RB2B Match Rate: What Share of Visitors Is Identified?

RB2B’s match rate is described in a competitor’s comparison as roughly 5 to 20 percent of US traffic at the person level, with non-US visitors not identified at all. In practice that means most visitors to your site will remain anonymous, whichever tool you choose.

The figure comes from BusinessMCP’s RB2B alternative page, which is a competing vendor’s description. We cannot verify it independently and neither should you rely on it for a business case. Use it to set a sceptical baseline, then measure.

What “match rate” really means

There are at least three different numbers people call a match rate:

Metric	Formula	Why it misleads
Fill rate	Identified visitors / all visitors	Includes non-US traffic the tool cannot cover
US match rate	Identified US visitors / US visitors	Only valid if you can isolate US traffic
Precision	Correct identifications / identified	A high fill rate with low precision is worse than useless

Vendors tend to quote the one that flatters them. A tool that matches 40 percent of visitors with half of them wrong is worse than one that matches 15 percent with 90 percent correct. Our guide to testing a visitor identification tool shows how to check precision.

Key takeaway: ask for precision as well as fill rate. A wrong name is not a lead; it is a risk.

Why most visitors stay anonymous

1. Geography. The comparison says RB2B’s person-level matching relies on US identity-graph data, so visitors from elsewhere do not resolve. See does RB2B work outside the US.
2. Privacy tools. Browsers, extensions and VPNs reduce the signals available.
3. Shared or mobile networks. Signals are weaker on carrier and home connections.
4. Audience mix. Students, job seekers and researchers are often logged out and unmatched.
5. Bots. Automated traffic inflates the denominator and never resolves.

For comparison, BusinessMCP says company-level identification resolves roughly 20 to 35 percent of B2B traffic regardless of price tier. Those are different units, people versus companies, so a company rate is not a better or worse version of a person rate.

What can move your rate

• Traffic mix. More US, desktop, logged-in business visitors raises it.
• Page types. Pricing and demo pages attract identifiable buyers; blog traffic tends to be anonymous.
• Script placement. A script that loads late or is blocked by consent tools misses visitors.
• Plan tier. The comparison lists Pro+ with “premium resolution” at $199 a month; ask the vendor exactly what that changes and test it. Plan details are in RB2B pricing.

How to measure it on your site

1. Pick a window of at least two weeks.
2. Export total unique visitors and the US share from your analytics.
3. Count unique identified visitors from the tool.
4. Compute fill rate and US match rate.
5. Sample 20 identified visitors. Verify them against LinkedIn or company pages and note how many are correct.
6. Segment by page: how many identified visitors viewed pricing or demo pages?
7. Repeat with a second tool on the same traffic.

A realistic way to use the number

If the vendor-reported range holds, a site with 3,000 US visitors a month might resolve a few hundred people, of whom only some fit your customer profile. That can still be worthwhile if each sale is large. It is not a replacement for form fills or good analytics. Pair it with intent signals such as pricing-page visits and repeat sessions.

Many teams add a company-level layer to capture more of the traffic the person-level graph misses. BusinessMCP describes this setup on its visitor identification page, including the option to ingest RB2B data via your own key. See also person-level versus company-level identification.

Warning signs in vendor claims

• A single rate quoted with no traffic mix.
• No mention of precision or false positives.
• Match rates that exceed what network and identity data can plausibly support.
• No free trial or data export.
• A refusal to explain data sources.

Bottom line

Expect most visitors to remain unknown, treat any rate as a range, and decide whether the identified minority is worth the monthly cost. The sensible next step is a short pilot with your own traffic and a hand-checked sample.

What to do with a low match rate

A low rate is not a failure. It is the normal state of this market. The question is what you do with the minority you can identify and how you serve the rest.

Prioritise intent over volume. A visitor who views pricing, returns twice in a week and reads a comparison page is worth more than twenty blog readers. Use identification to rank accounts by behaviour, not to build a mailing list.

Capture the rest another way. Strong forms, a chat prompt, a calendar link and gated resources produce first-party data with consent. These sources are lower volume but higher quality.

Add a second layer. Some teams add company-level identification for the traffic a person-level tool cannot see. BusinessMCP describes this approach in its RB2B alternatives comparison and is open about its own limits, including that person-level reveal needs your own key.

Use aggregates. Even anonymous traffic informs content and campaign choices: which pages convert, which sources bring buyers, which countries engage.

Tracking the metric over time

Match rates drift. Browser privacy changes, new consent banners and shifts in your traffic mix all move the number. Record fill rate and precision monthly in a simple sheet, along with the traffic mix. Note any site or banner changes beside each month. If the rate drops by half overnight, suspect a script or consent change before assuming the vendor failed. Bring that record to renewal discussions; vendors respond better to specific numbers than to vague dissatisfaction.

How match rate interacts with price

Per-person credit pricing means a higher match rate raises your bill, while flat plans do not. If your traffic is mostly US and well matched, credits may run out quickly, which is a good problem if the leads are good. If your match rate is low, a credit plan keeps costs proportionate. See RB2B pricing for the arithmetic, and consider whether a flat model suits you better if volumes are unpredictable.

Remember that no one outside the vendor can audit the underlying data. That is why a hand-checked sample is the only honest basis for trust.