Data Enrichment API Pricing: What You Actually Pay Per Record

Every enrichment vendor publishes a monthly price and almost none of them publish the number you need, which is what one usable record costs you. The plan says $98 a month. The question is whether that buys you three hundred emails or three hundred complete profiles with a phone number attached, and whether you pay for the ones the API could not find.

Those differences are not rounding. Between the cheapest and the most expensive way to enrich the same contact, the gap in this article is roughly thirty times, and the expensive option is not always the wrong one.

Here is how the pricing models work, what a credit buys at each vendor, and the arithmetic that turns a plan into a cost per record.

The three pricing models

Almost every vendor uses one of three shapes, and the shape matters more than the headline number.

Per credit, bought in advance. You buy a pool of credits and spend them as you enrich. The subtlety is that different fields cost different amounts of credits, so the pool empties at a rate that depends on what you ask for. Clay, Coresignal, Lusha and People Data Labs all work this way.

Per seat, with a credit allowance attached. The licence carries the credits. This is how sales platforms like Apollo sell enrichment: you are buying a seat and the data comes with it, which makes the data look free and the seat look expensive.

Per record, on success. You are only charged when the record is actually found. Dropcontact prices this way, and it removes the single most annoying line in an enrichment bill, which is paying for misses.

There is a fourth shape that matters at the top of the market: no published price at all. ZoomInfo and Cognism quote after a conversation, and the number depends on volume, contract length and what you are willing to commit to for a year.

What one credit buys

This is where the comparison usually goes wrong, because a credit is not a unit of anything standard.

• Lusha: an email costs 1 credit, a phone number costs 5.
• Hunter: finding an email costs 1 credit, verifying one costs half a credit.
• Coresignal: an employee or company record costs 10 to 20 credits, a job posting costs 1, and contact enrichment costs 20.
• Clay: data credits buy records from the provider marketplace, and separate action credits pay for the orchestration around them.
• HubSpot: one credit enriches one record, and credits are sold in packs.

So "credits included" tells you nothing until you know what you intend to ask for. A team that needs mobile numbers will burn a Lusha allowance five times faster than a team that needs work emails, on the same plan.

The real cost per record

The table below converts published plans into an approximate cost per enriched record. The arithmetic is simple division of the entry tier, so treat it as an order of magnitude rather than a quote, and check the vendor's page before you commit.

Two things stand out once the numbers sit next to each other.

The first is that the cheapest per-record prices come attached to something else you are buying. HubSpot credits are cheap because you already pay for the CRM. Apollo credits are cheap because you are paying per seat. Neither is a fair comparison with a standalone API unless you wanted the seat anyway.

The second is that the gap between an email and a phone number is the real cost driver, not the vendor. Mobile numbers are expensive everywhere, because they are hard to source and quick to go stale.

What the free tiers actually give you

Most vendors have one, and they are more useful for testing match rates than for running anything.

• People Data Labs: up to 100 credits a month, with contact data obfuscated until you upgrade.
• Lusha: 40 credits a month, enough for eight phone numbers.
• Hunter: 50 credits a month.
• Dropcontact: 50 credits on signup, with the pay-on-success model applying afterwards.
• Clay: 100 data credits a month on the free tier, plus a monthly action allowance.
• Apollo: 900 credits per seat per year on the free plan.

The right use of a free tier is a match-rate test, the same discipline as automating CRM data entry rather than buying more of it: take a hundred rows you already know the answer for, run them through two or three vendors, and count how many come back correct rather than how many come back at all.

Five things that change your real bill

Match rate. A vendor that finds 60 percent of your list at half the price is more expensive per usable record than one that finds 90 percent, and the only way to know your rate is to test on your own data. Coverage varies enormously by geography and company size.

Whether you pay for misses. A pay-on-success model like Dropcontact's changes the effective price by exactly your miss rate. On a list where a third of contacts cannot be found, that is a third off the bill.

Waterfall or single provider. Running several providers in sequence until one returns an answer raises the match rate and the cost, because you pay for the attempts. Clay's marketplace is built around this, and it is the reason waterfall enrichment appears on so many stacks: better coverage, more credits burned per record.

Re-enrichment cadence. Contact data decays at roughly a quarter of records a year as people change jobs, which is why lead enrichment is a recurring cost rather than a project, which means enrichment is a subscription to accuracy rather than a purchase. Deciding how often you refresh a record is a bigger budget lever than which vendor you pick, and it is the practical side of data decay.

What you enrich. Enriching every record in the database is the most common way to waste a year's credits, and it is the same instinct that produces bad list hygiene. Enriching the ones somebody is about to contact costs a fraction and produces the same outcome.

How to budget it

Work backwards from contacts you will actually touch, not from the size of the database.

1. Count the records you will contact this quarter. For most small teams that is a few hundred, not the twenty thousand in the CRM.
2. Decide which fields you genuinely need. Work email is cheap. Mobile numbers multiply the bill by five or more, and most outbound never uses them.
3. Test match rates on a hundred known rows before you commit to a year.
4. Multiply: records to contact, times fields needed, times the per-credit cost, divided by your measured match rate.
5. Add a refresh cycle for the records you keep, because the data you bought this quarter will be measurably wrong by the next.

That number, not the plan price, is what enrichment costs you.

Where enrichment sits in the stack

Enrichment is a supplier, not a strategy, and it lives on top of whatever CRM database you already keep. It fills the gaps in records you already have or in lists you have already decided to work, and its value is capped by what happens next: if nobody contacts the enriched record, you have bought a more accurate silence.

That is why the credits question matters beyond price. In a system where enrichment runs on the same budget as the rest of the automation, you can compare the cost of finding a phone number against the cost of writing and sending a follow-up to someone whose email you already had. The second one is usually the better trade, which is the argument behind an autonomous CRM: the data is only worth what the action on top of it produces.

For a team starting from a spreadsheet, the cheapest honest sequence is to get the record right first, contact the people you already know, and buy enrichment when the list of people you cannot reach becomes the actual constraint. A free CRM plan covers the first two steps, and CRM data quality is the discipline that decides whether the third one is worth paying for.