How to transcribe user research interviews and usability sessions

Consent and personal information

• Use a written consent form that says what you record, who will see it, where it's stored and how long you'll keep it. Then ask again at the start so the yes is on tape.
• Zoom shows participants a recording notice, but say it out loud too.
• In Zoom's recording settings you can turn off "Display participants' names in the recording" so names don't appear in the video. It isn't retroactive.
• Use participant codes (P01, P02) instead of names when you rename speakers. Strip emails, phone numbers and company names from quotes you share.
• If a session might touch health information, note that Dolcenota is not HIPAA-compliant, so don't use it for protected health information.

Only the audio is analyzed

Dolcenota transcribes the audio track of a video. It never looks at the screen: no clicks, no heatmaps, no screen analysis. For a usability test, the transcript captures what the participant said while thinking aloud; pair it with your own observations of what they did. Ask participants to narrate ("I'm looking for the shipping cost"), because what they say is what ends up in the transcript.

From transcript to themes

The summary and the answers to your questions come from the transcript, but they aren't synthesis. Patterns, severity and recommendations are your call. Use the AI to find things, and your judgment to decide what they mean.

Free and built-in options

Zoom, Meet and Teams include transcripts on some paid plans, and dedicated research platforms often transcribe inside their own tools. If you have those, use them. Dolcenota is useful when your plan doesn't include transcripts, when you also run in-person sessions, when interviews are in Spanish or switch between English and Spanish, or when you want a plain TXT or Word file you control.