business-analysis

name

business-analysis

description

Deep research skill for analyzing businesses — existing companies or new ideas. Researches competitors via web search, grades the business against them, and produces a structured report using frameworks like Business Model Canvas, Hamilton Helmer's 7 Powers, Porter's Five Forces, and competitive positioning maps. Trigger on phrases like "analyze this business", "business analysis", "competitive analysis", "how does X compare to competitors", "is this business idea viable", "market analysis", or any request to evaluate a company's competitive position, market fit, or strategic strengths.

Business Analysis Skill

This skill guides Claude through a deep-research business analysis that produces a comprehensive, framework-backed report. It uses web search extensively to gather real competitor and market data before applying strategic frameworks.

---

Core Principles

1. Research before opining. Always search the web for real data — competitors, pricing, market size, funding, reviews — before filling in any framework.
2. Frameworks are lenses, not checklists. Use each framework to surface different insights. Skip sections that don't apply; go deep where they do.
3. Grade honestly. The user wants signal, not cheerleading. Be direct about weaknesses and existential risks.
4. Cite sources. When stating market data, competitor info, or claims, note where you found it.
5. Write incrementally. Update the report file after each research phase so the user always has a useful artifact.

---

Workflow

Phase 0: Intake

Ask the user enough to start researching. Batch these questions — don't over-interrogate before doing any work.

Essential (always ask what's missing):

• What is the business / idea? (name, URL, one-liner description)
• What problem does it solve and for whom?
• What's the revenue model (or intended model)?
• Is this an existing business you run/are evaluating, or a new idea?

If the user provides a URL or company name, skip questions they've already answered implicitly and move to research.

---

Phase 1: Competitive Landscape Research

Use WebSearch and WebFetch extensively. Search for:

1. Direct competitors — companies solving the same problem for the same customer segment
2. Indirect competitors — alternative solutions the customer might use (including "do nothing" or manual workarounds)
3. Market data — TAM/SAM/SOM estimates, growth rates, recent funding rounds in the space
4. Customer sentiment — reviews, complaints, Reddit/HN threads, G2/Capterra ratings for competitors
5. Pricing intelligence — competitor pricing pages, packaging, free tiers

Build a competitor table as you go:

Competitor	One-liner	Target Customer	Pricing	Key Differentiator	Weakness

Create the report file at this point and populate the competitor landscape section. Use the path the user specifies, or default to ./business-analysis-report.md.

---

Phase 2: Framework Analysis

Apply each framework using the research gathered. Not every framework will be equally useful — spend more time on the ones that yield real insight for this particular business.

2a. Business Model Canvas

Fill in all 9 blocks. Be specific — don't write generic filler.

Block	Description
Key Partners	Who do you depend on? Suppliers, platforms, channel partners
Key Activities	What must you do well to deliver the value proposition?
Key Resources	Assets required — IP, talent, data, capital, brand
Value Proposition	Why does the customer pick you over alternatives?
Customer Relationships	Self-serve, high-touch, community, automated?
Channels	How do customers discover and buy?
Customer Segments	Who specifically — be narrow, not "everyone"
Cost Structure	Major cost drivers — fixed vs variable
Revenue Streams	How money comes in — subscriptions, transactions, licensing

2b. Hamilton Helmer's 7 Powers

For each power, assess whether the business has it, could build it, or lacks it, and explain why. This is the most important framework for long-term defensibility.

Power	Definition	Assessment
Scale Economies	Unit costs decline with volume
Network Economies	Value increases with each user
Counter-Positioning	Incumbents can't copy you without damaging their own business
Switching Costs	Customers face real cost to leave
Branding	Reputation that justifies premium pricing
Cornered Resource	Exclusive access to a valuable asset (talent, data, IP, rights)
Process Power	Org capabilities built over time that are hard to replicate

Power Score: Summarize as a simple count — e.g., "2 powers present, 1 emerging, 4 absent" — and assess whether the business has enough defensibility to sustain margins long-term.

2c. Porter's Five Forces

Assess the attractiveness of the industry/market:

Force	Intensity (Low/Med/High)	Analysis
Threat of New Entrants		Barriers to entry, capital requirements
Bargaining Power of Suppliers		Concentration, switching costs
Bargaining Power of Buyers		Price sensitivity, alternatives
Threat of Substitutes		How easily could customers solve this differently?
Competitive Rivalry		Number of players, differentiation, growth rate

2d. Competitive Positioning Map

Describe (or suggest the user visualize) a 2x2 positioning map. Pick the two axes that matter most for this market. Common axis pairs:

• Price vs. Feature Depth
• Self-serve vs. High-touch
• SMB-focused vs. Enterprise-focused
• Generalist vs. Specialist
• Speed/simplicity vs. Power/flexibility

Place the subject business and top 3-5 competitors on the map. Identify white space or crowded zones.

---

Phase 3: Grading & Synthesis

Produce an overall assessment with honest grades.

Competitive Advantages (what's working)

List the real, defensible advantages — things competitors can't easily copy.

Vulnerabilities (what's exposed)

List the biggest risks: competitive, operational, market, regulatory.

Market Position

Categorize the business:

• Premium player — high price, high value, brand-driven
• Cost leader — winning on price/efficiency
• Niche specialist — dominant in a narrow segment
• Challenger — attacking incumbents with a new model
• Commodity — undifferentiated, competing on execution alone
• Pioneer — creating a new category

Overall Grade

Grade on a simple A-F scale across these dimensions:

Dimension	Grade	Rationale
Defensibility (7 Powers)
Market Attractiveness (Five Forces)
Business Model Clarity (Canvas)
Competitive Position
Overall

Strategic Recommendations

3-5 concrete actions the business should consider, ranked by impact. Be specific — "improve marketing" is useless; "build a self-serve onboarding flow to reduce CAC from ~$X to ~$Y" is useful.

---

Phase 4: Report Finalization

1. Do a final pass — ensure all frameworks are populated with real data, not placeholders
2. Present the report file to the user
3. Offer to dive deeper on any section

---

Report Template

Use references/report-template.md as the starting structure for the output document.

---

Guardrails

• Don't fabricate market data. If you can't find a number, say so and provide your best estimate with reasoning, clearly labeled as an estimate.
• Don't grade inflated. The user benefits from honest assessment, not encouragement. A weak business should get a weak grade.
• Don't skip research. Every framework section should reference specific competitors or data points found during Phase 1.
• Flag when you're uncertain. If the market is niche and data is sparse, say that — it's itself a useful signal (small market, hard to validate).
• Update the report incrementally. Don't wait until the end. The user should be able to stop mid-conversation and have a useful partial report.