Market Map and Sourcing
SkillCommerce & financeBuilds the map of a sub-sector -- target universe, ownership status, EBITDA size bands and a shortlist ranked by thesis fit rather than availability -- when you need a pipeline you chose rather than one you were sent.
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; ahel provides instructions and does not run this skill.
Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
Then ask your AI: use the Market Map and Sourcing skill
What this skill tells your AI
The instructions your AI receives, as published by andreworia/claude-finance-skills in packs/private-equity/skills/market-map-and-sourcing/SKILL.md and read by ahel’s review.
When to use
Use this skill once a thesis exists and before any intermediary is called. Use it also as a diagnostic: if the pipeline is entirely banked processes, the map is being assembled from what happens to be for sale, and the fund is bidding on other people's timing.
What it does
Produces a market map of a defined sub-sector: the target universe with sources, ownership status and vintage, EBITDA size bands, a fit score derived from the thesis conditions, a separately held accessibility score, and a shortlist ranked by fit with an owner and cadence per name.
Method
Step 1 -- Build the universe before filtering it
Take the inclusion test from the thesis and count every company that meets it, using registry filings, licence registers, trade bodies and sponsor portfolio pages. Count first, filter second: a map with 40 names in a sector of 90 is a shortlist someone else assembled.
Step 2 -- Record ownership status and vintage
Record founder-owned, family, sponsor-owned with entry year, corporate subsidiary, or listed-group division. Ownership decides whether a transaction can happen at all: sponsor-owned assets in hold year four and beyond are the predictable supply of process, and founder-owned assets with no succession plan are the only genuinely proprietary pool.
Step 3 -- Band by EBITDA, not revenue
Band each name against the mandate: below, within, or above with co-invest required. Where EBITDA is undisclosed, estimate it from revenue times a sector margin benchmark and label it an estimate -- unlabelled, it becomes a fact by the third version of the map.
Step 4 -- Score fit against the thesis conditions
Build the criteria from the must-be-true list, not from generic quality. If the thesis requires an installed base, score installed base; if it requires licensing as a barrier, score the licence, each 1-5 with a line of evidence and a source.
Step 5 -- Score accessibility separately, and never blend the two
Accessibility -- process status, owner receptivity, relationship, timing -- is scored on its own axis, never averaged into fit. A blended score always promotes the available asset over the right one, which is how a map degrades into a list of whatever is on the market.
Step 6 -- Rank by fit, then assign an action per quadrant
High fit and high accessibility: engage now. High fit, low accessibility: cultivate on a multi-year clock under a named partner. Low fit, high accessibility: decline in writing and log why, so the name is not re-screened in six months. Low fit and low accessibility: off the map.
Step 7 -- Set the cadence and refresh on a clock
Give every shortlisted name an owner, a contact interval and a dated next action, then refresh quarterly and record why any name changed band, ownership or score.
Inputs
- Thesis document, including the must-be-true conditions and boundary
- Fund mandate: EBITDA range, geography, cheque size
- Registry, filing and trade-body sources for the sector
- CRM or contact history, including prior approaches and outcomes
- Observable ownership data, including sponsor entry years
Output format
- Universe count with the sources used to build it
- Ownership breakdown by category and vintage
- Size bands against mandate, with estimates labelled
- Fit scorecard in prose, criteria traced to thesis conditions, never a table
- Accessibility assessment, held separately from fit
- Shortlist ranked by fit, each with owner, cadence and next action
- Declined names with reasons logged, plus the refresh date
Example
Fictional worked example -- UK specialist veterinary referral clinics.
Universe: 48 companies meet the inclusion test. Ownership: 27 founder-owned, 9 sponsor-owned (5 in hold year four or later), 7 corporate subsidiaries, 5 listed-group divisions. Size: 19 below GBP 3M EBITDA and out of mandate, 21 in the GBP 3-10M band, 8 above it.
Of the 21 in band, 12 score 7 or better on fit: 4 are accessible now, 8 go to cultivation under a named partner on a quarterly interval. The two most available assets in the sector -- both sponsor-owned, hold year six -- score 4 and 5 on fit and stay off the shortlist, though they are the only two names any intermediary has offered this year.
Signals
- GitHub stars
- 20
- Forks
- 4
- Last commit
- Sep 2026
Advanced
- Item type
- skill
- Key
market-map-and-sourcing- Source
- github.com/andreworia/claude-finance-skills
github.com/andreworia/claude-finance-skills
Related picks
Skill · coreyhaines31
The pick for Companiesquantitative-screening
Skill · agentii-ai
The pick for Companiesparallel-monitor
Skill · parallel-web
The pick for Monitoryutori-api-monitor
Skill · yutori-ai
The pick for Monitorfirecrawl-build-scrape
Skill · firecrawl
The pick for Scrapescrape
Skill · davila7
The pick for Scrape