ggeoaeo
REFERENCE

The GEO & AEO glossary

The vocabulary of this space is half genuinely useful, half invented to make dashboards look clever. Here's the plain version of both — including which numbers to take with a pinch of salt.

Marcus TaylorBy Marcus TaylorUPDATED JUN 2026
Answer engine
Any tool that responds to a question with a direct answer rather than a list of links — ChatGPT, Perplexity, Google's AI Overviews, Gemini, Copilot, Claude.
GEO (Generative Engine Optimization)
Getting your brand named in the answers generative models write. The newest flavour of the discipline.
AEO (Answer Engine Optimization)
The broader, older idea of being the answer wherever answers get served. Used more or less interchangeably with GEO.
Citation
When an engine names or links a source in its answer. The thing you're ultimately trying to win — being cited is the AI-era equivalent of ranking.
Brand mention
Any reference to your brand in an answer, cited or not. A model can recommend you without linking you, which still counts.
Prompt tracking
Monitoring a fixed set of questions over time to see whether — and how — you appear. The core monitoring job every tool does.
Share of voice
Roughly, how often you appear in answers for your tracked prompts versus competitors. Useful as a trend; treat the absolute number with caution.
Sentiment
Whether a model talks about you positively, neutrally or negatively. Worth watching, easy to over-read.
Engine coverage
How many of the major answer engines a tool genuinely monitors. Coverage is fragmented — very few tools watch them all.
Prompt Query Volume (PQV)
A vendor estimate of how often a given prompt gets asked. Sounds precise, isn't — useful for prioritising, not for forecasting. A good example of a metric to hold loosely.
Visibility / interaction score
A composite number a platform invents to summarise your AI presence. Every tool has one, none agree, and they're not comparable across platforms. Track your own over time; don't compare it to anyone else's.
Hallucination
When a model states something untrue — including about your brand. Part of why monitoring matters: engines occasionally get your facts wrong.
Structured data
Machine-readable markup (schema.org) that helps engines understand a page. Mild but real help for being parsed cleanly into an answer.
Re-test cadence
How often a tool re-checks your prompts. Daily beats monthly — a tool that only refreshes once a month makes you wait a month to see if a change worked.