AI visibility glossary
Plain definitions of the terms used when measuring how AI assistants recommend brands.
- AI visibility
- How often, and how prominently, AI assistants name a brand when people ask them questions. It is measured across many questions and several engines, because a single answer varies with wording, engine and date.
- Buyer question
- A question a prospective customer asks an AI assistant while choosing a product, such as “best CRM for small agencies”. Buyer questions are unbranded: they describe a need and do not name your company.
- Unbranded query
- A query that does not contain a brand name. Unbranded queries show whether an engine recommends you unprompted, which is what happens in a real buying conversation.
- Branded query
- A query that names the brand, such as “what is Acme CRM?”. An engine will nearly always describe the brand in reply, so branded queries say little about whether it recommends you.
- Mention rate
- The share of buyer questions for which an engine's answer names your brand. If an engine names you in 9 of 24 answers, your mention rate on that engine is 37.5%.
- Named instead
- The competitors that appear in answers to your buyer questions, counted by how often each is named. It shows who takes the recommendation when you do not.
- Cited source
- A web page an AI engine links to or draws on when it answers. Cited sources show where an engine's view of a category comes from.
- Answer check
- One buyer question asked to one engine. Ten questions across three engines use 30 answer checks. Mentionary plans are sized in answer checks.
- Answer engine
- A system that replies to a question with a written answer rather than a list of links. ChatGPT, Gemini and Claude are answer engines, and Google AI Overviews is an answer engine inside search results.
- Answer engine optimisation (AEO)
- The practice of improving how often and how accurately answer engines mention and cite a brand. It starts with measuring current answers to buyer questions.
- Generative engine optimisation (GEO)
- Another name for work aimed at visibility in generative AI answers. In practice AEO and GEO describe the same goal.
- Model Context Protocol (MCP)
- An open protocol that lets AI assistants use external tools and data. An assistant that supports MCP can call a server's tools during a conversation.
- MCP connector
- A remote MCP server added to an assistant such as Claude or ChatGPT. The Mentionary connector, at https://platypus.mentionary.ai/mcp, gives an assistant 19 tools for measuring AI visibility.
Your buyers are already asking AI. Find out who it tells them to pick.
Add Mentionary to Claude and ask one question. Your first analysis across ChatGPT, Gemini and Claude is free, inside your assistant or in the web app.
Free plan: 10 buyer questions, 3 engines, 2 competitors. No card.