B2B lead enrichment: what the accuracy claims leave out.

Nobody has published an independent accuracy test of B2B lead enrichment providers. Every figure you have seen, the 95% and 97% email accuracy claims on the pricing pages, comes from the company selling the data or from a competitor trying to beat it. Forrester evaluated eleven marketing and sales data providers in its Q1 2026 Wave, and even that report ranks them on capability, not on how many of their records bounce.

That absence shapes every purchase in the category. A VP Sales comparing two providers is comparing two marketing claims, and the only way to get a real number is to run the test on your own list. This article covers what B2B lead enrichment is and what a record costs once match rates are counted. It also covers how fast enriched data decays by seniority and country, who ends up running the pipeline, and a 300-record test to run before signing anything.

What is B2B lead enrichment?

B2B lead enrichment is the process of adding missing or corrected data to a lead record from outside sources. The goal is a record that describes a real person at a real company well enough to route, score and contact them. A form gives you a name and an email address. Enrichment adds the job title, seniority, company size, industry, location, phone number, technology stack and whatever else your sales process reads before deciding what happens next.

It works by matching. The provider takes the fields you have, usually an email address or a name plus a company domain, and looks them up against a database assembled from public web pages, partner data, user-contributed address books and purchased lists. If the lookup matches, you get the fields back. If it doesn't, you get nothing, and whether you still pay for the attempt depends on the provider.

Two neighbouring terms get mixed up with it. Data enrichment is the wider job of filling any record, including accounts, products and existing customers; lead enrichment is the version pointed at new prospects. Lead scoring happens after enrichment and reads the fields enrichment added, which is why a scoring model fed blank fields scores everyone the same.

Which fields are worth paying for?

The fields that change a decision. For most B2B teams that is four: a verified email, a direct phone number where phone outreach is used, a job title with seniority, and a company headcount or revenue band. Everything past that is useful only if a rule somewhere reads it. A technology-stack field, for example, matters if you sell into a specific stack. It is a line item if you don't.

The test for any field: name the routing rule, scoring rule or sequence step that reads it. If no step reads a field, enriching it costs money and buys a column nobody opens.

How accurate is enriched lead data, really?

Nobody outside the vendors knows, because no one has measured it independently. Providers publish their own accuracy rates, usually above 90% for email, and every public comparison between providers was run by one of the providers or by a reseller. Those numbers describe the vendor's best markets. They say little about your list.

What you can measure is the outcome. Belkins analyzed 7.53 million cold emails sent through its client campaigns during 2025 and found an overall bounce rate of 1.71%. That figure comes from one agency's clients with a verification layer in front of sending, so it is a floor rather than a target. It still gives you a shape: a provider whose emails bounce at 8% on your list is handing you records nobody verified, whatever the pricing page says.

Accuracy also varies by field, and the fields fail differently. Email addresses fail hard: either the address exists or the message bounces. Job titles fail soft: the title comes back, it was correct two years ago, and the person now runs a different department. Soft failures are the expensive ones, because nothing flags them. The record looks complete, the routing rule fires, and an account executive calls a VP of Marketing about a problem they stopped owning in month four.

Why vendor accuracy rates don't predict your match rate

Accuracy and match rate are different numbers, and the second one is what costs you. Accuracy is how often a returned field is correct. Match rate is how often the provider returns anything at all. A provider can be 97% accurate on the 55% of your list it recognizes and blank on the rest, and the pricing page will show you the 97.

Match rates are driven by geography, company size and seniority, and they move with your ICP. Consider two lists. A list of US SaaS directors matches well almost everywhere. A list of plant managers at Israeli or German manufacturers does not, and no vendor's global average tells you which provider is better at that list. Only a test on that list does. That gap is why teams chain providers into a waterfall enrichment workflow, which raises the match rate and the bill together.

How fast does enriched data go stale?

About 1% a month for senior US contacts, faster in the UK, and faster again for junior roles, and the rate changes with who is in your list. Lusha tracked 148,000 US records for VP and C-suite sales leaders at 3, 12 and 24 months and measured 12.6% going stale within a year, with 25.7% changed at 24 months. The same measurement on UK senior contacts came out at 21.7% a year.

Those are a data vendor's own figures, published by a company that sells refreshed records, so read them with that in mind. They do line up with public labor data, though. The US Bureau of Labor Statistics put median employee tenure at 4.1 years in January 2026 (Employee Tenure release, September 2026), which means a meaningful share of any contact list changes employer inside a typical sales cycle plus one renewal period. Senior people stay longer, which is why their records decay more slowly than the all-contacts averages the industry likes to quote.

The practical reading is that a single refresh cadence is the wrong design. A US C-suite list enriched once a year will be roughly 87% right on the person-level fields. A list of SDR managers and marketing coordinators enriched once a year will not. Build the refresh around the role: senior accounts quarterly, operating roles monthly, and anything a sequence is about to touch re-verified the week it goes out.

Which fields decay and which don't?

Firmographics barely move; people move constantly. Company headcount band, industry and headquarters change slowly, and a two-year-old value is usually still right. Job title, email address, direct phone and employer are person-level fields, and they are the ones that carried the 12.6% to 21.7% churn in the Lusha cohort. Spend refresh budget on the person-level fields and leave the firmographics alone until an account changes hands or raises money.

