pareto, the Brazilian multi-model platform. Agents work across model providers behind one connection. - [TestGorilla Integration](https://flowrunner.ai/integrations/test-gorilla.md): Connect AI agents to TestGorilla skills assessments. Agents invite applicants to role-specific tests the moment they enter your ATS, pull scored results for side-by-side comparison, and browse the test library to assemble assessments for open roles. - [Testlify Integration](https://flowrunner.ai/integrations/testlify.md): Connect AI agents to Testlify, a skills assessment platform for hiring. Agents create assessments, invite candidates, and pull results so screening runs automatically and shortlists build themselves. - [Testomato Integration](https://flowrunner.ai/integrations/testomato.md): Testomato monitors websites with checks that assert things about your pages. Workflows create projects, run checks on demand, read results, uptime, and response-time history, and manage notification settings. - [Textbelt Integration](https://flowrunner.ai/integrations/textbelt.md): Connect AI agents to Textbelt, a deliberately simple SMS API for developers. Agents send SMS, run one time password verification flows, and check delivery status and remaining quota so lightweight text notifications work without any account setup. - [TextCortex Integration](https://flowrunner.ai/integrations/textcortex-ai.md): Draft, rewrite, and summarize with the TextCortex API, an OpenAI-compatible text generation service. Agents produce copy inside the workflow that already holds the customer context. - [Texter Integration](https://flowrunner.ai/integrations/texter.md): Connect AI agents to Texter, a unified business messaging platform for WhatsApp, Messenger, and Instagram. Agents create and send WhatsApp templates, manage and assign chats, apply labels, and react to incoming messages and status changes so customer conversations get routed and resolved fast. - [TextIt Integration](https://flowrunner.ai/integrations/textit.md): Connect AI agents to TextIt, a flow based messaging and chatbot platform. Agents send messages and broadcasts, start flows, manage contacts and groups, schedule campaigns, and react to resthook events so conversational outreach runs across channels without manual dispatch. - [Textline Integration](https://flowrunner.ai/integrations/textline.md): Connect AI agents to Textline, a shared business text messaging inbox. Agents message and resolve conversations, schedule messages and announcements, send NPS and CSAT surveys, and manage contacts and agent availability so customer texting stays fast and accountable. - [TextMagic Integration](https://flowrunner.ai/integrations/textmagic.md): Send business SMS through TextMagic. Agents text individuals, saved contacts, or entire lists, monitor inbound replies and opt-outs for follow-up, manage the address book, and check account balance before a campaign goes out. - [ThaiBulkMail Integration](https://flowrunner.ai/integrations/thaibulk-mail.md): ThaiBulkMail is the email product of Thai provider ThaiBulkSMS. Agents send templated email, deliver one-time passwords by email, manage attachments, and check remaining credit. - [ThaibulkSMS Integration](https://flowrunner.ai/integrations/thaibulksms.md): Connect AI agents to ThaibulkSMS, a Thai bulk messaging gateway with SMS, OTP, and transactional email. Agents send bulk SMS with link tracking, request and verify one time passwords, send templated emails, and check credit balances so Thai customer messaging runs across channels. - [Thanks.io Integration](https://flowrunner.ai/integrations/thanks-io.md): Send handwritten-style postcards, notecards, letters, and gift cards by post with Thanks.io and track each piece through delivery. Agents trigger physical mail off a milestone and see when it lands. - [Thankster Integration](https://flowrunner.ai/integrations/thankster.md): Create and post handwritten-style greeting cards with Thankster, in one call or step by step with a proof stage. Agents draft the card and hold it for a person to approve before it is mailed. - [The Bot Platform Integration](https://flowrunner.ai/integrations/the-bot-platform.md): Message employees, manage user attributes, and pull bot analytics with The Bot Platform, which builds chatbots for Microsoft Teams and Workplace. Agents reach staff in the tool they already have open. - [GoodAPI Integration](https://flowrunner.ai/integrations/the-good-api.md): Add verified environmental and social impact to a workflow with the GoodAPI Standalone API, planting trees and recording other verified actions. Agents tie a real contribution to a customer event. - [The Keys Integration](https://flowrunner.ai/integrations/the-keys.md): Connect AI agents to The Keys, a French smart lock system. Agents lock and unlock doors, create an accessory share, and read lock status, existing shares, and access logs. The connector exposes no share expiry and no revoke. - [theMarketer Integration](https://flowrunner.ai/integrations/the-marketer.md): Connect AI agents to theMarketer, the Romanian email, SMS, and loyalty marketing platform, through its official API. Agents run lifecycle messaging and keep loyalty balances driven by real purchases. - [TheHive Integration](https://flowrunner.ai/integrations/thehive.md): Connect AI agents to TheHive, the open-source Security Incident Response Platform. Agents create and manage cases, tasks, observables, and alerts, promote or merge alerts into cases, and query the incident record. - [Thiio Integration](https://flowrunner.ai/integrations/thiio.md): Thiio is a MarTech platform for direct response commerce with leads, orders, subscriptions and fulfillment. Agents create and enrich leads, read and fulfill orders, manage subscriptions, and react to order and subscription events through a realtime trigger. - [Thinkific Integration](https://flowrunner.ai/integrations/thinkific.md): Connect AI agents to your Thinkific school. Agents create student accounts and enroll them in the right course after a purchase, set activation and expiry dates for cohort access, and sync enrollment progress into reporting and follow-up tools. - [ThriveCart Integration](https://flowrunner.ai/integrations/thrivecart.md): Connect AI agents to ThriveCart. Agents verify API connectivity, look up products and pricing, retrieve customers with purchase summaries to enrich CRM records, and review recent transactions for reconciliation and follow-up messaging. - [Ticket Tailor Integration](https://flowrunner.ai/integrations/ticket-tailor.md): Read Ticket Tailor box office data with AI agents. Agents sync orders and attendee lists into a CRM or warehouse, look up individual orders and issued tickets for support and check-in, and report on sales across events and recurring series. - [TicketPAY Integration](https://flowrunner.ai/integrations/ticketpay.md): Connect AI agents to TicketPAY, a German event ticketing and event management platform. Agents read events, catalog items, sales, finance, campaign metrics, and accreditation records through an export API that creates and modifies nothing. - [TickTick Integration](https://flowrunner.ai/integrations/ticktick.md): Connect AI agents to TickTick. Agents create tasks from incoming events with due dates and priorities, complete and reschedule work in bulk, provision projects on demand, and pull a full project snapshot with open tasks and Kanban columns in one call. - [Tidely Integration](https://flowrunner.ai/integrations/tidely.md): Tidely is a German liquidity-planning and cash-flow platform with a Public API and a Partner API. Agents read bank accounts, balances, and transactions, manage invoices, plans, and categories, read scenarios, pull the cash-flow forecast, and onboard customer accounts. - [TidyCal Integration](https://flowrunner.ai/integrations/tidycal.md): Connect AI agents to TidyCal, the budget-friendly scheduling tool from AppSumo. Agents list available time slots, create and cancel bookings, and manage contacts so appointment scheduling runs without manual back-and-forth. - [Tiflux Integration](https://flowrunner.ai/integrations/tiflux.md): Connect AI agents to Tiflux, an IT service management and helpdesk platform. Agents open and resolve tickets end to end, communicate with the requester, and route between desks, stages, and agents. - [TikTok Integration](https://flowrunner.ai/integrations/tiktok.md): Publish to TikTok from AI agents. Agents post videos and photo carousels from a URL or stored file, poll until a post goes live, and pull the connected creator's profile stats and per-video engagement counts for reporting. - [TikTok Audiences Integration](https://flowrunner.ai/integrations/tiktok-audiences.md): Manage TikTok Ads custom audiences with AI agents. Agents create audiences, sync CRM segments and subscribers in by email, remove churned or unsubscribed users to keep targeting compliant, and audit audience size and status. Emails are SHA-256 hashed before they leave FlowRunner. - [TikTok Conversions Integration](https://flowrunner.ai/integrations/tiktok-conversions.md): Send server-side conversion events to TikTok from AI agents. Agents report purchases with order value and line items, track backend-only events like registrations and subscriptions, and deduplicate against browser Pixel events with a shared event ID. Customer identifiers are SHA-256 hashed automatically. - [TikTok Lead Forms Integration](https://flowrunner.ai/integrations/tiktok-lead-forms.md): Pull TikTok lead-generation form submissions with AI agents. Agents discover the lead-gen forms in an advertiser account, inspect each form's question schema before mapping fields, and retrieve leads by submission time window for fast routing to sales. - [TikTok Reports Integration](https://flowrunner.ai/integrations/tiktok-reports.md): Pull TikTok ad performance data with AI agents. Agents run integrated reports grouped by campaign, ad group, ad, or time, fetch spend, impressions, clicks, and conversions synchronously, or start async report tasks for large date ranges and poll for the finished file. - [Time Doctor Integration](https://flowrunner.ai/integrations/time-doctor.md): Connect AI agents to Time Doctor. Agents pull worklogs and application or website time-use reports over a date range, sync tracked time into billing and payroll systems, and create projects and discover user and task IDs for downstream automations. - [timeBuzzer Integration](https://flowrunner.ai/integrations/timebuzzer.md): timeBuzzer is the time tracking tool built around a physical desk buzzer. Agents create, filter, and bulk edit activities, manage the layer and tile hierarchy that categorizes them, assign tiles to users and groups, pull reports and charts, and react to webhook events. - [TimeCamp Integration](https://flowrunner.ai/integrations/timecamp.md): Connect AI agents to TimeCamp. Agents read and write time entries across a date range, build timesheets filtered by user, keep the project and task hierarchy in sync, and start or stop the live timer from event-driven flows. - [TimelinesAI Integration](https://flowrunner.ai/integrations/timelinesai-for-whatsapp.md): Automate WhatsApp for a whole team through a TimelinesAI workspace, sending messages, voice notes, and attachments and