B2B Data Providers: 51 Compared, Ranked and Explained

A B2B data provider is a company that sells business contact and company information: names, job titles, work emails, phone numbers, firmographics, technologies, and buying signals. You query it by domain, by person, or by filter, and it returns records you use for prospecting, routing, or CRM enrichment.

The label covers four very different businesses, and confusing them is the most common buying mistake. Some providers own a static database they resell. Some query live sources at the moment you ask. Some only verify data you already have. Some are marketplaces that resell other vendors' datasets. A static database is fast and ages badly; a live lookup is fresher and slower. Neither is universally better, but they fail differently, and your renewal will hinge on which failure mode you can tolerate.

Derrick belongs to the second family: it queries sources at the moment you ask, from the Google Sheet where your list already lives, and bills a credit only when a result comes back.

Providers are compared as if they sold the same product. They do not. There are five distinct data types, and most vendors are genuinely good at one or two.

• Contact data - the person: name, title, work email, mobile. The most perishable data on this list, because people change jobs. Anything older than 6 months is a guess.
• Firmographic data - the company: size, revenue, industry, location, legal identifiers. Slow-moving, cheap to keep current, and the base layer of any segmentation.
• Technographic data - the stack: what a company runs on its website and in its tooling. The sharpest filter for technical ICPs, and useless for everyone else.
• Intent data - the research: which accounts are reading about your category right now. Probabilistic by nature, and only actionable if your team can work an account-level signal.
• Signal data - the event: hiring, funding, a new office, a technology swap, a leadership change. Cheaper than intent and far more concrete, because the event either happened or it did not.

Write down which of these five you are actually missing before you look at a single vendor page. Most teams discover they have plenty of firmographics and no usable contact data, then buy another firmographic-heavy platform. Derrick covers contact, firmographic, technographic and signal data from one place - and the free plan gives you 100 credits per month to check which of them your list actually needs.

The B2B data market moves faster than your stack does. Coverage shifts as sources open and close, match rates drift, and multi-year contracts are priced on a snapshot of a market that no longer exists by renewal. If you signed a two-year deal on price-per-seat math, the assumptions behind that price have almost certainly changed.

The right answer is not picking one best provider, it is ordering the ones you use: cheapest first, expensive ones only on the rows still missing. That is how Derrick is built. You enrich in Google Sheets, pay per verified match rather than per seat, and only spend a credit when a result comes back. A row that resolves on the first pass never touches an expensive source.

1. Region match. A provider with 320M US records won't help your EU expansion. Look at your ICP region distribution first, then filter providers.
2. Field strength. Each provider is strong at 1-2 fields (email, phone, firmographic, intent). Match this to where you need lift, not where you have surplus.
3. Match rate on YOUR data. Vendor-reported match rates are theatre. Always benchmark on 500-1000 of your own contacts before signing.
4. Compliance posture. Opt-out and opt-in models change whether you can email cold in the EU, and every vendor documents its own posture: read that page, do not take a sales answer for it. Derrick's own answer is the simplest one to audit, because official registries are a first-class source in the product: SIRENE and NAF/APE data comes straight from the French public registry, with the identifiers you can trace back yourself.

Those four dimensions are why this page lists 51 B2B data providers in twelve categories rather than a shortlist of five. A shortlist assumes everyone has the same ICP. The matrix below assumes you don't, and lets you filter on the axis that actually constrains you.

Most comparisons put every vendor in one list and call it a ranking. That hides the only thing that matters: two providers in different categories are not substitutes, and two providers in the same category rarely both deserve a line in your budget. Here is how the 51 providers in the matrix split up.

• All-in-one platforms - one contract for contacts, companies and often sequences. Broadest, most expensive, most likely to be under-used. This is where Derrick sits, minus the seat-based pricing.
• Email finders - a domain or a name in, a deliverable address out. Cheap, narrow, easy to benchmark in an afternoon.
• Email verification - they do not find addresses, they tell you which ones will bounce. Always a second tool, never a first one.
• Phone data - mobile and direct dials, the field where coverage claims diverge most from reality.
• LinkedIn-based tools - extensions and exporters that read what is on the profile in front of you. Strong signal, hard ceiling: no LinkedIn, no data.
• Intent data - who is researching your category right now. Expensive, probabilistic, worth it only when your sales motion can act on an account-level signal.
• Buying signals - hiring, funding, partnerships, tech changes. Cheaper than intent and far more concrete.
• Tech stack detection - what a company runs on the web. The sharpest filter that exists for technical ICPs.
• Data APIs and bulk datasets - raw records for a team that has a data engineer. Powerful, and useless without one.
• Local business data - Maps listings, storefronts, opening hours. The right source for local B2B, the wrong one for enterprise decision-makers.
• Public registries - official, free or near-free, and traceable. Underused by almost every GTM team.
• Europe-focused providers - built for EU coverage and EU compliance rather than retro-fitted for it.

Derrick covers several of these categories from one place, which is the point: company enrichment, email finding, phone finding, LinkedIn imports, technology lookups, Google Maps imports and French public registry data (SIRET, SIREN, NAF/APE) all run in the same Google Sheet, on the same credits. Check what each one costs before you commit: the full feature list shows the credit cost per operation, and the free plan includes 100 credits per month at no cost.

Three columns in the table below are routinely misread, and each misreading costs money.

Record count is a claim, not a measurement. A vendor announcing 1.3 billion records and one announcing 100 million are not describing the same object: one counts every row it has ever seen, the other counts what it will stand behind. Record count tells you almost nothing about whether your 500 accounts are in there. Treat it as marketing until your own test says otherwise.

Entry price is not your price. The figure in the matrix is the public entry tier at the time of writing. What you pay depends on seats, on annual commitment, and on how much of your credit balance you burn on rows that return nothing. That last one is the silent cost: on a seat-based contract, a failed lookup costs the same as a successful one. On Derrick, it does not - you are billed per verified match, which is why the honest comparison is cost per usable row, not price per month.

Match rate depends on the list, not the vendor. The rates shown are indicative on a mixed US/EU tech dataset. Run the same 500 contacts through two providers and the ranking can invert - that is normal, and it is the entire argument for benchmarking before signing. Two hours of testing beats twelve months of regret.

Want this breakdown as it gets updated? Get the new Data Providers guides by email - one email per release, nothing else.

For most teams, the provider math breaks at ~150K enriched records/month. Below that, vendor stacking is cheaper than building. Above, the per-record cost makes hiring a junior data engineer competitive - especially if your ICP is narrow (a few EU sectors, a few US industries) and you can scrape + enrich from 2-3 high-quality public sources yourself.

Derrick is built for that break point, and it removes most of the reason to build. You get the pipeline without the engineer: sources chained in order, official registries included, results landing in the Google Sheet your team already works in. You pay per verified match instead of per seat, so an unmatched row costs nothing and a matched one costs the same whether you run 500 rows or 500,000. The feature catalogue covers company enrichment, email finding, phone finding, LinkedIn imports and French registry lookups from a single sidebar, and the same operations are available through the API and the MCP server when you do want to script them.

That is the honest answer to build vs buy at scale: you build when you need a pipeline nobody sells you. Derrick sells you the pipeline. Start free with 100 credits per month and benchmark it on your own list before you budget a single engineer-day.