What website visitor identification actually means

What are the two kinds?

Resolution tells you who the visitor is. A script records the visitor's IP address, the vendor looks it up against a database that maps address ranges to companies, and the live board says a person from a named company read your pricing page. The more aggressive versions go further and resolve to an individual, matching against identity graphs assembled from data the person handed to some other website years ago.

Behavioural intent tells you how interested they are. No name, no company, no bought data. It reads the things the browser already knows: which page, how long, how many pages, how many visits, what brought them, how far down they scrolled. A model turns those into a number, and the number ranks who to talk to first.

Both get called visitor identification in marketing copy. They are different purchases with different prices, different accuracy problems and different legal exposure.

Where does the identity data come from?

This is the question that separates the two, and it is the one a demo will not volunteer.

Company level resolution comes from IP allocation records: which ranges a business registered, which ranges an internet provider assigned to a corporate office. It works when somebody is at their desk on the corporate network. It degrades on home broadband, on mobile data, behind a VPN and behind the privacy relays now shipped by default on several platforms, and in each of those cases it either returns nothing or returns the provider rather than the employer.

Person level resolution comes from identity graphs: third party datasets that tie a device or a hashed email to a name. You are buying access to data somebody else collected from somewhere else. Whether the person whose name comes back ever agreed to that is not a question the vendor's dashboard shows you.

Is it legal?

Not a question anyone should answer for you in a blog post, and this is not legal advice. But the shape of the problem is worth knowing before you sit through the demo.

In the EU and the UK, an IP address is generally treated as personal data. The Court of Justice of the European Union held in Breyer (2016) that a dynamic IP address can be personal data in the hands of a website operator where the operator has legal means to identify the visitor. That puts reverse IP lookup inside data protection law rather than outside it, and person level resolution from bought data further inside it again.

In practice this means the question is not whether the technology works. It is whether you have a lawful basis to do it to the specific people visiting your site, and whether you can tell them you are doing it. Plenty of businesses can answer that comfortably. Ask your lawyer rather than the vendor.

Which one do you actually need?

It depends on what you do after you find out, and there are two honest answers.

If you sell to companies with long cycles and a sales team that researches accounts, resolution is worth paying for. Knowing that somebody at a named company read your pricing page three times gives an account executive something to act on a week later.

If somebody on your team could be talking to that visitor while they are still on the page, resolution mostly does not help. The useful question at that moment is whether this person is worth interrupting, and behaviour answers it. A name would not change what you say.

• Long sales cycle, named accounts, outbound follow up later: resolution earns its price.
• Someone available to answer now, purchase decided in one session: behaviour is the signal that matters.
• Consumer traffic rather than business traffic: company resolution returns the internet provider and nothing usable.
• Mostly mobile traffic: resolution accuracy drops sharply, whatever the vendor's landing page says.

What does IntellQ do?

Behaviour, and only behaviour. IntellQ stores no IP address at all. Country, region and city come from the headers the CDN already attaches to the request, they are coarse, and the address itself is never written down.

A visitor gets a name in exactly one case: your own site tells IntellQ who they are, by calling intellq.identify() after they sign in or fill in a form. That is your data about your customer, passed deliberately by you. There is no lookup, no identity graph, no third party dataset, and no reverse IP resolution anywhere in the product.

If reverse IP company resolution is the thing you came looking for, IntellQ is not it and no amount of configuration will make it so. Several products do that well and this page is not a reason to avoid them.

How do you tell which kind a vendor is selling?

Three questions, asked in the demo, in this order. Each one has a short true answer and a long evasive one.

• Where does the identity data come from? A resolution vendor names a data source. A behavioural product says it has none.
• What is the match rate on mobile traffic, and on visitors outside the United States? Resolution accuracy falls off in both, and the honest vendors say so with a number.
• Do you store the IP address, and for how long? This is the question that decides whether your privacy policy needs rewriting.