user-interview-synthesis

Synthesize user interview transcripts into evidence-graded findings and decision-ready research readouts.

1|Updated Jul 3, 2026
One-click install
npx skills add https://github.com/Nandansai08/skillz --skill user-interview-synthesis-nandansai08
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: user-interview-synthesis
Source: https://github.com/Nandansai08/skillz/tree/main/skills/product-pm/user-interview-synthesis
Command: npx skills add https://github.com/Nandansai08/skillz --skill user-interview-synthesis-nandansai08

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a batch of user interviews into trustworthy findings is hard: interpretation gets smuggled into evidence, vivid participants dominate themes, and small-sample opinions get dressed up as statistics. This Skill enforces a disciplined synthesis workflow that separates observation from interpretation and produces findings stakeholders can act on. ## Core Features & Use Cases - Two-layer evidence discipline: Extract verbatim, participant-tagged observations first, keeping interpretation strictly out of the evidence layer. - Bottom-up thematic coding: Cluster observations by affinity, tag each by tier (behavior, pain, preference, speculation), and keep participant IDs attached to every theme. - Bias hunting and graded findings: Actively surface contradicting evidence, grade each finding with n/N strength notation, and bridge every finding to a decision or follow-up question. - Use Case: After 9 interviews with stalled-trial users, code 214 tagged observations into five themes, discover the real churn driver is a permissions wall rather than onboarding confusion, and ship a one-page readout that rewrites the PRD problem statement. ## Quick Start Synthesize these interview transcripts into graded findings with a contradiction section and a one-page decision readout.

Frequently Asked Questions about user-interview-synthesis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I synthesize user interview notes into findings?▼

Start by extracting verbatim, participant-tagged observations with no interpretation, then cluster them bottom-up into themes. Grade each finding by evidence strength (behavior outweighs stated opinion) and attach a so-what implication linking it to a decision or follow-up question.

How many user interviews are enough for synthesis?▼

The workflow targets batches of 5-12 interviews, which is enough to establish that a pain exists and runs deep. Small-n findings must ship graded with n/N notation and never be presented as market-share percentages, which require survey-based quantification.

How do I avoid confirmation bias in research synthesis?▼

Actively list observations that contradict the team's hypothesis and give them their own section in the readout. Check whether repeated quotes are representative or merely quotable, and treat a readout that confirms everything the team believed as a warning sign requiring evidence.

When should I use a survey instead of interview synthesis?▼

Use surveys when you need to quantify how widespread a finding is, since interviews with small samples cannot support percentage claims. Interview synthesis establishes that a problem exists and its depth; survey design handles the quantification step downstream.

Why should observations be separated from interpretation in qualitative research?▼

Keeping the verbatim observation layer interpretation-free prevents downstream synthesis failures, since every major distortion comes from smuggled interpretation. Tagging observations to participant IDs also makes it mechanically visible when one vivid participant dominates a theme.