The demo goes well. A dashboard fills with company logos, someone says a big round number out loud, and for about ninety seconds it feels like your pipeline problem is solved. Then the trial starts and the list looks thinner than the slide did. This piece explains, in the plainest words possible, how to test website visitor identification software yourself — and how to work out which one is best for your traffic rather than somebody else's. Three things you measure in week one, before you sign anything.
So what does visitor identification software actually do?
Someone visits your site. Their connection has an IP address. If that address belongs to a company network rather than a private household line, it can be matched back to the company that owns it. You get a company name, not a person's name.
Think of it like recognising a delivery van by the logo on the side. You know which firm sent it. You don't know the driver, and you're not trying to find out. lead.box works at company level on purpose — that's a product decision and a legal one at the same time.
Visitor tracking software and visitor identification software are not the same thing
Most people searching for visitor tracking software already have tracking. Analytics tells you, in aggregate, roughly how many sessions happened and roughly which pages and channels they came from. It never tells you who.
Identification answers the "who" — at company level. Tracking counts, identification names. If a vendor blurs those two words in their marketing, that's your first small data point about how they talk.
Why every recognition rate you'll ever be shown is really about someone else's traffic
Here is the uncomfortable truth the category rarely says out loud: recognition rate is not a property of the software. It's a property of your visitors. The same tool on two different websites produces two very different numbers, and neither one is a lie.
What moves the number:
- Office versus home: people on a corporate network are identifiable, people on a home line or a phone usually are not
- Company size: large offices with their own IP ranges show up far more reliably than a five-person agency on a consumer contract
- Traffic source: branded search and industry newsletters skew toward desk-bound office traffic, paid social skews mobile
- Country and ISP structure: business connections are registered differently from market to market
- Bots and crawlers: if they aren't excluded, they inflate everything
So when you see "we identify X% of your visitors," the honest translation is: "on the sites we chose to measure, in the period we chose, counting the things we chose to count." That's not fraud. It's just not a promise.
The week-one test protocol
You don't need a procurement committee for this. You need a trial, two weeks of normal traffic, and a spreadsheet. Three measurements, in this order.
- Count what share of the "identified companies" list is made up of internet providers rather than businesses you could sell to
- Recalculate the recognition rate on your own traffic mix instead of the vendor's sample
- Confirm where the data is processed and get the data processing agreement in writing before you pay
Everything else — dashboards, integrations, scoring, alerts — is a preference. These three are facts.
Measurement 1: how much of the list is internet providers?
A large share of identified visits resolves to a telecoms or hosting company: Telekom, Vodafone, a cable provider, a cloud host. That's not a bug and it shouldn't be swept out of sight — it's an accurate description of who owns that IP address. It's also, for most sales teams, not a lead. Deutsche Telekom is only a prospect if you actually sell to Deutsche Telekom.
The question is not whether these entries exist. They will, in every tool in this category. The question is whether the vendor counts them inside the recognition rate they put on the website. That single accounting choice can double a headline number without a single extra company being found.
TIP — Ask one question in the demo: "Does the recognition rate in your marketing include records that resolve to internet or hosting providers?" A clear yes or no tells you more about the vendor than the whole feature list. A long answer is also an answer.
Measurement 2: recognition on your own traffic mix
Do the arithmetic yourself, because nobody will do it for you. Take two weeks. Divide the number of sessions where a company was identified by the total number of sessions. Then do it again with provider records counted in a separate column. Now you have two numbers: the marketing one and the working one.
Then split it by channel. This is where the useful insight hides. A tool can look mediocre overall and excellent on exactly the traffic you care about — say, organic search on your solution pages — while looking terrible on a mobile-heavy social campaign. Same tool, same week, opposite conclusions.
| Column | Where it comes from | Why you want it |
|---|---|---|
| Total sessions | Your analytics | The denominator nobody agrees on |
| Sessions with a company name | The trial dashboard | The raw, generous number |
| Of those: provider or hosting records | Sort the company list by name | The gap between headline and reality |
| Sessions with a company in your target profile | Filter by size, country, industry | The only number that predicts pipeline |
| Recognition by channel | Split the above per source | Tells you where the tool actually earns its keep |
Measurement 3: where does the data live, and who signs the DPA?
Company-level data is not automatically outside the scope of European data protection rules, and any vendor who tells you it obviously is should worry you slightly. IP addresses are involved. That means processing, and processing means paperwork.
Three things to have in hand before the card details: where the data is processed and stored, the current list of subprocessors, and a data processing agreement you can actually download and forward to whoever reviews these things in your company. If any of the three arrives "after we get the contract signed," you've learned something useful about the next two years.
lead.box processes in the EU and the DPA is available before you spend anything. That's not a differentiator we invented — it's the minimum, and you should hold every vendor on your shortlist to it, us included.
Five questions the demo never invites
Keep these on a sticky note during the call.
- What share of the identified records in a typical account resolves to internet providers?
- How do you exclude bots and crawlers, and can I see the excluded traffic?
- What happens to companies you cannot identify — do I see the volume of anonymous traffic, or does it just vanish from the report?
- How long is data retained, and what happens to it when I cancel?
- Is there a minimum contract term?
What "best website visitor identification software" really means
There is no best website visitor identification software in the abstract, and anyone publishing a ranking that ignores your traffic mix is publishing entertainment. Leadfeeder, Leadinfo, Lead Forensics, Dealfront, Albacross, lead.box — same category, different bets on price, region, workflow and how much they'll admit to.
Best, for you, is the tool that produces a defensible number on your own traffic, is honest about the visitors it cannot see, and doesn't punish you for changing your mind. Home office traffic stays anonymous. Mobile connections stay anonymous. Any vendor who promises otherwise is selling you a number, not a signal.
And a quiet last one: judge the tool by what happens on Tuesday morning, not by what happens in the demo. Can someone on your team open it, see three companies worth a call, and get on with their day? That's the whole trick in one sentence.
- Related: [The silent majority — what happens to the visitors who never fill in a form](/blog/the-97-percent-gap-anonymous-website-visitors)
- Related: [Who actually visits my website? A simple explanation](/blog/who-actually-visits-my-website-a-simple-explanation)
Published by
lead.box Team
More articles
See lead.box on your own traffic
Start free — no card, no sales call required. Or book a 20-minute walk-through if you want the guided tour.