organising chats. Agents keep shared WhatsApp coverage from depending on one person's phone. - [Timely Integration](https://flowrunner.ai/integrations/timelyapp.md): Timely is the automatic time tracking platform. Agents create and import time entries, control the running timer, manage projects, clients, tags, and planned tasks, approve and lock time for billing, pull reports, and react to time entry events in real time. - [TimescaleDB Integration](https://flowrunner.ai/integrations/timescaledb.md): Connect AI agents to TimescaleDB, the time-series database built on PostgreSQL. Agents run SQL, do CRUD without SQL, manage hypertables, run time_bucket rollups, and enforce chunk retention. - [TimeTonic Integration](https://flowrunner.ai/integrations/timetonic.md): TimeTonic is a French no-code database and collaboration workspace built around books containing smart tables. Agents list books and tables, read table values, create and update rows and cells, roll back changes, manage cell comments, send book messages and attach files. - [EARLY (Timeular) Integration](https://flowrunner.ai/integrations/timeular.md): EARLY, formerly Timeular, is the time tracking platform paired with a physical tracker. Agents start and stop live tracking, create time entries with tags and mentions, organize activities and folders, pull reports, manage team folder access, and handle leave and its approval. - [Timing Integration](https://flowrunner.ai/integrations/timing.md): Timing is the automatic time tracker for Mac. Agents create and update projects and time entries, control the running timer, pull reports, manage teams, and read the activity hierarchy Timing's automatic tracking produces. - [Tinify (TinyPNG) Integration](https://flowrunner.ai/integrations/tinify.md): Compress PNG, JPEG, and WebP images from a public URL, then resize or convert the result to modern formats like WebP and AVIF through the Tinify (TinyPNG) API. Every call reports monthly compression usage so flows stay inside plan limits. - [Tisane Integration](https://flowrunner.ai/integrations/tisane.md): Moderate content and run deep text analysis with the Tisane API, detecting abusive content with severity ratings and extracting sentiment. Agents screen user content and escalate borderline cases to a human moderator. - [TMetric Integration](https://flowrunner.ai/integrations/tmetric.md): TMetric is the work time tracking platform for teams. Agents control the running timer and time entries, manage tasks, projects, and clients, submit and approve timesheets, handle time off, pull reports, and create invoices and expenses. - [Tny Integration](https://flowrunner.ai/integrations/tny.md): Create and manage short links with Tny by Chief Tools on shared or custom domains, reading human and crawler click counts separately. Agents measure real engagement rather than bot traffic. - [Todoist Integration](https://flowrunner.ai/integrations/todoist.md): Manage Todoist tasks, projects, sections, labels, and comments from your flows, with natural-language due dates. Agents capture actionable messages as tasks and confirm before assigning work to a teammate. - [Together AI Integration](https://flowrunner.ai/integrations/together-ai.md): Run open models on Together AI from your flows: chat and text completions across Llama, DeepSeek, Qwen and hundreds more, plus embeddings, document reranking, FLUX image generation, speech synthesis, and audio transcription and translation through one OpenAI-compatible API. - [Toggl Track Integration](https://flowrunner.ai/integrations/toggl.md): Track time in Toggl Track from your flows: start and stop timers, log entries, and keep projects, clients, and tags in sync. Agents track routine work and gate bulk corrections that move billing behind a person. - [Toggl Plan Integration](https://flowrunner.ai/integrations/toggl-plan.md): Toggl Plan is visual team planning on a timeline, with a backlog for work that has no dates yet. Agents create and schedule tasks, manage projects, milestones and members, and move backlog items onto the timeline when they are committed. - [Toky Integration](https://flowrunner.ai/integrations/toky.md): Manage agents, calls, SMS, and numbers on the Toky cloud phone and call center system. Agents keep call activity attached to the record it belongs to. - [Tomba Integration](https://flowrunner.ai/integrations/tomba.md): Find and verify emails and enrich B2B records with the Tomba API, covering domain search, person and LinkedIn finders, and deliverability checks. Agents build a verified list before the first send. - [Toodledo Integration](https://flowrunner.ai/integrations/toodledo.md): Toodledo is the long-running task, note, outline, and list manager. Agents add and edit tasks with folders, contexts, goals, and locations, write notebook entries, maintain outlines and custom lists, and use account timestamps for incremental sync. - [Tookan Integration](https://flowrunner.ai/integrations/tookan.md): Connect AI agents to Tookan, the Jungleworks delivery management platform. Agents create and edit delivery tasks, find available agents, assign work to fleets, track task status, and pull customer records to automate last-mile dispatch end to end. - [TopMessage Integration](https://flowrunner.ai/integrations/topmessage.md): Connect AI agents to TopMessage, a business messaging platform that sends SMS and WhatsApp through one API. Agents send templated and personalized messages, list and retrieve message history, and verify one-time codes so customer notifications and verification flows run from a single route. - [Totalum Integration](https://flowrunner.ai/integrations/totalum.md): Totalum is an AI app builder that turns a natural-language brief into a deployed Next.js application with a managed database, auth, payments and hosting. Agents launch and update projects, drive the build agent, query and edit database records, deploy to production, manage domains and files and react to project events. - [Tour Agency App Integration](https://flowrunner.ai/integrations/tour-agency-app.md): Connect AI agents to Tour Agency App, a CRM and booking management platform for tour operators. The connector exposes a single Verify Connection action confirming credentials and the connected user, with business events delivered as outbound webhooks configured inside the product. - [TrackingTime Integration](https://flowrunner.ai/integrations/trackingtime.md): TrackingTime is the team time tracking and timesheet platform. Agents manage customers, projects, and tasks, control the running timer, create and import time entries in batches, invite users, set custom fields, and handle time off requests, policies, and balances. - [Trados Integration](https://flowrunner.ai/integrations/trados.md): Connect AI agents to RWS Trados Cloud, the platform behind Trados Enterprise and Trados Team. Agents create localization projects, upload source files, run analysis and quotes, assign tasks, and query translation memories and termbases. - [Trafft Integration](https://flowrunner.ai/integrations/trafft.md): Connect AI agents to Trafft, appointment scheduling and booking software. Agents create bookings, manage customers and coupons, and read the service catalog, and react to new appointments so client scheduling runs end to end. - [Trainual Integration](https://flowrunner.ai/integrations/trainual.md): Connect AI agents to Trainual, the employee training and SOP platform. Agents invite new hires, look up training subjects and topics, browse roles and groups, and pull completion records to keep onboarding and compliance reporting current. - [Trakt Integration](https://flowrunner.ai/integrations/trakt-tv.md): Trakt is the TV and movie watch-tracking platform with catalog, calendars, charts, and personal history. Agents search titles, read trending and recommendations, sync collection and watchlist, scrobble and check in, manage lists, and post comments for the connected account. - [Transistor Integration](https://flowrunner.ai/integrations/transistor-fm.md): Transistor.fm is podcast hosting with private podcasts, per-episode analytics, and webhooks. Agents create, update, and publish episodes, upload audio, manage private subscribers, pull show and episode analytics, and react to Transistor events in real time. - [Transloadit Integration](https://flowrunner.ai/integrations/transloadit.md): Kick off Transloadit Assemblies that import media, run it through encoding Robots to resize, transcode, or convert, then poll for the finished files and their URLs. Agents also manage the reusable Templates that hold Assembly Instructions, with every request HMAC-signed. - [Tranzila Integration](https://flowrunner.ai/integrations/tranzila.md): Tranzila is an Israeli payment processor for cards, Bit, and MASAV bank debits. Agents charge, verify, capture, and refund cards, complete 3DS flows, manage standing orders, create hosted-page handshakes and payment requests, and pull transaction reports and invoices. - [Travis CI Integration](https://flowrunner.ai/integrations/travis-ci.md): Manage Travis CI repositories, trigger and control builds and jobs, inspect logs, and configure environment variables from your flows over the Travis CI REST API v3. - [Trello Integration](https://flowrunner.ai/integrations/trello.md): The most comprehensive Trello integration available. 67 actions covering every Trello object: boards, cards, lists, checklists, labels, attachments, stickers, members, and notifications. - [Tremendous Integration](https://flowrunner.ai/integrations/tremendous.md): Send digital rewards through Tremendous: gift cards, prepaid cards, and charitable donations. Agents create and track orders, inspect individual rewards and their redemption status, and reconcile spend across funding sources, products, and campaigns. - [Trengo Integration](https://flowrunner.ai/integrations/trengo.md): Work Trengo's omnichannel inbox from your flows: agents list and triage tickets, assign conversations to the right teammate, reply to customers over email, WhatsApp, SMS, or live chat, and keep contacts in sync across every connected channel. - [TRIGGERcmd Integration](https://flowrunner.ai/integrations/triggercmd.md): Run commands on your own Windows, Mac, Linux, and Raspberry Pi computers from a flow through the TRIGGERcmd agent. Agents reach machines that sit outside any API, with the command set fixed in advance. - [Trint Integration](https://flowrunner.ai/integrations/trint.md): Send audio and video to Trint for transcription, wait for the transcript, and export it as captions, documents, or spreadsheets. Agents batch-ingest media from URLs with webhook callbacks, file transcripts into shared drives and folders, and machine-translate finished transcripts. - [Tripetto Integration](https://flowrunner.ai/integrations/tripetto.md): Tripetto studio is the hosted half of Tripetto's conversational form and survey product. Agents manage workspaces and forms, read definitions, styling, localization and connections, pull submitted responses, and react on each new response. - [Tripletex Integration](https://flowrunner.ai/integrations/tripletex.md): Tripletex is the Norwegian accounting platform. Agents manage customers, suppliers, contacts, and products, create orders, invoices, purchase orders, and