What Is a Good Match Rate for Visitor Identification?

A good match rate for website visitor identification in European B2B is roughly 30 to 40 percent of all visits resolved to a named company, measured over a full month with no traffic filtered out. Person-level identification is far lower, usually in single digits in the EU. Any figure above 60 percent almost always comes from a smaller denominator, not a better product.

That last sentence is the part most buyers miss. Match rate is the number that decides whether the whole category pays for itself, and it is also the number that is easiest to make look good without changing anything about the underlying data.

Same Website, Three Different Match Rates

Take a site with 10,000 visits in a month. 4,000 come from mobile devices, 1,500 are bots, crawlers or known non-business traffic, and of the 4,500 desktop business-hours visits that remain, 2,700 resolve to a company. Now watch what happens to the headline.

• Against all 10,000 visits, the match rate is 27 percent.
• Against the 8,500 visits left after removing bots, it is about 32 percent.
• Against the 4,500 qualifying desktop visits, it is 60 percent.

Nothing changed except the bottom of the fraction. All three numbers are true, and a vendor can put any one of them on a sales slide. If one tool quotes 27 percent and another quotes 60, the second one is not necessarily twice as good. It may be identical, just counted differently.

The Question That Cuts Through It

Ask every vendor the same thing: what share of all traffic, unfiltered, did you resolve to a named company over a named 30-day period on a site like ours? If they cannot answer in that form, run the trial and work it out yourself.

Each part of that question does a job. All traffic stops the denominator from shrinking. Named company stops ISPs, hosting providers and generic networks from being counted as matches. A named period stops someone cherry-picking a good week. A site like ours matters because a US enterprise software site and a Dutch manufacturing site get very different results from the same tool.

What Good Looks Like in Europe

Published benchmarks for company-level identification tend to cluster in that range for European B2B sites, with US-heavy traffic often running higher. Within Europe there is real variation by country. Markets where office workers sit on corporate networks resolve better than markets where most staff work from home on consumer broadband.

Person-level identification, meaning a named individual rather than a company, is a different conversation in the EU. GDPR limits what can lawfully be matched without consent, so expect a low single-digit share at best, and be wary of anyone promising more without explaining the legal basis.

Why Rates Drift Down

Most identification still starts with the IP address. When someone visits from an office network that is registered to their employer, the match is easy. Several things have chipped away at that over the past few years.

• Hybrid work puts a large share of visits on home broadband, where the IP belongs to the internet provider, not the employer.
• VPNs and privacy relays route traffic through shared addresses that point nowhere useful.
• Mobile traffic sits on carrier networks and rarely resolves at all.
• IP ranges change hands, so a database that is not refreshed slowly drifts out of date.

None of this means identification has stopped working. It means a match rate is a property of your audience as much as the tool, and a fair comparison runs every tool against the same traffic.

Accuracy

A high match rate made of wrong answers is worse than a modest one made of right ones. When you trial a tool, pull twenty identified companies at random and check them against what you know: existing customers, open deals, people who later filled in a form. If more than a couple are plainly wrong, the headline rate is not worth much.

Watch in particular for matches to large telecoms, cloud providers and universities. These are often the network the visitor was sitting on rather than the company they work for, and some tools count them to pad the number.

Is 30 Percent Enough?

Usually, yes, because the companies that resolve are not a random sample. Visitors on corporate networks during working hours skew towards people researching on the job, which is exactly the traffic a B2B team wants to see. A third of your visits, named, is typically more pipeline than your contact form produces from the whole site.

The useful test is not the percentage on its own. It is how many of the identified companies fit your ideal customer profile and showed real intent, such as a pricing page visit or a return visit within the week. That is the number to track month on month, and the one to compare tools on once the match rate question has been answered honestly.