← ClaudeAtlas

mi-collect-geoint-demointlisted

Pull the market's denominator from government statistics — establishment counts, occupations, wages, firmographics, trade flows. Use before sizing, ICP work, or persona localization.
Productside/Productside-Market-Intelligence-Skills · ★ 0 · AI & Automation · score 68
Install: claude install-skill Productside/Productside-Market-Intelligence-Skills
# GEOINT/DEMOINT Collection Sweep ## Purpose The cartographer's discipline. The terrain map, not troop movement. Government statistics are free intelligence that most Product Managers never open, and they are the backbone of every ICP, persona, and TAM that survives scrutiny. Establishment counts by industry code and size band are the *denominator* — the number of things that could possibly buy you. Everything sized without one is a vibe with a dollar sign in front of it. This is also the slowest-moving discipline, which cuts both ways: the data rots slowly, and it rots invisibly. Statistical releases lag, so every dataset here carries a **vintage**, not just a publication date. The sweep produces a labeled inventory, not a model. **Collection is not fusion**, and it is not sizing either — this run gathers the counts; `mi-size-tam-sam-som` builds the model on top of them. ## When to Use It Use it before any sizing exercise, before drawing ICP boundaries, and before localizing personas into a second country. Use it when two analyst reports disagree and you need a bottom-up number to referee them. Use it annually, because sizing rot is slow but real. Do not use it when: - **You need the model, not the counts.** This run produces the denominator; `mi-size-tam-sam-som` produces TAM, SAM, SOM, and sensitivity. - **You need a capture rate.** That comes from FININT — a competitor's revenue divided by their claimed customer count — not from statistics. - **The market is too