How to measure whether AI answer engines cite your brand

A prompt that names your brand, like "is BrandX good for Y", forces the model to echo the brand back. That is not visibility. It is the model repeating the question.

Real AI visibility is whether your brand appears unprompted on category questions: "best tool for Y", "how do I do Y", "what should I use for Y". Derive these from your strongest topic clusters, the questions a buyer would actually type.

For each prompt, record three things: whether the brand was mentioned at all, whether one of your pages was cited as a source, and which other domains were named. The gap between mention rate and citation rate tells you whether the model knows you but does not trust your pages, or does not know you at all.

Do not collapse this into one platform-wide percentage. Report it per prompt and per surface. A single number hides which questions you own and which a competitor owns.

For a new brand, near-zero mention rate is the expected result. The model has not seen the name enough to state facts about it. That is not a failure of measurement, and inflating it helps nobody.

What you can act on immediately is the retrieval side: whether your pages are clear, quotable and well-structured enough to be pulled in as a source when the model does search. The knowledge side comes from consistent presence and clear entity data over months.

For prompts where a competitor is cited and you are not, look at the page being cited. It is usually more direct, better structured, and more explicit in its claims than yours. Rewrite the relevant passage of your page to state the answer plainly near the top.

Semantyra runs your tracked prompts against configured surfaces, records mentions and citations per prompt, and links each result to the specific content and entity work that would move it.