supplier invoices, and react to signed webhook events as records change. - [Trustmary Integration](https://flowrunner.ai/integrations/trustmary.md): Collect reviews and testimonials with Trustmary, importing contacts, pushing reviews, and reading surveys and lists. Agents ask for a review at the moment the customer is happiest and route the answers. - [Trustpilot Integration](https://flowrunner.ai/integrations/trustpilot.md): Monitor and act on your Trustpilot reputation: agents resolve a domain to its business unit, pull TrustScores and star ratings, read public and private reviews, post or clear company replies, and send review invitations tied to order IDs. - [QuickBooks Time (TSheets) Integration](https://flowrunner.ai/integrations/tsheets.md): QuickBooks Time, formerly TSheets, is Intuit's time tracking and scheduling platform. Agents create and edit timesheets, clock users in and out, manage jobcodes, projects, and schedules, handle time off, work with geofences and locations, and pull the reports payroll needs. - [Tumblr Integration](https://flowrunner.ai/integrations/tumblr.md): Read public Tumblr blogs from your flows: agents fetch blog metadata, browse published posts, resolve avatars, read publicly shared likes, and discover content by tag across the network. - [TurboDocx Integration](https://flowrunner.ai/integrations/turbodocx-service.md): Automate document generation and e-signature with TurboDocx. FlowRunner agents generate documents from templates, route signing and review, download signed files, and track audit trails. - [29 Next Integration](https://flowrunner.ai/integrations/twentyninenext.md): 29 Next, now trading as Next Commerce, is an ecommerce platform built for direct response and subscription commerce. Agents react to paid orders in real time, run subscription dunning across gateways, manage the catalog, fulfillment and disputes, and drive external checkouts through the Campaigns API. - [Twilio Integration](https://flowrunner.ai/integrations/twilio.md): Add SMS and voice to your agent workflows. Twilio triggers fire on incoming calls and messages. Agents send SMS, make voice calls, and retrieve communication history. - [Twilio Verify Integration](https://flowrunner.ai/integrations/twilio-verify.md): Add phone and email verification to any flow with Twilio Verify. Agents start one-time passcode challenges over SMS, voice call, email, or WhatsApp and check the codes users submit, with Twilio generating, delivering, and validating every passcode. - [Twin Integration](https://flowrunner.ai/integrations/twin-web-agent.md): Create Twin AI web agents that drive a real browser to do work on sites with no API, then start and monitor their runs. Agents reach systems that were never given an interface. - [Twist Integration](https://flowrunner.ai/integrations/twist.md): Doist's threaded team communication app. Manage workspaces, channels, threads, comments, and direct-message conversations following the workspace, channel, thread, comment hierarchy. - [Twitch Integration](https://flowrunner.ai/integrations/twitch.md): Run a Twitch channel from your flows with 28 Helix API actions: update stream title and category, send chat messages, capture clips and stream markers, run polls, manage videos, and pull follower, subscriber, and Bits data. - [2Chat Integration](https://flowrunner.ai/integrations/two-chat.md): Run WhatsApp and SMS conversations through a 2Chat account, sending messages and media and managing contacts and WhatsApp groups. Agents handle routine threads and hand off the ones that need a person. - [Typebot Integration](https://flowrunner.ai/integrations/typebot.md): Connect AI agents to Typebot, the open source conversational form and chatbot builder. Agents list typebots, read submissions and their execution logs, and drive live chat sessions programmatically on Typebot Cloud or self-hosted instances. - [Typecast Integration](https://flowrunner.ai/integrations/typecast.md): Synthesize expressive speech, clone a voice from a sample, and get word-level timings with the Typecast API. Agents produce narration and voice prompts inside the workflow that needs them. - [Typeform Integration](https://flowrunner.ai/integrations/typeform.md): Build, manage, and process Typeform surveys and forms. 41 actions span form creation, response retrieval, theme management, translation, workspace administration, and analytics. - [Revid Integration](https://flowrunner.ai/integrations/typeframes.md): Revid, formerly Typeframes, generates finished videos with AI visuals, voiceover, captions, and music from a script, prompt, article URL, or audio file. Agents estimate credits, create and export videos, manage characters and cloned voices, and publish or schedule to TikTok, YouTube, and Instagram. - [Ubiqod by Skiply Integration](https://flowrunner.ai/integrations/ubiqod-by-skiply.md): Connect AI agents to Ubiqod by Skiply, the provisioning API behind Skiply IoT smart buttons and QR code data collection. Agents create and manage trackers, sites, badge lists, and PIN code lists in bulk, and read the dispatch destinations that carry collected data onward. - [Ublux Integration](https://flowrunner.ai/integrations/ublux-communications.md): Use the Ublux WhatsApp Business API for messaging, which this connector covers rather than the PBX or telephony side. Agents send and receive WhatsApp business messages as a workflow step. - [uClassify Integration](https://flowrunner.ai/integrations/uclassify.md): Score text against thousands of public classifiers with the uClassify REST API, covering sentiment, topic, and more. Agents route and prioritize inbound text without training a model first. - [UiPath Integration](https://flowrunner.ai/integrations/uipath.md): Drive UiPath Orchestrator from your flows: start and stop unattended RPA jobs, feed transactional queues and track item status, and enumerate the processes, robots, assets, and folders in a tenant so FlowRunner can trigger and monitor UiPath automations end to end. - [uTalk by Umbler Integration](https://flowrunner.ai/integrations/umbler-utalk.md): Automate WhatsApp customer service in a uTalk organization, sending messages and approved templates and scheduling follow-ups. Agents work the queue and escalate what needs a person. - [Unbounce Integration](https://flowrunner.ai/integrations/unbounce.md): Pull landing page and lead data out of Unbounce: agents list pages, pop-ups, and sticky bars, retrieve the leads each one captured, and inventory sub-accounts and custom domains for routing, reporting, and CRM sync. - [UnionBank Integration](https://flowrunner.ai/integrations/unionbank.md): UnionBank of the Philippines publishes its banking rails as an open API program. Agents read balances and transaction history, send InstaPay, PESONet, and RTGS transfers, pay bills and merchants, manage EON wallets and cards, and look up billers, branches, and forex rates. - [UniOne Integration](https://flowrunner.ai/integrations/unione.md): UniOne is a transactional email API with data centers in the EU and US. Agents send mail from templates, validate addresses, manage suppressions and unsubscribes, configure domains and webhooks, and pull event dumps for reporting. - [Unipiazza Integration](https://flowrunner.ai/integrations/unipiazza.md): Run an Italian customer loyalty programme with the Unipiazza Partner API, registering customers and recording receipts so points accrue. Agents award credit off the transaction rather than a manual entry. - [Unit Converter Integration](https://flowrunner.ai/integrations/unit-converter.md): Convert values between 142 units of measure across 13 dimensions entirely offline, with no API key or external call. Agents normalize units between systems that each chose a different one. - [Unleashed Software Integration](https://flowrunner.ai/integrations/unleashed.md): Read products, stock on hand, customers, suppliers, warehouses, and sales and purchase orders from the Unleashed cloud inventory platform, and create new sales orders. - [Unsplash Integration](https://flowrunner.ai/integrations/unsplash.md): Search Unsplash's library of free high-resolution photography from your flows: agents find photos and curated collections, fetch random on-topic images for dynamic heroes, pull full photo detail and renditions, and register downloads to stay compliant with attribution guidelines. - [Upgrade.Chat Integration](https://flowrunner.ai/integrations/upgradechat.md): Upgrade.Chat is the payments and membership platform for Discord and Telegram communities. Agents list and inspect orders, products, and users, and audit webhook configurations and their delivery log to keep community access in sync with what was paid. - [UpLead Integration](https://flowrunner.ai/integrations/uplead.md): Access UpLead's B2B contact and company database. Agents enrich people and companies, search for contacts at a target company by job function and seniority, and monitor remaining API credits. - [Uploadcare Integration](https://flowrunner.ai/integrations/uploadcare.md): Ingest files into Uploadcare from public URLs, manage the stored library with filters and pagination, inspect file and group metadata, and build on-the-fly CDN transformation URLs for resizing, cropping, and format conversion without an extra API round trip. - [uProc Integration](https://flowrunner.ai/integrations/uproc.md): Verify, validate, and enrich contact data through uProc's catalog of hundreds of processors across email, phone, company, IP, and geolocation. Agents verify deliverability, infer gender, search company data, and run any processor by name. - [UPS Quantum View Integration](https://flowrunner.ai/integrations/ups-quantum-view.md): Connect AI agents to UPS Quantum View, the UPS shipment visibility API. Agents retrieve manifest, origin, exception, and delivery events so shipment status feeds proactive alerts and reporting without manual tracking lookups. - [Upsales Integration](https://flowrunner.ai/integrations/upsales.md): Connect AI agents to Upsales, a Swedish sales and marketing CRM. Agents manage companies and contacts, move orders and opportunities forward, register campaigns for attribution, log calls, and track NPS responses. - [UptimeRobot Integration](https://flowrunner.ai/integrations/uptimerobot.md): Monitor uptime and manage alerting from your flows over the UptimeRobot API v2. Create HTTP, Keyword, Ping, Port, and Heartbeat monitors, manage alert contacts, maintenance windows, and status pages. - [UpViral Integration](https://flowrunner.ai/integrations/upviral.md): Run viral referral and giveaway campaigns with the UpViral API, adding entrants, crediting referrals, and awarding points. Agents keep rewards tied to verified actions. - [Urban Dictionary Integration](https://flowrunner.ai/integrations/urban-dictionary.md): Look up crowd-sourced slang definitions from Urban Dictionary, by term, by definition ID, or at random. Agents decode the language in user-generated content before acting on it. - [urlscan.io Integration](https://flowrunner.ai/integrations/urlscan.md): Connect AI agents to urlscan.io. Agents detonate suspicious URLs in a real browser, retrieve verdicts and contacted domains, search the historical scan database, and capture page screenshots and DOM