Issue 15 | Synthesis of User Interviews and Questionnaires (User Research Synthesis)

Updated on 4/5/2026

🎯 Learning Objectives

Upon completing this issue, you will master:

  1. How to use Claude Code to analyze user interview records across documents and extract core pain points
  2. Automate questionnaire data cleaning, classification, and thematic aggregation
  3. Generate structured user research reports
  4. Compress a full day's analysis work into a few minutes

📖 Core Concepts Explained

15.1 Pain Points in User Research Synthesis

Traditional Method (PM/UXR Manual):
  20 interview records × 30-60 minutes audio transcription each
  → Read and annotate each one
  → Manually categorize pain points
  → Create Affinity Diagram
  → Write research report
  Total time: 2-3 working days

Claude Code Method:
  Drop 20 documents into a directory
  → "Analyze all interviews, find Top 5 pain points"
  → Output complete report in 3-5 minutes

15.2 Analysis Methodology

Claude's User Research Analysis Workflow:

1. Batch Read (Glob + Read)
   Scan all interview documents / questionnaire CSVs

2. Thematic Coding
   Identify recurring keywords and sentiment

3. Affinity Clustering
   Group similar opinions under the same theme

4. Frequency Statistics
   Which pain points are mentioned most often? Which are mentioned by only 1 person?

5. Prioritization
   Output a priority matrix based on Frequency × Severity

6. Report Generation
   Output Markdown or HTML research report

💻 Simulate Claude TUI Interaction

Scenario: Automated Synthesis of 20 User Interviews

> There are 20 user interview text records in the interviews/ directory,
  Please analyze all of them to find core pain points and improvement directions

╭─ Claude ─────────────────────────────────────────────╮
│                                                      │
│  🔍 Tool: Glob → interviews/*.md                    │
│  Found 20 interview records                          │
│                                                      │
│  🔍 Tool: Read → (Reading 20 files one by one)      │
│                                                      │
│  📊 Analysis complete! Here is the comprehensive research report: │
│                                                      │
│  ══ Top 5 Core Pain Points (Sorted by Mention Frequency) ══ │
│                                                      │
│  1. 🔴 "Cannot find previous records" (mentioned by 17/20 people) │
│     Missing search function, users rely on manual searching │
│     Representative quote: "Every time I need to find last month's data │
│     I have to flip through several pages, it's too painful" — User #7 │
│                                                      │
│  2. 🔴 "Approval process is too slow" (mentioned by 14/20 people) │
│     Average wait time 3.2 days, users expect < 1 day │
│     Representative quote: "Urgent orders getting stuck in the approval process │
│     significantly impacts delivery" — User #12       │
│                                                      │
│  3. 🟡 "Poor mobile experience" (mentioned by 11/20 people) │
│     Tables and charts cannot be viewed properly on mobile phones │
│                                                      │
│  4. 🟡 "Too many and messy notifications" (mentioned by 8/20 people) │
│     Unable to distinguish important notifications from regular reminders │
│                                                      │
│  5. 🟢 "Hope to have data export" (mentioned by 6/20 people) │
│     Need to export reports to Excel                  │
│                                                      │
│  ══ User Persona Distribution ══                     │
│  • Heavy Users (Daily Active): 8/20 (40%)            │
│  • Moderate Users (Weekly Active): 9/20 (45%)        │
│  • Light Users (Monthly Active): 3/20 (15%)          │
│                                                      │
│  ⚙️ Tool: Write → research_report.md                 │
│  ✅ Complete report generated (incl. quote sources, frequency table, recommendations) │
│                                                      │
╰──────────────────────────────────────────────────────╯

💻 Code Demo

# ✅ Batch analyze interview records
claude "Read all files in the interviews/ directory,
  perform Thematic Analysis,
  output Top 10 user pain points and frequency statistics"

# ✅ Questionnaire data analysis
claude "Analyze survey_responses.csv,
  1. NPS score distribution and trends
  2. Keyword word cloud for open-ended questions
  3. Satisfaction differences grouped by user role"

# ✅ Competitor analysis synthesis
claude "Read user reviews in the competitor_reviews/ directory,
  compare review differences between our product and competitors'"

# ✅ Generate shareable research report
claude "Based on the analysis results,
  generate a 1-page research summary suitable for executives,
  including: core findings, priority matrix, recommended actions"

🔧 Involved Tools

Tool Research Stage Purpose
Glob File Discovery Scan interview document directory
Read Data Reading Batch read interview records
Bash Data Processing Run Python for statistical analysis
Write Report Generation Output Markdown/HTML research report

📝 Key Takeaways from this Issue

  1. Claude Code can extract valuable insights from batch unstructured text
  2. The key is to guide with correct methodological vocabulary: Thematic Analysis, Affinity Clustering, Frequency Statistics
  3. Always ask Claude to provide quote sources for verification
  4. Research reports can be directly output in Markdown and shared with the team
  5. Suitable for PM / UXR rapid research synthesis scenarios

🔗 References