Skip to main content
skill CX Codex

user-research-analysis — Claude Code skill: interview synthesis

A skill that turns interview transcripts and survey answers into themes, pain points and quotes you can act on.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add aj-geddes/useful-ai-prompts --skill user-research-analysis -a claude-code
Open on GitHub

What user-research-analysis does

This skill handles the messy middle of research work: raw transcripts in, structured findings out. The bundled Python example defines a ResearchAnalysis class with synthesize_interviews, identify_themes, extract_key_quotes, identify_pain_points and identify_opportunities. Themes get ranked by how often they show up, which keeps one loud participant from setting the agenda.

Four reference guides sit in the references/ directory: synthesis methods, affinity mapping, insight documentation and a research validation matrix. The skill also lists practices worth following — triangulate across sources, keep quotes and evidence attached, separate what you observed from what you concluded.

It is explicit about the failure mode too: don't over-read small samples, don't cherry-pick quotes that fit a hypothesis you already hold. Three interviews and a strong opinion is still three interviews. Runs in Claude, Codex, Cursor, GitHub Copilot, Gemini CLI and OpenCode.

What it helps with

  • collapse a stack of interview transcripts into recurring themes
  • surface user pain points from call recordings
  • keep findings separate from your own interpretation
  • back a design decision with quotes and evidence
  • check a hypothesis against more than one data source

Example prompts for the agent

  • synthesize these interview transcripts and rank the themes by frequency
  • pull the pain points out of these survey responses and cite the quotes
  • build a validation matrix from these findings

How to install user-research-analysis

1Skills CLI (Vercel)

npx skills add aj-geddes/useful-ai-prompts --skill user-research-analysis -a claude-code

2Manually: put the files in place

~/.claude/skills/user-research-analysis/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/user-research-analysis/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about user-research-analysis

How do I install user-research-analysis in Claude?

Run npx skills add aj-geddes/useful-ai-prompts --skill user-research-analysis -a claude-code. The skill file then lives at ~/.claude/skills/user-research-analysis/SKILL.md.

What does user-research-analysis do?

It analyzes qualitative and quantitative research data, extracting themes, quotes, pain points and opportunities, then turns them into design recommendations.

Which platforms does it support?

Claude, Codex, Cursor, GitHub Copilot, Gemini CLI and OpenCode. Each platform has its own install command and its own skills directory.

Source text

SKILL.md · MIT
---
name: user-research-analysis
description: >
  Analyze user research data to uncover insights, identify patterns, and inform
  design decisions. Synthesize qualitative and quantitative research into
  actionable recommendations.
---

# User Research Analysis

## Table of Contents

- [Overview](#overview)
- [When to Use](#when-to-use)
- [Quick Start](#quick-start)
- [Reference Guides](#reference-guides)
- [Best Practices](#best-practices)

## Overview

Effective research analysis transforms raw data into actionable insights that guide product development and design.

## When to Use

- Synthesis of user interviews and surveys
- Identifying patterns and themes
- Validating design assumptions
- Prioritizing user needs
- Communicating insights to stakeholders
- Informing design decisions

## Quick Start

Minimal working example:

```python
# Analyze qualitative and quantitative data

class ResearchAnalysis:
    def synthesize_interviews(self, interviews):
        """Extract themes and insights from interviews"""
        return {
            'interviews_analyzed': len(interviews),
            'methodology': 'Thematic coding and affinity mapping',
            'themes': self.identify_themes(interviews),
            'quotes': self.extract_key_quotes(interviews),
            'pain_points': self.identify_pain_points(interviews),
            'opportunities': self.identify_opportunities(interviews)
        }

    def identify_themes(self, interviews):
        """Find recurring patterns across interviews"""
        themes = {}
        theme_frequency = {}

        for interview in interviews:
            for statement in interview['statements']:
                theme = self.categorize_statement(statement)
                theme_frequency[theme] = theme_frequency.get(theme, 0) + 1

        # Sort by frequency
// ... (see reference guides for full implementation)
```

## Reference Guides

Detailed implementations in the `references/` directory:

| Guide | Contents |
|---|---|
| [Research Synthesis Methods](references/research-synthesis-methods.md) | Research Synthesis Methods |
| [Affinity Mapping](references/affinity-mapping.md) | Affinity Mapping |
| [Insight Documentation](references/insight-documentation.md) | Insight Documentation |
| [Research Validation Matrix](references/research-validation-matrix.md) | Research Validation Matrix |

## Best Practices

### ✅ DO

- Use multiple research methods
- Triangulate findings across sources
- Document quotes and evidence
- Look for patterns and frequency
- Separate findings from interpretation
- Validate findings with users
- Share insights across team
- Connect to design decisions
- Document methodology
- Iterate research approach based on learnings

### ❌ DON'T

- Over-interpret small samples
- Ignore conflicting data
- Base decisions on single data point
- Skip documentation
- Cherry-pick quotes that support assumptions
- Present without supporting evidence
- Forget to note limitations
- Analyze without involving participants
- Create insights without actionable recommendations
- Let research sit unused

© aj-geddes · MIT · aj-geddes/useful-ai-prompts

Similar skills

Everything for Science and research →

From the same set118

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor