GEO & LLMO glossary
- AEO (Answer Engine Optimization) Answer Engine Optimization: making information easy to find and use as a direct answer in search results, voice assistants and AI search. It overlaps with GEO and LLMO.
- AI crawler A bot that fetches web content for an AI service. Training crawlers, search crawlers and user-initiated fetches serve different purposes and have different access controls.
- AI Overviews AI-generated summaries that Google may show in search results, with links to supporting web pages. They provide another place for people to encounter a brand or its content.
- AI share of voice A brand’s share of measured brand appearances relative to a defined comparison group. It describes competitive presence, whereas visibility measures appearances across all sampled answers.
- Answer capsule A short, self-contained answer placed directly below a heading or question. It gives the essential explanation before the evidence, examples and qualifications.
- Citation A source reference shown with an AI answer, usually linking to a web page. A citation identifies a source; a mention names a brand in the answer text.
- E-E-A-T Experience, Expertise, Authoritativeness and Trustworthiness: aspects Google uses to describe content quality. Trust is central; E-E-A-T is not a single ranking factor.
- GEO (Generative Engine Optimization) Generative Engine Optimization: improving how content and brands appear in answers from generative AI services. It covers being found, cited and described accurately.
- Hyokiyure (表記ゆれ, spelling variants) hyōki-yure Different written forms of the same name, such as キヤノン, キャノン and Canon. Brand measurement needs to recognize valid variants without confusing them with other names.
- LLMO (Large Language Model Optimization) Large Language Model Optimization: improving how a brand and its information are represented in answers from large language models. The term overlaps with GEO and AEO.
- RAG (Retrieval-Augmented Generation) Retrieval-Augmented Generation: generating an answer using information retrieved from an external source, such as a document collection, database or web search.
- Sentiment analysis An assessment of the evaluative tone used about a brand, such as positive, neutral or negative. It describes the wording of an answer, not the feelings of customers.
- Structured data (JSON-LD) Machine-readable information describing a page’s content. Schema.org provides a vocabulary; JSON-LD is one format used to express it. Structured data does not guarantee AI citations.
- Visibility score The percentage of measured AI answers that mention a brand. Supasaito reports average position separately; position does not weight visibility.
- Zero-click search A search that ends without a click to an external website. Direct answers, featured snippets and AI summaries can satisfy a request on the search surface itself.