Can you track how many people find your business through ChatGPT, Claude, CoPilot and other AI tools?

Or in other words how visible is your brand to AI?

It is a question we are hearing more and more often from businesses who are beginning to think seriously about AI search:

“Can you tell us how many people found us through ChatGPT?”

It is an entirely reasonable question, especially at a time when AI tools are becoming part of how people research suppliers, compare services and make sense of the options available to them.

The honest answer, however, is that you cannot track it in the same way you might track traffic from Google, LinkedIn or an email campaign.

At least, not yet.

That can feel frustrating, particularly when there are now plenty of tools and platforms talking confidently about AI visibility, AI search tracking and brand performance inside large language models. Some of these tools can be useful, and there is certainly value in understanding how your business is appearing across AI-assisted search environments, but it is important to be clear about the difference between useful visibility insight and true attribution.

At the moment, no tool can show you every prompt that real users are typing into ChatGPT, every answer those users are seeing, or every lead that has come as a direct result of your business being included in an AI-generated response.

The important phrase here is real users.

You can test prompts. You can run simulations. You can look for patterns. You can monitor whether your brand is appearing in certain types of answer. But you cannot yet see the full journey from an individual asking an AI tool a question through to that person becoming a visitor, enquiry or customer.

What AI visibility tools can actually tell you

Most AI visibility tracking tools work by testing sets of prompts that are likely to be relevant to your business, your customers and your competitors.

For example, a professional services firm might want to know whether they appear when someone asks:

• “Who are the best firms for [specific service]?”

• “What should I look for when choosing a [type of provider]?”

• “Which companies help with [specific problem]?”

• “What are the alternatives to [competitor name]?”

• 
  

By running these kinds of questions through ChatGPT, Perplexity, Gemini or other AI search tools, you can start to build a picture of whether your brand is being mentioned, how it is being described, which competitors are appearing alongside you, and which sources seem to be influencing the answers.

This is useful information, particularly when it is repeated over time, because it can help you understand whether your positioning is clear, whether your content is answering the right questions, and whether your credibility is being reinforced in the places AI tools are likely to draw from.

However, this kind of testing should be treated as an indicator rather than a complete reporting dashboard. It gives you a way to understand how AI tools may interpret your brand, but it does not provide the same kind of attribution that many marketers are used to seeing in platforms such as Google Analytics.

In simple terms, it can help answer the question, “Are we visible in the kinds of AI answers our customers might see?”

It cannot reliably answer the question, “How many customers came to us because ChatGPT mentioned us?”

Why AI search attribution is still limited

One of the reasons this area is so difficult to measure is that AI search does not behave in exactly the same way as traditional search.

With Google, you are often dealing with a search results page, a visible link, a click and then a visit to your website. Even then, attribution is not perfect, but there is at least a more familiar trail of data to follow.

With AI tools, the experience can be much less linear. A person might ask a broad question, refine it several times, compare options, ask for recommendations, read a summary, visit a website later, search for the brand separately, or mention the company to a colleague before making an enquiry days or weeks afterwards.

That journey is much harder to capture.

There may be some AI-related signals in platforms such as Google Search Console, especially where more conversational or question-led queries start appearing in your data. Regex can help you identify longer, more natural-language searches that may suggest a shift in how people are researching. However, this does not neatly confirm that a search came from Google’s AI Mode, ChatGPT or any other AI tool.

So while there are signals worth paying attention to, we should be careful not to overstate what the data can prove.

Does AI visibility still matter?

Yes, it does.

Even though the measurement is imperfect, visibility in AI search still matters because your customers may already be using AI tools to explore their options, understand your market and decide which businesses are worth considering.

If your organisation is not appearing in those conversations, or if it is being described in a way that feels vague, outdated or inaccurate, that is something worth knowing.

The encouraging part is that improving your chances of being understood and cited by AI tools does not require you to abandon everything you already know about good marketing, SEO and brand building. In fact, many of the fundamentals remain reassuringly familiar.

AI tools need clear, credible and well-structured information to work with. They need to understand what your business does, who you help, what you are known for, and why you are a trustworthy source or recommendation.

That means the same elements that already mattered for SEO and brand reputation are still important:

• Clear positioning on your own website

• Useful pages that answer genuine customer questions

• A logical and well-structured site

• Strong case studies, testimonials and proof points

• Mentions in credible third-party places

• Reviews, interviews, bylines, directories and trade press coverage

In other words, you need to make it easy for both people and machines to understand your business.

What businesses should focus on now

Rather than getting too distracted by the promise of perfect AI attribution, the more useful question for most businesses is:

Are we giving AI tools enough clear, credible and useful information to understand and recommend us?

That starts with your own website. If your services are difficult to understand, if your pages are thin or generic, or if your content does not reflect the questions your customers are actually asking, AI tools may struggle to place you confidently in relevant answers.

It also extends beyond your website. AI systems do not only look at what you say about yourself. They may also draw on what others say about you, which is why reviews, case studies, directories, media mentions, partnerships and industry commentary can all play a role in strengthening your overall visibility.

This does not mean chasing every new AI search tactic or rewriting your entire website around ChatGPT. It means doing the strategic work that helps your brand become clearer, more useful and more credible across the web.

That work was already important.

AI has simply made it harder to ignore.

The tools are new, but the fundamentals are not

AI visibility tools can be helpful. They can show you how your business appears in simulated searches, where competitors are being mentioned, what sources are being referenced, and whether your brand is being described in a way that feels accurate.

Used well, they can give you insight.

But they should not be treated as a perfect measurement system, because there is not yet a Google Analytics equivalent for AI search.

For now, the best approach is to combine sensible testing with strong marketing fundamentals. Track what you can, look for patterns, pay attention to the questions your customers are likely to ask, and make sure your website and wider online presence give clear answers.

Whether someone finds you through Google, ChatGPT, Perplexity, a trade publication or a recommendation from a real person, the underlying question is still the same:

Do they understand what you do, who you help and why they should trust you?

If the answer is yes, you are in a much stronger position to be found, understood and recommended — by people, by search engines and by AI tools.

Need help? Get in touch

Who are The Polka? Find out here