market-research

name	market-research
description	Sizes markets (TAM/SAM/SOM), maps competitive landscapes, and synthesizes industry trends into structured investment or go-to-market recommendations. Trigger phrases: "market size", "TAM SAM SOM", "competitive analysis", "market opportunity", "industry analysis", "addressable market", "market landscape", "Porter's Five Forces", "go-to-market research". Do NOT use for: single-company financial analysis (use financial-tracker-ops), building spreadsheets from raw data (use spreadsheet-analysis), one-off fact lookups with no synthesis, or competitor UX teardowns (use competitor-teardown).
license	Apache-2.0
compatibility	{"clients":["openai-codex","gemini-cli","opencode","github-copilot"]}
metadata	{"owner":"codex","domain":"market-research","maturity":"draft","risk":"low","tags":["market-sizing","competitive-analysis","industry-research","strategy"]}

Purpose

Produce structured, evidence-backed market analyses that size an opportunity, map the competitive landscape, identify trends and risks, and deliver an actionable recommendation. Every claim must trace to a source with an explicit quality tier.

When to use this skill

• The user asks to size a market, estimate TAM/SAM/SOM, or evaluate a market opportunity.
• A task requires competitive landscape mapping — identifying direct, indirect, and substitute competitors.
• The user needs a Porter's Five Forces analysis or an industry structure assessment.
• A go-to-market plan, pitch deck, or business case needs supporting market data.
• A repo contains business-plan artifacts (e.g., docs/market-analysis.md) that need updating with fresh research.

Do not use this skill when

• The task is a single-company financial deep-dive — use financial-tracker-ops.
• The user wants to build or transform spreadsheet data — use spreadsheet-analysis.
• The request is a UX/product feature comparison of a single competitor — use competitor-teardown.
• The question is a quick factual lookup ("What is Company X's revenue?") with no synthesis needed.
• The user is scouting domain names or digital assets — use domain-scouting.

Operating procedure

Step 1 — Define the research question and scope

• State the specific decision the research must inform (e.g., "Should we enter the US pet-insurance market?").
• Define geographic scope, time horizon, and customer segment boundaries.
• Agree on the level of precision required: order-of-magnitude vs. bottom-up model.

Step 2 — Gather and tier sources

Apply the source quality hierarchy to every piece of evidence:

Tier	Source type	Example	Reliability
1	Government / regulatory data	Census Bureau, SEC filings, Eurostat	Highest
2	Industry association reports	IBIS World, Gartner, IDC	High
3	Analyst estimates / equity research	Morgan Stanley blue papers, CB Insights	Medium-high
4	Company disclosures (non-audited)	Press releases, investor decks	Medium
5	Journalistic / blog sources	TechCrunch, industry blogs	Low — corroborate before citing

Distinguish primary research (surveys, interviews, proprietary data) from secondary research (published reports, databases). Note which type backs each claim.

Step 3 — Size the market (TAM / SAM / SOM)

1. TAM (Total Addressable Market): Estimate the total revenue opportunity if 100% market share were achieved globally (or within the defined geography). Use top-down (industry report total × relevant segment %) AND bottom-up (unit price × total potential buyers) approaches; compare the two and explain variance.
2. SAM (Serviceable Addressable Market): Narrow TAM by the segments the product can actually reach given current capabilities, channels, and geography.
3. SOM (Serviceable Obtainable Market): Estimate realistic capture within SAM based on competitive position, go-to-market capacity, and comparable company benchmarks. SOM should never exceed 5–15% of SAM for an early entrant without extraordinary differentiation.

State every assumption explicitly. Provide low / base / high scenarios.

Step 4 — Map the competitive landscape

Classify every identified competitor:

• Direct competitors: Same product, same customer segment.
• Indirect competitors: Different product, same customer need.
• Substitute competitors: Different category that eliminates the need entirely.

For each competitor capture: name, estimated revenue or funding, primary value proposition, key differentiator, and weakness.

