Market Research Skill
Professional market research and competitive analysis using McKinsey/BCG/Bain methodologies. Generates hypothesis-driven research with PPTX presentation deliverables.
Overview
This Claude skill implements proven consulting methodologies for market research, structured as a four-level pyramid analysis:
YOUR COMPANY (Level 1)
Strategic Fit & Positioning
▲
COMPETITIVE LANDSCAPE (Level 2)
Positioning & Differentiation Analysis
▲
CUSTOMER INSIGHTS (Level 3)
Needs, Motivations, Barriers, Behaviors
▲
MARKET & SEGMENT ANALYSIS (Level 4)
Size, Growth, Trends, Development Opportunities
Features
✅ Hypothesis-driven methodology - MECE principle, Issue Trees, 7-step consulting process ✅ Four-level pyramid analysis - Systematic bottom-up market research ✅ CRAAP source validation - Professional-grade source quality assessment ✅ Multi-source triangulation - Validate critical findings with 2-3 independent sources ✅ Free sources optimization - Designed for government, academic, and free industry data ✅ PPTX generation - Professional presentations via document-skills/pptx integration ✅ Quick scan timing - Optimized for 1-2 day research projects
Quick Start
1. Installation
# Clone or copy this skill to your Claude skills directory
cp -r market-research ~/.claude/skills/
# OR for project-specific installation
cp -r market-research .claude/skills/
2. Initialize New Research Project
cd .claude/skills/market-research
bash scripts/init-research.sh "acme-corp" "healthcare"
This creates:
- Project structure with templates
- Pre-populated research brief
- Pyramid analysis framework
- Source bibliography tracker
3. Execute Three-Phase Workflow
Phase 1: Research Philosophy (30 min)
- Complete
research-brief.md - Define hypothesis tree using MECE
- Identify source strategy
Phase 2: Pyramid Analysis (1-2 days)
- Level 4: Market & segment data collection
- Level 3: Customer insights research
- Level 2: Competitive landscape analysis
- Level 1: Company positioning assessment
- Document in
pyramid-analysis.md
Phase 3: PPTX Generation (1-2 hours)
- Update
slide-deck.htmlwith findings - Run:
node scripts/generate-slides.js deliverables/slide-deck.html - Invoke
pptxskill for final conversion
Directory Structure
market-research/
├── SKILL.md # Main skill definition
├── LICENSE.txt # MIT License
├── README.md # This file
├── references/ # Methodology documentation
│ ├── MBB_METHODOLOGY.md # 7-step, MECE, Pyramid Principle
│ ├── FREE_SOURCES_GUIDE.md # Tier 1-3 data sources
│ └── VALIDATION_FRAMEWORKS.md # CRAAP test, triangulation
├── templates/ # Project templates
│ ├── research-brief.md # Phase 1 hypothesis setup
│ ├── pyramid-analysis.md # Phase 2 data collection
│ ├── source-bibliography.md # Source tracking
│ └── slide-deck.html # Phase 3 PPTX template
├── scripts/ # Helper automation
│ ├── init-research.sh # Project initialization
│ ├── generate-slides.js # PPTX generation wrapper
│ └── validate-sources.py # Automated CRAAP scoring
└── assets/ # Visual resources
├── color-scheme.json # Professional color palette
└── pyramid-diagram.svg # Pyramid visualization
Methodology Details
MBB 7-Step Process
- Define Problem - SMART criteria, success metrics
- Structure Problem - MECE decomposition, hypothesis trees
- Prioritize - Impact vs. ease, 80/20 rule
- Work Plan - Timeline, sources, responsibilities
- Execute Analysis - Systematic testing, triangulation
- Synthesize Findings - Pyramid Principle, "Day 1 drafts"
- Develop Recommendations - Actionable with ownership
Source Quality Hierarchy
Tier 1 (Highest Credibility):
- Government agencies (Census, BLS, Eurostat)
- Peer-reviewed academic journals
- Public company SEC filings
- Established market research firms
Tier 2 (High Credibility):
- Industry associations
- Premium databases (IBISWorld, Euromonitor)
- Think tanks and research centers
