User interview synthesis with evidence-traced findings
Synthesises notes or transcripts from several user interviews into findings that each cite the participants and quotes behind them, with strength ratings, contradictions and the limits of what was learned.
Prompt in Research by AISkillHub Library
What it does
Interview synthesis is where research quietly turns into opinion: one vivid quote becomes "users want X". This prompt keeps every finding tied to the participants who support it, rates its strength by how many independent participants back it, and reports contradictions and gaps rather than smoothing them over.
**Use it when** you have notes or transcripts from roughly 4 to 15 interviews on a defined research question and need a synthesis you can share with product, design or leadership.
**What you get** - A short summary answering the research question as far as the evidence allows. - Findings ranked by strength, each with the participant IDs supporting it, two or three verbatim quotes, and the implication for the decision you are trying to make. - A participant-by-theme matrix so readers can check coverage. - Contradictions and outliers, with possible explanations drawn from participant info. - What the research did not answer, and suggested follow-up questions.
**What it will not do** - It will not quote anything that is not in the notes. Paraphrases are labelled as paraphrases. - It will not generalise to percentages of your user base. Interview counts are reported as counts ("3 of 5 participants"), not as population estimates. - It will not include names or identifying details; use participant IDs in your notes.
**Strength scale used:** strong (most participants, including different segments), moderate (several participants or one segment), weak (one or two participants). This is a reading aid, not a statistical measure.
The example file shows a filled run on onboarding research with five participants.
How to use it
## How to use
1. Open `prompts/user-interview-synthesis.md` and copy everything below the title line. 2. Replace each `{{variable}}` with your own content. Paste long material (notes, transcripts, tables) directly in place of the variable; keep the surrounding tags so the model can tell inputs apart. 3. Start a new chat in ChatGPT, Claude or Gemini and paste the filled prompt as your first message. 4. Read the output against the quality checks at the end of the prompt before you use or send it.
## Variables
- `{{research_question}}`: the question the interviews were meant to answer. - `{{interview_notes}}`: notes or transcripts for each interview, each starting with a participant ID like `P1`. - `{{participant_info}}`: relevant segment details per participant ID (role, plan, company size, how they signed up). No names. - `{{decisions}}`: the product or business decision this research should inform.
If you do not have a value for an optional variable, write `none` rather than deleting it, so the prompt still reads correctly.
## Example
`prompts/user-interview-synthesis.example.md` shows the prompt with every variable filled and a trimmed sample of the output, so you can see the expected shape before you run it on real material.
## Reusing it
To reuse it often, save the prompt (with the variables left in) as the instructions of a Claude Project, a custom GPT or a Gemini Gem, then paste only the inputs in each new chat.
Files
Save each file at the path shown, relative to your project root.