Website visitor identification is software that tries to name the company behind anonymous traffic on your site, and sometimes the person. A script on your pages matches each visit to a company and pushes the result to Slack, email or your CRM.
Leadfeeder and RB2B have free tiers. Entry paid plans on Leadfeeder, RB2B and Albacross cost under €100 or $100 a month (Leadfeeder and Albacross on annual billing), at the time of writing. Warmly starts at $10,000 a year, and ZoomInfo sells by quote. Person-level data costs more and covers fewer visitors.
Is it worth acting on? Only when three things hold at once:
- Your target market is too large to simply contact all of it.
- A real share of your traffic comes from companies in that market.
- Your follow-up stays human and never mentions the visit.
If one of the three is missing, put the same money into the offer and the list instead. Fix that first. Then buy the pixel.
What does a visitor ID tool actually return?
Analytics counts visits; identification tries to name who made them. If you came here for a traffic checker, heatmaps or session recordings, those are different tools.
Identification comes in two levels:
- Company-level returns an organization, such as "a logistics firm in Manchester read three pages". You still have to find the person who buys.
- Person-level returns a named individual, usually as a LinkedIn URL or a business email.
Most paid tiers bundle identification with enrichment, contact data from the vendor's own database, so part of what you pay for is a list.
When a founder asks me why nobody answers cold emails, I ask for the list before I read a single message, because the rows decide the result (how I take a lead list apart). An identified visitor is one more row.
When does visitor identification make sense for a small team?
I use intent data when the total market is large. Take 2,000 target companies with signals on about 10 of them. I skip the signals there and build one offer that works for all 2,000. Intent data is one tool among several. Chasing 10 signals is a coin flip on timing: any of them can be busy, unready or uninterested this month.
Visitor ID is first-party intent data, so the same rule applies. For the paid third-party kind, see the wider intent-data picture for small teams.
"Real B2B traffic" means visits from companies that fit your ideal customer profile (ICP): right industry, size and region. Students, job seekers, suppliers, competitors and existing clients get identified too. None of them is a lead.
The free tiers make a test cheap. Run one for 30 days, sort the output against your ICP, then decide:
| What you find | Decision |
|---|---|
| Your market is small enough to contact in full | Skip it and contact the market directly |
| Large market, but only a handful of ICP companies visited | Skip it, your traffic is not B2B enough |
| Large market, steady ICP visits, few replies to your outreach | Fix the offer before you pay for data |
| Large market, steady ICP visits, outreach already converts | Test a paid tier for one quarter |
The third row matters most. When good companies visit and nobody books, you may not have a lead gen problem at all. You may have a packaging problem: what you sell and how it reads to a stranger. Prospects check your LinkedIn and your website before they ever talk to you, and that is where they decide whether you are legit.
What do the main visitor ID tools cost?
Each vendor meters something different: identified companies, resolutions (visitors matched to a company or a person), contact credits or licenses. Figures come from each vendor's pricing page at the time of writing (Leadfeeder, RB2B, Albacross, ZoomInfo, Warmly).
| Tool | Free option | Paid entry | Output at entry | Higher tiers |
|---|---|---|---|---|
| Leadfeeder | Lite, €0, 100 identified companies a month | Discover, €79 a month billed annually | Company names pushed to your CRM | Activate from €369 adds verified emails, phone and intent filters; Scale from €599 |
| RB2B | Free, $0, 150 resolutions a month, company names to Slack, no person-level data or emails | Starter, $79 a month, 300 resolutions | LinkedIn URLs of US visitors to Slack or Teams | Pro from $149 adds business emails and all integrations; Pro+ from $199 adds premium resolution |
| Albacross | 14-day trial, no card needed | Starter, €59 a month billed yearly or €84 monthly | 10 verified email and 5 verified phone credits | Professional, €149 yearly or €213 monthly; a custom plan on request, annual billing only |
| Warmly | No free plan | AI Web-Deanonymization, from $10,000 a year on annual billing, or $4,875 a quarter | Contacts and companies, ICP filtering, real-time alerts | Other plans on the vendor's pricing page |
| ZoomInfo | No public prices | Quote only | Website visitors listed in Copilot Advanced | Each export uses one credit; add-ons such as Streaming Intent change the price |
For a small team, the line that matters is where business emails start: two tiers up on Leadfeeder (Activate), at Pro on RB2B. Budget for that tier. If what you really want is named decision-makers, a contact database gives you those without waiting for anyone to visit.
What match rate should you expect?
No tool names every visitor. RB2B's own coverage figures at the time of writing: company-level identification works worldwide at 15 to 20% on Free, Starter and Pro, and 35 to 45% on Pro+. Contact-level identification is US only, at the lower band from Starter and the higher band on Pro+.
Even on the top tier, most visits stay unnamed at person level.
My team's campaigns mostly aim at buyers in the USA, the UK and Europe. For a seller whose buyers sit in London or Amsterdam, a US-only person layer shrinks to a company name.
Leadfeeder counts a company once per month however often it returns, and a company name still gives you nobody to email. Between the name and a booked meeting sits normal outbound work, so judge any tool by cost per booked meeting from the same ICP over the same months and ignore vendor demo rates.