snapshots. - [URL to Text Integration](https://flowrunner.ai/integrations/urltotext.md): Turn any web page into clean, model-ready text with the URLtoText API, returning plain text, Markdown, or structured content. Agents ground a model on a live page without parsing HTML themselves. - [Uscreen Integration](https://flowrunner.ai/integrations/uscreen.md): Uscreen is a video membership and monetization platform. Agents invite and update customers, grant or revoke content access, create and cancel subscriptions, manage group plans, control content visibility, read invoices, pull viewing analytics, and react to new or updated customers. - [UseINBOX Integration](https://flowrunner.ai/integrations/useinbox.md): UseINBOX bundles INBOX for email marketing, INBOXNotify for transactional email, and INBOXIys for filing consent records with Turkey's national message registry. Agents manage subscribers and campaigns, send transactional messages, and record consent, all from one account. - [Userback Integration](https://flowrunner.ai/integrations/userback.md): Userback is a visual website feedback and bug reporting tool. Agents list and update feedback, add comments and screenshots, move items through workflows, and review projects, members, and session recordings. - [UserCheck Integration](https://flowrunner.ai/integrations/usercheck.md): Detect disposable and temporary email addresses with the UserCheck API, checking an email or a domain. Agents block throwaway signups before they consume a trial or a credit. - [User.com Integration](https://flowrunner.ai/integrations/usercom.md): Sync contacts into User.com and keep them current: agents create or update users matched on email, update individual profile attributes, look up contacts by any field, and send custom events that power User.com automations and segmentation. - [Userlist Integration](https://flowrunner.ai/integrations/userlist.md): Userlist is lifecycle email and in-app messaging for SaaS companies. Agents push users and companies, link them together, record behavioral events that drive campaigns, and send transactional messages through the Push API. - [Uspacy Integration](https://flowrunner.ai/integrations/uspacy.md): Connect AI agents to Uspacy, a Ukrainian CRM and collaboration workspace. Agents create and update CRM entities, assign tasks, and read the company structure behind each record. - [Vaimero Integration](https://flowrunner.ai/integrations/vaimero.md): Connect AI agents to Vaimero, an AI meeting booker that contacts warm leads over email, SMS, voice, or webhook until a meeting is set. Agents pick up the booked meeting and prepare whoever is taking it. - [ValidEmail Integration](https://flowrunner.ai/integrations/validemail-co.md): Validate email addresses in real time with ValidEmail.co, returning a deliverability verdict per address. Agents screen an address at capture rather than discovering the problem on send day. - [Valuecase Integration](https://flowrunner.ai/integrations/valuecase.md): Connect AI agents to Valuecase, a digital sales room platform for B2B deals. Agents create buyer-facing spaces, set custom properties per deal, and read or write the forms a buyer completed inside the room. - [Vapi Integration](https://flowrunner.ai/integrations/vapi.md): Run voice AI operations on Vapi with 47 actions: agents place and analyze phone calls, build assistants and multi-assistant squads, provision phone numbers, manage tools and knowledge files, drive outbound calling campaigns, and pull call cost and duration analytics. - [Vayne Integration](https://flowrunner.ai/integrations/vayne.md): Scrape LinkedIn Sales Navigator data with the Vayne API, extracting leads and accounts and scraping profiles and job listings. Agents build target lists from live hiring and headcount signals. - [VBOUT Integration](https://flowrunner.ai/integrations/vbout.md): Sync ecommerce activity and read marketing data with the VBOUT API, covering carts, orders, and transactions. Agents trigger recovery and lifecycle campaigns off what shoppers actually did. - [VdoCipher Integration](https://flowrunner.ai/integrations/vdocipher.md): VdoCipher is DRM-protected video hosting with encrypted streaming, dynamic watermarking, and per-viewer playback authorization. Agents import and upload videos, generate playback OTPs, organize folders and tags, add posters and captions, track bandwidth, and react to VdoCipher events. - [Lightspeed Retail X-Series Integration](https://flowrunner.ai/integrations/vend.md): Lightspeed Retail X-Series, formerly Vend, is a cloud point of sale for retailers. Agents sync an ecommerce catalog into the store, push outside orders in as sales, manage inventory, consignments and purchase orders across outlets, work customers, loyalty and gift cards, and react to webhooks. - [Vercel AI Gateway Integration](https://flowrunner.ai/integrations/vercel-ai-gateway.md): Reach hundreds of models across OpenAI, Anthropic, Google, Meta, and more through the Vercel AI Gateway's single OpenAI-compatible endpoint. Agents run chat completions with automatic provider failover, route by cost or latency, generate embeddings, and track exact per-request spend. - [Verifi.Email Integration](https://flowrunner.ai/integrations/verifi-email.md): Validate single addresses or whole lists in real time with Verifi.Email, returning a deliverability verdict with supporting detail. Agents clean a list before a send rather than after a bounce. - [Verificare TVA Integration](https://flowrunner.ai/integrations/verificare-tva.md): Check the VAT position of Romanian companies against official sources by tax identification number and reference date. Agents confirm a counterparty's VAT status before invoicing them. - [Verify by Tailr Made Integration](https://flowrunner.ai/integrations/verify.md): Connect AI agents to Verify by Tailr Made, an AI resume verification and fraud-detection tool for recruiters. Agents submit resume text for analysis and receive a risk level, summary, and structured red-flag report so screening happens before an interview is ever scheduled. - [Veriphone Integration](https://flowrunner.ai/integrations/veriphone.md): Validate, format, and look up carriers for phone numbers worldwide with the Veriphone API, checking against national numbering plans. Agents confirm a number is reachable before a message is sent. - [Vero Integration](https://flowrunner.ai/integrations/vero.md): Connect AI agents to Vero. Agents identify users, update profiles and tags, track behavioral events, manage subscription state, and reidentify anonymous users after signup to drive behavioral email. - [Vidalytics Integration](https://flowrunner.ai/integrations/vidalytics.md): Vidalytics is video hosting and analytics built for conversion, with a customizable player, timed calls to action, and per-second retention stats. Agents upload and publish videos, manage thumbnails, pause screens, and CTAs, organize folders, and pull drop-off and timeline statistics. - [VideoAsk Integration](https://flowrunner.ai/integrations/videoask.md): Pull VideoAsk data into your flows: agents list interactive video forms and their steps, read respondent answers complete with video, audio, and transcription media, build contact lists from respondents, and enumerate organizations for reporting. - [VIES API Integration](https://flowrunner.ai/integrations/vies-api.md): Validate EU VAT identification numbers against the European Commission VIES registry through viesapi.eu. Agents confirm a VAT number is genuinely registered before applying a zero-rated invoice. - [ViewDNS Integration](https://flowrunner.ai/integrations/view-dns.md): Query live DNS records and trace domain ownership and hosting history with the ViewDNS research toolkit. Agents investigate a domain before trusting an inbound signup or a supplier. - [Vimeo Integration](https://flowrunner.ai/integrations/vimeo.md): Manage a Vimeo account from your flows with 26 actions: upload videos from a URL and poll transcoding, update metadata and privacy, set thumbnails from a frame, organize folders and showcases, search public videos, and automate comments and likes. - [Vincario VIN Decoder Integration](https://flowrunner.ai/integrations/vincario-vindecoder.md): Connect AI agents to Vincario, a VIN decoding and vehicle data service. Agents turn a 17-character VIN into structured vehicle data, estimate market value, and run a stolen-vehicle check. - [Viral Loops Integration](https://flowrunner.ai/integrations/viral-loops.md): Run referral and giveaway campaigns with the Viral Loops API, registering participants and returning their referral codes. Agents credit referrals off confirmed conversions. - [Virtuagym Integration](https://flowrunner.ai/integrations/virtuagym.md): Connect AI agents to Virtuagym, fitness club and coaching management software. Agents create club members and employees, book them into events, assign credits and workouts, register visits, and raise club invoices. - [Virtual Buttons Integration](https://flowrunner.ai/integrations/virtual-buttons.md): Press a Virtual Button to start an Alexa Routine from a flow, using the Alexa Smart Home skill that exposes numbered buttons. Agents reach smart home and physical-space automations that expose no API. - [Vision6 Integration](https://flowrunner.ai/integrations/vision6.md): Vision6 is an Australian email and SMS marketing platform, now part of Constant Contact. Agents manage lists and contacts, create and send messages, send transactional email, and read the full set of send reports. - [Visma eAccounting Integration](https://flowrunner.ai/integrations/visma-eaccounting.md): Visma eAccounting, also sold as Bookkeeping & Invoicing, is the Nordic accounting platform. Agents manage customers, suppliers, articles, and price lists, draft and send customer invoices, record supplier invoices, and handle orders and quotes through a scoped OAuth2 connection. - [Visma.net ERP Integration](https://flowrunner.ai/integrations/visma-net-erp.md): Visma.net ERP is the Nordic enterprise resource planning platform. Agents manage customers, suppliers, and contacts, create sales orders, shipments, and customer invoices, and issue credit and debit notes across the finance modules. - [Vista Social Integration](https://flowrunner.ai/integrations/vista-social.md): Schedule social posts through Vista Social: agents create and schedule posts across Facebook, Instagram, X, LinkedIn, TikTok, and more, target whole profile groups at once, and audit the content calendar by listing scheduled, published, draft, or failed posts. - [Vitally Integration](https://flowrunner.ai/integrations/vitally.md): Connect AI agents to Vitally, the customer success platform, through 30 actions. Agents upsert accounts, users, and organizations by external ID, log conversations and NPS responses, create tasks and notes at renewal or risk milestones, and read traits and custom objects for health-based automations. - [Voice Partner Integration](https://flowrunner.ai/integrations/voice-partner.md): Send voice message campaigns through Voice Partner, the French voice messaging platform, depositing pre-recorded audio. Agents reach a list by voice and hold the send behind an approval. - [Voicenter Integration](https://flowrunner.ai/integrations/voicecenter.md): Work