Moderated usability testing, run by an AI moderator.
Hear the moderator before you read about it. Press the mic and it will probe you the way it probes a participant who gives a thin answer. It is not a human researcher, and this page is specific about where that matters.
You want to cancel the subscription on this account. Start from the homepage and do what you would normally do.
You paused there. What were you expecting to see?
Hear the moderator before you trust it with a task.
This demo is an interview, with no screen share and no task. It shows the thing that decides whether a session is worth reading: how the moderator probes an answer that stopped short.
- About 3 minutes
- Uses your microphone
- An interview, with no task
Always available. Never going to redesign your study mid-session.
Unmoderated tests scale, then go quiet at the exact moment something surprising happens. Moderated tests catch the surprise, but each one costs an hour of a researcher’s day, so you get as many as fit in the week.
This sits between the two. The AI moderator works through your task list, prompts people to think aloud, and probes a thin answer. It will not notice that the study is asking the wrong question, or that participant five is the wrong person entirely. That part is still yours.
Tasks in. Evidence out.
A usability test is a question type, so it sits in the same study as your screeners and your post-test questions.
Write the tasks
A scenario for each task.
Give each task a scenario and a starting URL, break it into steps if you want, and add a warm-up task if the flow needs one.
Run the session
Share screen, think aloud.
Respondents share their screen and talk through the task. The moderator prompts them to think aloud and probes, and it is told not to give hints.
Review the evidence
A clip for each task.
Each task returns its own clip, transcript, and ease rating, with synthesized pain points sitting on top of them.
Start from a task list someone already wrote.
Navigation, checkout, onboarding and account settings. Each has editable scenarios, post-task questions and a final debrief.
What it actually does during a session.
The moderator reads your scenarios in order, or randomized if you switch that on, and stays out of the way while someone works. After 20 seconds of silence it asks them to think aloud, and it probes thin answers.
It is briefed never to hint. A participant who cannot find the button is data; a participant who was told where the button is, is not.
- Reads your task scenarios in order, or randomized
- Prompts for think-aloud when someone falls silent
- Probes a thin answer instead of accepting it
- Asks a 1 to 7 ease rating after each task
- Told never to hint at the answer or the next step
Where screen share works.
On desktop, respondents share through the browser’s own screen capture, with nothing to install. On Android, the QuestionPunk app uses native screen sharing.
To control where people take the study, set Respondent surface in the survey settings to Web only, Android app only, or both, before you send the link.
- Desktop: the browser’s screen capture, no install
- Android: native screen sharing in the QuestionPunk app
- Screen-share sections set to Computer ask phone users to switch to a desktop browser
- The Respondent surface setting decides where a study can be taken
Where you still want a human in the room.
Use the AI moderator for the parts that repeat: the same five tasks, run thirty times, each transcribed and rated. That is the work that makes human moderation expensive and is the least interesting part of a researcher’s day.
Keep a human for sessions where the next question depends on judgment you haven’t written down yet. Be honest with yourself about which of your studies those are.
- Exploratory studies where the research question is still moving
- Sessions a stakeholder is watching live and will want to redirect
- Participants who need accommodation a script cannot anticipate
- Anything where the next probe depends on an unwritten judgment call
Evidence at the level of a single task.
Per-task video clips
Each task is cut into its own clip automatically when the recording stops.
Pain points with evidence
AI-detected pain points, each with a timestamp in the session it came from, plus themes that recur across participants.
Ease ratings and navigation
The 1 to 7 ease rating for every task. Sessions in the Android app also count revisits and back navigation.
Built for the people who watch the recordings.
Watch someone actually use it.Start on the Free plan.
Write the tasks, send the link, read the clips. Free includes 20 visible responses a month across your account, with no credit card.