Webmetic
Sales IntelligenceConnect 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.
What This Integration Enables
Webmetic is a German B2B visitor identification platform, and the most useful thing to be clear about is what it identifies. It resolves the organization behind a website session. It does not name the person. Nothing in this connector returns an individual visitor's identity, and a flow built on it is reasoning about companies, not people. That single fact separates this integration from the contact-finding tools it sits beside in the catalog, and it changes both what you can do with the output and what can go wrong with it. The risk here is not exposing someone's personal mobile number. The risk is confidently acting on a wrong inference.
The connector splits into three surfaces. The visitor data reads are the raw feed: Get Company Info returns identified company profiles across one or more of your tracked domains with pagination, Get Company Sessions returns the fullest export by joining sessions to their identified companies for a single domain, Get Session Info returns the technical side of each visit including device, browser, referrer, and engagement, and Get Session User Data returns the page-level interaction detail. Get Company NACE Codes attaches standardized economic activity classifications, which matters more in a European B2B context than a generic industry string does.
The visit segments are pre-computed views rather than segments you define. Get Intensive Visits returns companies that engaged for a meaningful duration, roughly thirty seconds or more. Get Returning Visits returns those that came back across at least two sessions. Get New Visits returns those whose earliest recorded session falls inside the range. These are reads: the connector does not create, name, or store a segment definition of your own. All three are rate limited to one request per second, so a flow that fans them out in parallel will get throttled.
The enrichment surface is the part that writes. Enrich Company is a synchronous lookup by domain returning firmographics, NACE codes, employee count, revenue, address, contact details, and branding, with a status of processing when the domain has not been crawled yet, which is a retry rather than a failure. Around it sits a small list system: Create Enrichment List, Add Domains To List for up to 5000 domains at a time, Get Enrichment List to poll the total, done, and pending counters, Get Enrichment List Rows to read results as JSON filtered by status, and Export Enrichment List to pull the whole thing as CSV. These enrichment operations authenticate with a separate Enrichment API Key when one is configured, and fall back to the main key when it is left empty.
There are no triggers on this connector. Webmetic does not wake a flow up, and there is no webhook subscription surface here either. Every read takes a date range, which accepts ISO dates or relative values such as -30 days or -1 hour, with timestamps in the Europe/Berlin timezone. The pattern is a scheduled pull with a moving window, and the timezone is worth getting right on the first day rather than the third.
Finally, the vendor is explicit about this and so are we: identified company, session, and enrichment data should be treated as business personal data and handled in accordance with GDPR and your own privacy policy. That is a live consideration for a German platform reading traffic on European visitors, and it is a reason to keep the outbound side of any flow built here deliberate rather than automatic.
Without FlowRunner
With FlowRunner
Use Case Scenarios
A morning shortlist instead of a daily company dump
A scheduled flow calls Get New Visits for yesterday on the tracked domain and gets the companies seeing the site for the first time. On its own that is a list, not a signal. The agent then calls Get Session User Data for the same window and joins the page-level detail onto each company, so an account that read one blog post is visibly different from one that read pricing, the security page, and an integration page in a single session. It checks each domain against Pipedrive and drops existing customers and open opportunities, then calls Enrich Company on what remains to attach employee count, revenue band, and NACE code. What reaches Slack is a handful of accounts with their evidence, which is a thing a human can act on before lunch, rather than four hundred rows nobody opens twice.
Return visits as the signal worth watching
Get Returning Visits is the most underrated read in this connector, because a second visit is a far stronger indicator than a long first one. A weekly flow pulls returning companies over a rolling window, cross-references them against accounts already in a nurture sequence in Brevo, and separates two groups: accounts being nurtured that came back, which tells the marketing owner the sequence is landing, and accounts nobody is contacting that came back anyway, which is the more interesting list. Both go to Google Sheets with session counts and NACE codes from Get Company NACE Codes attached, so the segmentation argument is settled with data rather than with instinct. Remember the one request per second limit on the segment endpoints when the flow covers several tracked domains.
Backfilling an account list without doing it by hand
Sales has a list of 3,000 target domains with nothing but the domain. The agent calls Create Enrichment List, then Add Domains To List in chunks, since a single call takes up to 5000 domains, and Webmetic enriches them in the background. Known domains typically complete quickly and unknown ones are crawled on demand and take longer. Rather than guessing when it is done, the agent polls Get Enrichment List and reads the total, done, and pending counters, then calls Get Enrichment List Rows filtered to completed rows and writes them into HubSpot. Export Enrichment List gives the same data as CSV when the destination is a spreadsheet or a one-off import instead. The advantage over a loop of single Enrich Company calls is not speed, it is that the progress is observable and a partial result is readable.
