The GEO & GEA glossary

How to read this glossary

Each entry is a self-contained 1–2 sentence definition designed for both human readers and AI extraction. Where a term has its own dedicated page on this site, the term links to it. Use the alphabetical list below or jump straight to What is GEO? for the pillar overview.

A–Z of GEO and AI search

GEO (Generative Engine Optimization)

The practice of structuring a brand's content and authority so AI systems cite, summarise or recommend it inside generated answers.

GEA (Generative Engine Advertising)

Paid placement of brand content inside generative AI answers - the advertising counterpart to GEO.

GSO (Generative Search Optimization)

Vendor synonym for GEO, emphasising visibility inside generative search experiences such as Google AI Mode.

AEO (Answer Engine Optimization)

Optimisation for concise direct answers in answer engines and AI Overviews; in practice a subset of GEO.

LLMO / AI SEO

Vendor labels used interchangeably with GEO. As of early 2026 there is no consensus academic definition separating them. (Source: Wikipedia, ‘Generative engine optimization’.)

Retrievability

How efficiently an AI can access, understand and reuse a piece of content inside its generated answer. The fourth pillar after crawlability, indexability and ranking.

Share of voice (AI)

The percentage of relevant AI answers in which a given brand is mentioned, measured across a tracked set of prompts.

Citation rate

The percentage of AI answers that explicitly cite a brand or source via a link or named reference.

Fan-out query

When an AI silently expands one user prompt into several sub-queries to gather and synthesise its answer.

Entity

A uniquely identifiable thing - a company, product, person or place - that LLMs reason about as a node, not as a string of characters.

Knowledge graph

A structured network of entities and the relationships between them; used by search engines and increasingly by LLMs to ground answers.

RAG (Retrieval-Augmented Generation)

An LLM architecture that retrieves external documents at query time and uses them to ground its answer, increasing accuracy and enabling citations.

AI Overviews

Google's AI-generated answer block shown above traditional results; cites a small set of sources, 93.67% of which link to a top-10 organic result. (Source: Seer Interactive, 2025.)

AI Mode

Google's full conversational search experience powered by Gemini, in which results are AI-synthesised rather than listed.

E-E-A-T

Experience, Expertise, Authoritativeness, Trustworthiness - Google's quality framework, which also strongly correlates with LLM citation behaviour.

Mention rate

How often a brand is named - with or without a link - across a tracked set of AI answers.

Zero-click search

A search interaction that ends without a click to any website, because the answer is delivered in-place by the engine.

LLM (Large Language Model)

A machine-learning model trained on very large text corpora to predict and generate language; the engine behind ChatGPT, Gemini, Claude, Perplexity and others.

Grounding

The process of anchoring an LLM's answer in verifiable external sources at query time, typically through retrieval.

Hallucination

When an LLM generates plausible but false content - invented facts, figures or sources - usually because grounding is weak.

Schema.org / structured data

A shared vocabulary of types (Organization, Article, FAQPage…) embedded as JSON-LD; LLMs grounded in structured data show up to 300% higher factual accuracy than raw text. (Source: Data World benchmark, via elementera.com.)

Why a glossary matters for GEO

LLMs cite the sources that name and explain entities most clearly. A precise, well-structured glossary is itself a retrievability asset: it gives models compact, copy-ready definitions to lift into answers, and it anchors your brand's vocabulary in the wider topic graph. If your team is standardising terms internally, link this page from your style guide and from every article that uses these terms.