Citations

Now — everything you've just seen depends on one thing. Are we asking the AI assistants the right questions? Because if we tracked questions nobody actually asks, all those numbers would be meaningless. So this is where the real work happens.

The platform starts by looking at the brand's website to understand what it actually sells and who it sells to. Then it finds the real questions people ask before they buy something like it. Not made-up questions — the ones people are genuinely asking.

And here's a deliberate choice that's worth understanding. Notice these questions don't include the brand's name. That's on purpose. If you ask an AI 'tell me about [brand]' — it'll say something nice. Of course it will. You've learned nothing.

The questions that matter are the ones a customer asks before they've ever heard of you. Things like 'what's the best option for someone in my situation', or 'how do I choose between these'. That's the moment where an AI either says your name — or says someone else's. That's what's being measured here.

And the questions get sorted by where the customer is in their decision. Some people are just learning about the category. Some are comparing options. And some are ready to buy right now — those are the ones sitting closest to your revenue.

There's one more layer here, and it's an important one.

People don't talk to AI assistants the way they talk to a search box. Nobody types three words. They write a whole paragraph — with their budget, their situation, what they've already tried, what they're worried about.

So the platform takes each of those core questions and expands it into the many different ways a real person might actually phrase it — the short version, the long detailed version, the one where they're worried about cost, the one where they're comparing two options.

Because a brand can be recommended in one phrasing and completely absent from another. If you only checked one, you'd get a badly misleading picture.

And you can go all the way down to the individual answer. This is the receipt.

Here's the exact question that was asked. Here's which AI answered it. Here's whether the brand was mentioned, where in the answer, and how it was described. So this isn't a summary or a score any more. This is the actual thing that happened, and you can read it yourself.

And this part is the most useful thing on the screen. When an AI answers a question, it reads specific pages to build that answer. This shows you which pages it read.

That tells you something you can act on immediately: if the AI is building its recommendation out of review sites, comparison articles, and directories — then being mentioned on those is what gets you into the answer. It's not always about your own website at all. Sometimes the most valuable thing you can do is make sure you're properly represented on the pages the AI is already reading.