with the Voicenter cloud PBX for call records, live extension and queue monitoring, agent status, and click-to-call. Agents see real queue load and connect the right person to the caller. - [VoiceDrop Integration](https://flowrunner.ai/integrations/voicedrop.md): Run ringless voicemail campaigns through VoiceDrop, sending AI-voice or pre-recorded voicemails straight to the recipient's box. Agents reach a list without ringing a phone. - [Voiso Integration](https://flowrunner.ai/integrations/voiso.md): Run agents, queues, campaigns, and omnichannel messaging on the Voiso cloud contact center. Agents staff against live demand and keep conversations attached to the right record. - [Voluum Integration](https://flowrunner.ai/integrations/voluum.md): Pull Voluum performance reports grouped by campaign, offer, country, or device, and manage the tracking entities behind them: agents create campaigns, offers, landers, and flows, log offline conversions via server-to-server postbacks, and reconcile traffic sources and affiliate networks. - [Vonage Integration](https://flowrunner.ai/integrations/vonage.md): Send SMS and multichannel messages, run two-factor authentication, and validate phone numbers with Number Insight through Vonage (formerly Nexmo). - [VosFactures Integration](https://flowrunner.ai/integrations/vosfactures.md): VosFactures is the French online invoicing platform. Agents issue billing documents of every kind, manage contacts and their delivery addresses, maintain products, and send and track documents from a workflow. - [Voxloud Integration](https://flowrunner.ai/integrations/voxloud.md): Manage the shared phonebook, read call records, and place click-to-call calls on the Voxloud cloud phone system. Agents connect a person to a caller straight from the record. - [VOYP Integration](https://flowrunner.ai/integrations/voyp.md): Place outbound phone calls with a VOYP AI voice assistant by supplying a number and the context of the call. Agents make the routine call and return a tracking reference and the outcome. - [vPlan Integration](https://flowrunner.ai/integrations/vplan.md): vPlan is visual work planning and scheduling for manufacturing and project teams from a Dutch vendor. Agents create and schedule cards, manage boards, resources and collections, and keep the plan current with orders and capacity from other systems. - [VTEX Integration](https://flowrunner.ai/integrations/vtex.md): Connect AI agents to your VTEX commerce cloud. Agents list and inspect orders, look up products, SKUs, inventory, and pricing, and push price updates directly through the VTEX APIs. - [Vtiger CRM Integration](https://flowrunner.ai/integrations/vtiger.md): Connect AI agents to Vtiger CRM. Agents query any module with Vtiger's query language, create and update contacts, leads, organizations, and other records, and inspect module schemas to work with custom fields. - [Vybit Integration](https://flowrunner.ai/integrations/vybit.md): Turn automation events into personalized push notifications with their own custom sound through Vybit. Agents make an alert recognizable without a person looking at the screen. - [Vyfakturuj.cz Integration](https://flowrunner.ai/integrations/vyfakturuj-cz.md): Vyfakturuj.cz is the Czech online invoicing platform. Agents create billing documents and their items, manage the address book, run templates and recurring invoices, and maintain tags, number series, and payment methods. - [Vyte Integration](https://flowrunner.ai/integrations/vyte.md): Connect AI agents to Vyte, a meeting scheduling platform for individuals and teams. Agents create and confirm events, find free slots across users and teams, and manage invitees so group scheduling converges without long email threads. - [Wachete Integration](https://flowrunner.ai/integrations/wachete.md): Create and manage Wachete monitors that watch web pages for changes and raise alerts when they happen. Agents react to a competitor, policy, or price change the moment it is detected. - [Wappalyzer Integration](https://flowrunner.ai/integrations/wappalyzer.md): Give AI agents instant technographics: look up the full technology stack behind any website with Wappalyzer, browse the technology and category taxonomy, and track remaining API credits. Useful for enriching leads and qualifying prospects by the tools they already run. - [Wasabi Integration](https://flowrunner.ai/integrations/wasabi.md): Let AI agents manage hot cloud storage on Wasabi. Agents create and delete buckets, upload, download, copy, and delete objects, check metadata and existence, and generate presigned URLs through Wasabi's S3-compatible API. - [IBM watsonx.ai Integration](https://flowrunner.ai/integrations/watsonx-ai.md): Bring IBM watsonx.ai foundation models into your workflows. Agents run text generation, chat, and question answering against Granite and other hosted models, create embeddings, tokenize text, and list available foundation models. - [Wave Integration](https://flowrunner.ai/integrations/wave.md): Connect AI agents to Wave accounting. Agents manage customers and products, create, approve, and send invoices, read chart-of-accounts and transaction data, and run arbitrary GraphQL queries against the Wave API. - [Waze Deep Links Integration](https://flowrunner.ai/integrations/waze-deep-links.md): Build Waze deep links entirely offline, with no API key, account, or external call. Agents attach turn-by-turn navigation to dispatch records, tickets, and customer messages. - [Wealthbox Integration](https://flowrunner.ai/integrations/wealthbox.md): Connect AI agents to Wealthbox, the CRM built for financial advisors. Agents create contacts, tasks, events, and notes, start workflows against a contact, project, or opportunity, and read the comment history on each record. - [Weavely Integration](https://flowrunner.ai/integrations/weavely.md): Weavely is an AI form builder that turns a plain language brief, a Figma design, a PDF or a screenshot into a themed multi-page web form. Agents generate forms on demand, create and edit them from a specification, read field structure, pull responses, and react the moment a form is submitted. - [Weaviate Integration](https://flowrunner.ai/integrations/weaviate.md): Open-source vector database for agent memory. Manage collections and objects, and run vector, semantic, keyword, and hybrid searches, with a raw GraphQL escape hatch for advanced retrieval. - [Web Data Forms Integration](https://flowrunner.ai/integrations/web-data-forms.md): Web Data Forms is a no-code platform for embeddable, branded web forms with automation and reporting on top. Agents read and write submissions, move each one through status, priority, assignee and comments, and subscribe to webhooks to react the moment a form is filled in. - [WebCrawlerAPI Integration](https://flowrunner.ai/integrations/webcrawlerapi.md): Crawl and scrape websites at scale with WebCrawlerAPI, starting an asynchronous job from a seed URL and collecting every page. Agents gather source material that no API exposes. - [Cisco Webex Integration](https://flowrunner.ai/integrations/webex.md): Integrate Cisco Webex messaging, spaces, memberships, people, teams, and meetings. Post to spaces or people, manage spaces and members, look up people, schedule meetings, and manage Webex-side webhooks. - [Webflow Integration](https://flowrunner.ai/integrations/webflow.md): Automate CMS content and e-commerce workflows. 3 triggers fire on form submissions and order events. 16 actions manage collections, items, fields, and publishing. - [WebinarJam Integration](https://flowrunner.ai/integrations/webinarjam.md): Let AI agents work your webinar funnel: list scheduled WebinarJam webinars, pull a webinar's details and schedules, and register attendees straight from your workflows. - [WEBLUCY Integration](https://flowrunner.ai/integrations/weblucy.md): Manage contacts, members, store, and marketing data on a WEBLUCY website, covering contacts, members, products, and orders. Agents keep site data aligned with the systems of record behind it. - [Formstack Documents Integration](https://flowrunner.ai/integrations/webmerge.md): Formstack Documents, formerly WebMerge, generates PDF, Word, Excel, and PowerPoint files from templates and delivers them where the template says. Agents merge data into documents, fan one payload across a bundle with data routes, and combine, convert, encrypt, or split files. - [Webmetic Integration](https://flowrunner.ai/integrations/webmetic.md): Connect AI agents to Webmetic, a German B2B website visitor identification platform. Agents identify which companies visited the site, segment those visits, and enrich the company behind each one. - [GP webpay Integration](https://flowrunner.ai/integrations/webpay.md): GP webpay is the Global Payments Europe card gateway used across Czechia, Slovakia, and central Europe. Agents check payment status and detail, capture, reverse, and refund, close batches, charge recurring and card-on-file subscriptions, and manage tokens and PUSH payment links. - [WebWork Time Tracker Integration](https://flowrunner.ai/integrations/webwork.md): WebWork Time Tracker is the time tracking and workforce monitoring platform. Agents manage workspaces, members, and teams, create projects, contracts, and tasks, read the live tracker and time entries, approve timesheets, leave, and time requests, pull attendance and activity reports, and react to webhook events. - [WeChat Integration](https://flowrunner.ai/integrations/wechat.md): Connect AI agents to your WeChat Official Account. Agents send template and customer service messages, manage followers, tags, and remarks, build custom menus, upload media, and generate parameterized QR codes. - [WeChat Scraper Integration](https://flowrunner.ai/integrations/wechatscraper.md): Collect public data from WeChat official accounts and their articles, finding accounts by name and reading profile and publication details. Agents monitor Chinese-market publishing without an account on the platform. - [weclapp Integration](https://flowrunner.ai/integrations/weclapp.md): Connect AI agents to weclapp, a cloud ERP and CRM for small and midsize businesses. Agents manage parties and articles, accept quotations into sales orders and raise invoices from them, open opportunities, and read warehouse stock. - [Weekdone Integration](https://flowrunner.ai/integrations/weekdone.md): Weekdone is the OKR and weekly status reporting tool. Agents post plans, progress, and problems with comments and likes, create and update objectives and key results, log KPIs, and read the weekly report for any team or user. - [Wekan Integration](https://flowrunner.ai/integrations/wekan.md): Manage Wekan boards, lists, cards, swimlanes, and checklists on your own server from your flows. Agents turn inbound tickets into cards on the right board and confirm before an irreversible board teardown. - [WeSupply Integration](https://flowrunner.ai/integrations/wesupply.md): WeSupply is the post-purchase layer of a store: tracking, returns and delivery notifications. Agents push orders in so they become trackable and returnable, move shipments through their statuses, run the returns flow, enroll customers in SMS updates, and quote delivery dates. - [wflow