agentleFS
nisarthnisarth/claude-skills-store/skills/design/research-synthesis/SKILL.md

research-synthesis

Skill1 starsChanged 7 months agoMITView on GitHub

Synthesize user research into themes, insights, and recommendations. Use when you have interview transcripts, survey results, usability test notes, support tickets, or NPS responses that need to be distilled into patterns, user segments, and prioritized recommendations. Trigger phrases: "synthesize research", "research synthesis", "analyze user research", "find themes in", "research report", "distill findings", "pattern analysis", "insights from interviews", "usability findings", "research debrief"

28 other projects use a version of this, 6 of them edited.

Adopted by

Most recently taken up by

Used with

Projects with it also use competitive-analysis and brainstorming.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • AnthropicSkill

    competitive-analysis

    Framework for building competitive landscape decks — market positioning, competitor deep-dives, comparative analysis, strategic synthesis. Use when the user asks for a competitive landscape, competitor analysis, peer comparison, market positioning assessment, strategic review, or investment memo deck. Also triggers on "who are the competitors to X", "benchmark X against peers", "build a market map", or any request to systematically evaluate competitive dynamics across an industry.

    Also in 3 projects that use this

  • OpenAISkill

    brainstorming

    You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 26 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read through each source and extract atomic observations:

    line 39
  2. Distinctive:

    Use a predefined set of topic tags (or develop them as patterns emerge)

    line 52
  3. Distinctive:

    Let the data speak first: Start with observations, then themes, then insights.

    line 176
  4. Distinctive:

    Include counter-evidence: If most users had a problem but some did not, the exception is as informative as the rule.

    line 179
  5. Distinctive:

    Acknowledge what you do not know: Every study has limitations.

    line 181

“Common” is measured: kept by at least 6 in 10 of the 26 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Research Synthesis

Synthesize qualitative and quantitative user research data into structured themes, insights, and actionable recommendations. Process interview transcripts, survey results, usability tests, and feedback data.

Workflow

Step 1: Organize the Raw Data

Catalog and prepare the research inputs:

### Research Inventory

| Source | Type | Volume | Participants | Date |
|--------|------|--------|-------------|------|
| Interviews | Qualitative | [N] transcripts | [Segments] | [Range] |
| Usability tests | Qual + Quant | [N] sessions | [Segments] | [Range] |
| Survey | Quantitative + Open-ended | [N] responses | [Segments] | [Range] |
| Support tickets | Qualitative | [N] tickets | [All users] | [Range] |
| NPS comments | Qual + Quant | [N] responses | [Segments] | [Range] |

Before synthesizing, note:

  • What user segments are represented and which are missing?
  • What is the research objective or question being answered?
  • Are there any known biases in the data collection?

Step 2: Extract Observations

Read through each source and extract atomic observations:

### Observation [#]
**Quote/Data**: "[Verbatim quote or data point]"
**Source**: [Interview 3, Survey Q7, Ticket #4521]
**Participant**: [Segment, role, tenure]
**Topic**: [Category tag]
**Sentiment**: [Positive / Negative / Neutral / Mixed]
**Intensity**: [Strong / Moderate / Mild]

Code observations consistently:

  • Use a predefined set of topic tags (or develop them as patterns emerge)
  • Include both positive and negative observations
  • Preserve the user's language -- do not prematurely interpret

Step 3: Cluster into Themes

Group related observations into themes using affinity mapping:

### Theme: [Descriptive Name]
**Frequency**: Mentioned by [X] of [Y] participants ([%])
**Segments**: [Which user types]
**Severity**: [Critical / Major / Minor]

**Pattern**: [2-3 sentence summary of what you see across observations]

**Supporting evidence**:
1. "[Quote]" -- [Source, Segment]
2. "[Quote]" -- [Source, Segment]
3. "[Data point]" -- [Source]

**Counter-evidence**: [Any contradicting observations]
**Confidence**: [High / Medium / Low]

Prioritize themes by:

  1. Frequency x Impact: How often does it come up, and how much does it matter?
  2. Segment breadth: Does it affect one segment or many?
  3. Actionability: Can the team realistically address this?

