How Does Leadfeeder Work? Visitor Identification Explained

Tools like Leadfeeder work by running a tracking script on your site, matching each visitor’s network address to an organisation in a company database, and attaching firmographic details such as industry and size. It names a company only when the visitor connects from that company’s own network.

The three layers in plain terms

Visitor identification products are built from the same basic layers, even when branding differs. The description below follows the structure in BusinessMCP’s guide to how visitor identification works, which is vendor-authored but describes the general technique.

1. Capture. A script on your pages records visits: which pages, in what order, for how long, and from which network address.
2. Resolution. The network address is looked up against databases that map address ranges to organisations. This is probabilistic. A match means “this range belongs to Acme Ltd”, not “an Acme employee is definitely here”.
3. Enrichment. The matched company domain is enriched with industry, size, location and sometimes technology information. The guide notes these are snapshots that can go stale and vary in quality by company segment.

Why the IP layer is a filter, not a census

The guide’s core caveat is that home internet providers, mobile carriers and VPNs resolve to the provider rather than the visitor’s employer. A remote worker at a kitchen table looks like a consumer connection. A visitor on a corporate office network, or a corporate VPN with company-owned address space, looks like their employer.

That is why the commonly quoted company-level range is about 20 to 35 percent of B2B traffic, according to the same guide, and why enterprise and mid-market traffic resolves better than small business or consumer traffic. Geography matters too, with stronger coverage in North America and Western Europe.

Key takeaway: a company name next to a visit is a lead for investigation. It tells you a network, not a person, and certainly not a buying decision.

Layer four: stitching with email

The second method is deterministic. When a visitor fills out a form or clicks a tracked link in an email, their business email domain confirms the company. According to the guide, stitching is retroactive: when a visitor identifies, earlier anonymous sessions can be attached to the new contact.

This is the bridge between anonymous traffic and known accounts. It is also why your forms matter. The guide says about 98 percent of visitors never fill out a form, so every identified domain from a form is worth far more than an unconfirmed IP match.

What you see in the dashboard

Leadfeeder’s output, as described in BusinessMCP’s comparison, is a dashboard and CRM records. Typical elements across tools in this category include:

Element	What it tells you	Caveat
Company name and domain	Who owns the network	May be a provider or cloud host
Pages viewed	What interested them	One visitor may not speak for the firm
Visit count and recency	Repeat interest	Multiple people can share a network
Firmographics	Fit with your ICP	Data can be stale
Source	Which channel brought them	Depends on tracking quality

Where cookies come in

The comparison page describes Leadfeeder as cookie-based, requiring consent in the EU, and BusinessMCP as cookieless. Cookies help recognise a returning browser across sessions, but they also trigger consent requirements, and visitors who decline are invisible. A cookieless tracker resolves company from the network signal at request time, per the guide. Which approach is right for you is partly a legal question, so we cover it in GDPR-compliant visitor tracking and recommend taking advice for your situation.

A quick way to see resolution for yourself

You can test the idea without a subscription. Take a handful of IP addresses from your logs, paste them into a free tool such as BusinessMCP’s reverse IP lookup and read what comes back. You will notice that many addresses belong to broadband providers and cloud platforms. Those results are not failures. They are the tool honestly telling you it cannot see the employer.

For a fuller method on measuring this across a day of logs, read how to see which companies visit your website.

Limits worth remembering

• Shared networks. Many people share one office address, so one visit does not tell you who read your pricing page.
• Stale data. Address ranges change owners, and company records age.
• Wrong attribution. A consultant visiting from a client’s network is attributed to the client.
• Bot and cloud traffic. Crawlers and cloud-hosted tools resolve to hosting companies, which should be filtered out.
• No intent guarantee. A visit can be research, a competitor or a student.

What to do with this knowledge

When you evaluate any tool in this category, ask how it handles non-business networks, whether it separates company from provider results, how often its data refreshes and what it does with visitors who decline consent. Then judge the answers by testing on your own traffic using the approach in our accuracy guide.

Understanding the mechanism protects you from the two classic mistakes: expecting names for most of your traffic, and treating a name as a contact. Used as a prompt to look closer, the technique is genuinely helpful. Used as a verdict, it misleads.

A walk-through of one visit

Picture a visitor from a mid-sized firm opening your pricing page from the office network. The script records the page, the time and the network address. Resolution finds that the address range is registered to that firm, and enrichment adds its industry, size and location. A minute later the visit shows in the dashboard under the firm’s name with the pages viewed.

Now picture the same person at home that evening. The address belongs to a broadband provider, so the tool either shows the provider or discards the visit as non-business. The same person produces one named visit and one anonymous visit. That is why repeat behaviour and email stitching add so much: they let one identified moment explain several anonymous ones.

Different vendors vary in how they handle the second case, and in how clearly they label it. BusinessMCP’s comparison page is one place to see how a competitor describes its own approach, including the caveat that match rates for any vendor are roughly 20 to 35 percent of B2B traffic.