Field note
How to read your own AI-visibility audit
You can check what AI says about you in an afternoon. The hard part isn't running it — it's knowing what you're looking at.
Checking what AI says about you is easy to start and easy to misread. Anyone can ask an assistant about their company; the skill is in reading the answer. Four questions turn a vague impression into something you can act on.
Are you named — without being prompted? Ask your category's real buyer question ("best tool for X"), not your own name. If you only appear when you spell yourself out, you're not yet recognised as part of the category.
Is what it says true? Read the description as a sceptical buyer would. Outdated pricing, the wrong category, a feature you don't have — confident and wrong is the costly kind, because there's no second link to correct it.
Whose sources is it using? If the assistant shows citations, look at whose they are. For many companies the cited pages aren't their own — they're third parties describing you. That tells you who actually owns your narrative.
Who shows up instead of you? Note the competitors it names and how it frames them. That's the comparison a buyer sees — made for them, without you in the room.
Two cautions. Do it signed out, with memory off — logged in, the assistant flatters you with your own history. And check more than one engine on more than one day: they disagree, and they drift. One answer is an anecdote; the pattern across engines is the finding.
Want the full step-by-step? Audit your AI visibility in an afternoon. Reading it well is the start; changing it is the method.
New here? Start with the overview → · By Denes Csaszar · written in the open