Human-in-Loop Highlight
The write actions here are mild. The consequential moment sits one system later, and pretending otherwise would be dishonest about where the risk actually lives. Get Intensive Visits returns companies that engaged with your site for roughly thirty seconds or more, and it is enormously tempting to wire that straight into an outbound sequence, because the whole product exists to make that possible. The reason to put a person in front of it is that IP-to-company identification is an inference, and its failure modes are systematic rather than random. A shared corporate proxy attributes a subsidiary's visit to its parent. A coworking space, a university network, or a large ISP surfaces as an organization that never visited anything. A VPN attributes a visit to whichever company operates the exit node. None of these arrive flagged; they arrive as a company name in the same field as a correct one, and the endpoint returns all fields with missing values as null rather than telling you the confidence was thin. Send outbound on a wrong identification and you have written to a stranger claiming to know they visited your pricing page, which is a specific and memorable kind of bad first impression, and one that a German recipient is entitled to ask pointed questions about under GDPR. So the agent gathers the evidence and stops before the send. It pairs the intensive-visit companies with Get Session User Data for the pages actually viewed, adds employee count and NACE code from Enrich Company, and asks the demand generation owner: "Four companies crossed the intensity threshold yesterday. Deutsche Bahn AG, 3 sessions, pricing and security pages, 4 minutes. Two others read a single blog post at length. The fourth resolves to a regional ISP with 40 unrelated page views, which reads as shared network rather than a prospect. Approve outbound on which of these?" The mechanical part, pulling, joining, enriching, and deduplicating against the CRM, runs unattended every morning. The human-in-the-loop step is the one that decides an inference is solid enough to act on a person's behalf.
Agent Capabilities
15 actionsVisitor Data
5- Get Company Info Retrieves identified B2B company profiles for one or more of your tracked domains within a date range, returning firmographic details such as company name, industry, size, and location for organizations whose sessions Webmetic de-anonymized. Supports pagination and multiple domains in a single call, which makes it the base read for a multi-site account.
- Get Company Sessions Returns a full export of visitor sessions joined with their identified company data for a single tracked domain and date range. The most detailed feed in the connector, combining firmographics with per-session engagement. Use the narrower reads when you need only one side of that join.
- Get Company NACE Codes Retrieves NACE industry classification codes for the companies identified on a tracked domain within a date range. Standardized economic activity codes segment a European B2B list far more reliably than a free-text industry field does.
- Get Session Info Retrieves technical session records for a tracked domain within a date range, including device, browser, referrer, and engagement details for each visit. Session-level data without company firmographics; pair it with Get Company Info when you need both.
- Get Session User Data Retrieves granular per-session interaction data for a tracked domain within a date range, such as the pages viewed and the actions taken during each visit. This is the evidence layer, and it is what turns an identified company name into something a human can judge.
Visit Segments
3- Get Intensive Visits Retrieves companies that engaged with a tracked domain for a meaningful duration, roughly thirty seconds or more. Returns all fields, with missing values as null. Rate limited to one request per second.
- Get Returning Visits Retrieves companies that visited a tracked domain more than once, indicating sustained interest across at least two separate sessions. Returns all fields, with missing values as null. Rate limited to one request per second, and often the stronger signal of the three.
- Get New Visits Retrieves companies visiting a tracked domain for the first time, where their earliest recorded session falls within the given range. The read for surfacing freshly identified accounts. Returns all fields with missing values as null, and is rate limited to one request per second.
Company Enrichment
7- Enrich Company Performs a synchronous CRM enrichment lookup for a single company by domain, returning firmographics including name, industry, NACE codes, employee count, revenue, address, contact details, and branding. A domain not yet crawled returns status processing, which should be retried in about fifteen seconds rather than treated as a failure; an unknown domain returns not_found. Uses the Enrichment API Key.
- Get Enrichment Lists Retrieves all CRM enrichment lists in the account, each list being a batch of company domains enriched in the background. Use it to find a list ID before adding domains, polling progress, or exporting rows. Uses the Enrichment API Key.
- Create Enrichment List Creates a new CRM enrichment list to which company domains can later be added, returning the created list including its ID. Uses the Enrichment API Key.
- Get Enrichment List Retrieves a single enrichment list by ID including its progress counters for total, done, and pending. Poll it after adding domains so a flow knows when background enrichment has actually finished instead of guessing at a wait time. Uses the Enrichment API Key.
- Add Domains To List Adds one or more company domains to an existing enrichment list, creating rows as pending and enriching them automatically in the background. Known domains typically complete within a minute; unknown domains are crawled on demand and take longer. Accepts up to 5000 domains per call. Uses the Enrichment API Key.
- Get Enrichment List Rows Retrieves the rows of an enrichment list as JSON, one row per domain with its enrichment status and enriched company data, optionally filtered by status so a flow can read only what has completed. Uses the Enrichment API Key.
- Export Enrichment List Downloads an enrichment list as CSV, returning raw CSV text with one row per domain and its enriched company columns. The right read when the destination is a spreadsheet or a bulk CRM import rather than a per-record write. Uses the Enrichment API Key.
Frequently Asked Questions
What can FlowRunner do with Webmetic?
FlowRunner agents can run Get Company Info, Get Company Sessions, and Get Company NACE Codes in Webmetic, plus 12 more actions.
Does connecting Webmetic to FlowRunner require OAuth?
No. Webmetic connects to FlowRunner with an API key, no OAuth flow required.
Can Webmetic trigger a FlowRunner workflow automatically?
Webmetic doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.
Start building with Webmetic
$100 in credits. No card required. Connect in minutes.