What lead enrichment costs per usable record.

The price on the page is the price of an attempt. Clay's published pricing lists data credits from $0.05 each on paid plans and charges nothing when an enrichment returns no result. Top-up credits carry a 30% premium over the plan rate (source: Clay pricing page, October 2026). Most providers in the category use a version of that model: a monthly allotment of credits, one or more credits per field per record, and a premium when you run out mid-month.

Match rate is what turns the list price into a real cost. Divide the price per credit by the share of records that actually come back. We ran that arithmetic on the $0.05 list price at three match rates:

The match rates are illustrative, since no independent measurement exists, and the real cost is higher than the table shows. A record that comes back with a stale title still counts as a match. Rerunning no-matches through a second provider raises the match rate and the cost at the same time.

Then there is the cost that never appears on a vendor invoice. Salesforce's 2024 State of Sales survey of 5,500 sales professionals found reps spending 70% of their time on non-selling tasks, and only 35% saying they completely trust their organization's data accuracy. Half-right records create both numbers at once: a rep who does not trust the title field checks LinkedIn before every call, and that check is the non-selling time.

Validity's 2025 survey of 602 CRM users and stakeholders found 37% reporting revenue lost directly to poor data quality. That is the number to put next to the credit price. The question to ask is whether $0.08 buys a record someone will act on.

Who keeps the enrichment pipeline running?

Someone has to, and in most companies nobody was assigned. Enrichment is a pipeline with six moving parts, and a contract buys only the lookup. It starts with a trigger when a lead arrives and a lookup against one or more providers. Then come the rules for which fields may overwrite what the person typed on the form, a de-duplication step, a routing step that reads the new fields, and a refresh job that reruns all of it on a schedule. Each piece is small. Together they form a system that breaks the week a provider changes its API, a form adds a field, or the CRM admin renames "Job Title" to "Title".

The common pattern is that a RevOps lead or a motivated SDR builds it in a spreadsheet tool or a workflow builder, and it works until that person is busy or gone. Nobody owns the overwrite policy, so a form entry of "Head of Growth" gets replaced by a three-year-old "Marketing Manager". The credit budget has no owner either, so retries burn the month's allotment by the 20th. Nobody owns the refresh, so the one-time enrichment from the quarter you signed is the one the team is still routing on a year later. The sales ops version of this problem has its own article.

There are three ways to staff it. Give the pipeline to a RevOps person as a standing responsibility, with hours attached, and accept that the hours come out of reporting and forecasting. Hire a development shop to build it and own the code afterwards, which is the right choice if you want an asset you can hand to the next team. Or buy it as a running service. That is the Uplift model. You describe the routine in plain language, and we build the AI agent that runs it, keep it running as providers and apps change, and fix it when it breaks. The price is flat, with no per-seat licenses and no credit meter of our own. The cost of that model is just as plain: the code never becomes yours, and when the plan ends the automation ends with it. If you want to own the code at the end, the development shop is the better purchase.

Whichever you choose, write down before you sign who owns four decisions: overwrite rules, refresh cadence, credit budget, and what happens to a lead the enrichment cannot match. An unmatched lead that sits in a queue for a week is a lost deal at a cost no credit model records.

Is B2B lead enrichment GDPR compliant?

It can be, and the compliance work falls on you as the buyer. The UK Information Commissioner's Office is explicit in its direct marketing guidance (section on buying or renting information from other sources, updated April 2026). A buyer of third-party data "must undertake proportionate checks and due diligence before you get the information." And further: "it is not enough to simply accept a third party's assurances that the information they are supplying to you is compliant." A provider's compliance page is a claim, in the same way its accuracy rate is a claim.

In practice that means three questions in writing before the contract. Where did the records come from? What lawful basis does the provider rely on for business contacts in each country you sell into? And how does a person's objection or deletion request reach your CRM once you have imported the record? The last one is where enrichment pipelines fail quietly. The provider removes a contact from its database, your copy stays, and the next refresh has no way of knowing it should delete it. If nobody on your side can answer the third question, ask the provider to answer it in the contract rather than in a sales call.

A 300-record test you can run before signing anything.

Take 300 records from your own CRM that match your ICP, including the hard ones, and run them through each shortlisted provider before you sign. The number is large enough to show differences between providers and small enough to check by hand in an afternoon.

Measure five things:

• Match rate: how many of the 300 came back with the fields you actually route on.
• Email validity: send nothing; run the returned addresses through a verification service and count the invalid ones.
• Title accuracy: pick 50 at random and check them against LinkedIn the same day.
• Overwrite damage: for records that already had a title or phone, count how many the enrichment would have replaced with something worse.
• Cost per usable record: credits consumed, divided by the records that passed the first three checks.

Then do the arithmetic the vendor will not do for you. If provider A matches 70% at $0.06 a credit and provider B matches 55% at $0.04, A's usable record costs about $0.086 and B's about $0.073. B is cheaper only if the 15 points of leads it missed are leads you did not need, and usually they are the ones that needed enrichment most.

The test takes an afternoon and answers the question no analyst report or pricing page can. What it does not settle is who runs the pipeline afterwards, how enriched fields feed your CRM workflows, and whether that is a job for your team or for someone who builds and runs it for you. Decide that on the same afternoon, while the person who ran the test is still in the room.