ICP Analysis

Use this skill when: user asks to "analyze ICPs", "identify target customers", "who should [company] sell to", or any request to understand buyer profiles for a business.

Input Gathering

Before analyzing, collect:

1. Company name (required)
2. Website (required - primary research source)
3. Product/service description (optional - will research if not provided)
4. Existing customers (optional - helpful for pattern recognition)

If the user doesn't provide website or description, use web search to find the company first.

Research Phase

Step 1: Website Analysis

Fetch the company website and extract:

• Products/Services: What exactly do they sell?
• Value proposition: What problem do they solve?
• Pricing signals: Enterprise, SMB, consumer?
• Case studies/testimonials: Who are their current customers?
• Industries mentioned: What verticals do they serve?
• Company sizes: Startup, mid-market, enterprise?
• Job titles: Who do they market to?

Step 2: Market Context

Search for:

• Competitors and their customers
• Industry reports or analyses
• Reviews and customer mentions
• LinkedIn company page (employee count, industry)

Step 3: Pattern Recognition

Look for signals:

• Direct fit: Companies that obviously need this product
• Adjacent fit: Companies with similar needs but different context
• Expansion opportunities: New markets or use cases
• Anti-patterns: Who would NOT be a good fit

ICP Framework

Determining Number of Tiers

The number of tiers should match the business reality. DO NOT default to 4 tiers—analyze the situation first.

Questions to determine tier count:

1. How many distinct buying personas exist? (Each major persona type = potential tier)
2. How different are sales cycles? (If cycles vary 3x+, likely need separate tiers)
3. Are there natural market segments? (Industry, size, geography, use case)
4. What's the product complexity? (Simple = fewer tiers, complex = more)
5. How wide is the TAM? (Narrow = 2-3 tiers, broad = 4+)

Tier Template

For EACH tier you create, define:

Tier [N]: [Descriptive Name] ([Priority Level])

• Profile: Who are they? (size, industry, stage, characteristics)
• Need intensity: Immediate/strong/moderate/emerging
• Buying behavior: Self-serve, sales-assisted, enterprise sale
• Sales cycle: Days/weeks/months
• Success indicators: What signals a good fit?
• Decision makers: Titles to target

Example Tier Structures

2-Tier Example (Developer Tool):

• Tier 1: Individual Developers — Self-serve, immediate need, low friction
• Tier 2: Engineering Teams — Sales-assisted, team licenses, longer cycle

3-Tier Example (Marketing SaaS):

• Tier 1: Growth-Stage Startups — Immediate need, fast decision, price-sensitive
• Tier 2: Mid-Market Companies — Strong need, committee buying, budget available
• Tier 3: Enterprise — Strategic deals, long cycle, high value

5-Tier Example (Enterprise Platform):

• Tier 1: Innovation Leaders — Early adopters, direct fit, fast cycle
• Tier 2: Fast Followers — Proven ROI needed, strong fit
• Tier 3: Industry Specialists — Vertical-specific needs, customization required
• Tier 4: Cost Optimizers — Replace existing solution, price-driven
• Tier 5: Strategic Partnerships — Platform/integration deals, complex

Priority Mapping

Note: Not all tiers need to exist for every business. Use only what makes sense.

Exclusion Criteria

Define who NOT to target:

Automatic Exclusions

Exclusion Keywords

List specific keywords/patterns that indicate non-fit:

• Industry terms that suggest wrong vertical
• Company name patterns (e.g., "consulting" if selling to product companies)
• Product categories that don't overlap

Scoring System (0-5)

Scoring Modifiers

Decision-Maker Titles

For each tier, identify relevant titles to target:

Executive (strategic deals)

• CEO, Founder, President, Owner
• CTO, CIO, COO, CFO
• VP of [relevant function]

Manager (operational deals)

• Director of [function]
• Head of [function]
• [Function] Manager

Practitioner (bottoms-up adoption)

• Senior [role]
• Lead [role]
• [Function] Specialist

Output Formats

Option 1: Methodology Document

Generate a markdown document like:

# ICP Methodology for [Company]

## Overview
[Brief description of the company and their product]

## Why [N] Tiers
[Explain the reasoning for the chosen number of tiers based on business model, sales motion, and market structure]

## Tier Definitions

### Tier 1: [Name] ([Priority Level])
**Score Range: [X-Y]**

[Description of who belongs here]

| Criteria | Examples |
|----------|----------|
| [Signal 1] | [Company types] |
| [Signal 2] | [Company types] |

**Why This Tier:** [Business rationale]
**Sales Motion:** [Self-serve / Sales-assisted / Enterprise]
**Decision Makers:** [Titles to target]

### Tier 2: [Name] ([Priority Level])
...

[Continue for each tier - may be 2, 3, 4, 5, or more]

## Exclusion Criteria
[Categories and keywords to skip]

## Scoring System
[Customized scoring with modifiers]

## Enrichment Priority
[Which tiers to enrich first and what to look for]

Option 2: Structured Data

Output as structured data for CSV import:

tier,tier_name,description,criteria,examples,decision_makers,priority_score_range
1,Direct_Buyers,Companies with immediate need,[criteria],[examples],[titles],4-5
2,Strong_Fit,Adjacent use cases,[criteria],[examples],[titles],3-4
...

Option 3: Both

Generate methodology doc AND structured data.

Quality Checklist

Before delivering:

• Tier count justified (explained why 2, 3, 4, or more)
• Each tier has clear, distinct criteria (no overlap)
• Exclusion criteria are specific (not generic)
• Scoring modifiers are relevant to this business
• Decision-maker titles match the buying process
• Examples are concrete, not theoretical
• Output format matches what user requested

Example Analyses

Example 1: Developer CLI Tool (2 Tiers)

Why 2 tiers: Simple product, binary fit (devs use it or don't), self-serve model.

Tier 1: Power Users — Individual developers already using similar tools, active on GitHub, blog about tooling Tier 2: Team Adoption — Engineering teams at companies >20 devs, standardizing toolchain

Exclusions: Non-technical roles, companies without engineering teams

Example 2: B2B SaaS Project Management (3 Tiers)

Why 3 tiers: Clear market segments by company size, different sales motions.

Tier 1: Growth Startups — VC-backed (50-200 employees), scaling fast, modern stack, hiring PMs Tier 2: Mid-Market — Established companies (200-1000), consolidating tools, budget approved Tier 3: Enterprise — Large orgs (1000+), complex procurement, long cycle, high ACV

Exclusions: Solo consultants, <20 employees, specialized industries (construction, film)

Example 3: Enterprise Data Platform (5 Tiers)

Why 5 tiers: Complex product, multiple buyer types, varying use cases, long sales cycles.

Tier 1: Data-First Companies — Tech companies with dedicated data teams, immediate need Tier 2: Digital Transformers — Traditional companies modernizing, strong executive buy-in Tier 3: Regulated Industries — Finance/healthcare with compliance needs, specific requirements Tier 4: Cost Consolidators — Companies replacing legacy tools, price-sensitive Tier 5: Platform Partners — ISVs/consultancies building on the platform

Exclusions: Companies <$10M revenue, no data team, industries we can't serve (defense)

This skill adapts to any business model. Let the customer's reality dictate the tier structure.