Integration](https://flowrunner.ai/integrations/wflow.md): wflow is the Czech invoice approval and document workflow platform. Agents file incoming and outgoing documents with their attachments, drive approvals, read comments, manage tags and links, and administer access rights and custom fields. - [Whapi.Cloud Integration](https://flowrunner.ai/integrations/whapi-cloud.md): Connect AI agents to Whapi.Cloud, a WhatsApp API for developers. Agents send text, media, and poll messages, administer groups and labels, manage contacts and media files, and react to inbound messages and delivery receipts so WhatsApp conversations run inside automated workflows. - [WhatConverts Integration](https://flowrunner.ai/integrations/whatconverts.md): Give AI agents full control of WhatConverts lead tracking. Agents create, update, and manage leads, accounts, and tracking profiles, and pull attribution data that shows which campaigns drive calls, chats, and form fills. - [WhatsApp Business Integration](https://flowrunner.ai/integrations/whatsapp.md): Reach customers and teams on the channel with the highest open rates. Agents send text, images, documents, templates, and location via WhatsApp Business API. - [360Messenger Integration](https://flowrunner.ai/integrations/whatsapp-360messenger.md): Connect AI agents to 360Messenger, a WhatsApp HTTP API. Agents send text and file messages to numbers and groups, read sent and received message queues, administer groups and labels, and manage chats and contacts so WhatsApp outreach and follow up run automatically. - [WhatsApp Number Validator Integration](https://flowrunner.ai/integrations/whatsapp-number-validator.md): Check in real time whether phone numbers are registered on WhatsApp and whether the account is a WhatsApp Business account. Agents confirm reachability before a campaign spends a message on a dead number. - [WhatsScale Integration](https://flowrunner.ai/integrations/whatsscale.md): Connect AI agents to WhatsScale, a WhatsApp automation platform with a built in CRM. Agents send messages and media, run paced broadcasts to tagged audiences, manage CRM contacts and groups, and react to inbound messages so WhatsApp campaigns scale without manual sending. - [When I Work Integration](https://flowrunner.ai/integrations/when-i-work.md): Connect AI agents to When I Work employee scheduling. Agents manage users, create and review shifts, pull time entries, and look up locations and positions to keep hourly teams staffed. - [Whereby Integration](https://flowrunner.ai/integrations/whereby.md): Let AI agents spin up video meetings on demand. Agents create and delete Whereby rooms, fetch meeting details, and review room session history through the Whereby Embedded API. - [WHMCS Integration](https://flowrunner.ai/integrations/whmcs.md): Connect AI agents to WHMCS, the billing and automation platform for hosting businesses. Agents manage clients, create and pay invoices, place and accept orders, work support tickets, record transactions, and call any other WHMCS API action directly. - [WhoisFreaks Integration](https://flowrunner.ai/integrations/whoisfreaks.md): WhoisFreaks is a WHOIS, DNS, and domain intelligence API. Agents run live and historical WHOIS and DNS lookups, check domain availability, find typosquats and subdomains, inspect SSL certificates, and score IP reputation. - [Wild Apricot Integration](https://flowrunner.ai/integrations/wild-apricot.md): Connect AI agents to Wild Apricot membership management. Agents manage contacts and members, browse membership levels, track events and registrations, and pull invoices to keep your association's records current. - [Windy Integration](https://flowrunner.ai/integrations/windy.md): Retrieve numerical weather forecasts for any coordinates and reach Windy's worldwide webcam network. Agents factor real conditions into dispatch, scheduling, and field work. - [Wise Integration](https://flowrunner.ai/integrations/wise.md): Connect AI agents to Wise (formerly TransferWise). Agents manage profiles, quote currency conversions at the mid-market rate, create recipient accounts, send and fund international transfers end to end, read multi-currency balances and exchange rates, and track or cancel transfers through the Wise Platform API. - [WiseOCR Integration](https://flowrunner.ai/integrations/wise-ocr.md): Turn receipt and invoice images, PDFs, or raw text into structured JSON with the WiseOCR API, with confidence returned on every extraction. Agents post the confident results and route the rest to a person. - [WiseWand Integration](https://flowrunner.ai/integrations/wisewand.md): Generate ranking articles in 83 languages with WiseWand, the French AI SEO article generator, and publish them onward. Agents route every draft to an editor before it goes live under your byline. - [Wishpond Integration](https://flowrunner.ai/integrations/wishpond.md): Work with the Wishpond marketing automation API across leads, anonymous visitors, contact lists, and user and event tracking. Agents keep campaigns driven by real on-site behavior. - [Wistia Integration](https://flowrunner.ai/integrations/wistia.md): Put AI agents in charge of your Wistia video library. Agents upload media from URLs, update and organize videos into projects, and pull play and engagement stats to see how content performs. - [Wix Integration](https://flowrunner.ai/integrations/wix.md): Connect AI agents to your Wix site. Agents manage contacts and labels, read and write CMS data collections, run your store's products, orders, fulfillments, and coupons, draft blog posts, and look up site members and properties. - [Wiza Integration](https://flowrunner.ai/integrations/wiza.md): Reveal and enrich LinkedIn prospect data at scale. Agents build prospect lists, reveal individual contact information, retrieve validated emails, and manage credits for lead generation workflows. - [WiziShop Integration](https://flowrunner.ai/integrations/wizishop.md): WiziShop is a French hosted ecommerce platform, with Dropizi as its dropshipping edition. Agents push orders into a warehouse, ERP or 3PL and write tracking back, keep SKU stock in step with a supplier feed, manage products, customers and reviews, and react to store webhooks in real time. - [WizyChat Integration](https://flowrunner.ai/integrations/wizychat.md): Query a WizyChat chatbot trained on your own content and use its answer inside a workflow. Agents answer from an approved corpus rather than from general model memory. - [WooCommerce Integration](https://flowrunner.ai/integrations/woocommerce.md): Automate a self-hosted WooCommerce store from FlowRunner over the WooCommerce REST API. Agents manage products, variations, categories, attributes, orders, notes, refunds, customers, and coupons, bulk-sync catalogs in a single request, and react in real time to order, product, and customer events. - [Woodpecker Integration](https://flowrunner.ai/integrations/woodpecker.md): Let AI agents run your cold outreach lists in Woodpecker. Agents add prospects to campaigns, update prospect data, stop follow-ups when a reply lands, and monitor campaign status. - [Woosmap Integration](https://flowrunner.ai/integrations/woosmap.md): Connect AI agents to the Woosmap location platform to manage store and asset data and search it by proximity or attribute. Agents route customers and work to the nearest capable location. - [WordPress Integration](https://flowrunner.ai/integrations/wordpress.md): Connect AI agents to a self-hosted WordPress site through the REST API. Agents manage posts, pages, categories, tags, users, media, comments, settings, and custom types, and react to new published posts and new comments. - [Words API Integration](https://flowrunner.ai/integrations/words-api.md): Look up English words through Words API, retrieving definitions, synonyms, antonyms, examples, rhymes, syllables, and pronunciation. Agents check and vary language before content is published. - [Workable Integration](https://flowrunner.ai/integrations/workable.md): Connect AI agents to Workable, a recruiting and applicant tracking system. Agents list and read jobs, create and move candidates through pipeline stages, log comments and ratings, read the account roster, and watch for new candidates via access token. - [Workast Integration](https://flowrunner.ai/integrations/workast.md): Workast is task and project management built around spaces, with a task search that reads like a query builder. Agents create and update tasks, assign and comment, manage spaces and lists, and search tasks by any combination of fields. - [Workday Integration](https://flowrunner.ai/integrations/workday.md): Connect AI agents to Workday HCM. Agents look up workers and their direct reports, browse organization roles and reference values, and pull any custom report through Report-as-a-Service for data no standard endpoint exposes. - [Workiom Integration](https://flowrunner.ai/integrations/workiom.md): Workiom is a no-code database and app builder where apps hold lists, fields, records and views, plus forms, dashboards and automations. Agents create apps from templates, manage lists and fields, read and write records, and run automations and approval requests. - [WOZTELL Integration](https://flowrunner.ai/integrations/woztell.md): Connect AI agents to WOZTELL, an omnichannel conversational platform for WhatsApp, Messenger, Instagram, and web chat. Agents send responses, manage members and audiences, run broadcasts, administer channels and WhatsApp templates, and react to channel events so conversations flow between bots and human teams. - [WPForms Integration](https://flowrunner.ai/integrations/wpforms.md): Give AI agents read access to WPForms on your WordPress site. Agents list forms, inspect form structure, and retrieve submitted entries so downstream workflows can qualify, route, and respond to every submission. - [Wrike Integration](https://flowrunner.ai/integrations/wrike.md): Connect AI agents to Wrike. Agents create folders and tasks, update status and assignments, post and read comments, and look up contacts to keep project work moving across your workspace. - [WS Form Integration](https://flowrunner.ai/integrations/ws-form.md): WS Form is a drag and drop form builder plugin for WordPress. Agents create, import, export and style forms, read and manage submissions, and react to new submissions on your own WordPress site. - [Wufoo Integration](https://flowrunner.ai/integrations/wufoo.md): List Wufoo forms and fields, read and create form entries, work with reports, and manage submission webhooks through the Wufoo REST API. - [X (Twitter) Integration](https://flowrunner.ai/integrations/x-twitter.md): Post content to X programmatically. Agents create text posts and image posts as part of content distribution workflows. Automate social publishing without manual platform access. - [xAI Grok Integration](https://flowrunner.ai/integrations/xai-grok.md): Run xAI's frontier Grok models: chat completion, real-time answers via Live Search across web, X, and news, vision, image and video generation, and structured output. - [Xama Integration](https://flowrunner.ai/integrations/xama-onboarding.md): Connect AI agents to Xama, an AML and KYC client onboarding platform for accountants and professional services firms. Agents create