The Warm AI Blog

It is software that tells you which companies, and in some regions which people, visited your website without filling in a form. It matches the anonymous visit against data sets that link web activity to real businesses.

When someone lands on your site, a small script checks signals from that visit against a data set the tool has built or bought. Where it finds a confident match, it returns the company, and in some regions the person, along with the pages they viewed.

You can see which companies visit, and in the US and Canada often the individual. In the UK and EU you mostly see the company first, because of privacy law. You will not see everyone, because the data behind these tools only covers part of the web.

Analytics counts sessions and pages anonymously. Visitor identification adds the missing layer: which business or person was behind the visit, so you can act on it.

Yes. That is the whole point. It identifies visitors who never fill in anything, which is the large majority of B2B traffic.

Usually the company name, location, industry, and size, plus the pages they viewed and how often they came back. Person-level detail, where it is legally available, can include name, role, and LinkedIn.

Less reliably. People working from home connect through their home internet provider, not a company network, so IP-only matching struggles. Tools that combine several signals and data sources recover more of this than single-signal tools.

Accuracy varies by tool, traffic, and region, so treat any single headline number with caution. Company-level identification on business-heavy traffic tends to be more reliable than person-level. The honest test is a trial on your own site.

There is no universal figure, because match rate depends on your traffic and region. The only match rate that means anything is the one you measure on your own traffic over a couple of weeks.

Each provider only holds data on part of the web, so it can only match the visitors inside its own data set. A tool that queries several providers catches more, because the gaps rarely line up.

Usually because the promised number came from someone else's traffic, or from a single data source. Your real rate depends on how much of your traffic is business traffic, and which regions it comes from.

Often, yes. A flat number quoted before anyone has seen your site is a guess dressed up as a fact. Ask for a trial and judge the real, useful identifications.

Only if the matches are accurate and useful. A big number padded with weak guesses is worse than a smaller set you can trust.

Pricing ranges widely, from around a hundred dollars a month at the low end to enterprise contracts in the tens of thousands a year. Watch for long lock-ins, and prefer terms that let results decide.

It is worth it if your site gets real business traffic and you act on what it shows. It is wasted if you install it and never follow up, or if you blast everyone it identifies.

Compare the cost against the pipeline you create from visitors you would otherwise never have known about. Run a trial, count the genuinely useful identifications, and see how many turn into conversations.

No. Month-to-month, usage-based pricing lets you judge a tool on real results instead of committing before you have any.

Analytics shows you anonymous traffic for free, but not identities. Identification is a paid layer, though many tools offer a free trial so you can test it on your own traffic first.

Identifying the companies that visit your site is generally legal, because company-level information usually is not personal data. Identifying named individuals is more restricted and depends on having a lawful basis and being transparent.

It can be. Company-level identification typically relies on legitimate interest and does not require consent. Personal data about a named individual needs a lawful basis, transparency, and an easy opt-out.

Generally yes. Knowing the business behind a visit usually sits outside personal-data rules, which is why company-level identification is widely used in the UK.

Only in limited ways. A named, identified person is personal data, so you need a lawful basis, transparency, and an opt-out. Many tools handle this by resolving to the company first.

If the tool uses non-essential cookies or similar tracking, then in the UK and EU you need consent before it runs, under PECR. Identification that does not rely on such cookies can sit on a different footing.

No. Consent under PECR is about setting the cookie. A lawful basis under GDPR is about processing any personal data you collect. You can need both.

It is the lawful basis many B2B tools rely on, where your interest in identifying business visitors is weighed against the person's rights, you are transparent about it, and you make opting out easy.

Yes. The US allows person-level identification more freely than the UK and EU, which is why many US tools lead with named individuals. State laws still apply, so it is not unlimited.

Company-level tells you which business visited. Person-level tells you which individual. Which one you get is decided mostly by law and region.

US and Canada traffic can often be identified at person level. UK and EU traffic is mostly company level first, with personal detail added only where there is a lawful basis.

Yes. You sell to the business, so knowing which company is evaluating you, and what they read, is enough to act on through your existing channels.

Usually because they are built around US person-level data, or they are overselling. If a tool promises named individuals on UK or EU traffic, read the small print.

The best one identifies the most of your traffic, in your regions, without crossing privacy law. The biggest real difference between tools is how many data sources sit behind them, so judge that, not the feature list.

Lead Forensics relies on a single data source and often a long contract. Warm queries several providers at once and returns the best match, and is GDPR and UK GDPR native.

Leadfeeder leans on its own data and behaviour tracking. Warm draws on several providers at once for broader coverage, and keeps the best match per visitor rather than relying on one source.

RB2B is one of the data sources the multi-provider approach can draw on. Warm sits a layer above single sources, querying several at once so you are not limited to any one provider's coverage.

Clearbit, now part of HubSpot, is one provider's data. Warm is multi-provider by design, so it is not capped by a single source, and it is built around UK and EU compliance.

You can, but you will only ever see that provider's slice of your traffic. Querying several and keeping the best match catches the visitors any single source misses.

You can, but it means managing several contracts, stitching the data together, and writing your own matching logic. A multi-provider tool does that for you in one place.

It is all the buying activity you cannot see in your analytics: research on Google, reviews, peer conversations, social lurking, and un-tracked website visits. Most B2B buying happens here.

Because most buyers research you quietly long before they fill in a form, if they ever do. The form gets the credit, but the real work happened where you could not see it.

Not all of it. You can recover parts, like identifying the companies visiting your website, but private research and conversations stay out of view.

Patterns, not single views: repeat visits, time on pricing or demo pages, several people from one company in a short window, and quick return visits.

It is a hint. It becomes meaningful when paired with repeat visits or several colleagues from the same company in a short window.

No. They raise the odds that someone is in-market, but they are probability, not proof. Use them to prioritise outreach, not to guarantee a sale.

About two minutes. You paste one line of script into your site before the closing body tag. No engineering project.

Not for the basic setup. Adding the script is a copy and paste. Connecting the data into your CRM or Slack may need a little help, depending on your stack.

Yes. The point of identification is to get the visitor into your workflow, so most tools sync to your CRM, push alerts to Slack, or export to CSV.

A well-built tracking script is lightweight and loads in the background, so the impact on page speed should be minimal. Check the script size if you are unsure.

In real time, or close to it. Visitors start appearing in your dashboard shortly after you install the script.

Reaching out to people or companies who already showed interest, like visiting your website, instead of cold contacts who have never heard of you.

See who visited, spot the ones showing real intent, route them to the right person, and reach out like a human who noticed. Keep it relevant, not a blast.

Look for intent: repeat visits, time on high-value pages, or several people from one company in a short window. A single bounce is not worth chasing.

It can, if you treat a visit as permission to spam. Handled as a relevant, human nudge to genuinely interested accounts, it lands far better than cold outreach.

No. A visit is a signal of interest, not a request to be contacted. Use it to be relevant and well timed, not pushy.