Prompt library
Synthesize user interview notes
Requiring two sources per theme and quoting verbatims keeps the synthesis grounded in what people said rather than what the model expects users to say. Keeping contradictions visible instead of averaged is what makes the document trustworthy in a debate.
Last reviewed July 17, 2026
The prompt
Synthesize the interview notes below into findings.
{{notes}}
Method:
1. Extract every distinct observation as a one-line note tagged with the participant it came from.
2. Cluster the notes into themes. A theme needs at least 2 participants; single-source observations go in an "outliers" list, not a theme.
3. For each theme: name it in the participants' vocabulary, count supporting participants, quote the single strongest verbatim line, and state the design or business implication.
4. List contradictions between participants explicitly; do not average them away.
5. End with the 3 questions the next round of interviews must answer.
Do not propose solutions; this is a findings document.Run in idaptOpens a new chat with the prompt prefilled. Nothing sends until you press send.
Fill in the variables
| Variable | What it is | Example |
|---|---|---|
| {{notes}} | Raw notes or transcripts from all interviews | [paste notes from each interview, labeled P1, P2, ...] |