What a business acquisition database is
At its core, a business acquisition database is a searchable collection of private companies, organized so a buyer can filter by industry, geography, and size, and learn enough about each one to decide whether it is worth investigating further. The best ones go beyond a static list and add estimated financials, ownership signals, and some way to save and track companies over time, since sourcing off-market businesses is rarely a one-session task.
It is worth being clear about what these databases are not. None of them are a listing service, and finding a company in one does not mean that company is for sale. A good database tells a buyer which companies exist and fit a thesis, and gives enough signal to decide where to spend research time next. Whether an owner wants to sell is something only a real conversation can answer.
The three tiers of the market
At the top, enterprise platforms such as Grata and Sourcescrub serve institutional buyers, private equity firms and investment banks with dedicated sourcing teams. They offer broad coverage, relationship-mapping features, and integrations with CRM systems, priced accordingly at roughly 10,000 to 40,000 dollars a year per seat.
In the middle sit pay-per-record list vendors, who sell contact and company lists, often filtered by industry or size, charged per record or per export. These can be useful for a narrow, one-time list-building exercise but get expensive fast if a buyer needs to search and re-search a large universe repeatedly, which is exactly what a real search fund sourcing process requires.
At the bottom, in cost only, are free public-data tools built on government and registration records: business registrations, licensing data, and released datasets such as the SBA’s Paycheck Protection Program records, covered on the PPP loan company lookup page. These tools rarely match the relationship-mapping depth of enterprise platforms, but for coverage of the small and lower-middle-market private company population, the gap is often smaller than the price difference would suggest.
What to evaluate before choosing one
Coverage matters most: how many companies does the database actually include, and does that number match your thesis’s target size and geography. Freshness matters nearly as much, since a database built on stale registration data will miss recent changes and closures. It is also worth asking how a provider handles industries and geographies outside their core coverage; a database that is excellent for one region or sector and thin everywhere else can look comprehensive in a demo and prove disappointing once you point it at your actual thesis.
Financial estimates deserve particular scrutiny. Ask whether the provider shows a single confident-sounding number or an honest range with a confidence label. A tool that presents an estimated SDE as a precise figure is hiding its own uncertainty, not removing it. Also check export limits and pricing structure, since a database that is technically comprehensive but caps exports at a few hundred rows, or charges per record, will not support a real thesis-driven search.
Where SIFT fits
SIFT sits in the free public-data tier by price, with a company universe of 3.85 million businesses, growing toward roughly 10 million, that puts it closer to enterprise-tier coverage for small and mid-sized private companies specifically. Financial figures are always shown as ranges with a confidence label, never as a single invented number, and every company can be saved into a universe, a list, or a watch for ongoing tracking. Export is free, up to 10,000 rows per export, and buyers who want an LLM-assisted read on a listing can bring their own Gemini or Claude API key to run a scan, rather than paying SIFT a markup on someone else’s model.
Building your stack
Most serious buyers end up combining tools rather than picking exactly one. A free tool like SIFT can cover the bulk of universe-building and initial qualification, described step by step in how to find off-market businesses, while a narrow, paid list purchase or a broker relationship fills in gaps for a specific sub-market. The right stack depends on budget and how much of the sourcing workload you are prepared to do yourself. Start with the free tier and see how far it gets you on SIFT before paying for anything else.
Frequently asked questions
What is the best database for buying a business?
There is no single best answer; it depends on budget and how much of the sourcing work you want done for you. Institutional buyers with dedicated analysts often use enterprise platforms. Independent buyers and searchers typically get more value from free public-data tools, since the coverage of small and mid-sized private companies is broadly similar, and the price difference is substantial.
How much do acquisition databases typically cost?
Enterprise sourcing and market intelligence platforms commonly charge somewhere from about 10,000 to 40,000 dollars a year per seat. Pay-per-record list vendors charge per contact or per export, which adds up quickly at any real scale. Free public-data tools, including SIFT, charge nothing for search and standard export.
What kind of data do acquisition databases actually contain?
Most combine business registration records, industry classification codes, employee and location data, and, where available, estimated financials. Few contain verified private-company financials, since most small and mid-sized businesses do not disclose them; any database claiming otherwise for a broad universe of private companies should be treated skeptically.
Can I export lists of companies I find?
It depends on the provider and plan. Many enterprise and list-vendor platforms cap exports or charge per record exported. SIFT allows CSV export up to 10,000 rows per export at no cost, across saved universes, lists, and watches.