Stay found
AI citation tracking
SEO tells you if Google can read you. This tells you the newer thing: when a customer asks an AI assistant, does the answer actually name you?
The problem
Your customers increasingly ask an assistant instead of searching — and you have no idea whether it mentions you or a competitor. You can't improve what you can't see.
How it works
Track the real questions your customers ask — 'best accessibility ramp company near me' — with the brand, domain, and any aliases that count as a mention. Run a check and each question is put to a model; the answer is scanned for your name, and the result is recorded: cited, or not, on which term.
Detection is deliberately plain — a case-folded match on your brand, domain, or alias — because 'did it say our name' is a yes/no a business can act on, not a similarity score it has to trust. Every check is stored, so you watch the trend as you improve your content, your llms.txt, and your structured data.
It's the companion to the SEO/AEO audit: that one checks whether the crawlers can get in; this one checks whether the answers that result actually cite you.
- Track the questions that matter, with brand / domain / alias matching
- Each check asks a model and records cited-or-not, with the answer snippet as evidence
- History per question, so you can see the trend as you fix your AEO
- Metered and injected: point it at your own model client; no key means a clean skip
npm install @adminigloo/aeoBook a callAll features