- Consulting firm public research
Tier 3 (Requires Verification):
- Company white papers
- Industry blogs and media
- Third-party databases
- Must be cross-validated with Tier 1-2
CRAAP Test Framework
Every source evaluated on 5 dimensions (1-5 scale):
- Currency - Publication date and timeliness
- Relevance - Fit to research question
- Authority - Author credentials and reputation
- Accuracy - Methodology and verification
- Purpose - Intent and potential bias
Quality thresholds:
- 22-25: Excellent (use as primary)
- 18-21: Good (use confidently)
- 14-17: Acceptable (use with caution)
- <14: Questionable or reject
Dependencies
Required
- Claude Code - This skill runs within Claude environment
- Bash - For init-research.sh script (Unix/Linux/Mac)
- Python 3.6+ - For validate-sources.py script
- Node.js 14+ - For generate-slides.js script
Optional (for PPTX generation)
- document-skills/pptx - From Anthropic skills repository
# Install as git submodule
cd .claude/skills
git submodule add https://github.com/anthropics/skills document-skills
Note: PPTX generation can also be done by invoking the pptx skill directly in Claude Code without local installation.
Usage Examples
Example 1: Quick Market Sizing
User: "Research the US plant-based protein market size and growth"
Claude (with market-research skill):
Phase 1: Creating hypothesis tree for market sizing
- TAM estimation (top-down from food industry)
- SAM focus on plant-based segment
- Growth drivers identification
Phase 2: Data collection
- Level 4: Census food manufacturing data, USDA reports
- Level 4: Academic papers on plant-based trends
- Level 4: Industry association (GFI) reports
Phase 3: Synthesis
- Market size: $7.4B (2024)
- CAGR: 11.4% (2020-2024)
- Key trends: Health consciousness, sustainability
- Sources: USDA, Good Food Institute, Euromonitor
Deliverable: PPTX with market sizing analysis
Example 2: Competitive Analysis
User: "Analyze top 5 competitors in enterprise SaaS project management"
Claude (with market-research skill):
Phase 1: Hypothesis - Market dominated by established players
Phase 2: Analysis
- Level 4: $6.5B market, 12% CAGR
- Level 3: Customer need for integration (78% priority)
- Level 2: Top 5 identified (Monday, Asana, Jira, Clickup, Smartsheet)
- Level 2: Positioning map created
- Level 1: Gap analysis vs. client capabilities
Phase 3: PPTX with competitive landscape and positioning recommendations
Example 3: Customer Insights
User: "Research B2B SaaS buyer decision criteria"
Claude (with market-research skill):
Phase 1: Hypothesis tree on decision factors
Phase 2: Research
- Level 3: Academic papers on B2B buying behavior
- Level 3: Industry surveys (Gartner, Forrester previews)
- Level 3: Case studies and analyst reports
Findings:
1. ROI justification (87% cite as critical)
2. Integration capabilities (76%)
3. Support/training (68%)
4. Security/compliance (65%)
Deliverable: Customer insights deck with decision journey map
Best Practices
1. Start with Hypothesis
❌ Wrong: "Let me collect all available data on healthcare market" ✅ Right: "Hypothesis: Telehealth market will grow 15%+ due to aging population and tech adoption"
2. Apply MECE Ruthlessly
❌ Wrong: Overlapping segments (SMB, Mid-market, Enterprise, B2B) ✅ Right: Mutually exclusive (B2B SMB, B2B Enterprise, B2C)
3. Triangulate Critical Findings
❌ Wrong: Single source for market size: "$4B market" ✅ Right: Three sources converge: $4.0B, $4.2B, $4.1B → High confidence: $4.1B
4. Document Everything
❌ Wrong: "Market growing fast" (no source) ✅ Right: "Market growing 12% CAGR (Source: IBISWorld 2024, CRAAP: 23/25)"
5. Continuous Synthesis
❌ Wrong: Collect all data first, then analyze at end ✅ Right: "Day 1 draft" immediately, refine continuously
Common Pitfalls
Pitfall 1: Accepting Sources at Face Value
