Qualtrics

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.

10 actions API key available

Qualtrics website ↗ Platform Documentation ↗ Capability data verified 2026-07-28

A support case closes in the service desk

Create Contact adds the customer to the XM Directory with their case reference as external data

The agent verifies the customer has not been surveyed within the fatigue window before queueing any invitation

The CX owner reviews the week's queued distribution, audience size and message, before it is scheduled

Create Distribution schedules the survey invitation to the mailing list with the approved send time

Start Response Export begins the pull of completed responses on the reporting schedule

Get Export Progress polls to completion and Download Export File retrieves the archive

Low scores route to the account owner within the same workflow run

What This Integration Enables

Qualtrics is where enterprises hear their customers, and the recurring failure is not in the listening, it is in the plumbing: responses trapped behind manual exports, invitations sent to whoever a list happened to contain, insights arriving after the moment to act. This connector puts the whole loop under FlowRunner orchestration. Agents manage XM Directory contacts, schedule distributions, run the asynchronous export flow to completion, and inject responses collected in external systems. The export flow deserves its own mention because it is genuinely three-step: Start Response Export returns a progressId, Get Export Progress polls to a fileId, Download Export File retrieves the archive. Async patterns like this are exactly what breaks casual automation and exactly what an orchestration engine handles natively, with waiting, retry, and failure branches built into the flow rather than bolted onto a script. - Response exports run on schedule with date filters, and the data lands in reporting tools the day it exists - Directory contacts are created from CRM, support, and commerce events with external references preserved - Distributions are scheduled with audience and timing decided by workflow logic and approved by a person - Survey definitions are readable at authoring depth, so downstream systems can map QIDs to meaning - External form submissions become Qualtrics responses via Create Response, unifying feedback in one dataset Survey invitations spend a finite resource, respondent goodwill, and that spend is where FlowRunner's [human-in-the-loop](/concepts/human-in-the-loop) gate sits.

Without FlowRunner

Survey data leaves weekly, by hand Someone exports responses to CSV every Friday, and every downstream report inherits that lag

Sampling is uncontrolled The same customers get surveyed after every interaction, and response rates decay from fatigue

Detractors wait in the dataset A scathing response sits unread until the monthly review, long after the save window closed

With FlowRunner

Exports run themselves The start-poll-download flow executes on schedule, and responses land in reporting systems the same day

The directory enforces the sampling rules Agents check contact history before inviting, so the fatigue window is policy, not hope

Responses become events Each export is scanned as it lands, and detractor scores page the account owner immediately

Use Case Scenarios

The relationship survey with a managed audience

Customers hitting their renewal quarter flow from the CRM into the XM Directory via Create Contact, each carrying an external reference back to their account. The agent checks the fatigue window, assembles the quarter's mailing list, and presents the audience to the CX owner. On approval, Create Distribution schedules the invitation. Responses export on schedule and reconcile back to accounts by reference. The program runs quarterly without anyone rebuilding lists, and no customer gets surveyed twice because two teams both wanted feedback.

The detractor who got a call before lunch

The morning export completes, Start Response Export through Download Export File, and the agent parses the archive. Scores below threshold trigger immediate action: the response summary and account context post to [Slack](/integrations/slack) for the account owner, and a follow-up task is created in [HubSpot](/integrations/hubspot). The rest of the responses append to [Google Sheets](/integrations/google-sheets) for trend reporting. The dataset still feeds the quarterly review, but the customers who needed a call today got one today.

One dataset, no matter where the answer was given

Feedback arrives through channels Qualtrics never sees: a [Typeform](/integrations/typeform) embedded in the product, a kiosk form, a partner portal. The agent maps each submission's fields to the survey's question IDs, discovered once via Get Survey Definition, and records it with Create Response, tagged with embedded data naming its source. Analysis runs on one dataset with one schema instead of three exports stitched in a spreadsheet, and the source tag keeps every record auditable.

