Findings go where the work happens.
Send a study, or the findings you choose, to the tools your team already uses. Each finding takes its evidence with it, and the tracker’s status comes back.
- Nine destinations
- Jira, Linear, Notion, Confluence, Miro, Figma, Slack, Microsoft Teams and signed webhooks.
- Evidence travels with it
- Each finding from a study carries up to three captioned screenshots, steps to reproduce and acceptance criteria.
- Status comes back
- A Jira or Linear issue shows its state on the finding it came from: UX-3 · In Progress.
- Coding agents over MCP
- An agent in your editor fetches the findings, claims the ones it takes and reports back.
What leaves, and what comes back
Pick a capability to see it drawn from a sample study.
Documents to Notion or Confluence, issues to Jira or Linear, a summary to Slack or Teams, a board to Miro, a comment to Figma. Each tool gets the part it is for.
Where a finding goes next
A tracker, a design file, or an editor.
Findings land in Jira under an epic, screenshots attached
Choose the findings and send once. Each becomes an issue under the study version’s epic, with its pictures, steps to reproduce and acceptance criteria. A finding sent before is not made twice; when its numbers move, the issue gets a comment saying what changed.
A comment pinned to the frame where it happened
In a study of a Figma prototype, a finding about one screen can go to Figma as a comment on that frame: the basis, the headline, where people stopped and a link back.
An agent claims the fixes and reports back
Connect your editor once from Settings, then ask it to fix the findings from your latest study or audit. It claims them, makes the change, and records a fix submitted with its commit — or a dismissal with its reason.
Findings that keep their evidence
- UX-3 · In Progress ↗
Status without switching tabs
The tracker’s state sits on the finding, so the results page says what has moved.
- 3 screenshots · steps · criteria
Evidence that travels
At most three pictures a finding, each captioned with its number, so the issue argues for itself.
- claim_findings
No fix done twice
An agent leases what it takes. Another agent sees it claimed and moves on.
Nine destinations and your editor
Each connection is a token or a webhook URL you make in the tool itself.
Trackers
Jira and Linear. Issues under an epic or project per study version, with their pictures, and their status back on the finding.
Documents
Notion and Confluence. A page per study version with the summary, numbers, findings and heatmaps. Notion also keeps a findings database.
Boards and design
Miro gets a frame with a sticky note per finding and its heatmaps. Figma gets a comment pinned to the frame.
Chat
Slack and Microsoft Teams. A study summary when you send one, and alerts when a study closes or a run finishes.
Anything else
A signed webhook carries study events and the findings you chose, with their evidence, to any endpoint that accepts one.
Coding agents
Claude Code, Claude Desktop, Cursor, Codex, Zed, Gemini CLI and VS Code, over MCP. Cursor has an “Add to Cursor” link.
Where the findings come from
- Evidence and reportsFindings ranked, every claim tied to the moment it happened.Present · PDF · share link
- UX AuditOne screen, one goal — ranked findings, a Markdown work order and CSV.No testers required
- Performance testingAPI suites, load tests and page speed, with alerts to Slack, Teams or a signed webhook.API · load · page speed
Before you connect anything
How do I connect a tool?
Paste a credential you create in the tool itself: a webhook URL for Slack, Teams or your own endpoint, a token or API key for Jira, Linear, Notion, Confluence, Miro or Figma. It is stored sealed and never sent back to a browser. Jira and Linear issues are created under the account of whoever connected them.
What does a finding carry?
Its basis (“8 of 24 sessions”), the problem, up to three captioned screenshots, steps to reproduce, a suggested fix, acceptance criteria and a link back. No tester names or emails.
Can a coding agent fix findings?
Yes. Make a token in Settings → Coding agents and add uTestMe as an MCP server. The agent reads UX Audit and study findings, claims the ones it takes, and reports each as a fix submitted or dismissed. Findings from a Figma study can be read but not claimed — there is no code behind them.
My tracker isn’t listed.
Connect a webhook. It receives a signed event with the findings you chose and their evidence, so anything that accepts a webhook can take it from there.
Test from your pocket
The uTestMe tester app runs a study on a phone. Taps and screens recorded, consent first, results in the same view.
- Tap tracking
- Screen recording
- Consent first