AEO & SEO
glossary.

The vocabulary of AI search, in plain English. 25 terms, defined the way we would explain them to a client.

Answer Engine Optimisation (AEO)
The practice of structuring and writing content so AI answer engines such as ChatGPT, Google AI Overviews, Perplexity and Gemini can extract it, trust it and cite it as a source. Where SEO earns a ranked link, AEO earns a mention inside the answer itself.
Search Engine Optimisation (SEO)
The practice of earning visibility in the classic list of search results, through relevant content, a fast and crawlable site, and authority built over time. It is the foundation that AEO is built on, not a separate discipline.
AI Overview
The AI-generated summary Google places above the classic blue links for many searches, with links to the sources it drew from. It is assembled from pages already in Google’s index, so normal indexing is the entry ticket.
Answer engine
A system that responds to a question with a synthesised answer rather than a list of links. ChatGPT, Perplexity, Google AI Overviews, Gemini and Copilot are answer engines, most of which cite the sources they used.
Large language model (LLM)
An AI model trained on large amounts of text to understand and generate language. Answer engines use LLMs to read retrieved sources and write the answer a user sees, which is why clear, well-structured source pages matter.
Citation
In AI search, a reference an answer engine gives to a source it used, usually your brand name and a link to your page. Being cited is the AEO equivalent of ranking: the answer is built partly from your content.
Share of voice
How often your brand is cited across answer engines for the questions that matter to you, measured against competitors. It is the main way to track AEO progress over time, rather than a single ranking position.
E-E-A-T
Experience, expertise, authoritativeness and trustworthiness: the qualities Google uses to judge content. Google says trust is the most important, and rewards clear authorship, demonstrated expertise and transparent sourcing.
Structured data (schema markup)
Machine-readable code that describes a page’s content to search and answer engines. It does not guarantee a rich result, but it helps engines understand what a page is, who wrote it, and how it relates to other things.
Keyword cannibalisation
When two or more of your own pages compete for the same search term, so neither ranks as well as a single strong page would. Google cannot pick a clear winner and the intent gets split across the pages.
Crawler (bot)
An automated program that fetches web pages so an engine can index them or answer with them. Each answer engine runs its own, such as GPTBot, OAI-SearchBot, PerplexityBot and Google-Extended, and you control access through robots.txt.
robots.txt
A file at the root of a site that tells crawlers which parts they may access. Blocking an answer engine’s crawler in robots.txt prevents that engine from using your content as a source.
Zero-click search
A search that ends without the user clicking through to a website, because the answer appears directly in the results. AI Overviews and answer engines increase zero-click behaviour, which is why being cited matters more than the click alone.
Entity
A distinct thing an engine recognises, such as a brand, person or product, and understands in relation to others. Being described consistently across the web helps engines recognise your entity and trust it.
Retrieval-augmented generation (RAG)
A technique where an AI model retrieves relevant documents and uses them to write its answer, rather than relying on memory alone. Most answer engines retrieve, which is why crawlable, well-structured pages can be cited.
Grounding
Tying an AI model’s answer to real sources it can cite, so the response reflects retrieved content rather than the model’s own recall. Grounded answers are the ones that carry citations back to websites.
llms.txt
A proposed file for telling AI models about a site’s content. The major answer engines do not use it to choose citations, and Google says you do not need AI text files or special markup, so it is not a shortcut.
Anchor text
The visible, clickable words of a link. It gives search engines a hint about the linked page’s topic. Natural, varied anchor text is healthy; stuffing exact-match keywords into anchors across many links looks manipulative and can backfire.
Nofollow
A link attribute, rel="nofollow", that tells search engines not to pass ranking signals through the link. Google also offers rel="sponsored" for paid links and rel="ugc" for user-generated content. Nofollow links can still bring traffic and brand visibility.
Search intent
The goal behind a search: what the person actually wants. The main types are informational (to learn), commercial (to compare), transactional (to buy) and navigational (to reach a specific site). Matching your content to the intent is what makes it rank.
Long-tail keyword
A longer, more specific search phrase, usually lower in volume but clearer in intent and less competitive. "Running shoes" is a head term; "best trail running shoes for wide feet" is long-tail, and far easier to win and to convert.
Keyword mapping
Assigning one primary target keyword or intent to each page, written down, so every page has a clear job. It is the simplest way to prevent keyword cannibalisation, where two of your pages compete for the same term.
SERP
The search engine results page: the page Google returns for a query. It holds the classic blue links plus features like AI Overviews, featured snippets, the local pack, images and "people also ask". Reading the SERP reveals a term’s intent and competition.

New to this? Start with what Answer Engine Optimisation is, or browse all guides.

Ready to ship more
without hiring more?

Start the 7-day trial: the full Professional plan, no credit card. Add a client and put the whole workspace to work, from AEO content and AI visibility tracking to Google Ads, site audits and reporting.