AI shopping and AI visibility glossary
Plain-English definitions of the terms CPG and ecommerce teams run into when AI assistants and shopping agents start recommending products: from AI Visibility Score and AEO to query fan-out, agentic commerce and llms.txt.
19 terms · Updated · By Akhil Suresh Nair
- Agentic commerce
- AI citation
- AI crawler
- AI product visibility
- AI share of voice
- AI shopping agent
- AI Visibility Score
- Answer engine optimization
- Buyer-intent prompt
- Content gap
- Fix Queue
- Generative engine optimization
- Google AI Overviews and AI Mode
- llms.txt
- Model Context Protocol
- Query fan-out
- Structured data
- Unbranded prompt
- Visibility Score signals
Agentic commerce #
Agentic commerce is shopping in which an AI agent searches, compares and sometimes buys or reorders products for a person, inside a retailer’s app or across the web.
Why it matters for CPG brands: When an agent fills the cart, it may choose the brand for generic list items, so product data has to be clear enough for software to pick it.
AI citation #
Also called: Cited source
An AI citation is a link or source an AI answer shows as the basis for what it says. Citations reveal which domains and pages an engine relies on for a topic.
Why it matters for CPG brands: If your product pages are never cited, the engine is describing your products from someone else’s pages, such as retailers, reviewers or competitors.
AI crawler #
Also called: AI bot, AI user agent
An AI crawler is an automated client that fetches web pages for an AI company, to train models, to build a search index or to answer a user’s request live. Examples include OpenAI’s GPTBot and OAI-SearchBot, PerplexityBot and Anthropic’s ClaudeBot; Google-Extended is a robots.txt token that controls use of content for Google’s Gemini models.
Why it matters for CPG brands: If robots.txt or a firewall blocks these clients, the engines cannot read your current product pages.
AI product visibility #
AI product visibility is how often, and how prominently, AI assistants such as ChatGPT, Google Gemini, Perplexity, Microsoft Copilot and Google’s AI answers name a specific product when shoppers ask for recommendations. It is measured per product (SKU), not per brand.
Why it matters for CPG brands: A brand can look healthy overall while many of its products never appear in AI answers. Product-level measurement shows which SKUs are missing.
AI shopping agent #
Also called: AI shopping assistant
An AI shopping agent is an assistant that acts on a shopper’s behalf: it answers product questions, builds carts and can complete or manage orders. Examples covered on this blog include Walmart’s Sparky and Amazon’s Alexa for Shopping.
Why it matters for CPG brands: Retailer agents often recommend from their own catalog and bundles, so visibility work includes retailer listings, not just your own site.
AI Visibility Score #
Also called: Visibility Score
An AI Visibility Score is a 0–100 measure of how likely AI engines are to name and recommend a specific product when a shopper asks. Searchpanel calculates one for every SKU, refreshes it daily and splits it by engine and market.
Why it matters for CPG brands: One number per SKU makes it possible to rank products, spot drops quickly and see which engine to work on first.
Answer engine optimization (AEO) #
Also called: AEO
Answer engine optimization (AEO) is the practice of making your content easy for AI assistants and answer features to find, understand, quote and cite, so your product or brand is named in the answer itself rather than in a list of links.
Why it matters for CPG brands: AI answers name only a few products. Clear, specific, well-structured pages are more likely to be the ones an engine relies on.
Buyer-intent prompt #
Also called: Prompt tracking
A buyer-intent prompt is a question a shopper might ask an AI assistant while deciding what to buy, such as “best gluten-free crackers for kids.” Prompt tracking runs a fixed set of these prompts against each engine on a schedule and records which products are named.
Why it matters for CPG brands: AI answers change from week to week, so a single check is not a measurement. Repeating the same prompts shows the trend.
Content gap #
A content gap is information a shopper asks about, such as a spec, ingredient, size, use case or comparison, that is missing or unclear on your product pages, so AI answers draw on other sources or skip your product.
Why it matters for CPG brands: Filling content gaps is usually the most direct fix for a low mention rate.
Fix Queue #
The Fix Queue is Searchpanel’s ranked list of changes for a product, covering content gaps and technical and schema issues, so teams work on the fixes most likely to improve AI visibility first.
Why it matters for CPG brands: It turns a visibility report into a work list, then the score is re-checked after changes are published.
Generative engine optimization (GEO) #
Also called: GEO
Generative engine optimization (GEO) is the practice of improving how often and how prominently content appears in answers written by generative AI search engines. The term was introduced in the 2023 research paper “GEO: Generative Engine Optimization” (Aggarwal et al.).
Why it matters for CPG brands: GEO and AEO are often used interchangeably. Both measure success by whether you are named or cited in the answer, not by a page ranking.
Google AI Overviews and AI Mode #
Also called: AI Overviews, AI Mode
AI Overviews are AI-generated summaries that Google shows at the top of some search results, with links to the sources used. AI Mode is Google Search’s conversational mode, where people ask follow-up questions and get AI-written answers.
Why it matters for CPG brands: Both can recommend specific products, so a product can be visible in classic results and still be missing from Google’s AI answers.
llms.txt #
llms.txt is a proposed plain-text Markdown file at the root of a website (/llms.txt) that gives large language models a short summary of the site and links to its most useful pages. It was proposed by Jeremy Howard in September 2024.
Why it matters for CPG brands: It is a low-effort way to hand AI tools an accurate, current summary of who you are and what you sell. Searchpanel publishes one, plus a full-text version.
Model Context Protocol (MCP) #
Also called: MCP
The Model Context Protocol (MCP) is an open standard for connecting AI applications to outside tools and data. Anthropic introduced it in November 2024, and retailers now use it to let outside AI agents work with their stores; Walmart described its Agent Gateway as MCP-based in October 2026.
Why it matters for CPG brands: MCP-style connections let agents read product data and take actions directly, without scraping web pages.
Query fan-out #
Also called: Fan-out queries, Background searches
Query fan-out is when an AI assistant turns one user prompt into several background web searches, then writes its answer from the results. The searches it chooses, including any brand names in them, shape which products can appear.
Why it matters for CPG brands: If an assistant already has brands in mind when it searches, being mentioned on cited pages may not be enough to get named.
Structured data (schema.org) #
Also called: Schema markup, JSON-LD
Structured data is machine-readable markup, usually JSON-LD using the schema.org vocabulary, that labels facts on a page: a Product’s name, brand, price, availability, reviews and FAQs.
Why it matters for CPG brands: It removes guesswork for crawlers and AI tools reading your product pages, and it is one of the technical checks behind a product’s visibility.
Unbranded prompt #
Also called: Non-branded query
An unbranded prompt is a shopping question that does not name a brand, for example “what’s a good electrolyte drink?” It shows which brands an AI assistant recommends on its own.
Why it matters for CPG brands: Unbranded prompts are where shoppers discover new products, so they are where smaller brands have the most to gain or lose.
Visibility Score signals #
The Searchpanel Visibility Score is built from six signals: mention rate, average position, citations, consistency, freshness and location visibility.
Why it matters for CPG brands: Each signal points to a different fix. A low mention rate is a content problem; weak citations usually mean your own pages are not among the sources an engine reads.
See these terms on your own products
Searchpanel scores how ChatGPT, Gemini, Perplexity, Copilot and Google’s AI answers recommend each SKU. The free SKU audit needs no account and no credit card.