Definitions
AEO glossary
Sixteen terms from AI search visibility, defined precisely — and each one paired with the term it is easily confused with, because that confusion is where the reporting errors come from. A mention and a citation happen at two different steps, so a brand can be named in an answer built entirely from somebody else’s pages.
Updated 14 October 2026 · 8 minute read
How the terms fit together
Each measurement belongs to a different step of an answer, which is why they move separately. A mention and a citation happen at two different steps, so a brand can be named in an answer built entirely from somebody else’s pages.
The same steps, in words
- Someone asks an answer engine a question. If you track it, it is part of your prompt set.
- The engine fetches live pages and writes from them: grounding.
- It credits a handful of sources. Each is a citation; your slice is your citation share.
- It names a few brands. Whether you are among them is your mention rate; where is your position; your slice of all mentions is your share of voice.
- Those are weighted into one visibility score, re-run on a schedule as prompt tracking.
- The reader may need nothing more: a zero-click answer, which is why none of this shows in your analytics.
The engines, and the work
What answers the question, and what it means to be found in it.
Answer engine
A system that responds to a question with a written answer rather than a list of links. ChatGPT, Perplexity, Claude, Google AI Overviews and Microsoft Copilot are all answer engines.
Why it matters: an answer names a short list of brands, so visibility is closer to binary than ranking is. How they choose sources.
Answer engine optimization (AEO)
Making a brand easy for an answer engine to find, trust and quote: your own pages, how they are structured, and what the rest of the web says about you. What is AEO.
Local trap: in the Philippines, AEO usually means Authorized Economic Operator, a Bureau of Customs accreditation. AI search visibility in the Philippines.
Generative engine optimization (GEO)
A near-synonym for AEO, introduced in a 2023 research paper of the same name. GEO stresses being cited by generative models; AEO also covers featured snippets and voice results.
Confused with: AEO. In practice the work is identical. GEO vs AEO.
AI Overview
Google’s generated summary above conventional search results, previously branded Search Generative Experience.
Why it matters: it has no public consumer API, and Search Console does not break its impressions out. A vendor reporting AI Overviews numbers is checking by hand or scraping — ask which. How to choose an AEO vendor.
Grounding
Retrieving live documents at answer time and writing the answer from them, rather than from the model’s training data alone. Also discussed as retrieval, or RAG.
Why it matters: publishing a page now can affect an answer within weeks, and crawlability matters exactly as much as it does in search. AI crawlers and robots.txt.
What gets counted
Six measures, each paired with the one it is easily confused with.
Citation
A source an engine credits beneath or within its answer.
Confused with: a mention, and a click. An answer can cite your page without naming you, or name you while citing a competitor’s page about you. Neither is a visit. What Perplexity actually cites.
Citation share
The percentage of cited sources, across a fixed question set, that point at your own domain. Formula and worked example.
Our own figure: 0.0% — in September 2026, none of the 90 sources cited across our eight questions pointed at bungad.com. Published at our score.
Mention rate
The share of analysed answers in which your brand is named at all. Computed over answers, where citation share is computed over citations.
Confused with: citation share. A brand discussed widely on review sites can be named often and cited never.
Share of voice
Your proportion of all brand mentions across the question set, counting competitors alongside you.
The trap: it rises when a competitor falls. A share-of-voice chart going up while your mention rate is flat is not an improvement.
Position
Where your brand falls in the order of brands named in an answer.
The subtlety: it is only defined when you were named. Our own audit reports no average position, because we were never named. In our formula first place counts fully and fifth counts for nothing: the curve.
Visibility score
A single 0–100 figure combining the measures above. Ours weights mention rate at 50%, citation share at 30% and position at 20%: the arithmetic.
Do not compare across vendors. Weightings and question sets differ. A score is only meaningful against itself, on a fixed prompt set, over time.
Why the score tops out below 100
Because citation share carries three tenths of it, a brand named first in every answer with no citation pointing at its own site tops out at 70 rather than 100. What to make of a number once you have one: what is a good AI visibility score.
The questions you track
The fixed set every number is counted against, and the habit of asking it again.
Prompt set
The fixed list of questions a measurement runs against, also called tracked questions. It is the denominator of every metric: change it and the series resets.
A report that does not print its prompt set cannot be verified. Ours is stored verbatim inside every run record.
Prompt tracking
Asking a prompt set to one or more engines on a schedule and recording who is named and what is cited.
What it is not: a census. It samples what an engine says when asked your questions. Who tracks AI citations · How often to check.
A tracked question containing your own brand name guarantees a mention, so it can never be lost and cannot measure anything. We ask such questions, keep the answers and exclude them from every figure. Choosing the questions to track.
What your analytics cannot see
Three terms about what happens off your site, and what an engine reads before it answers.
Zero-click answer
An answer that satisfies the question without anyone visiting a website.
Why it is invisible: analytics cannot show a visit that never happened. The only instrument that detects it is asking the engines directly. AEO vs SEO.
llms.txt
A proposed plain-text file at a site’s root summarising it for language models. A community proposal, not a standard. What is llms.txt.
Status: we know of no major engine that has publicly confirmed it reads one. We publish ours and have no evidence it has changed an answer.
Entity consistency
Describing your business identically wherever it appears: the same legal name, address, category, founding date and service description on your own site, your listings and your profiles.
Why it is worth a dull afternoon: faced with three different descriptions of you and one consistent description of a competitor, an engine names the competitor. Why your brand is not in AI answers.
Terms we deliberately avoid
Three phrases in this category assert something nobody can currently support. If a vendor uses them, ask what they mean.
| Phrase | Why we avoid it | What we say instead |
|---|---|---|
| AI ranking | There is no ranked list to be placed in. An answer names a few brands or none, and the next run may name different ones. | Mention rate, and position among the brands named |
| All AI platforms | It hides the difference between an engine queried through an API and one a person checked by hand. Google AI Overviews and Copilot have no public consumer API. | The engines named individually, and how each is queried |
| AI traffic | It implies visits. The characteristic outcome here is a zero-click answer, where the reader is satisfied and nobody is visited. | Citations, mentions, and the referral visits your own logs actually show |
More guides
Read next
Put the vocabulary to work
Enter your domain. We ask Perplexity three of your buyers’ questions and show you whether you were named and which domains the engine cited.
3 questions on Perplexity, free. A domain checked in the last week is served from that stored run rather than asked again, and the daily free allowance resets at midnight UTC.
