Reference

GEO Glossary

A reference glossary of the terms that define Generative Engine Optimization (GEO). Each entry is a short, citable definition designed to be quoted by AI models and human readers alike.

What is the GEOMOND glossary and why it exists

The GEO glossary is the GEOMOND reference list of the terms that define Generative Engine Optimization (GEO) — the discipline that makes a brand visible inside the answers written by ChatGPT, Perplexity, Gemini, Claude and Google AI Overviews. Each of the 27 entries below is a short, self-contained definition designed to be quoted verbatim by an AI engine or a human reader, and is exposed as a schema.org/DefinedTerm inside a single DefinedTermSet.

Why a citable glossary matters: Gartner predicts that traditional search engine volume will drop 25% by 2026 as users move queries to AI chatbots. OpenAI reports ChatGPT reached 800 million weekly active users in 2025, and Google rolled out AI Overviews to U.S. Search in May 2024 and to more than 100 countries thereafter. In that landscape, definitions become the unit of citation: an LLM that needs to explain "GEO", "Presence Index" or "AI Overviews" will quote the page that gives it the cleanest, most structured definition.

The glossary is bilingual (this English version mirrors the Spanish glossary) and is linked from every comparison and case study, so the underlying terminology stays consistent across the GEOMOND site. Cross-reference: see also the public definition of Generative Engine Optimization on Wikipedia.

GEO (Generative Engine Optimization): Digital marketing discipline that optimises a brand's presence so it consistently appears in the answers of generative engines like ChatGPT, Perplexity, Gemini or Claude. Unlike SEO, the goal is not to rank a URL in a list of results, but to get the model to mention the company directly when a user asks about its sector.
AEO (Answer Engine Optimization): Optimisation for direct answer engines (voice assistants, snippets, AI Overviews). It is a subset of GEO focused on structuring content as short question-and-answer pairs so it can be cited as a single response.
AIO (AI Overviews): Generative answer block that Google shows at the top of its search results, powered by its Gemini model. Appearing inside an AI Overview multiplies brand visibility and reduces clicks to the website.
LLM (Large Language Model): Large-scale language model trained on billions of web pages, books and documents. ChatGPT, Gemini, Claude and Llama are LLMs. GEO works to make these models associate your brand with the relevant questions in your sector.
Generative model: Artificial intelligence system capable of producing new text, images or code from learned patterns. In GEO the term refers to the chatbots and assistants that generate natural-language answers recommending companies and services.
Presence Index: GEOMOND's proprietary metric, on a 0–10 scale, measuring how often and how relevantly the main LLMs (ChatGPT, Gemini, Perplexity, Claude) mention a company when users ask about its sector. It is the only AI visibility metric in the Spanish-speaking market.
Entity: A unique, identifiable concept (a company, person or place) that AI models recognise as a node in their knowledge graph. Working on the entity means consolidating name, description, headquarters, founders and services so the AI knows precisely who you are.
Knowledge Graph: Structured database of entities and relationships used by Google and the main LLMs to answer factual questions. Appearing in the Knowledge Graph increases the probability of being cited by generative AI.
Schema.org: Standardised semantic markup vocabulary (JSON-LD, Microdata, RDFa) that helps search engines and LLMs understand a page's content. In GEO we use types like Organization, Service, FAQPage, Article, Person and DefinedTermSet.
JSON-LD: JSON-based format for embedding Schema.org structured data in HTML. It is the format Google recommends and the easiest one for LLM crawlers to parse.
E-E-A-T: Acronym for Experience, Expertise, Authoritativeness and Trustworthiness. Quality framework that Google and LLMs use to assess source reliability. It is one of the factors that most influence an AI's decision to cite your brand.
Citation: Explicit mention of a brand, author or source inside an LLM's generated answer, usually with a source link. Citations are the main currency of GEO: more citations means more visibility.
Sampling: Measurement technique of asking the same question to an LLM N times to capture the variability of its answer. It enables computing the real appearance probability of a brand and building the Presence Index.
Prompt: Instruction or question that the user submits to the AI model. Commercial prompts ("best lawyer in Madrid", "marketing agency in Barcelona") are the battleground of GEO.
RAG (Retrieval-Augmented Generation): Architecture that combines an LLM with a real-time search engine. Perplexity and Google AI Overviews are RAG systems: they query the web before generating the answer. That is why fresh, well-structured content is decisive in GEO.
GPTBot, PerplexityBot, ClaudeBot, Google-Extended: Official crawlers from OpenAI, Perplexity, Anthropic and Google that browse the web to feed and refresh their models. Allowing them in robots.txt is a basic requirement for doing GEO.
llms.txt: Emerging convention — analogous to robots.txt — that gives LLMs a curated map of a site's most relevant URLs and content. GEOMOND publishes it by default on every project.
AI Sitemap: Variant of sitemap.xml that prioritises educational content (articles, guides, glossaries, FAQs) over transactional pages, so LLM crawlers reach the most semantically dense information first.
Structured data: Information marked up with a recognisable schema (Schema.org, OpenGraph, microformats) that lets machines understand content unambiguously. It is one of the technical pillars of GEO.
Topical Authority: Cumulative perception by AI models that a brand is the expert reference on a given topic. It is built by deeply covering a cluster of related content.
Brand Mention: Mention of a brand in any source that LLMs crawl (media, forums, podcasts, social networks). Even without a link, it counts as an authority signal for generative models.
Hallucination: Made-up or factually incorrect answer generated by an LLM. A solid GEO strategy reduces the risk that AI invents information about your company by providing clear, consistent sources.
Embedding: Vector representation of a text that LLMs use to measure semantic similarity. The better structured and topically coherent your content is, the closer it sits to the embedding of the relevant queries.
5 GEO Pillars: GEOMOND's proprietary methodology: Entity, Authority, Context, Structure and Amplification. Each pillar groups the actions that move the Presence Index in measurable ways.
Amplification: Coordinated distribution of content and mentions in the sources LLMs read (industry media, Wikipedia, vertical directories, academic repositories, topic forums). It is the fifth pillar of GEO.
Zero-click: A search where the user gets the answer directly from the search engine or chatbot without clicking any result. Zero-click makes the brand mention inside the answer more valuable than the click itself.
AI Share of Voice: Percentage of mentions a brand obtains versus its competitors in LLM answers about a given sector. It is the generative equivalent of advertising market share.

