What actually happens when you press search
A dozen independent data layers, queried in parallel across your territory. One reconciled list of real businesses. Every account scored against your brand, and the best of them handed to you with the reason to walk in the door already written. Here is the whole pipeline, stage by stage.
A dozen data layers, queried at the same time
Drop a pin, set a radius between one and ten miles, and Account Finder fans out across every source registered for your industry at once. For a wine & spirits territory that is eight independent layers on the account side alone, with four more behind distributor search. Every industry runs on at least six. They are deliberately different kinds of data, because each one sees businesses the others miss.
Location and place data
Google Places and OpenStreetMap, queried per venue type and by keyword — restaurants, wine bars, hotels, bottle shops, grocers, cafés, salons, whichever categories your industry actually sells into. Plus a nationwide places extract behind them.
Licence and permit registries
Federal TTB alcohol permits, state ABC retail and wholesale licensees, state professional-licensing boards, and statewide food-service licences. This is the layer that confirms a business is real, operating and licensed for your category — not a listing someone typed into a map once.
Municipal open data
Live business-licence feeds published by major-city open-data portals, selected automatically by proximity to your search point. Fresh, local, and maintained by the city itself rather than by a directory scraper.
Federal business records
SBA 7(a) and 504 borrowers, USDA SNAP-authorised retailers, the USDA farmers-market directory, and the national provider registry — layers that surface owner-operated businesses which never bothered with a map listing.
Institutions and member organisations
Schools, hospitals, and IRS 501(c) records for clubs, chambers, fraternal halls and veterans’ posts. Real volume and real standing orders, in a channel most reps never think to prospect.
Curated trade data
Hand-maintained industry award and recognition data, and a curated roster of fine-wine importers and distributors — the judgment layer that no public registry contains and no scraper can produce.
A dozen sources, one clean list
Run those layers side by side and the same restaurant turns up five times under five slightly different names at three slightly different addresses. Account Finder reconciles them before you ever see them: venues are matched on a geographic grid combined with name tokens, so “Jack’s Wine Bar”, “Jacks Wine Bar LLC” and the licence record on the same corner collapse into a single account carrying the best of all three records. Companies are matched on name and state, because a wholesaler’s registry address and its map listing rarely agree.
What reaches your screen is one row per real business, with every source that confirmed it attached — so you can see exactly why it made the list. Each source also fails soft: if one layer is slow or quiet, the search completes on the rest rather than breaking.
See the whole pipeline run on your own territory
Pick a city, set a radius, and watch every layer report in live. Your first ranked call list lands in about a minute.
Every account scored against your brand, not against an average
Before any AI touches the list, every single account is scored by a deterministic brand-fit model — the same inputs always producing the same score, across every result rather than a sample. It weighs price tier, rating and review depth on a curve that stops a four-thousand-review chain burying a three-hundred-review neighbourhood wine bar, on- versus off-premise channel, the favour and exclude keywords you set, and curated industry accolades weighted by tier.
Your brand profile is the lens
Category, price tier, style, and the keywords that describe your ideal door. Change any of them and the whole list re-orders instantly in your browser — no new search, no waiting, no credits spent.
State liquor law, built into the score
In states where a retail channel cannot legally sell your category, those accounts sink instead of ranking like real prospects. A chain drugstore that can’t carry your product never wastes a slot near the top of your list.
Distributors ranked on their own terms
Wholesalers have no review counts or price levels, so distributor search scores what matters for a trade partner instead: portfolio fit, portfolio breadth, federal permit status, and licence type.
The AI writes the opener. It never decides who exists.
This is the deliberate line down the middle of the product. Discovery and ranking are deterministic and evidence-based — every account on your list came out of a real record in a real source. Only then does a language model read the top of the ranked list and write, for each account, a refined fit score and a one-sentence rationale in your brand’s language: why this door, what to lead with, what the programme looks like. Accounts are enriched in parallel batches racing a shared deadline, so one slow response can never hold up your search, and anything that misses the deadline keeps its deterministic score. No account is ever left unscored.
What you get per account
Name, address, channel type, price tier, rating and review depth, the sources that confirmed it, any curated accolades, a brand-fit score, and a call opener written for your brand rather than a generic template.
Why the split matters
A model asked to invent a prospect list will invent prospects. Ours is never asked to. It only ever writes about accounts a registry, a licence record or a live listing already confirmed — which is why the list holds up when you drive to it.
The data refreshes itself, every morning
Federal permit and state licence data rebuilds daily on a schedule, and the heavy national datasets rebuild on their own cycle — more than a million business records across every US state. Every refresh passes a sanity gate before it publishes: if a source returns a suspiciously small file after a change upstream, the refresh is blocked and yesterday’s good data stays live. Fresh data reaches your searches with no release and no downtime. You never re-pull anything, and you never work from last quarter’s export.
Out to a call list in one click
Work the list on the map or in the ranked table, then export to CSV or straight into a Google Sheet, or push it into your CRM. Import your existing accounts first and every search renders new prospects against your current coverage, so you can see where a territory is thin before you plan the week.
Run the pipeline on your own market
Start with a 7-day free trial — card up front, $29/month after, cancel any time before day 7.