Step 5 — Conduct Porter's Five Forces analysis

Assess the industry across all five forces, rating each Low / Medium / High:

1. Threat of new entrants — barriers to entry, capital requirements, regulatory hurdles.
2. Bargaining power of suppliers — supplier concentration, switching costs, input uniqueness.
3. Bargaining power of buyers — buyer concentration, price sensitivity, switching costs.
4. Threat of substitutes — availability of alternatives, relative price-performance.
5. Competitive rivalry — number of competitors, industry growth rate, exit barriers.

Summarize overall industry attractiveness.

Step 6 — Identify trends, risks, and timing

• List 3–5 macro trends affecting the market (regulatory, technological, demographic, behavioral).
• Identify the top 3 risks with likelihood and impact ratings.
• Assess market timing: is the window opening, peak, or closing?

Step 7 — Synthesize recommendation

Produce a clear, directional recommendation tied to the original decision question. State confidence level (High / Medium / Low) and list the top 3 things that would change this recommendation if new evidence emerged.

Decision rules

• Never cite a Tier 5 source as the sole evidence for a quantitative claim; corroborate with Tier 1–3.
• If top-down and bottom-up TAM estimates diverge by more than 3×, investigate and explain the gap before proceeding.
• SOM estimates must include an explicit comparables basis — cite at least one analogous company's market-share trajectory.
• If fewer than 3 independent sources confirm a market-size figure, flag it as "low-confidence estimate."
• When data is older than 2 years, apply a stated growth rate to extrapolate and mark the figure as "projected."
• Separate fact from inference: use "data shows" for sourced claims and "this suggests" for interpretive conclusions.

Output structure

Every market research deliverable must include these sections in order:

1. Market Definition — Decision question, scope boundaries, customer segment, geography, time horizon.
2. Size Estimate (TAM / SAM / SOM) — Three-tier sizing with methodology, assumptions, low/base/high scenarios, and source citations.
3. Competitive Map — Table of direct, indirect, and substitute competitors with key attributes.
4. Industry Structure (Porter's Five Forces) — Force-by-force ratings with supporting evidence.
5. Trends & Timing — Macro trends, market-timing assessment.
6. Risks — Top risks with likelihood × impact matrix.
7. Recommendation & Confidence — Directional recommendation, confidence level, reversal conditions.

Anti-patterns

• Vanity metrics: Citing the largest possible TAM number without narrowing to SAM/SOM to make an opportunity look bigger than it is. Always size down to the obtainable slice.
• Confirmation bias: Selectively citing sources that support a predetermined conclusion while ignoring contradictory data. Present the strongest counter-evidence explicitly.
• Extrapolating from anecdote: Treating a single customer interview or case study as representative of the entire market. Require quantitative corroboration for claims about market behavior.
• Stale data presented as current: Using 5-year-old figures without adjustment or caveat. Always note data vintage and apply growth-rate extrapolation where needed.
• Ignoring substitutes: Mapping only direct competitors and missing the substitute products that could collapse the category entirely.
• Precision theater: Presenting a TAM as "$4.237B" when the underlying data supports only an order-of-magnitude estimate. Match precision to evidence quality.

Related skills

• competitor-teardown — Deep-dive on a single competitor's product, pricing, and positioning.
• financial-tracker-ops — Financial modeling and projection once market sizing is complete.
• spreadsheet-analysis — Structuring raw data into analysis-ready formats.
• domain-scouting — Evaluating digital asset opportunities adjacent to a market entry.
• business-idea-evaluation — Applying market research findings to a go/no-go decision on a specific idea.

Failure handling

• If the market is too nascent for reliable sizing data, state that explicitly, provide analogous-market comparisons, and lower confidence to "Low."
• If competitive data is sparse (e.g., private companies with no disclosures), note gaps in the competitive map and suggest primary-research methods to fill them.
• If the decision question is ambiguous, restate it as a concrete yes/no or rank-order question and confirm scope before proceeding.
• If the task is better served by a single-competitor deep-dive, redirect to competitor-teardown.