Set up a study, interview users, review evidence and Findings, then decide what to send to delivery. This illustrative example follows trial users who stop before creating their first report.
The Study defines what to learn. Interviews capture what users say and do. Evidence links to those moments; a Finding summarizes the pattern. Your team reviews the Finding before sending it to a tracker.
You → your agent
Choose one question about your users. For example: why do trial users stop before creating a report?
The brief, as you would type it
Trials keep stalling before anyone creates a report. Find out why — with evidence, not guesses.
MCP · CLI · Dashboard
Your agent drafts a study: the goal, how to meet users — silent observation, a task to try, or an open conversation — and the debrief questions it plans to ask. You review the plan before anything goes live, and you can see exactly what the participant is asked to agree to.
Study plan — ready for your review
Inside your product · Widget
With consent, the interviewer records voice and product interactions. Supported desktop browsers also allow screen recording; mobile interviews continue with voice and in-page events. During observation, the interviewer stays quiet.
Captured session — minute one of six
While context is fresh
After observation, a planned debrief asks about specific moments from the task and what the participant expected.
Planned debrief
Interviewer
You clicked New report three times, then left the page. What did you expect to happen?
Participant
I thought the report would show up in that list. Nothing changed, so I assumed the trial did not include it.
Source-linked
Everything the run learned comes back attached to its source — the participant's words, the observed behavior, or both — with the page path, the playback moment, and a confidence level. Participant words and interpretation stay in separate fields. If a card reads wrong, the recording is one click away.
Evidence card
Nothing changed, so I assumed the trial did not include it.
/reports · playback 00:19–00:52 · interview #14
Draft → reviewed → send
UserTold groups related Evidence into a draft Finding. You or your agent check the sources and summary, then mark it reviewed. Sending it to a tracker is a separate decision.
Draft Finding — under review
Report creation gives no feedback, reads as a trial limit
Pushed to Linear with sources attached
Linear or GitHub
Send a reviewed Finding to Linear for product triage or to GitHub for delivery. The issue includes the Finding and its Evidence links. In Linear, your team decides whether to accept it into delivery.
Closing the loop
When a linked Linear issue completes, UserTold resolves its linked Evidence. Later interviews may surface similar Evidence for review. GitHub handoff does not include completion sync.
Recurrence watch
Evidence resolved when the Linear issue completed
New interviews checked for the same stall
It interviews people who are already using your product, on the pages where you installed the script. You choose whom to invite; responses are not guaranteed.
During observation nothing interrupts, hints, or navigates for the user. Stuck moments are preserved as evidence, then explained in the debrief.
Review the Finding and its Evidence, then choose whether to send it to Linear for product triage or GitHub for delivery.
Review the quotes or observed behavior behind a Finding to check the interpretation.
The docs cover study design, the widget embed, methodology, and the agent surfaces.
Prepare a study in the dashboard or through your agent. Try the interview yourself before inviting users.