Frequently asked questions about the GEO glossary

What is GEO (Generative Engine Optimization)?

GEO (Generative Engine Optimization) is the discipline of optimising a brand's visibility inside the answers generated by AI engines such as ChatGPT, Perplexity, Gemini, Claude and Google AI Overviews. Where SEO targets the 10 blue links of a search results page, GEO targets the answer the engine writes for the user, measured by whether the brand is named, cited or linked.

How is GEO different from SEO?

SEO optimises pages to rank in the 10 blue links; GEO optimises pages to be cited inside generative answers. SEO measures clicks, rankings and CTR; GEO measures presence, citations and links inside ChatGPT, Perplexity, Gemini and AI Overviews. They are complementary disciplines, not substitutes — most GEOMOND clients run both in parallel.

How do you measure brand presence in ChatGPT, Perplexity and Gemini?

Through a Presence Index that runs the same prompt set on each engine and scores three dimensions: whether the brand appears (mention), whether it is cited as a source (citation) and whether the answer links back to the brand site (link). The result is a 0–10 score per engine and a weighted total, refreshed on a monthly cycle. See the Presence Index entry below for the full definition.

Why does schema.org matter for GEO?

Schema.org structured data (Article, FAQPage, BreadcrumbList, DefinedTerm, HowTo) is one of the strongest signals an AI engine uses to extract a verbatim claim from a page. A page with rich, valid JSON-LD is much more likely to be quoted as a source than the same content without schema, because the engine can locate the answer without parsing prose.