Resources  /  Glossary

AEO Glossary

The vocabulary of Answer Engine Optimization. Every term here is one you will run into the first week you take AI visibility seriously.

Answer Engine Optimization (AEO)
The practice of structuring your website, content, and authority signals so that AI answer engines like ChatGPT, Claude, Perplexity, and Gemini cite or recommend you when users ask questions in your domain. AEO is to AI search what SEO was to traditional search.
AI Citation
When a generative AI assistant attributes a piece of information to a specific source in its response — typically with a link. Citations are stronger than mentions because they pass attention back to the source and influence which sources future models train on.
AI Mention
When a generative AI assistant refers to a brand or source by name in a response without linking to it. Mentions build awareness but do not pass referral traffic the way citations do. Tracking the citation-to-mention ratio is a leading indicator of AEO health.
Answer Engine
A search interface that returns a synthesized answer instead of a list of blue links. ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews are all answer engines. Each builds its answer by retrieving and condensing source documents at query time.
Retrieval-Augmented Generation (RAG)
The architecture most answer engines use: a retrieval step pulls a small set of relevant documents from the open web (or a private index), then a language model generates the response grounded in those documents. AEO targets the retrieval step — if you are not retrieved, you cannot be cited.
llms.txt
A proposed convention (analogous to robots.txt) where a site publishes a plain-text manifest of its most important content at /llms.txt. Designed to give language models a curated entry point to a site rather than crawling the full HTML. Adoption is still early but growing.
Schema.org Structured Data
A vocabulary for marking up content with machine-readable type information — Article, Product, FAQPage, HowTo, Review, Organization, etc. Encoded as JSON-LD in a page’s head. Improves the odds a retrieval system parses your content correctly and surfaces it in rich results.
JSON-LD
JavaScript Object Notation for Linked Data. The recommended format for embedding schema.org structured data on a web page. Goes inside a `<script type="application/ld+json">` block in the document head.
Citation Velocity
How quickly your domain is accumulating new AI citations week over week. A health metric specific to AEO — flat or declining velocity signals you are being outpaced by competitors even if total citation count is rising.
Authority Surface
The set of third-party places where your brand or experts publish — Reddit threads, Quora answers, podcast transcripts, conference talks. Language models train on these surfaces, so presence there compounds your visibility independent of your own domain.
Crawler Allowlist
The subset of user-agent strings explicitly permitted in your robots.txt — typically GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and similar. If an AI crawler is not on the list, it may treat your site as opted-out of training and retrieval.
Answer-First Formatting
A content pattern where the direct answer to a likely question appears in the first sentence of a section, followed by supporting context. Optimizes for retrieval systems that extract the lead sentence as the response candidate.
Semantic HTML
HTML that uses meaning-bearing elements — `<article>`, `<section>`, `<header>`, `<nav>`, `<main>`, `<h1>` through `<h6>` — instead of generic `<div>` soup. Retrieval systems use semantic structure to identify the canonical heading, main content, and supporting sections of a page.
AI Visibility Score
A composite metric measuring how often your brand is cited or mentioned across major AI assistants for a defined set of prompts. RankRush reports it per assistant (ChatGPT, Claude, Perplexity, Gemini) and as a weighted aggregate so you can see where you are strong and where you are absent.
Prompt Universe
The defined set of prompts a brand wants to be cited for — typically the buying-intent questions, comparison queries, and category-defining questions in its market. Selecting the right prompt universe is the foundation of any AEO program; tracking against a generic set produces meaningless scores.

Next step

Now that you have the vocabulary, run the AI-visibility checklist against your own site to find the gaps worth fixing first.