Jorge LewisSomeone from a target account read your pricing page this morning, and your analytics filed the visit under Direct with no name attached. Working out how to identify companies visiting your website is really two problems: what technology can put a company name on an anonymous visit, and what you should do with that name once you have it. This guide covers both, including the match-rate numbers most vendors keep in the footnotes.
We come at this from the outreach side. Our parent agency, Referral Program Pros, has run more than 4,000 outbound campaigns and booked over 7,000 meetings for clients, and GTM Bud was built on that playbook. Website visits are among the signals those campaigns act on, which is exactly why the limits of visitor identification matter to us: a misread signal becomes a badly targeted campaign.
How does website visitor identification work?
Website visitor identification is the process of matching an anonymous website visit to the company it came from, using the visitor’s IP address. A small tracking script on your site records each visit, a reverse lookup matches the IP against databases that map address ranges to the organizations that own them, and first-party cookies stitch repeat visits into one account’s trail. The output is firmographic: company name, size, industry, and the pages that company viewed. It is not a person, an email address, or a phone number. That company-level ceiling is stated plainly across the category’s own explainers, from Leadfeeder’s guides to ZoomInfo’s 2026 tool roundup, and it shapes everything downstream: the tool tells you an account showed interest, and finding the human inside that account remains your job.
Google Analytics will not do this for you, and it used to come close. Until February 4, 2020, Universal Analytics exposed Service Provider and Network Domain dimensions that marketers mined for corporate network names, until Google removed both, per Seer Interactive’s analysis of the change, and GA4 was rebuilt without them. Analytics now shows you counts, channels, and pages. The entire visitor identification category exists to fill the company-name hole Google left.
One boundary before going further. Your own website traffic is first-party intent: the strongest and cheapest corner of the wider intent-data market, whose paid third-party tiers we cover in our guide to buyer intent data. And acting on signals like these is its own discipline, signal-based outreach, with rules about signal tiers and decay that apply here in full. This article stays on the one signal you already own and nobody can sell you.
How much of your traffic can actually be identified?
A minority, and anyone who tells you otherwise is selling the tracking script. The published ranges disagree with each other, which is itself the honest finding: ZoomInfo’s 2026 visitor-tracking guide puts typical company-level identification at 15 to 30 percent of B2B traffic, Common Room’s comparison guide claims 30 to 65 percent for US B2B traffic at the company level and just 5 to 20 percent at the person level, and 2026 explainers from Storylane and Abmatic cluster around 20 to 35 percent. Leadfeeder’s marketed figure, identifying up to 45 percent of visiting companies, is vendor-stated and appears with that framing even in rival roundups. Every one of these numbers comes from a company with a product to sell. What they agree on is the shape: under half of your traffic resolves, often far under, and European traffic resolves at lower rates still because consent banners gate the tracking script before it fires.
The spread also explains the standard sales trick, which Coffee’s 2026 accuracy guide calls out directly: a headline match rate is usually a best case on ideal traffic, or a blend of company-level and person-level numbers into one flattering figure. The only match rate that matters is the one a pilot produces on your own traffic, checked against accounts you can verify.
The blind spots every vendor admits
The unresolved majority is not random noise. It falls into specific, predictable holes, and the vendors’ own guides concede each one:
- Home and mobile traffic is invisible. An employee browsing from home or a phone resolves to Comcast or Vodafone, not to their employer. Remote work widened this hole permanently, and category guides from Abmatic to Coffee date the decline in pure IP-based match rates to 2020 for exactly this reason.
- It identifies companies, not people. A visit from a 400-person company could be your champion, an intern, a competitor doing research, or someone prepping for a job interview. The data cannot distinguish them.
- Shared IPs misattribute. Coworking spaces, multi-tenant office buildings, and VPN exit nodes put many organizations behind one address, so the name in your dashboard can simply be the wrong company.
- Coverage skews large. Enterprises own registered IP ranges that resolve cleanly. A 10-person company on a consumer ISP plan often resolves to nothing, which matters if small companies are your ICP.
