Not a logo wall, not a round number. One question, asked from every one of 33,791 US ZIP areas: is there a real, licensed professional near me — or an honest handoff to the regulator? 2,867,681 people are in the index, and 99.8% of the map now answers within 25 miles. The aggregate is the easy number, so below it is the one that actually matters: the same question asked specialty by specialty, where some answers are still thin.
Snapshot 2026-08-04 — monotonic, recomputed as the index grows.
A coverage number is meaningless without its denominator. So here are the three quantities, kept separate on purpose. The trick the industry plays is quoting one and implying another.
UEveryone licensed to practice — the true denominator. ~2M insurance producers; ~735k advisor individuals. Public counts, attributed below.
the whole population
IThe verified listings we actually hold and serve. A strict subset of U. This is what most sites inflate; we publish it raw.
what we hold
q(z, r)From ZIP z, is there a verified listing within radius r? Coverage is the fraction of the map for which q returns true. This is the number that matters.
what we optimize
The insight worth the whole page: spatial recall decouples from inventory recall. We hold 3.20% of the advisor universe, yet already answer 92.9% of ZIP queries within 50 miles — because advisors cluster where people live, and one verified advisor covers a whole metro of queries. You do not need the population to cover the map.
RIAs, IARs, CFP professionals, and FINRA series-licensed representatives.
22,558 registered adviser firms live — 58.6% of all ~38,500 in America, from the SEC Form ADV public-domain compilation.
Every person licensed to sell any line of insurance in America.
719,265 licensed producers live from official state rosters (TX, FL, LA, IA, OH, NV) — combined with curated listings: 37.8% of ZIPs within 25 mi, 54.0% within 50.
Two very different numbers, side by side, on purpose. Insurance is early — a deep first market (King County) and a national map still filling in. Advisors are far along on the axis that matters, the map. Neither number is rounded up.
Coverage is not a spreadsheet; it is the output of a pipeline. Every query flows through the same five stages, and the last guarantee holds at 3% recall and at 99%.
22,558 adviser firms from the SEC feedPublic-domain sources only. The SEC Form ADV compilation; state Department-of-Insurance producer rosters as public records.
O(log n) ZIP resolutionEvery listing geocoded and placed on a spatial index over all 33,791 US ZIP areas. Lookups are nearest-first by great-circle distance.
answer < 25 / 50 miA ZIP query returns the nearest verified listings, distance-ranked. This is the only number that matters to a person searching: is someone near me, verified?
0 fabricated resultsNo verified result within 50 miles? A deterministic handoff to the authoritative regulator by name. Never a dead end. Never an invented profile.
owned, removable, theirsThe consent-first write path. A professional creates, claims, verifies, updates, or removes their own listing, checked against the regulator at claim time.
All 33,791 ZIP queries return an answer. Zero are fabricated. Recall only ever goes up.
A verified listing when we have one; an honest, by-name handoff to the SEC, FINRA, the CFP Board, NIPR, or your state Department of Insurance when we do not. There is no fourth outcome, and there is never an invented profile.
The largest part of the directory is not firms — it is people. Licensed professionals, from the public record, in the neighborhoods where they actually practise. Measured on exactly the same axis as everything above: can a person standing in a given ZIP reach one?
The 99.8% aggregate is the flattering number. Measured separately, against the same 33,791-ZIP denominator at 25 miles, the everyday professions are close to answered and the specialised ones are not — Real Estate Salesperson sits at 15.1%. That gap is real, it is ours to close, and hiding it inside an average would be the easiest lie on this page. Below are the 40 specialties a person can filter for today with at least a thousand people behind them, out of 368 distinct credentials in the index.
ELC, 70, 10— and a code tells you nothing. But that is a labelling problem, never the licensee’s fault, so we went and got the regulator’s own dictionary: Connecticut publishes the code beside its plain-English name, and 465 of 518 resolve to exactly one credential. Electricians, contractors, engineers and EMTs are in the directory because of it.ELC covers both Electrical Journeyperson and Electrical Contractor), and states whose dictionary we do not yet hold. Picking the most common meaning would put a wrong profession on tens of thousands of real people. Being listed wrongly is worse than not being listed.Getting here was never a data problem — the public record already holds far more people than we serve. It was a bytes problem, and naming it correctly was most of the work. Depth costs about 126 MB for 1,633ZIP areas; the same depth everywhere would cost roughly 2.6 GB, which is not a thing you can ship in a container image. But answering “is someone near me” only needs one reachable professional — so coverage and depth get bought separately, and coverage turned out to be cheap.
National ZIP resolution over all 33,791 ZIP areas, distance-ranked, with a by-name handoff to the regulator wherever the index is still thin. Nobody hits a dead end, and nothing is ever invented to fill a gap.
The source was never the constraint: the CMS NPPES registry is public domain and publishes providers in nearly every populated ZIP. So every ZIP outside the deep tier got a shallow shard — about 30 listings, phone-first, one of every specialty before two of any. 28,923 ZIP areas, 572,239 listings, 62 MB. Shards group by 3-digit ZIP prefix, which is both cheaper to build and better to serve, because a proximity query fans out to neighbours and neighbours share a prefix.
A thin shard spreads across specialties, so common professions land everywhere and rare specialists rarely make the cut — Real Estate Salesperson is still at 15.1% while the aggregate reads 99.8%. The fix is not more generic depth; it is a targeted pull per specialty per region, which is a different query and a much smaller one. This is the honest next number to move.
The per-ZIP cap is gone — every one of the 2,295,442 people we have captured is listed, because someone who is in the index but hidden behind a truncation is not in the directory in any sense that matters to them. That costs 251 MB inside the image, which works today and will not work at ten million. So depth stops living in the image: deep shards become object-store artifacts fetched per request behind a small in-process cache, versioned by content hash so a deploy can never serve a half-swapped index.
Healthcare is the vertical with a clean public-domain national file, which is why it is first, not because it matters most. Insurance producers (state DOI rosters, NIPR), securities professionals (SEC IAPD, FINRA), and the building trades all need either a code dictionary or a licensing agreement. Each is a named piece of work with a named blocker — never a scrape, never a guess.
Size is not the goal; being trusted is. Every professional can claim their one listing, verified against the regulator at claim time, and remove it in a single step with no retention flow. A directory that grows faster than it earns consent is the thing we are replacing.
Coverage figures are measured against 33,791 Census ZCTAs and committed as a dated snapshot: professions last measured 4 August 2026, the people index 28 July 2026. They move when we re-run that measurement, not on every deploy. We never bulk-copy restricted registries (NIPR, FINRA BrokerCheck, CFP Board); firm-level facts are the SEC Form ADV public domain, attributed and dated. A directory, not advice — confirm licensure with the regulator.