There are hundreds of tools for finding a VP of Engineering at a software company. There is almost nothing for finding the 1,284 hair salons in Lyon — and that is where most of the world's businesses actually are.
Every serious sales data product is built on the same assumption: that a business is a company with a website, a LinkedIn page, a funding round and an org chart. That assumption holds for a few million companies and fails for the hundreds of millions of businesses that are a shop, a clinic, a workshop or a van.
A plumber in Manchester has no LinkedIn presence worth the name. A bakery in Porto has no CRM. Neither will ever appear in a database built by scraping professional networks — and both buy software, insurance, equipment, marketing and accounting every year. They are the largest under-served market in B2B, and they are almost entirely absent from the tools people use to sell.
What they do have is a map listing. Google, Apple and Bing have each spent a decade building the most complete census of physical commerce that has ever existed — and none of them will hand you a spreadsheet. So we built the index that will.
FindAndClose started because of a specific, boring frustration: pulling a list of local businesses meant either paying an agency four figures for a one-off file that was stale on arrival, or writing a scraper and then maintaining it forever. Both are bad answers to a question that gets asked every week.
Not a scraper you rent. An index we maintain, that you query.
Most tools in this category open a headless browser the moment you press search and scrape a results page while you wait. That is why they are slow, why they break, and why they cannot give you a country. We do the opposite, and it changes what the product can be.
Our crawlers work through Google Maps continuously, country by country and category by category, whether or not anyone has asked for that market. By the time you search for veterinarians in Valencia, they are already indexed. Nothing you do triggers a crawl. Apple Maps and Bing Maps are the next two crawlers on the roadmap.
public listings only · no logged-in surfaces
The same restaurant is listed more than once, with slightly different names, two phone numbers and one wrong address. We match on name, coordinates, phone and domain, merge them into a single record, and keep the disagreements internally so we can tell which source was right when they diverge again. The record is keyed for all three maps, so Apple and Bing merge into the businesses you already have rather than duplicating them.
deduplicated · conflicts resolved by recency
A map listing gives you a website, rarely an email, and never a social profile. So we visit the business's own site — the one it published on its own listing — read the contact points it puts there, and follow the social links it chose to display. Email addresses are checked for a valid domain, MX record and a clean SMTP handshake before the verified flag is set. We never guess an address from a name.
published contact points only · no pattern guessing
Restaurants and salons churn; notaries and funeral directors do not. Re-crawl
frequency follows how fast a category actually changes, so effort goes where the data
decays. Rather than hide behind an average, every record carries a Link
to the listing it was read from, so any row can be checked against the live source in
one click.
rolling schedule · per-record source link
Because the work happened in advance, your search is a database query. A city returns instantly, a country is a queued file rather than an impossibility, and two customers asking for the same market do not cause two crawls. Politeness to the sources and speed for you turn out to be the same design decision.
seconds for a city · minutes for a country
We know who they are because we read every support email. If you are not one of these three, the trial will tell you faster than we can.
Web design, SEO, paid ads, reputation management. You need every dentist in a metro who has a website but no Instagram, or every restaurant rated under 4.0 with more than 200 reviews. That is a filter combination, not a list you can buy.
Field sales for POS systems, insurance, food distribution, equipment leasing. Your territory is a shape on a map, not a set of counties, and you need it re-pulled every month with only what changed. Geo Search and scheduled exports exist for you.
People wiring lead flow into a product, a CRM or an agent. You want an API with cursor pagination and idempotent exports, a Parquet file, and rate limits published in a header rather than discovered by being throttled.
This industry has a deserved reputation problem. The only useful response is to state the boundaries publicly, so you can hold us to them and so we cannot quietly move them later.
Everything in the index is information a business published about itself so it could be contacted: its map listing, its own website, the social profiles it links to. Nothing comes from behind a login, a paywall or a purchased broker file.
Each record carries a Google ID and a Link to the public
listing it was read from, and every address found on a website carries the URLs it was
found on. If your counsel asks where a phone number came from, the answer is a column,
not a support ticket. Traceability is what makes an Article 14 notice possible at
all.
A business asking to be delisted is suppressed within 30 days, and its identifiers go on a suppression list so the next crawl cannot reinstate it. A deletion that undoes itself on the following pass is not a deletion. Suppressed records cost nobody a credit.
If a business publishes no email address, the field comes back empty. A plausible guess that bounces costs you sender reputation and costs a stranger their inbox. We would rather ship a gap than a fabrication.
We index businesses. We do not build consumer profiles, we do not append demographic or intent data from third parties, and we do not sell personal inboxes. There is a large market for those things. We are not in it and do not plan to be.
Deriving firstname.lastname@company.com from a name and a domain is
standard practice at several well-known vendors. It manufactures personal data about a
person who never published it. We will not do it, on any plan, at any price.
We do not publish a single headline accuracy percentage, because across 4,000 categories and every market we index one number would be marketing rather than measurement — and every vendor grades their own homework. We do not claim real-time data, because we crawl ahead of demand and say so. We do not claim that using this data is automatically legal for your campaign, because that depends on your jurisdiction and your method; the legality section of the FAQ is specific about where the lines fall in the EU and the US, and it says plainly that it is not legal advice.
FindAndClose is built by a small founding team of engineers who got tired of writing the same scraper for the fourth time. There is no sales department, no offshore support tier and no press office. The person who answers your email has commit access.
That has consequences you should know about before you buy. Support runs on European business hours rather than around the clock. Feature requests are answered honestly, including "no" and "not this year". And when something breaks, you will hear about it from us with a cause rather than a status page that goes green without explanation.
We would rather be a company that a thousand people rely on and can get hold of than one with a logo wall and a ticket queue. If you want to talk to whoever built the part of the product you are asking about, that is usually arrangeable — write to hello@findandclose.com.
Fourteen days, 100 leads, no credit card. Pull a list from a market you already know and see whether any of the above is true.