None of this makes the signal worthless. It makes the signal partial, which changes how much weight any single visit can carry.
Is it legal to identify companies visiting your website?
The category’s working posture under GDPR is company-level identification only, and the careful version of that position comes from the vendors themselves: Leadfeeder’s own compliance guide states that IP-based identification still involves personal data, so GDPR applies and you need a documented legitimate-interest assessment plus a privacy-policy disclosure, not nothing. Some rival guides claim company data falls outside personal data entirely; between the two readings, the vendor admitting its own product needs a lawful basis is the one to trust. Person-level identification, the de-anonymizing of individual visitors, is a different animal: vendors market it almost exclusively for US traffic and acknowledge it needs consent in the EU. Whichever tier you deploy, the responsibility is yours, not the vendor’s, and a tool’s compliance page is not a substitute for your own legal review. This is not legal advice; it is the map of where the questions live.
The visitor identification tool landscape in 2026
You do not need a roundup’s false precision to navigate this market, because the tools share one mechanism and differ mainly in packaging. The names below are the ones that dominate the category’s own search results, observed November 2026:
- Leadfeeder (Dealfront) is the category’s best-known name, and 2026 reviews from Sondero and SyncGTM report the Dealfront platform rebranded back to Leadfeeder in March 2026, so both names refer to the same product. Third-party 2026 pricing guides describe a free tier of up to 100 identified companies with seven days of history, and paid plans from roughly 99 per month on annual billing, priced by identified companies rather than seats; the guides disagree on whether that list price is in dollars or euros, so confirm before budgeting. The up-to-45-percent match figure is vendor-stated.
- Lead Forensics is one of the longest-standing names and works at the same company level, per rival-published tool lists. We found no independently verifiable pricing for it in November 2026 search listings, so budget for a sales conversation.
- Clearbit no longer exists as a standalone product: HubSpot acquired it in 2023 and folded its Reveal feature into Breeze Intelligence, with the legacy free Clearbit tools sunset in April 2025, per ZoomInfo’s and rival 2026 reviews. Reported pricing conflicts by two orders of magnitude across those reviews, from 45 dollar monthly credit packs to annual contracts near 20,000 dollars, so treat every figure as unverified and get the quote from HubSpot.
- Happierleads, Leadinfo, Albacross, and a cluster of newer challengers market variations on the same mechanism, some claiming person-level resolution for US traffic. We could not verify their specific match-rate or pricing claims from third-party sources, which is not a criticism, just a reason to pilot before you pay.
The selection method matters more than the selection: start on a free tier, run it against your own traffic for a few weeks, and measure the share of resolved names that are genuinely new target accounts. That number, not any vendor’s headline, is what you would be paying for.
From anonymous visit to booked meeting
Identification is the cheap half of the problem. Across the 4,000+ campaigns our agency has run, a signal only ever mattered when it changed who got contacted, when, or with what, so here is the workflow that turns a dashboard row into a conversation.
First, filter the noise. Strip ISPs and telecom names (unresolved home traffic), competitors, existing customers, and careers-page traffic. What survives is a short list of accounts worth scoring.
Second, score by visit behavior. Not all visits are equal, and the page trail is the signal strength:
| Visit behavior | Signal strength | What to do |
|---|---|---|
| Pricing page plus repeat sessions across days | Strong | Prioritize now: find the buyer, launch a campaign |
| Comparison or case-study pages, single session | Moderate | Check ICP fit, queue for the next campaign |
| One blog post from search, then gone | Weak | Leave it alone; that is content marketing working |
| Careers pages | Noise | Exclude: job seekers, not buyers |
| ISP or telecom name | Unresolved | Exclude: home or mobile traffic the tool cannot place |
Third, turn the account into people. The tool hands you a company, and the mechanics of getting from a company name to the two or three humans with buying authority are a separate skill, covered in our guide to finding decision makers in a company.
