ICP research skill
Generate ideal customer profiles for B2B SaaS clients through systematic research and structured output.
Report structure
The final ICP report follows this numbered section order:
| Section |
Purpose |
| Header |
Research date, website, category, confidence score (1-5) |
| 1. Executive summary |
High-level synthesis of findings and strategic recommendations |
| 2. TAM analysis |
Market sizing with targeting strategy per layer (TAM/SAM/SOM/ICP) |
| 3. Firmographics analysis |
Geographic, industry, company segment patterns, and technographics |
| 4. Roles and personas |
Core use case, Champion deep-dive, Economic Buyer deep-dive, buying journey |
| 5. Negative ICP |
Who is NOT a fit, disqualification criteria, and red flags |
| 6. Customer proof points |
Named customers, outcomes, and evidence with URLs |
| 7. Voice of customer synthesis |
Language patterns, pain points, and outcome terminology |
| 8. ICP segment definitions |
Scoring matrix, in-market signals, segment deep-dives |
| 9. Intent signals and buying triggers |
Observable signals indicating purchase readiness |
| 10. Recommendations |
Prioritization and messaging by segment |
| 11. Data gaps |
Missing information and follow-up suggestions |
| 12. Source appendix |
All sources with access dates, URLs, and confidence levels |
Confidence score calculation: Count High/Medium/Low data points. Score 5 if >70% High, Score 4 if >50% High, Score 3 if mixed, Score 2 if >50% Low, Score 1 if >70% Low.
Sorting rules
Apply consistently across all tables:
| Dimension |
Sort order |
| Decision role |
Champion → Economic Buyer → User → Influencer |
| Company size |
Enterprise → Mid-market → SMB → Startup |
| Frequency |
Very high → High → Medium → Low |
| Confidence |
High → Medium → Low |
| Customer concentration |
High → Medium → Low |
| Priority |
1 → 2 → 3 → 4 |
| Industry presence |
Strong → Moderate → Emerging |
Workflow
The research runs in 3 phases. Read the premium reference for the full step-by-step.
Phase summary:
- Data extraction — discover key pages (customers, case studies, solutions, pricing, integrations, G2, LinkedIn) → extract raw data with URL+date per source → normalize attributes (geography, industry, company size, team size, tech stack)
- Analysis and synthesis — identify patterns per segment → build Champion + Economic Buyer deep-dives → identify negative ICP + intent signals → collect proof points → document technographics → calculate TAM with targeting strategy → identify ICP as highest-priority segment below SOM
- Structured output — generate the 12 numbered sections, apply sorting rules, include rich descriptions with URLs and dates
Input requirements
Required
- Website URL — primary company website
Optional (improves quality)
| Input |
Purpose |
| Case studies URL |
Direct link to case studies page |
| Testimonials URL |
Direct link to testimonials |
| Market context |
Category, competitors, GTM approach |
| Sales call notes |
Win/loss context, objections |
| Existing ICP docs |
Validate or expand current understanding |
Anti-hallucination guardrails
- Never invent customer names. Only cite publicly referenced customers.
- Quote verbatim. Use exact customer language in quotes.
- Mark confidence levels. Tag data as High/Medium/Low confidence.
- Cite sources with URLs and dates. Include URL and access date for every claim.
- Acknowledge gaps. Explicit "Not available" for missing data.
| Confidence |
Definition |
| High |
Direct from official source, verifiable |
| Medium |
Third-party source, multiple signals |
| Low |
Single indirect source, inferred |
Quality
Pre-delivery checklist (coverage / personas / segments / evidence): the premium reference.
1---2name: icp-research3description: "Builds an ideal customer profile from a company's own website: case studies, testimonials and solutions pages become personas, segments, pains and buying triggers, each backed by evidence. Use when asked for ICP research, customer personas or who a company sells to."4metadata:5 source: https://skillmd.com/skills/matteotitta/icp-research6---78# ICP research skill910Generate ideal customer profiles for B2B SaaS clients through systematic research and structured output.1112## Report structure1314The final ICP report follows this numbered section order:1516| Section | Purpose |17|---------|---------|18| Header | Research date, website, category, confidence score (1-5) |19| 1. Executive summary | High-level synthesis of findings and strategic recommendations |20| 2. TAM analysis | Market sizing with targeting strategy per layer (TAM/SAM/SOM/ICP) |21| 3. Firmographics analysis | Geographic, industry, company segment patterns, and technographics |22| 4. Roles and personas | Core use case, Champion deep-dive, Economic Buyer deep-dive, buying journey |23| 5. Negative ICP | Who is NOT a fit, disqualification criteria, and red flags |24| 6. Customer proof points | Named customers, outcomes, and evidence with URLs |25| 7. Voice of customer synthesis | Language patterns, pain points, and outcome terminology |26| 8. ICP segment definitions | Scoring matrix, in-market signals, segment deep-dives |27| 9. Intent signals and buying triggers | Observable signals indicating purchase readiness |28| 10. Recommendations | Prioritization and messaging by segment |29| 11. Data gaps | Missing information and follow-up suggestions |30| 12. Source appendix | All sources with access dates, URLs, and confidence levels |3132**Confidence score calculation:** Count High/Medium/Low data points. Score 5 if >70% High, Score 4 if >50% High, Score 3 if mixed, Score 2 if >50% Low, Score 1 if >70% Low.3334---3536## Sorting rules3738Apply consistently across all tables:3940| Dimension | Sort order |41|-----------|------------|42| Decision role | **Champion → Economic Buyer → User → Influencer** |43| Company size | Enterprise → Mid-market → SMB → Startup |44| Frequency | Very high → High → Medium → Low |45| Confidence | High → Medium → Low |46| Customer concentration | High → Medium → Low |47| Priority | 1 → 2 → 3 → 4 |48| Industry presence | Strong → Moderate → Emerging |4950---5152## Workflow5354The research runs in 3 phases. Read the premium reference for the full step-by-step.5556Phase summary:57581. **Data extraction** — discover key pages (customers, case studies, solutions, pricing, integrations, G2, LinkedIn) → extract raw data with URL+date per source → normalize attributes (geography, industry, company size, team size, tech stack)592. **Analysis and synthesis** — identify patterns per segment → build Champion + Economic Buyer deep-dives → identify negative ICP + intent signals → collect proof points → document technographics → calculate TAM with targeting strategy → identify ICP as highest-priority segment below SOM603. **Structured output** — generate the 12 numbered sections, apply sorting rules, include rich descriptions with URLs and dates6162---6364## Input requirements6566### Required67- **Website URL** — primary company website6869### Optional (improves quality)7071| Input | Purpose |72|-------|---------|73| Case studies URL | Direct link to case studies page |74| Testimonials URL | Direct link to testimonials |75| Market context | Category, competitors, GTM approach |76| Sales call notes | Win/loss context, objections |77| Existing ICP docs | Validate or expand current understanding |7879---8081## Anti-hallucination guardrails82831. **Never invent customer names.** Only cite publicly referenced customers.842. **Quote verbatim.** Use exact customer language in quotes.853. **Mark confidence levels.** Tag data as High/Medium/Low confidence.864. **Cite sources with URLs and dates.** Include URL and access date for every claim.875. **Acknowledge gaps.** Explicit "Not available" for missing data.8889| Confidence | Definition |90|------------|------------|91| High | Direct from official source, verifiable |92| Medium | Third-party source, multiple signals |93| Low | Single indirect source, inferred |9495---9697## Quality9899Pre-delivery checklist (coverage / personas / segments / evidence): the premium reference.