Something changed quietly in how people search. A question that used to produce ten blue links now produces a paragraph. The user reads the paragraph, gets what they needed, and never visits a website. If a brand is named in that paragraph, it wins. If it isn’t, it may as well not exist for that query — regardless of where it ranks in traditional search results.
This is the gap we kept running into with clients. Organic traffic was flat or softening, rankings hadn’t moved much, and nobody could explain the disconnect. The explanation was that a growing share of the demand was being satisfied upstream, inside ChatGPT, Gemini, and AI Overviews, and we had no way to see any of it.
Rank tracking doesn’t measure this
A rank tracker tells you position three for a keyword. It doesn’t tell you whether an AI assistant cites you when someone asks the question that keyword represents. Those are different questions with different answers, and the second one is increasingly the one that decides whether a customer ever hears your client’s name.
There’s also no single ranking to check. Ask the same question twice and you may get two different sets of brands mentioned. The answer depends on the phrasing, the model, and the moment. Spot-checking by hand tells you almost nothing, because a single result is a sample of one.
What AI visibility tracking actually does
Instead of a keyword and a position, you work with a set of prompts — the real questions a customer would ask before buying — and you measure how often a brand appears in the answers to them. Run that on a schedule and you get something you can compare over time: which prompts surface your client, which surface a competitor instead, and whether that’s moving in the right direction.
We use cloro for this. The part that mattered for us wasn’t a feature so much as the shift from anecdote to baseline. Before, the answer to “are we visible in AI search?” was a shrug and a screenshot. Now it’s a number, with the underlying answers there to show a client what the assistant actually said about them.
Why this matters if you’re a brand
Three practical reasons to look at this now:
It’s measurable, which means it’s improvable. You can’t optimise what you’ve never observed. A baseline is the prerequisite for everything else.
Your competitors may already own the answer. Being absent from an AI response isn’t neutral — the assistant still names someone. Often it’s a competitor, and often the client has no idea.
Content strategy depends on it. Once you know which questions you’re missing, you know what to publish. It turns “we should do more content” into a specific list.
Where to start
Pick fifteen to twenty questions a real customer would ask on the way to buying from you — not keywords, actual questions. Check how often your brand appears in the answers, and note who appears when you don’t. That single exercise usually tells a brand more than a month of rank reports.
If you’d rather we ran it for you, get in touch. We do this as part of our reporting for clients, and the first check usually surfaces something nobody expected.
