Website visitor identification match rate: why UK numbers fall

A website visitor identification match rate is the share of visits a tool can attribute to a named organisation, and it is lower than most buyers expect. The reasons are not mysterious: hybrid working moves visits onto home broadband, mobile networks share addresses between many customers, and IPv6 changes which address a visit arrives from. This piece explains each factor with public UK figures, so you can judge any vendor's number, including ours.

What is a match rate, and why do vendors' numbers differ?

A match rate is a fraction, and the denominator decides everything. The same tool on the same site can report very different rates depending on what it divides by:

Vendors rarely use the first. Happierleads' 2026 benchmark report, for example, defines match rate as a share of "total qualified sessions, after bot and consumer-ISP filtering".¹ That is a legitimate choice, but it means a headline such as "we identify 60% of visitors" usually describes 60% of a filtered subset, not 60% of everything in Google Analytics.

When you compare tools, ask each vendor the same three questions: what is the denominator, which traffic is excluded before the calculation, and does a match include low-confidence guesses from broadband providers?

How much B2B traffic comes from home?

A lot, and disproportionately from the people B2B websites want to reach. The Office for National Statistics reported that "more than a quarter (28%) of working adults in Great Britain hybrid worked" between January and March 2025. Among workers earning £50,000 or more the figure was 45%, compared with 8% of those earning under £20,000.²

Senior, well-paid staff are the buyers and influencers most B2B sites exist to attract, and nearly half of that group split their week between office and home. When they read your site from home, the visit arrives from their broadband provider's network, not their employer's. No IP-based tool can honestly turn a residential connection into a company name. The best it can do is recognise the provider and say so.

This is why ISP names appear in raw visitor lists. Our explainer on why ISPs show up as website visitors covers the mechanics.

What does IPv6 change?

Nothing about the principle and a fair amount about the data. IPv6 is the newer internet addressing scheme, with vastly more addresses than IPv4. A company can still be identified from an IPv6 address if the tool's data covers the ranges that company uses.

Adoption has crossed a symbolic line. The Internet Society reported that access to Google over native IPv6 "has for the first time exceeded 50% (50.10% on 28 March 2026, to be precise)".³ APNIC reported Google's measurement reaching 50% on 23 April 2026, while APNIC's own weighted measurement put worldwide IPv6 capability at 42% on the same date. APNIC suggests treating "the two data sets together, as they effectively bracket the likely range of actual IPv6 capability".⁴

In the UK the picture is uneven by provider. Sky said in September 2016 that "over 90% of its broadband customers are now IPv6 enabled".⁵ Virgin Media is the counter-example: an independent tracker of its status says it has "not yet deployed IPv6", with the most recent update in August 2025.⁶ Google does not publish a UK-only figure in a form we can cite, so we will not guess one.

What this means for identification:

• A visit may arrive over IPv4 on one day and IPv6 the next, depending on the network.
• An IPv6-only match is only possible if the tool's company-to-network data covers IPv6 ranges.
• Home broadband on IPv6 is still home broadband. A new address format does not reveal an employer.

What is CGNAT, and why does it blur mobile visits?

Carrier-grade NAT (CGNAT) is a way for internet providers to share scarce IPv4 addresses. Wikipedia describes it as "a type of network address translation (NAT) used by Internet service providers (ISPs) in IPv4 network design", and notes the technique was first used in mobile networks to meet their need for large numbers of addresses.⁷

Router maker DrayTek puts the effect simply: "A single public IP address is shared between multiple customer's routers", and "Hundreds of other customers will be sharing that same public address." It also says "CGNAT is most widely used on mobile phone networks".⁸

So when someone reads your pricing page on a phone over 4G or 5G, your site sees an address shared by many of that carrier's customers. It resolves to the carrier, never to the employer. Happierleads' benchmark estimates that mobile B2B traffic is identified at rates roughly 30 to 50% lower than desktop, because carrier addresses are shared by many consumers and "rarely resolve to a single employer".¹ Treat that as a vendor estimate rather than a measured UK figure, but the direction is not in doubt.

What about VPNs and privacy relays?