Is website visitor identification legal in the UK and EU?
It can be, with consent and a lawful basis in place. This is not legal advice.
The GDPR covers "any information relating to an identified or identifiable natural person". Processing it needs one of the six lawful bases in Article 6, and if you rely on consent, you must be able to show it.
Person-level output is personal data. Company-level output is safer ground. But the lookup still runs on IP addresses and cookies. Tie a visit history to one named employee and it becomes personal data.
The UK regulator works the same way. The ICO's guidance on identifiers says a name like "John Smith" may not identify anyone on its own. Add other details, such as where that person works, and it can become personal data.
In the EU, Article 5(3) of the ePrivacy Directive requires consent before storing or reading information on a user's device, unless it is strictly necessary for a service the user asked for.
In the UK, regulation 6 of PECR covers storing or accessing information on a user's device. The ICO's guidance requires you to say which cookies you set and why, and to get consent before any non-essential one.
Consent needs an unambiguous positive action, such as ticking a box; carrying on browsing does not count. Similar technologies are covered too.
The current PECR text lists its exceptions in Schedule A1. Strictly necessary storage for a service the user asked for needs no consent (the ICO's examples: basket contents, banking security, load balancing). Statistics about how your service is used and website appearance are exempt too, if users get clear information and a simple, free way to object.
A script that names visitors for a sales team does not look like strictly necessary storage, and its purpose goes beyond statistics. Ask your lawyer, plan on consent, and expect your match rate to shrink to consenting visitors.
What you do next has rules as well. My team does not cold email companies in Germany, Austria or the Czech Republic without a compliant basis; we reach them on LinkedIn instead. For the outreach side, see whether scraping LinkedIn and cold emailing are legal. For how vendors describe their own position, see what B2B data providers say about GDPR.
How should you act on identified visitors without getting creepy?
Never open with "I saw you on our pricing page". It tells a stranger they were watched. Buyers check the sender before they answer. Across two companies, one message earned a 2% reply rate at the first and 7.5% at the second, and I watched it happen. A creepy opener fails that check in seconds.
Run each identified company through five questions, in this order:
- Is it in your ICP? If not, do nothing. Most identified companies stop here.
- Does a rep already own it? If it sits in a live sequence, tell that rep and start nothing new. Two people writing to one buyer kills the thread.
- Did the visit show a buying question? Pricing, a case study from their industry, a service page. A single blog read only earns the account a place on your list.
- Who actually decides? Find the real title for that niche, not a generic "manager".
- Which channel first? I put a LinkedIn touch first and send the email after it.
In a first message, sell the meeting, never the service. The visit decides who you write to and when. It never decides what you say. Here is a made-up example from a temporary fencing supplier to a procurement lead at a civil engineering contractor that read its rail case study:
Subject: fencing moves on live rail sites
Hi Sarah, on rail jobs the fencing rarely stays put for long. As work moves along the line, someone has to relocate it, often every week. We supply and move temporary fencing for civil contractors on live rail sites across the North of England. If you have a rail package starting next year, would a 15-minute call help you see whether our phasing fits your schedule?
It names her situation and asks for a short call. Nothing in it reveals the visit. Judge it inside your whole outbound program: count conversations started, aiming for 50+ a month across all outreach, and let booked meetings come out of those.
The verdict: when to buy and when to skip
Buy when all three conditions from the top hold. Then the tool tells you which accounts to work first this week.
Skip it when the market is small, when most traffic will never buy, or when you hope a pixel will rescue a weak offer. Reach the whole market with a sharp offer before you chase the ten companies that happened to visit. Volume negates luck. A pixel does not. Visitor data can wait until that works, and free trigger signals cost nothing in the meantime.
If you are weighing a visitor identification pixel and want a view on whether your market is big enough to justify it, book a call and we can run the test together.
Questions people ask about visitor identification
Can a small B2B team use visitor identification without a CRM?
Yes, but the data goes stale fast. Without a CRM, Monday's visitor is out of view the following week. Set one up first; which one matters far less than using it.
Does an anonymous visitor counter still fall under UK cookie rules?
Yes. The ICO says PECR applies to all cookies, even when the data collected is anonymous, so "we only count visits" does not take a cookie outside the rules. At best it fits the statistics exception above.
Once the data is not anonymous, UK GDPR and the Data Protection Act apply on top. The ICO is also consulting on revised cookie guidance at the time of writing.
What if most of my traffic is outside my target market?
Filter before you judge the tool. Albacross lets you filter visitors on revenue, location, industry, employee count, technology used and UTM parameters. On Leadfeeder's free plan, every unique company identified counts toward the monthly cap, so suppliers, universities and competitors can use up your allowance before a single buyer appears.
What does a one-month paid test cost?
More per month than annual billing, without the year-long commitment. Leadfeeder's Discover plan costs €113 on monthly billing, about 43% more than its annual rate. RB2B's paid plans include a 7-day full-featured trial, and Starter charges $0.45 for each resolution beyond its monthly allowance. Run the free 30-day test first, then pay monthly through the paid quarter and switch to annual only if it pays off.