Experiment design before the build
researchWrite the hypothesis, the metric, and the analysis before the test exists.
Interview synthesis without invented consensus
Turn several interview notes into themes that keep who said what.
When to use it
After a round of user interviews, before a product decision meeting.
The prompt
Copy into your assistant
You are synthesising interviews.
Input: notes or transcripts, labelled by participant.
Rules:
- Every theme must cite at least two participants; single-participant observations go in a separate list.
- Keep disagreement visible: if two participants contradict each other, report both.
- Do not merge quotes into a single narrative voice.
- Do not infer motivations that were not stated.
Output: Themes (with participants), Contradictions, Single-participant observations, Questions for the next round.
If a transcript is partial, say which topics cannot be assessed for that participant.What to change
Paste the interview guide too; themes that map back to the guide are easier to defend in a meeting.
Why it works
Synthesis tends to manufacture consensus by averaging. Requiring a participant count per theme keeps minority views visible, which is usually where the useful signal is.
More research prompts
- Experiment design before the build Write the hypothesis, the metric, and the analysis before the test exists.
- Source quality triage Rank a source list by authority and recency before writing from it.
- Competitor teardown from public pages Structure what a competitor says about itself, with quotes, and without speculation.