accounts and contacts, initiate company, onboarding, and AML reports, manage subscriptions, and read the risk assessments and documents each onboarding produces. - [Xano Integration](https://flowrunner.ai/integrations/xano.md): Wire AI agents into your Xano backend. Agents query, create, update, and delete records through your API groups and call any custom endpoint you have built, with full control over method, path, and payload. - [Xata Integration](https://flowrunner.ai/integrations/xata.md): Let AI agents work your Xata serverless Postgres data. Agents insert, update, and bulk-load records, run typed queries, full-text search across tables and branches, aggregations, summaries, and raw SQL. - [Xero Integration](https://flowrunner.ai/integrations/xero.md): Real-time accounting automation with 8 event triggers and 37 actions. Agents react to invoices, payments, and bank transactions the moment they happen. - [XLS Tools Integration](https://flowrunner.ai/integrations/xls-tools.md): Create and read spreadsheet files inside FlowRunner with no API key, vendor account, or outbound call beyond fetching the file itself. Agents build and parse workbooks as a native workflow step. - [XMP Tools Integration](https://flowrunner.ai/integrations/xmp-tools.md): Read, write, and strip XMP metadata on media files inside FlowRunner with no API key or vendor account. Agents clean or stamp media metadata before an asset is published or archived. - [Yay! Forms Integration](https://flowrunner.ai/integrations/yay-forms.md): Yay! Forms is an AI powered form and survey builder from Brazil. Agents manage workspaces, folders, forms and questions, read responses with AI enrichment on every answer, work the response kanban, manage themes and media, and drive the built in AI assistant to edit a form. - [Ycode Integration](https://flowrunner.ai/integrations/ycode.md): Connect AI agents to Ycode, a no-code website builder with a built-in CMS. Agents create, update, and delete CMS collection items and read or create form submissions so site content and lead capture stay synced with the rest of the stack. - [Yeastar Contacts Integration](https://flowrunner.ai/integrations/yeastar-contacts.md): Manage contacts, phonebooks, and extensions on a Yeastar P-Series PBX. Agents keep the company directory in step with the CRM and the HR roster. - [Yeeflow Integration](https://flowrunner.ai/integrations/yeeflow.md): Yeeflow is the no-code workflow and application platform. Agents create and update list items with fields and files, start workflows and complete their tasks, set delegations and variables, manage users, departments, groups, and positions, and react to webhook events. - [Yelp Integration](https://flowrunner.ai/integrations/yelp.md): Give AI agents local business intelligence from Yelp Fusion. Agents search businesses by keyword, location, or phone number, pull business details and reviews, and autocomplete queries for lead lists and market research. - [YepCode Run Integration](https://flowrunner.ai/integrations/yepcode-run.md): YepCode Run executes JavaScript and Python with npm and PyPI dependencies on a serverless platform. Workflows create and publish processes, run them synchronously or asynchronously, read execution logs, and manage schedules, modules, variables, and object storage. - [YeshInvoice Integration](https://flowrunner.ai/integrations/yeshinvoice.md): YeshInvoice is the Israeli invoicing and CRM platform. Agents issue tax invoices, receipts, quotes, orders, and credit notes, manage customers, suppliers, and products, and log expenses. - [YNAB Integration](https://flowrunner.ai/integrations/ynab.md): Connect AI agents to YNAB budgets. Agents read budgets, accounts, categories, and payees, and create and update transactions so every dollar gets recorded the moment it moves. - [Yodiz Integration](https://flowrunner.ai/integrations/yodiz.md): Yodiz is agile project management and issue tracking: epics, user stories, tasks, issues, sprints and releases. Agents create and update stories, tasks and issues, plan sprints and releases, and keep the backlog in step with what is happening elsewhere. - [YOOBIC Integration](https://flowrunner.ai/integrations/yoobic.md): Connect AI agents to YOOBIC, a frontline execution platform for retail and hospitality teams. Agents publish campaigns that create missions, assign and validate them across sites, track visits, and export the results. - [Yotpo Loyalty Integration](https://flowrunner.ai/integrations/yotpo-loyalty.md): Put AI agents behind your Yotpo Loyalty program. Agents create and update loyalty customers, adjust point balances, record custom actions, and surface redemption and referral options to reward repeat buyers automatically. - [You.com Integration](https://flowrunner.ai/integrations/you.md): Ground AI agents in live web data with You.com. Agents run web searches, fetch page contents, launch and poll deep research tasks, and pull finance-focused research to answer questions with current sources. - [YouCanBookMe Integration](https://flowrunner.ai/integrations/youcanbookme.md): Connect AI agents to YouCanBookMe scheduling. Agents list and inspect bookings across your booking pages, create a booking on a customer's behalf, cancel bookings during rescheduling flows, and read profile and account settings for reporting. - [Maybe Your AI Integration](https://flowrunner.ai/integrations/your-ai.md): Drive the Your AI assistant inside the Maybe social marketing platform, running conversations through assistants. Agents handle social marketing conversations grounded in the platform's own data. - [YourGPT Chatbot Integration](https://flowrunner.ai/integrations/yourgpt-chatbot.md): Build, train, and operate no-code AI support agents with the YourGPT Chatbot API and drive conversations through it. Agents keep support answers grounded in current documentation. - [YOURLS Integration](https://flowrunner.ai/integrations/yourls.md): Connect over a signature token to a self-hosted YOURLS URL shortener to shorten and expand links on your own branded domain and read per-link and instance-wide click statistics. - [YourTextGuru Integration](https://flowrunner.ai/integrations/yourtextguru.md): Pull semantic SEO guides, content scores, and AI writing from YourTextGuru through its V1 API. Agents score a draft against the target query before anyone publishes it. - [YouTube Integration](https://flowrunner.ai/integrations/youtube.md): Manage YouTube channels, videos, playlists, comments, captions, subscriptions, and live streams, and pull analytics and reporting for the authenticated account via the YouTube Data and Analytics APIs. - [Yunique Integration](https://flowrunner.ai/integrations/yunique.md): Connect AI agents to Yunique by Opens, a Brazilian omnichannel contact center that unifies WhatsApp and webchat. Agents read and send messages, open cases, log calls and activities, and tag conversations for routing. - [Z-API Integration](https://flowrunner.ai/integrations/z-api.md): Connect AI agents to Z-API, a Brazilian REST API that connects a WhatsApp account to your automations. Agents send text, media, and interactive messages, manage chats, contacts, groups, and communities, and administer the WhatsApp Business profile and catalog so WhatsApp operations run programmatically. - [Zadarma Integration](https://flowrunner.ai/integrations/zadarma.md): Use Zadarma VoIP telephony and cloud PBX for balance, SIP accounts, numbers, call statistics, recordings, and SMS. Agents provision lines and attach call evidence to the right record. - [Zagomail Integration](https://flowrunner.ai/integrations/zagomail.md): Zagomail is an email marketing platform for growing lists and sending campaigns. Agents manage lists, subscribers and tags, send campaigns, read form submissions, and react to subscriber events in real time through webhooks. - [Zammad Integration](https://flowrunner.ai/integrations/zammad.md): Connect AI agents to Zammad, the open-source helpdesk and customer support system. Agents manage tickets, articles, users, organizations, groups, and tags on a self-hosted or hosted Zammad instance over its REST API. - [Zamzar Integration](https://flowrunner.ai/integrations/zamzar.md): Convert files between hundreds of document, image, audio, and video formats with Zamzar. Agents start conversion jobs from uploaded files or public URLs, poll them to completion, download converted output into FlowRunner storage, and check supported formats and credit costs before converting. - [Zapier Integration](https://flowrunner.ai/integrations/zapier.md): Bridge FlowRunner into Zapier through the Webhooks by Zapier Catch Hook trigger. A flow posts data to a Catch Hook URL, so agents reach the long tail of consumer apps Zapier already lists. - [ZapSign Integration](https://flowrunner.ai/integrations/zapsign.md): ZapSign is a Brazilian electronic signature platform used across Latin America. Agents send documents for signature from a file, a Word template, or Markdown, set signing order, manage signers and templates, read the activity log, and register webhooks. - [Zapster API Integration](https://flowrunner.ai/integrations/zapster-api.md): Connect AI agents to Zapster API, a Brazilian WhatsApp platform. Agents provision and manage WhatsApp instances, send text, media, and template messages immediately or on a schedule, manage groups and participants, and react to inbound WhatsApp events so number operations and messaging run hands free. - [Zencal Integration](https://flowrunner.ai/integrations/zencal.md): Connect AI agents to Zencal, an online booking and meeting scheduling tool. Agents calculate available dates, schedule and confirm meetings, generate one-time scheduling links, and manage team members so booking operations run without manual steps. - [Zendesk Integration](https://flowrunner.ai/integrations/zendesk.md): Connect AI agents to Zendesk. Agents create, update, search, and comment on tickets, manage end users and organizations, look up agents and groups, and react to tickets being created or updated in real time. - [Zendesk Sell Integration](https://flowrunner.ai/integrations/zendesk-sell.md): Connect AI agents to Zendesk Sell, the sales CRM formerly known as Base. Agents run full CRUD across leads, contacts, deals, tasks, and notes, advance deals through pipeline stages, attach catalog products as line items, and read pipelines, sources, tags, and users to route work correctly. - [Sunshine Conversations Integration](https://flowrunner.ai/integrations/zendesk-sunshine.md): Run omnichannel messaging on Zendesk Sunshine Conversations, covering users, conversations, messages, and attachments with a realtime webhook trigger. Agents handle the routine thread and escalate what needs a person. - [New Zenler Integration](https://flowrunner.ai/integrations/zenler.md): Connect AI agents to New Zenler, an online course creation and marketing platform. Agents manage users, enroll them in courses and funnels, register attendees for live classes and webinars, and pull sales and progress reports so your school runs itself between launches. - [ZenRows Integration](https://flowrunner.ai/integrations/zenrows.md): Scrape any web page through the ZenRows Universal Scraper with rotating proxies and anti-bot handling built in. Agents