All products

Analyse user interviews with AI

Trace a product recommendation back to the people who described the problem. Keep recurring findings, isolated observations and gaps in the sample visible.

Analyse user interviews with AI after a round of research has left you with more notes than your team can read in one sitting. Suppose several customers mention an invitation email. One could not find it, another opened it but did not understand what to do. With discover-interview-synthesis, you can keep those observations distinct while investigating a common difficulty with onboarding.

The finished report gives readers a way back to the evidence. Quotations retain their wording and carry participant IDs. Recommendations describe which finding supports each action and how confident the team can be, given who was interviewed.

What it does

  • Group observations across interviews around the same research question.
  • Find quotations that illustrate a theme in the participant's own words.
  • Show how many different people mentioned an experience.
  • Distinguish an observed difficulty from an interpretation of its cause.
  • Identify missing perspectives before commissioning another round of research.

Install and use

  1. Open the skill in your agent. Supply the research question and authorised notes from at least three participants, using IDs and omitting identifying details.
  2. Ask “Which parts of the invitation flow caused trouble, and who reported each one?” Check the quotations against your notes.
  3. Review the proposed actions with your team. Include limitations alongside the findings when you share the report.

FAQ

Source

Made by Jonathan Prisant (Product on Purpose). Original skill: discover-interview-synthesis, published on GitHub under the Apache-2.0 licence.

View on GitHub