Live Google Maps + SERP read, graded on the playbook's real criteria. Search volume shown but not scored (unreliable for local).
🧠 Why two steps, two tools. Niche fit is a judgment call — brand loyalty, ticket size, whether an incumbent is truly unbeatable — that's nuance, so it goes to AI. Scoring a market has to apply the exact same rule to every city, every time, with zero drift — that's mechanical, so it goes to code. Curated judgment → ask Claude. Deterministic, repeatable execution → run the app.
🛑 Step 1 — decide the niche with Claude first. Don't scrape a niche/location willy-nilly. Ask Claude "is [niche] good for rank-and-rent?" (or read the criteria below) and agree on the niche before you touch this page. Then come here for Step 2: find the city.
Step 2 — once you have a niche, this decides if a market is worth building into
Niche — pick this yourself, before scoring a city
No brand loyalty / no unbeatable incumbent. "Nobody knows a concrete brand to call" = good. A 90-year incumbent winning on word-of-mouth alone = unbeatable, skip.
High ticket + real demand — at least ~10 local businesses in the area to eventually sell leads to.
Avoid red oceans already dominated by giants pulling branded/direct search (plumbing, HVAC, big-metro roofing).
Market score — what this tool actually computes, live
Map-pack opportunity (30%, the biggest lever) — LOW review counts on the top listings = good. 200-1,200+ reviews = a "review moat," score tanks.
Demand (20%) — business count + population, 100k-300k sweet spot.
Quickness (10%) — how fast the job closes.
🎯 Bottom line: this tool wants a LOW benchmark — an open market with no established review moat. An established competitor with 200+ reviews is a reason to walk away, not an opportunity. (The opposite business, CoreFlux reputation marketing, wants the opposite — see coreflux-reputation-prospector — never mix the two up.)
~$0.006/market (Maps + SERP + volume). Up to 40 per run. Saved to Supabase if the table exists. Full walkthrough: SOP.md