Solution: CRAAP test every source, score ≥18 for primary evidence
Pitfall 2: Cherry-picking Data
Solution: Document comprehensive search, explain exclusions
Pitfall 3: Ignoring Original Context
Solution: Understand data collection purpose and methodology
Pitfall 4: Outdated Information
Solution: Check publication dates, apply growth rates to update
Pitfall 5: Unclear Methodology
Solution: Reject sources without transparent methodology
Integration with document-skills/pptx
HTML → PPTX Requirements
The slide-deck.html template follows html2pptx format:
✅ Correct dimensions: 720pt × 405pt (16:9)
✅ Semantic HTML: All text in <p>, <h1>-<h6>, <ul>, <ol>
✅ Professional colors: No purple gradients (#6366F1, #8B5CF6)
✅ Proper structure: No text directly in <div> or <span>
PPTX Generation Workflow
- Complete pyramid-analysis.md with findings
- Update slide-deck.html with data from analysis
- Run validation:
node scripts/generate-slides.js deliverables/slide-deck.html - Invoke pptx skill with HTML file
- Review generated PPTX for quality
Quality Checklist
Before delivering research to client:
- Research brief demonstrates clear hypothesis-driven approach
- All four pyramid levels analyzed with CRAAP-validated sources
- Critical findings triangulated with 2-3 independent sources
- Every quantitative claim backed by cited source
- Source bibliography complete with validation scores
- Pyramid analysis document complete
- PPTX slides generated and reviewed
- Professional design standards met (no AI slop)
- Slide narrative flows logically from market to company
- Recommendations are specific, actionable, and prioritized
- Final deliverable matches McKinsey/BCG/Bain quality standards
Troubleshooting
Issue: Scripts don't run
Solution:
# Make scripts executable
chmod +x scripts/*.sh
chmod +x scripts/*.py
# Ensure dependencies installed
python3 --version # Should be 3.6+
node --version # Should be 14+
Issue: PPTX generation fails
Solution:
- Verify HTML has correct dimensions (720pt × 405pt)
- Check all text is in semantic tags (
<p>,<h1>, etc.) - Ensure document-skills/pptx is available
- Remove # prefix from colors when using PptxGenJS
Issue: Sources score low on CRAAP
Solution:
- Prioritize Tier 1 sources (government, academic)
- Verify publication dates within 3 years
- Check methodology transparency
- Cross-validate with additional sources
Issue: Can't find data for market sizing
Solution:
- Try bottom-up estimation (components → total)
- Use proxy markets and apply ratios
- Leverage Census data and industry associations
- See FREE_SOURCES_GUIDE.md for discovery strategies
Contributing
To improve this skill:
- Fork and make changes
- Test with real research projects
- Submit pull request with:
- Description of improvement
- Example usage
- Documentation updates
Support
- Documentation: See
references/for detailed methodology guides - Issues: Report bugs or feature requests on GitHub
- Questions: Consult MBB_METHODOLOGY.md for process questions
License
MIT License - See LICENSE.txt for details
This skill implements publicly documented methodologies from top consulting firms. No proprietary client work or confidential information is included.
Acknowledgments
Based on methodologies from:
- McKinsey & Company (MECE, Pyramid Principle, 7-step process)
- Boston Consulting Group (Growth-Share Matrix, frameworks)
- Bain & Company (Results-oriented approaches)
Inspired by:
- "The McKinsey Way" by Ethan Rasiel
- "The Pyramid Principle" by Barbara Minto
- "Research Methods for Business Students" by Saunders, Lewis & Thornhill
- ESOMAR Guidelines on Secondary Data (2021)
- MRS Code of Conduct (2023)
Integration with:
- Anthropic document-skills/pptx for presentation generation
Version: 1.0 Last Updated: 2025 Status: Production Ready
For latest updates: Check repository for new versions and enhancements.

