When should I use a live prototype block?

Live prototype opens a working prototype from a public browser URL so the respondent can use it during the study. You set a task; screen recording is built in, and you can add a time guide and follow-up questions.

When to reach for it

Reach for this when you want people to use a whole prototype, not react to a still image or click one screen. It works with public browser URLs from tools such as Lovable, Replit, v0, and Bolt.

If you only want to know where people click on a single screen, use a First-click test. If you want reactions to an image, video, or audio clip rather than a hands-on session, use a Design survey.

Good to know

Set the public prototype URL and task in the editor. Recording is on by default. Before the task starts, Fiuto asks for recording consent. If the respondent declines or capture cannot start, they can continue without a recording. You can also add a time guide and follow-up questions. Check the URL in a private browser window before launch so you know respondents can open it.

Read what respondents said

A transcript lets you read and search what respondents said instead of scrubbing through the video. It appears in recording playback in results, only when the respondent agreed to it in the recording consent and a microphone was captured. It usually appears within two hours of upload. Some recordings show “A transcript isn’t available for this recording”, for example when the study’s language is not supported. Icelandic is the one study language not supported today.

Captions show on the video as it plays. To hide them, turn them off in the player’s captions control. The transcript panel groups what was said by task, and selecting a line jumps to that moment. Use “Download text” for a .txt file or “Download captions” for a .vtt file.

The in-app agent and MCP clients can read transcripts too, with the same access as playback: study owners and people with owner access. Editors and viewers cannot.

A transcript is deleted with its recording, including at the 12-month limit. Our speech-to-text provider processes the audio in the EU.