fetch raw HTML, extract structured JSON with CSS selectors or autoparse, render JavaScript-heavy pages, geo-target through residential proxies, and capture PNG screenshots. - [Zep Integration](https://flowrunner.ai/integrations/zep.md): Long-term memory for AI agents. Store conversation history in per-user threads, build a per-user knowledge graph, and pull relevant memory back into LLM prompts for persistent, cross-conversation recall. - [ZeroBounce Integration](https://flowrunner.ai/integrations/zerobounce.md): Validate email addresses in real time with ZeroBounce before they enter your lists. Agents verify single addresses or batches of up to 200, pull engagement activity for a contact, and monitor remaining credits and API usage to protect bounce rates and sender reputation. - [0CodeUtil Integration](https://flowrunner.ai/integrations/zerocodeutil.md): Use the 0CodeUtil data utility toolkit for text encoding and reformatting, sentiment scoring, and multi-pattern regex work. Agents handle the transformation chores between the systems that matter. - [Ziflow Integration](https://flowrunner.ai/integrations/ziflow.md): Connect AI agents to Ziflow creative proofing. Agents create and update proofs, invite reviewers, collect comments and review decisions, and read folders and users, so approval status flows into your project and messaging tools without manual chasing. - [Zigpoll Integration](https://flowrunner.ai/integrations/zigpoll.md): Zigpoll is an on-site and post-purchase survey widget best known as a Shopify app. Agents manage surveys, questions and participants, read responses, captured emails and AI insights, run email and SMS campaigns, set Action Logic, and react through signed webhooks. - [zistemo Integration](https://flowrunner.ai/integrations/zistemo.md): zistemo, formerly MoneyPenny, is the Swiss time tracking, project, and invoicing platform. Agents manage clients, vendors, projects, and tasks, run timers and timesheets, approve tracked time, and record payments against invoices. - [Zite DB Integration](https://flowrunner.ai/integrations/zite.md): Zite DB is the database behind Zite's no-code AI app builder, with databases containing tables containing records, a SQL read layer and webhooks. Agents manage databases, tables and fields, create, update and upsert records in bulk, aggregate and query with SQL and react to database events. - [Zixflow Integration](https://flowrunner.ai/integrations/zixflow.md): Connect AI agents to Zixflow, a CRM and omnichannel engagement platform. Agents create records in any collection, manage list entries, log activities, send messages, and read reports. - [Zoho Books Integration](https://flowrunner.ai/integrations/zoho-books.md): Automate Zoho Books accounting. Agents manage contacts, invoices, estimates, bills, expenses, and payments, run the order-to-cash and procure-to-pay cycle, and react to accounting events in real time. - [Zoho Calendar Integration](https://flowrunner.ai/integrations/zoho-calendar.md): Connect AI agents to Zoho Calendar, the calendar app in the Zoho suite. Agents create and update calendars and events and check free/busy availability so schedules across the Zoho suite stay coordinated automatically. - [Zoho Campaigns Integration](https://flowrunner.ai/integrations/zoho-campaigns.md): Connect AI agents to Zoho Campaigns. Agents add and unsubscribe mailing list subscribers, read list membership to keep audiences in sync, and list, inspect, and send email campaigns as the final step of a launch workflow. - [Zoho Cliq Integration](https://flowrunner.ai/integrations/zoho-cliq.md): Send messages from AI agents into Zoho Cliq channels, group chats, and direct conversations, with rich cards, buttons, and attachments. Agents also create and manage channels and look up users and contacts to route notifications to the right destination. - [Zoho Creator Integration](https://flowrunner.ai/integrations/zoho-creator.md): Connect AI agents to your Zoho Creator low-code apps. Agents add records through forms, read, update, and delete records through reports, filter with Creator criteria expressions, and discover an app's forms and reports to keep external systems in sync. - [Zoho CRM Integration](https://flowrunner.ai/integrations/zoho-crm.md): Connect AI agents to Zoho CRM. Agents create and update records across standard and custom modules, convert leads, manage notes, tags, and related records, and deduplicate with upsert. - [Zoho Desk Integration](https://flowrunner.ai/integrations/zoho-desk.md): Run help desk operations in Zoho Desk from AI agents. Agents create, search, assign, update, and close tickets, add comments, send email replies on ticket threads, and maintain the contacts, accounts, agents, and departments behind them. - [Zoho Forms Integration](https://flowrunner.ai/integrations/zoho-forms.md): Connect AI agents to Zoho Forms. Agents push new form entries from other systems, read a report's submission records, fetch a single entry by ID, and discover an application's forms and reports for downstream routing. - [Zoho Inventory Integration](https://flowrunner.ai/integrations/zoho-inventory.md): Automate Zoho Inventory order and stock management. Agents manage items, sales and purchase orders, invoices, payments, packages, and shipments, and monitor low-stock items across warehouses. - [Zoho Invoice Integration](https://flowrunner.ai/integrations/zoho-invoice.md): Automate accounts receivable in Zoho Invoice with AI agents. Agents create and email invoices and estimates, record customer payments, manage customers, items, expenses, and credit notes, void or mark invoices as sent, and pull invoice PDFs into FlowRunner storage. - [Zoho Mail Integration](https://flowrunner.ai/integrations/zoho-mail.md): Connect AI agents to Zoho Mail business mailboxes. Agents send email, search and read full message content, and keep inboxes organized by marking messages read or unread, archiving, flagging spam, and moving them between folders. - [Zoho Notebook Integration](https://flowrunner.ai/integrations/zoho-notebook.md): Zoho Notebook is the note taking app in the Zoho suite. Agents create notebooks, add and update notecards with tags and reminders, read versions, search, read and empty the trash, and sync changes across an account. - [Zoho People Integration](https://flowrunner.ai/integrations/zoho-people.md): Connect AI agents to Zoho People HR. Agents look up and list employees, read and write records on any People form, apply for and list leave, and pull attendance entries for reporting and timesheet automation. - [Zoho Projects Integration](https://flowrunner.ai/integrations/zoho-projects.md): Manage Zoho Projects from AI agents across the portal, project, and task hierarchy. Agents create projects and tasks, assign and update work as it progresses, and list portals, milestones, task lists, and users to drive reporting and notifications. - [Zoho Recruit Integration](https://flowrunner.ai/integrations/zoho-recruit.md): Automate Zoho Recruit applicant tracking. Agents manage candidates, job openings, applications, and interviews, parse resumes, advance applicants through stages, and react to hiring events in real time. - [Zoho SalesIQ Integration](https://flowrunner.ai/integrations/zoho-salesiq.md): Read Zoho SalesIQ live chat activity from AI agents. Agents pull conversations filtered by status, department, or operator, fetch full transcripts, track website visitors by portal view, and look up operators and departments for routing and reporting. - [Zoho Sheet Integration](https://flowrunner.ai/integrations/zoho-sheet.md): Zoho Sheet is the online spreadsheet in the Zoho suite. Agents create and share workbooks, read and write cells, ranges, and tabular records by criteria, build tables, pivots, and charts, apply formats and validation, and run merge templates and jobs. - [Zoho Sign Integration](https://flowrunner.ai/integrations/zoho-sign.md): Automate e-signature workflows in Zoho Sign. Agents send documents for signing, launch requests from reusable templates, remind stalled signers, recall documents sent in error, and download the completed signed PDF into FlowRunner storage. - [Zoho TeamInbox Integration](https://flowrunner.ai/integrations/zoho-teaminbox.md): Zoho TeamInbox is Zoho's shared inbox for teams. Agents list and search conversations, assign, archive, snooze and close threads, send email from a team address, and create contacts across your organization's inboxes. - [Zoho WorkDrive Integration](https://flowrunner.ai/integrations/zoho-workdrive.md): Connect AI agents to Zoho WorkDrive cloud storage. Agents upload and download files, browse folder contents, create, rename, move, and trash files and folders, and enumerate teams and team folders to keep shared workspaces organized. - [Zoho Writer Integration](https://flowrunner.ai/integrations/zoho-writer.md): Zoho Writer is the word processor of the Zoho suite, and its API centers on mail merge. Agents merge a template with records to produce documents one by one or in batch, store them in WorkDrive, send them for signature through Zoho Sign, or generate fill-in links. - [Zoho ZeptoMail Integration](https://flowrunner.ai/integrations/zoho-zeptomail.md): Send transactional email through Zoho ZeptoMail from a verified sending domain. Agents deliver single HTML or plain-text messages, send template emails with merge data, and dispatch batch template sends with per-recipient personalization. - [Zonka Feedback Integration](https://flowrunner.ai/integrations/zonka-feedback.md): Zonka Feedback runs NPS, CSAT and CES surveys over email, SMS, WhatsApp, web, kiosks and in-app SDKs, with AI scoring for sentiment and urgency. Agents read responses and their scores, upsert contacts, send survey links and distributions, manage surveys, locations and devices, and react to new feedback. - [Zoom Integration](https://flowrunner.ai/integrations/zoom.md): Create and manage Zoom meetings, register attendees and share their join links, retrieve cloud recording download links, and pull past-meeting participant reports. - [ZoomInfo Integration](https://flowrunner.ai/integrations/zoominfo.md): Enrich prospects with ZoomInfo B2B intelligence. Agents search for people and companies, enrich records with verified emails, direct and mobile phones, and firmographics such as industry, employee count, and revenue, and monitor API credit usage against an Enterprise API license. - [Zulip Integration](https://flowrunner.ai/integrations/zulip.md): Send stream and direct messages, manage streams and users, react to and edit messages, upload files, and register event queues in a Zulip organization over its REST API. - [Zuora Integration](https://flowrunner.ai/integrations/zuora.md): Automate subscription billing in Zuora from AI agents. Agents create customer accounts, pull consolidated account summaries with recent invoices and payments, manage subscriptions against the product catalog, browse products, and run ad-hoc ZOQL queries when no dedicated action exists. - [Zyllio Integration](https://flowrunner.ai/integrations/zyllio.md): Zyllio's SaaS API runs a software business on the no-code platform. Workflows clone an application per customer, configure and rename it, keep it in step with the template, and publish it.