Exhibit A · Sourcing at register scale
A sourcing engine for buy-and-build. It reads the entire UK Companies House register and returns a ranked target universe, including the companies no keyword search would ever surface.
A search fund or a buy-and-build thesis lives or dies on sourcing. The dream is the easy part; the grind is reading through thousands of companies to find the few that fit. The standard tools all start from a keyword, which means they can only return companies that describe themselves the way you happen to have phrased it.
The UK register is free, public and structured. Roughly five million live companies, about 400 columns each. Nobody screens it systematically because the raw file is unusable on its own: SIC codes are self-reported and often stale, most rows carry no website, and small companies file abbreviated accounts with no revenue line at all.
Two pipelines run over one register. The first decides what each company is. The second recovers what it is worth. Every stage caches its output, so a failed run resumes rather than restarting, which matters when a full pass costs real money in API calls.
SIC filtering is coarse and imperfect, but it runs on a laptop in seconds and removes most of the register before anything bills per company. Ordering the pipeline by cost per row is the difference between a run that is affordable and one that is not.
A company's own description of itself, scraped from its site, is far more reliable than a code it picked at incorporation and never revisited. So the engine spends real effort finding sites that were never filed anywhere machine-readable, then judges the company on what it says it does.
Claude is asked only the genuinely ambiguous cases. Rules handle the obvious ones deterministically, which keeps the classification reproducible, cheaper, and auditable. A four-axis taxonomy (stack layer, function, business model, vertical) means a company can be described precisely instead of dropped into one blunt bucket.
Small companies do not file a revenue line. The engine triangulates from what is disclosed in iXBRL filings and labels the result as an estimate throughout. A number presented with false confidence is worse than no number.
A single deduped master table: the target universe, ranked, with estimated financials and defensibility scores attached. Small enough for a partner to read, and evidenced enough to argue for in an investment committee.
A second lightweight entry point runs a fast SIC-only first cut over the same bulk register, for when you want a rough sector map rather than a full classification pass.