Every measurement product has a boundary, and most of them keep it in a footnote. Ours is worth reading before you trust a single number on the dashboard, so here it is in full.
An AI answer is not a document sitting on a server that we fetch. It is generated, once, for whoever asked. To measure it at all you have to ask the question yourself — and every decision about how you ask changes what you can honestly claim afterwards.
We sample the APIs, not the app
Answers come from the models' own APIs through a single gateway. That is not the same thing a person sees in a chat app: no account history, no memory of previous conversations, no personalisation, no shopping cards, none of the product-specific furniture a model renders around the text.
What it buys is a controlled measurement. The same prompt, the same market, the same model, asked on a schedule — so when a number moves you can rule out the interface having quietly changed underneath you. That trade is the right one for tracking, and it is still a trade. A screenshot of the consumer app will sometimes disagree with us, and when it does, the app is not wrong.
Citations are parsed out of the answer text
Links are extracted from the response deterministically and reduced to a canonical form first, because one answer will happily cite the same page as both https://x.com and https://www.x.com/, and counting that as two citations of two sources is wrong twice over.
Running a model over every answer to find its links would cost more than the answer did, and would introduce a second thing that can hallucinate into a pipeline built to catch the first. The cost of the deterministic route is that sources a model shows in its own UI panel, rather than writing into the text, are not represented. If a number here looks low against what you see in the app, that gap is usually why.
Named and cited are two different lists
An answer that links your domain without ever writing your name has cited you; it has not mentioned you. We keep those apart, down to the detail of ignoring brand names that appear only inside a URL — the characters around "acme" in https://www.acme.com/ are dots, so a naive match reads the domain as prose and credits you with a mention nobody made.
A brand the answer only linked to is on the citation list. A brand the answer recommended by name is on the other one. The difference is the whole point of the measurement.
The models we sample, and the ones we don't
ChatGPT, Gemini, AI Overviews and Google AI Mode are sampled on every pass, and the product records which model produced each answer it stores. Perplexity is not covered — the gateway does not carry it. Search surfaces like AI Overviews have no chat-completion form at all, so no gateway will ever serve them, and we would rather say that than list them as "coming soon" forever.
Nothing before the day you start
There is no archive of what ChatGPT said about your category last March. Nobody has one — not us, not the tools that imply otherwise. Tracking begins when a prompt is added, which is the least convenient honest answer and the only one available.
Find out what AI says about your brand
Add a domain and the first pass runs while you read the next one.