AI-moderated usability testing with real participants
Run Figma and website tasks with real participants in Versive, capture think-aloud feedback and screen recordings, and use AI follow-up questions.
AI-moderated usability testing lets real participants attempt a task on their own time while an AI interviewer asks the follow-up questions a researcher would. You get the screen recording, the participant's explanation, and adaptive probing, without scheduling a single session.
In Versive, a usability study combines a task, a quick rating, and an AI Question that digs into the experience. This guide covers testing with real people. To try a design with AI personas first, see Test a Figma prototype with AI personas or Run a usability test on a live website.
Pick the task type
| What you're testing | Question type | What you get |
|---|---|---|
| Figma prototype | Prototype Task | The screens participants visited, where they clicked, and whether they reached your goal screen |
| Live website or published prototype | Website Task | A recording of the attempt and the participant's think-aloud |
| Static design or concept | Image or video attached to a question | Reactions and ratings |
Prototypes built with Figma Make, v0, Lovable, Bolt, and similar tools are websites once published, so test them with a Website Task.
Write a realistic task
Give participants a goal, not instructions. For example:
You manage a small team's subscription. Find out how to add a colleague and what it would cost. Stop before you make a purchase.
Avoid naming the buttons or menus they should use. For a Figma prototype, set the screen that counts as success, and Versive tells you who completed the task.
Capture the screen and the think-aloud
Versive can record the participant's screen and voice while they work through the task, so you see where they hesitated and hear why. Let participants know what will be recorded, and use a test account with sample data if the task touches anything sensitive.
Take the study yourself before recruiting to make sure the task loads and the recording looks right.
Follow up while the experience is fresh
After the task, ask a short rating question such as "How easy or difficult was that?" Then add an AI Question:
Walk me through any point where you weren't sure what to do next.
The AI interviewer follows up on each answer, asking which screen caused trouble, what the participant expected, and what they tried instead. Add interviewer notes to steer the probing toward what your team needs to learn. See AI follow-up questions.
Recruit and review
Send the study to your own users or recruit through the built-in panels. See how to recruit research participants.
In Results, review each recording alongside the participant's rating and explanation. Insights pulls out recurring themes across participants, each linked to the quotes behind it. For the readout, pair each issue with the task, what you saw, a supporting quote, and a suggested fix.
Frequently asked questions
Can I run a usability test without a human moderator?
Yes. In Versive, participants complete a task on a Figma prototype or website, and the AI interviewer follows up on what happened: what confused them, what they expected, and what they did instead. There are no sessions to schedule.
Can Versive record the screen and what participants say?
Yes. Versive can capture the participant's screen and voice during a task, so you can watch what happened and hear them think aloud.
Is this the same as testing with synthetic users?
No. Here, real people complete the tasks. Versive also offers AI tests, where AI personas try your design before you recruit anyone. Many teams use both: AI personas first, then real participants to confirm what matters.
Full reference
Prototype testing
Keep reading
AI follow-up questions: probing without a moderator
How AI follow-up questions work in a Versive AI-moderated survey: how the interviewer probes, how to steer it, and when to cap follow-ups.
How to write screener questions that actually screen
Screener questions examples, the principles behind good ones, and how to wire screen-out logic in Versive so screeners actually screen.
Run exploratory AI interviews from a research goal
Turn a research goal into a focused exploratory interview in Versive, with a sample brief, moderator instructions, a pilot, and evidence review.