A VPN routes traffic through another network before it reaches your site, so the visit shows the VPN's exit address. A corporate VPN that exits through the company's own network can help identification. A consumer VPN or a cloud-hosted gateway hides it. Browser privacy features such as Apple's iCloud Private Relay work in a similar way for a subset of Safari users.

We cover those in detail in Private Relay, Chrome and the future of IP matching. The short version: they matter less than hybrid work and mobile, but they are another reason no tool reaches every visit.

So what is a realistic UK match rate?

There is no single honest number, because it depends on your audience. A site read mainly by office-based engineers at large firms will match far more visits than one read by consultants on phones.

Industry sources give a range to start from. Happierleads' 2026 benchmark, a vendor report blending public benchmarks, its own customer cohorts and competitors' reported figures, puts the United Kingdom at 35 to 55% company-level identification of qualifying B2B traffic.¹ That is their figure and their denominator, not ours, and it excludes consumer ISP traffic before it starts.

Treat any single number, including a vendor's, as a claim until you have run a trial on your own traffic. We do not publish a Scout match rate for this reason. What we do publish is how sure each match is: every identified company gets a confidence score from 1 to 99 with a band and a one-line reason, and ISP, residential, mobile, VPN and data-centre traffic is flagged and held in the low band. A tool that shows you its uncertainty is easier to judge than one that quotes a high rate.

How to get more from the visits you can identify

A lower match rate is not a failure if the matches you do get are reliable and you use them well.

• Link Google Analytics 4. Session depth and pages read turn a bare company name into a visit you can judge. Scout reads these from GA4 for each identified company.
• Use search attribution. Google Search Console can show the search queries associated with each company, which often says more about intent than the pages themselves.
• Weight return visits. A company that comes back from its office network three times in a fortnight is a stronger signal than one visit, and it gives the tool more chances to match.
• Keep forms for the rest. Visitors on mobile or home broadband who want to talk will still fill in a form or book a call. Identification complements lead capture. It does not replace it.

The pillar guide to website visitor identification covers the full method, and the pricing page explains how plans are measured in identified visits per month.

The takeaway

Match rates fall for structural reasons: nearly half of higher earners in Great Britain hybrid work, mobile networks share addresses through CGNAT, and IPv6 only helps if a tool's data covers it. Ask every vendor for their denominator, prefer tools that tell you how confident each match is, and measure on your own traffic.

Run Scout on your own site for 14 days and see your real number. Start a free trial, no card needed.

Frequently asked questions

Does IPv6 make visitor identification impossible?

No. A company can be identified from an IPv6 address in the same way as from IPv4, provided the tool's data covers the IPv6 ranges that company uses. IPv6 does not reveal the employer behind a home broadband connection, though, so it does not solve the hybrid-work gap.

Will match rates keep falling?

Nobody can say for certain. Hybrid work and mobile browsing are established, so a large share of B2B visits will keep arriving from home and mobile networks. The practical response is to use confidence scoring, return visits and search data to get more from the visits that can be identified.

Why is desktop traffic identified more often than mobile?

Desktop visits more often come from an office network registered to the company. Mobile visits on 4G or 5G usually pass through carrier-grade NAT, so many customers share one public address and the visit resolves to the carrier. A vendor benchmark from Happierleads estimates mobile identification rates at roughly 30 to 50% lower than desktop.

What match rate should I expect in the UK?

It depends heavily on your audience and how the rate is calculated. One vendor benchmark puts the UK at 35 to 55% of qualifying B2B sessions after filtering out bots and consumer ISPs. The only figure that matters for you is the one from a trial on your own site.

Sources

1. Happierleads: 2026 B2B website visitor identification benchmark report happierleads.com
2. ONS: Who has access to hybrid work in Great Britain? ons.gov.uk
3. Internet Society Pulse: 18 years later, IPv6 reaches majority pulse.internetsociety.org
4. APNIC Blog: Google hits 50% IPv6 blog.apnic.net
5. Sky Group: Sky completes roll out of IPv6 skygroup.sky
6. Have Virgin Media enabled IPv6 yet? havevirginmediaenabledipv6yet.co.uk
7. Wikipedia: Carrier-grade NAT en.wikipedia.org
8. DrayTek: What is Carrier Grade NAT (CGNAT)? draytek.co.uk