← All work

For service businesses

Customers from public data.

A cleaning company needed customers. The machine read every public review of Denver’s short-term rentals, applied four rules, and returned the hosts whose own guests were asking for a better clean, with the evidence attached.

A short cut. Watch the full film on YouTube ↗

354,741public reviews read
807 → 39flagged listings narrowed to 39 qualified hosts
4 rulesall four or nothing: a low cleanliness score, repeated complaints, a recent one, a real whole-property rental
$19,731/moestimated cleaning spend those listings already have

It reads. It doesn’t just match.

A keyword matcher keeps any review that says “dirty.” A reader understands the context and rejects the ones that don’t fit, so the rejection log came out longer than the lead list. That’s the point.

Each lead came as a dossier: the listing, the guests’ own words and the spend floor. The finished list was given, free, to a real Denver cleaning company.

  • A four-rule qualifier
  • A 13-page lead list with guest quotes
  • A 710-line rejection log
  • An outreach email to a real cleaning company

Your market is already talking.

Reviews, competitor reviews, permits, listings and call notes: the machine reads the public evidence around your business and turns it into a list you can act on.

Send me your business ↗

Your next move

Send me your business.

Tell me what you sell and what you want to be better. I’ll come back with a direction.