Fourth, move fast. A pricing-page visit means the account is evaluating this week, not this quarter, and the same decay that governs inbound speed to lead governs visit signals: the value of the signal falls daily while the prospect’s shortlist hardens.
Fifth, never open with the visit. “I noticed your company was on our site” is the one message this data makes tempting and the one it least supports: the identification is account-level, so you are often writing to someone who never visited, and even when you guess right it reads as surveillance. Use the visit to decide who enters the campaign and when it launches. Write the message about the problem you solve, as if the visit never happened.
That last step is where signal-driven outreach tooling splits. Signal-native tools, with Coldreach the best-known name in that niche, watch for signals like website visits and trigger sequences off them, but mostly stay email-only; we map that trade-off and the LinkedIn-inclusive route in our Coldreach alternative breakdown. GTM Bud sits downstream of whichever identification layer you choose: it is not a visitor-ID tool and puts no script on your site, it takes the accounts your signals surface, researches the people, writes the messages, and runs the LinkedIn and email sequences.
Frequently asked questions about identifying website visitors
Can you see exactly who visited your website, not just the company?
Usually not. The standard mechanism resolves an IP address to an organization, so the output is a company name, never a person. A newer class of person-level tools claims to de-anonymize individual visitors, but vendors market it almost exclusively for US traffic, it requires consent under GDPR per the vendors’ own compliance guides, and even the friendliest vendor figures, like the 5 to 20 percent person-level match rate in Common Room’s 2026 comparison guide, cover a small slice of traffic. Treat person-level identification as a US-only add-on with its own legal review, not the default.
Does website visitor identification work for low-traffic websites?
The mechanism works, but the economics often do not. If your site gets 1,000 visits a month and tools resolve somewhere between 15 and 45 percent of traffic to a company, per the vendor ranges above, you are looking at a few hundred company hits, and after removing ISPs, bots, competitors, job seekers, and existing customers, the net may be a handful of genuinely new target accounts. That can still be worth a free tier. It is rarely worth a paid annual contract until your traffic, or your deal size, makes each resolved account valuable.
Do consent banners and ad blockers affect visitor identification?
Yes. Visitor identification runs from a tracking script on your site, so anything that stops the script, including a declined consent banner or a script blocker, stops the identification. This is one reason vendor guides consistently report lower match rates on European traffic, where consent banners gate tracking by default. Factor your audience region into any match-rate estimate a vendor shows you.
Can you email someone just because their company visited your website?
A company-level visit is not consent, and it does not even give you an address: the tool names an organization, and you still have to find the right person yourself. Once you have, normal cold outreach rules apply as if the visit never happened, including lawful-basis requirements for email in regulated regions. In practice a LinkedIn connection request is the softer first touch for a visit-sourced account, because it asks permission before it pitches, which is the motion LinkedIn outreach automation runs at scale.
How long should you trial a visitor identification tool before paying?
Long enough to compare the tool’s output against accounts you can verify, which usually means a few weeks on a free tier. Check three things: what share of resolved names are real target accounts rather than ISPs and noise, whether accounts you know visited (from demo calls or replies) actually appear in the feed, and whether anyone on your team acted on a single alert. If nothing in the feed changed what you did, the paid tier will not either.
Turn the visits you already earn into pipeline
Identifying companies visiting your website is a solved mechanism with honest limits: IP-to-company matching plus first-party cookies resolves a minority of traffic, names organizations rather than people, misses home and mobile entirely, and occasionally names the wrong tenant behind a shared IP. Used with those limits in view, it is still the cheapest buying signal you will ever get, because the accounts it surfaces came looking for you.
The return lives entirely in the follow-through: filter the noise, score the page trail, find the people, move inside the week, and never mention the visit. If the identification layer is running and the bottleneck is everything after it, that execution gap is what GTM Bud closes. It turns the accounts your signals surface into researched prospects, personalized messages, and running LinkedIn and email sequences through automated lead generation.