Step 4: Derive Insights

Transform each theme into an actionable insight:

### Insight [#]: [Statement]

**What we observed**: [The pattern in the data]
**What it means**: [The interpretation and implication]
**Why it matters**: [Business or product impact]
**Confidence level**: [High / Medium / Low]
**Evidence base**: [N sources, N participants]

Insight quality checklist:

  • Grounded in data, not assumption
  • Specific enough to act on
  • Includes the "so what" for the business
  • Acknowledges uncertainty where it exists
  • Distinguishes between what users said and what they did

Step 5: Formulate Recommendations

### Recommendations

| Priority | Recommendation | Based On | Effort | Expected Impact |
|----------|---------------|----------|--------|----------------|
| P0 | [Action] | Insight 1, 3 | S/M/L | [Metric change] |
| P1 | [Action] | Insight 2 | S/M/L | [Metric change] |
| P2 | [Action] | Insight 4, 5 | S/M/L | [Metric change] |

For each recommendation:

  • Connect it to specific insights (traceability)
  • Estimate effort and expected impact
  • Suggest how to validate before full investment
  • Note any risks or trade-offs

Step 6: Compile the Synthesis Report

## Research Synthesis: [Study/Topic Name]
Date: [Date] | Lead Researcher: [Name]

### Executive Summary
[3-5 sentences: key findings, most impactful insights, top recommendation]

### Research Overview
- Objective: [What we wanted to learn]
- Methods: [How we gathered data]
- Participants: [Who, how many, segments]
- Timeline: [When conducted]

### Key Insights (Ranked)
1. **[Insight title]**: [One sentence]
2. **[Insight title]**: [One sentence]
3. **[Insight title]**: [One sentence]

### Detailed Findings
[Theme sections with evidence]

### Recommendations
[Prioritized action items]

### User Segments Identified
[Any emergent user segments or personas]

### Open Questions
[What we still do not know]

### Methodology and Limitations
[How data was collected, biases, gaps]

### Appendix
[Raw data summaries, participant demographics, full observation log]

Examples

Example 1: Usability Test Synthesis

User: "Synthesize results from 8 usability tests of our checkout flow." Response: Extract observations per task (cart review, shipping, payment, confirmation). Note task completion rates and time-on-task. Cluster issues by flow stage. Prioritize by severity (blocked vs. struggled vs. minor friction). Recommend fixes for each stage.

Example 2: Interview Synthesis

User: "I have 15 interview transcripts about how teams use our reporting feature." Response: Code each transcript for themes around usage patterns, pain points, workarounds, and desired features. Identify user segments (power users vs. occasional). Derive insights about unmet needs. Recommend features that address the highest-frequency, highest-impact gaps.

Example 3: Mixed-Method Synthesis

User: "Combine our NPS survey, 6 interviews, and support ticket data." Response: Normalize across sources (NPS is broad but shallow, interviews are deep but narrow, tickets skew negative). Triangulate: findings supported by multiple sources get higher confidence. Flag where sources disagree. Weight recommendations toward findings with the strongest cross-source support.

Best Practices

  1. Let the data speak first. Start with observations, then themes, then insights. Do not start with a hypothesis and look for confirmation.
  2. Distinguish observation from interpretation. "5 of 8 users clicked the wrong button" is an observation. "The button placement is confusing" is an interpretation. Present both, but label them.
  3. Quantify qualitative data. "Several users mentioned X" is weak. "7 of 12 participants (58%) independently mentioned X" is strong. Count and report frequencies.
  4. Include counter-evidence. If most users had a problem but some did not, the exception is as informative as the rule. Understanding why some users succeeded reveals the root cause.
  5. Use direct quotes. They bring findings to life and help stakeholders empathize with users. Paraphrase loses emotional weight and specificity.
  6. Acknowledge what you do not know. Every study has limitations. State them explicitly so stakeholders can calibrate confidence appropriately.
  7. Make it scannable. Stakeholders will not read a 30-page report. Lead with the executive summary, use clear headers, and put details in the appendix.
  8. Close the loop. Track whether recommendations are implemented and whether they have the expected impact. This builds credibility for future research.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.