Skip to content
September 2026

Account-Level vs Person-Level Visitor ID: One Is US-Only

One tells you a company visited, the other names a human. Every vendor selling the second one restricts it to American traffic, and that restriction is the answer to a question buyers rarely ask.

Here's the trap. A vendor tells you they can identify your website visitors, you picture a name and a job title, and you buy. What arrives is a list of companies. Or you picture a list of companies, and what arrives is named individuals, which is a different conversation with your legal team entirely.

These are two different products sharing one phrase. Account-level identification maps the visitor's network back to a company. Person-level identification matches the browser to a named human through an identity graph. The signals category sells both under the same words.

There is one fact that separates them faster than any feature list. Every vendor we checked that does person-level identification restricts it to United States traffic. Account-level has no such restriction.

So this page covers what each one returns, what each actually costs per identification, why the published match rates disagree by a factor of three, and what to do if your traffic is not American.

What does each one actually tell you?

Account-level is a reverse lookup on the visitor's IP address, resolving the network to an organisation. The output is that someone at a named company looked at your pricing page. You do not learn who, and on a large company that is a weak signal, because the someone could be a procurement lead or an intern.

Person-level matches the browser against an identity graph assembled from other data sources, and returns a named individual with a title and usually an email address. The output is that Sarah Chen, who runs operations, looked at your pricing page. That is genuinely actionable, which is why it costs more and why it comes with conditions.

Why is person-level identification US-only?

Because the vendors have decided it should be. This is what our records show, all verified in August 2026: RB2B limits contact-level identification to the US. Vector geofences identification to the US. Common Room's identification, which is supplied by a partner, confines contact-level results to US visitors.

Three separate products, three separate identity graphs, the same border. We are not going to tell you what European law requires, because that is a question for your counsel and not for a directory. What we can tell you is that the companies selling this have all drawn the same line in the same place, and that is worth more than a paragraph of legal speculation.

The practical consequence is simple. If your traffic is mostly European, person-level identification is not a product you can buy, whatever the homepage implies. Account-level works everywhere, and our notes on compliant sales tooling cover the neighbouring questions.

How accurate is it, actually?

Between 10 and 75 percent, and the spread is not because some vendors are much better. It is because two different numbers are both called the match rate: how much of your traffic gets identified at all, and how often the identification is correct.

Vector demonstrates this on its own properties. Its help centre puts the contact match rate at about 10 percent, rising to roughly 30 percent if you count company-level fallback as a match. Its chief executive's blog post says 25 to 35 percent at contact level. Same company, figures about three times apart, and the difference is which question is being answered.

VendorPublished match rateWhose engine
RB2B15 to 20% basic, 35 to 45% premiumIts own, and it publishes it
VectorAbout 10% contact level, ~30% with company fallbackIts own; CEO's blog gives 25 to 35%
Common Room60 to 75%A partner's product, not its own engine
WarmlyNone publishedNot stated

RB2B is the only vendor in the category publishing a match rate for its own resolution engine, and it publishes an unflattering one. Common Room's 60 to 75 percent is the highest figure on this page and belongs to the partner supplying its identification, which is more than Common Room publishes about its own engine, for which no match rate exists anywhere.

A figure of 30 to 75 percent for realistic match rates against 90 percent claimed circulates widely in this category. It comes from a vendor write-up rather than an independent test, so treat it as a competitor's characterisation. The related claim that 96 to 98 percent of B2B visitors never fill in a form appears at three different values across vendor blogs with no traceable origin, and we would not build a business case on it.

What does one identification cost?

For three of the four, you cannot work it out before signing. That is the practical finding.

  • RB2B starts at $79 a month and publishes both an allowance and an overage rate. Its $149 tier covers 600 resolutions, which is about 24.8 cents each, and overages are published at 25 cents on the higher plans and 45 cents on Starter. You can budget this before you buy.
  • Vector starts at $399 a month priced by identified visitor. Purchases are final with no refunds, and cancelling requires thirty days' notice.
  • Warmly has removed its free tier and now starts at $10,000 a year, metering credits it never defines. If a credit is one identified visitor and you use the whole allowance, that lands near eight cents an identification, but that is our arithmetic on an assumption the vendor does not confirm. It has also been acquired by HubSpot, which already owns a de-anonymisation product.
  • Common Room is $2,500 a month flat with 1,000 de-anonymisations included, but dividing one by the other would be wrong because the fee buys a whole signal platform rather than the identifications. Its entry price has risen 150 percent in ten months.

Watch for products that sound like this and are not. Our record for Unify puts it plainly: at $20 to $60 a seat you are buying an outbound sequencer with third-party signals attached, not visitor de-anonymisation.

So what should you do?

  • Check where your traffic comes from before you take the demo. If it is not majority US, person-level is not available to you and the conversation is about account-level tools instead.
  • Ask which number the match rate refers to. The question that gets a straight answer is: of a hundred visitors, how many come back with a named person and an email, excluding company-only fallbacks? Anything above 45 percent should be asked about twice.
  • Ask whose identity graph it is. At least one vendor here resells a partner's identification and publishes the partner's accuracy figure as its headline. That is not dishonest, but you are buying the partner's engine and should price it accordingly.
  • Price it per identification, not per month. Only one of these four publishes enough for you to do that, which is itself a reason to start there when testing the category.
  • Talk to your legal team before switching person-level on, not after. Our own record on RB2B says the same thing, and it is the vendor being straightforward rather than us being cautious.

The difference between the two products is whether you get a company or a person, and the US-only restriction on the second one is the clearest signal in the category about which of those two things is complicated to sell.

Questions

What is the difference between account-level and person-level visitor identification?
Account-level resolves the visitor's IP address to a company, so you learn that someone at a named organisation visited. Person-level matches the browser against an identity graph and returns a named individual with a title and usually an email. They are sold under the same phrase and are different products at different prices with different availability.
Can I identify individual website visitors in Europe?
Not from the vendors we checked. RB2B, Vector and Common Room all restrict contact-level identification to United States traffic. Account-level company identification is not restricted the same way. What European law requires is a question for your counsel, but the product decision the vendors have made is consistent across all three.
What is a realistic visitor identification match rate?
For person-level, somewhere between 10 and 45 percent depending on the vendor and the tier. RB2B publishes 15 to 20 percent on basic resolution and 35 to 45 on premium. Vector's help centre says about 10 percent at contact level, roughly 30 percent if company-level fallbacks count. Higher numbers usually mean fallbacks are being counted as matches.
Why do two figures from the same vendor disagree?
Because two different things are called the match rate: what proportion of traffic gets identified, and how often the identification is right. Vector's help centre and its chief executive's blog give figures about three times apart for this reason. Ask specifically about named-person matches excluding company fallbacks.
Which visitor identification tool is cheapest per identification?
RB2B, at roughly 25 cents a resolution, and it is the only one of the four that publishes enough to let you calculate that before buying. Vector prices by identified visitor from $399 a month, Warmly meters credits it does not define from a $10,000 annual floor, and Common Room's $2,500 monthly fee covers a whole platform rather than the 1,000 identifications it includes.

Tools mentioned

All tools

Sources

Source interests are labelled. Almost everything published about this subject is written by someone selling into it.

More from the blog

All posts