Human-in-Loop Highlight

Create Distribution is the operation that spends what CX programs can least afford to waste: the customer's willingness to answer. A scheduled distribution sends real email to a real mailing list, and a wrong list, wrong survey, or duplicate schedule is not undone by apology, it is paid for in response rates for quarters. Create Response carries a subtler danger: it writes records directly into the research dataset, and a misfiring loop quietly corrupts the numbers that leadership treats as the voice of the customer. So FlowRunner gates the send and audits the write: every distribution's audience, message, and timing is approved by the CX owner before scheduling, and every Create Response carries a source tag in embedded data so injected records are forever distinguishable from organic ones. Exports, contact creation, and routing run at machine speed. The decision to ask thousands of customers for their time stays with a person who owns the relationship.

Agent processes routinely

Detects exception requiring judgment

Clear match Continues automatically

Ambiguous Routes to human via email

Human decides

Agent resumes with decision

Agent Capabilities

10 actions

Surveys

List Surveys Lists all surveys accessible to the API token, returning each survey's id, name, ownership, active status, and last modified date. Results are paginated by Qualtrics; pass the offset extracted from the previous response's nextPage URL to retrieve subsequent pages.
Get Survey Retrieves the metadata and summary configuration for a single survey by its ID, including name, status, expiration settings, and question count. For the full question/flow definition use Get Survey Definition instead.
Get Survey Definition Retrieves the complete survey definition, including all questions, blocks, survey flow, embedded data, and survey options. Use this when you need the full authoring-level structure of a survey rather than just its summary metadata.

Responses

Start Response Export Starts an asynchronous export of a survey's responses and returns a progressId. Because Qualtrics generates exports in the background, this is step one of a three-step flow: (1) Start Response Export, (2) poll Get Export Progress with the progressId until status is "complete" to obtain a fileId, then (3) Download Export File with that fileId. Supports optional date range filtering and label/recode formatting.
Get Export Progress Checks the progress of an in-flight response export using the progressId returned by Start Response Export. Returns percentComplete and status ("inProgress", "complete", or "failed"); when status is "complete" the response also includes a fileId to pass to Download Export File. Poll this until completion before downloading.
Download Export File Downloads the completed response export file identified by the fileId returned once Get Export Progress reports status "complete". Qualtrics returns a zip archive containing the responses in the requested format; this operation returns the raw file bytes. This is the final step of the start-poll-download export flow.
Create Response Records a new survey response directly via the API by supplying a map of question/embedded-data field IDs to answer values. Use this to programmatically submit answers (for example from an external form). The values object keys are Qualtrics question IDs (e.g. "QID1") or embedded data field names.

Contacts (XM Directory)

List Mailing Lists Lists the mailing lists in an XM Directory, returning each list's id, name, contact count, and owner. Mailing lists group directory contacts for distributions. Use the directory ID from your Qualtrics XM Directory (starts with POOL_).
Create Contact Creates a new contact in an XM Directory with name, email, and an optional external reference. The contact is added to the directory contact pool and can subsequently be added to mailing lists and used in distributions. Provide the directory ID (starts with POOL_).

Distributions

Create Distribution Creates an email distribution that sends a survey invitation to a mailing list from an XM Directory. Requires the survey ID, the sending message, the recipient mailing list, and a send time. Returns the created distribution's ID for tracking. Provide the message library, message, and sender details from your Qualtrics account.

Frequently Asked Questions

What can FlowRunner do with Qualtrics?

FlowRunner agents can run List Surveys, Get Survey, and Get Survey Definition in Qualtrics, plus 7 more actions.

Does connecting Qualtrics to FlowRunner require OAuth?

No. Qualtrics connects to FlowRunner with an API key, no OAuth flow required.

Can Qualtrics trigger a FlowRunner workflow automatically?

Qualtrics doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with Qualtrics

$100 in credits. No card required. Connect in minutes.

Start Building Free Talk to us