User Interview Synthesis Report

A structured synthesis report that transforms raw interview notes into clear themes, quotes, and actionable next steps.

Best For

•A UX researcher wrapping up a round of 10 customer interviews needs themes with supporting quotes instead of a pile of raw notes nobody on the team will read.
•A product manager building a case for a roadmap change needs unmet needs and contradictions clearly separated from the researcher's own interpretation.
•A design lead preparing a readout for stakeholders needs surprising findings called out explicitly, so the team doesn't just confirm what it already believed.

Prompt Template

Synthesize the following user interview transcripts or notes into a structured research report. Raw interview data: [INTERVIEW_DATA, paste notes or transcripts]. Number of interviews conducted: [NUMBER_OF_INTERVIEWS]. Research objective: [RESEARCH_OBJECTIVE]. Structure the synthesis as: (1) Methodology snapshot, participants, interview format, and dates, (2) Top themes, identify [NUMBER_OF_THEMES, e.g., 4–6] recurring themes, give each a clear name, and support each with 2–3 direct quotes from participants (anonymized as P1, P2, etc.), (3) Key tensions or contradictions, points where participants disagreed or expressed conflicting needs, (4) Unmet needs, specific needs that no current solution addresses well, (5) Surprising findings, 2–3 things that contradicted assumptions, (6) Recommended next steps, 3 specific research or design actions. Use neutral, objective language. Do not interpret beyond what the data supports.

Pro Tip

Instruct the model to flag any theme supported by only one participant, since synthesis reports quietly lose credibility when a single anecdote gets presented with the same weight as a pattern seen across five interviews.

Example Output

A sample of what this prompt produces once you fill in the placeholders.

Methodology snapshot: 8 interviews, 30 minutes each, conducted over two weeks, remote video format. Top theme: 'Trust in automation' appears in 6 of 8 interviews. P3: 'I still double-check anything the tool flags automatically.' P6: 'I'd trust it more if I could see why it made that call.' Tension: P2 wants more automation while P5 wants full manual override, both describing the same feature. Unmet need: no participant has a way to audit past automated decisions. Recommended next step: prototype an explainability panel and retest with the same cohort.