AI Search & GEO Glossary

    This glossary defines the terms used in Generative Engine Optimization (GEO) and AI search monitoring — from AI visibility and citations to query fan-out and llms.txt. Each definition links to a guide that covers the topic in depth.

    AI crawler

    An AI crawler is a bot that fetches web pages for an AI company, either to collect training data or to answer a user's question in real time; examples include GPTBot, OAI-SearchBot, PerplexityBot and ClaudeBot. Most AI crawlers do not execute JavaScript, so content that only appears after client-side rendering can be invisible to them.

    AI hallucination

    An AI hallucination is a confident statement in an AI answer that is false or unsupported, such as a wrong price, a discontinued product or an invented feature. Monitoring what AI engines say about a brand is the only way to catch hallucinations before customers act on them.

    Learn more: AI Search Analytics

    AI positioning map

    An AI positioning map plots a brand and its competitors by how visible they are in AI answers and how positively they are described. It shows at a glance who AI engines treat as the category leader and who is visible but poorly perceived.

    Learn more: GEO Competitive Analysis

    AI sentiment

    AI sentiment is the tone in which an AI answer describes a brand, from very negative to very positive. Aspect-based sentiment goes further and scores individual topics, such as price, quality or support, so a team can see which perception to fix.

    Learn more: AI Search Analytics

    AI share of voice

    AI share of voice is a brand's share of all brand mentions in AI answers for a defined set of prompts, compared with its competitors. It shows whether a brand is gaining or losing ground in AI search even when its absolute mention count stays the same.

    Learn more: GEO Competitive Analysis

    AI visibility

    AI visibility is how often, and how prominently, a brand appears in AI-generated answers to the questions its customers ask. It is usually measured by running a fixed set of prompts across AI engines and recording whether the brand is mentioned, in what position and with what sentiment.

    Learn more: How to Measure GEO Success

    Answer Engine Optimization (AEO)

    Answer Engine Optimization is an older, closely related term for structuring content so that search features and assistants can lift a direct answer from it, for example featured snippets and voice answers. In practice AEO and GEO overlap heavily; GEO is the term more often used for generative AI answers.

    Learn more: SEO vs. GEO: Complete Comparison

    Answer-first writing

    Also known as: BLUF, inverted pyramid

    Answer-first writing puts the direct answer to a page's main question in its first two or three sentences, before any background. AI engines extract these opening passages more often than conclusions buried at the end of an article.

    Learn more: Optimizing Existing Content

    Brand mention (in AI answers)

    A brand mention is any occurrence of a brand's name in an AI-generated answer, whether or not the answer links to the brand's website. Mentions shape what users remember and shortlist, so they are tracked separately from citations.

    Learn more: Why Your Brand Disappears from AI Answers

    Citable content

    Citable content is content written so that an AI engine can lift a passage and quote it accurately: a direct answer near the top, short self-contained paragraphs, explicit subjects instead of pronouns, and specific facts with their sources.

    Learn more: How to Write Citable Content

    Citation (AI citation)

    Also known as: source link

    An AI citation is a link to a source web page that an AI engine shows alongside its answer to indicate where information came from. Being cited sends referral traffic and signals that the engine treats the page as trustworthy.

    Learn more: How AI Search Engines Select Sources

    Content gap

    A content gap is a question that AI engines answer by citing competitors or third parties because the brand has no page that answers it well. Closing content gaps is usually the fastest way to gain new AI citations.

    Learn more: GEO Competitive Analysis

    E-E-A-T

    E-E-A-T stands for Experience, Expertise, Authoritativeness and Trustworthiness, the quality signals described in Google's search quality rater guidelines. Named authors with real credentials, original data and citations from reputable sites strengthen the same trust signals that AI engines rely on when choosing sources.

    Learn more: 7 Pillars of GEO Optimization

    Entity

    An entity is a uniquely identifiable thing, such as a company, product, person or place, that search engines and AI models track as a single concept regardless of how it is named. Consistent names, descriptions and profiles across the web help models attach the right facts to a brand's entity.

    Learn more: Link Building for GEO

    FAQPage schema

    FAQPage schema is structured data that marks up a list of questions with their answers. Question-and-answer pairs map directly onto how people prompt AI assistants, which makes FAQ sections one of the most reusable formats for AI answers.

    Learn more: FAQ and Q&A Content

    Generative Engine Optimization (GEO)

    Also known as: AI search optimization

    Generative Engine Optimization is the practice of making content more likely to be retrieved, cited and recommended in answers generated by AI systems such as ChatGPT, Perplexity, Gemini and Google AI Overviews. Where SEO competes for a ranked position in a list of links, GEO competes for inclusion in a single synthesized answer.

    Learn more: What is GEO Optimization

    Google AI Overviews

    Google AI Overviews are AI-generated summaries shown at the top of some Google search results, with links to the pages they draw on. They answer the query on the results page itself, which changes how much traffic the linked pages receive.

    Learn more: Zero-Click Future

    Grounding

    Grounding is the process of tying an AI answer to specific source documents so that its claims can be checked. Grounded answers are the ones that show citations, and they are where GEO work has the most direct effect.

    Learn more: How AI Search Engines Select Sources

    LLM Optimization (LLMO)

    LLM Optimization is another name for GEO that emphasises large language models rather than search engines. It covers both how a brand is represented in a model's training data and how it is found by models that search the web at answer time.

    Learn more: What is GEO Optimization

    llms.txt

    llms.txt is a proposed convention for a markdown file at a website's root that gives language models a concise, curated map of the site's most important pages. It complements, rather than replaces, robots.txt and sitemap.xml.

    Query fan-out

    Query fan-out is the way an AI search engine splits one user question into several related sub-queries, searches for each, and combines the results into one answer. A page can therefore be cited for a question it never mentions verbatim, as long as it answers one of the sub-queries well.

    Learn more: How AI Search Engines Select Sources

    Retrieval-augmented generation (RAG)

    Retrieval-augmented generation is a technique in which an AI model first retrieves relevant documents, for example through a web search, and then writes its answer based on them. It is why fresh, well-structured and easily retrievable pages can be cited even if they were published after the model was trained.

    Learn more: How AI Search Engines Select Sources

    Structured data (schema markup)

    Structured data is machine-readable markup, usually JSON-LD using the schema.org vocabulary, that states explicitly what a page contains: an article, its author and dates, a product and its price, or a list of questions and answers. It helps search and AI systems interpret a page without guessing.

    Learn more: Schema Markup for GEO

    Tracking prompt

    A tracking prompt is a question sent to AI engines on a schedule to measure AI visibility, written the way a real customer would ask it. A good prompt set mixes category, comparison, problem and brand questions so the results reflect real buying journeys.

    Learn more: Mastering AI Search Prompts: How to Track What Matters

    Training data vs. live retrieval

    Training data is the fixed body of text a language model learned from, while live retrieval is the web search it may run when answering. A brand's representation in training data changes only when a new model is released; its presence in live retrieval can change as soon as its content changes.

    Learn more: Why Your Brand Disappears from AI Answers