Glossary
AI search, in plain English.
Definitions for the terms that matter in AI search: AEO, GEO, LLMO, answer engines, AI Overviews, citations, entities, llms.txt and more. Written to be understood on first read, without the acronym soup most vendors hide behind.
- Answer Engine
- Any system that returns a synthesised answer instead of a list of links. ChatGPT, Claude, Perplexity, Gemini and Google AI Overviews are all answer engines. The defining trait is that the user gets a conclusion, not a set of options to evaluate.
- Answer Engine Optimization (AEO)
- The practice of making your business the one an AI answer engine names when a customer runs a search — for example “who is the best plumber in Austin, TX” — and gets one or two recommendations instead of a page of links. Because the customer clicks straight through from the answer, organic and paid listings see far less of that traffic. AEO is done by adding the elements answer engines rely on — an llms.txt file, structured schema, answer pages and FAQ pages — each customized around what AI is actually saying about your company right now. If you are not the answer, you never get the chance to compete.
- Generative Engine Optimization (GEO)
- The practice of building the authoritative online presence a Large Language Model requires before it will recommend your business inside a direct conversation — when a customer asks ChatGPT, Gemini or Grok to help them do, explain, create or recommend something. It is the production work that changes what generative engines say about you: publishing expert-cited, answer-first content on your own domain, seeding the community and forum sources models draw on, and maintaining freshness so the citation persists. GEO is built on top of properly formatted, customized AEO files.
- LLMO (Large Language Model Optimization)
- An umbrella term used interchangeably with AEO and GEO for optimising how large language models represent and recommend a brand. Different vendors use different labels for substantially overlapping work.
- Findability Score
- MyFast.ai’s 0–100 metric summarising how often AI assistants name your business across 100+ real buyer prompts in your category. It is built from five equally weighted parts — Foundation, Brand Recognition, Authority, Content Coverage and Ranking Share — and provides the baseline that makes visibility work measurable rather than anecdotal.
- AI answer engine
- The AI layer inside a web search that returns a synthesised recommendation above or instead of the organic results. It typically names one or two businesses, and the customer clicks directly through to that business, bypassing the traditional scroll through web results.
- Generative engine
- A Large Language Model a customer converses with directly — ChatGPT, Gemini, Grok, Claude or Copilot — rather than searching. Generative engines can give extremely detailed recommendations about a business, but only if they perceive it as having an authoritative online presence.
- Developer bundle
- The set of ready-to-install files MyFast.ai generates from your own AEO results — a custom llms.txt, structured schema, answer pages and FAQ pages — which you or your web developer add to your website host. Once installed, MyFast.ai tracks the changes so the gains are measurable.
- Swipe file
- A shortened version of a published article, written for posting to Reddit and related forums. It carries the same evidence off your own domain, into the community sources generative engines weigh when judging authority.
- AI Overviews
- Google's generated summary that appears above traditional search results. Being cited in an AI Overview places your business in front of the user before they reach any organic listing.
- Citation
- A reference an answer engine attributes its statement to. Citations are the currency of AI visibility: being cited puts your brand in the answer itself, not merely in a source the user might click.
- Zero-click search
- A search where the user gets their answer without visiting any website. As answer engines improve, zero-click share rises, which shifts the value of search marketing from traffic toward being named in the answer.
- Entity
- A distinct thing a knowledge graph can identify — a company, product, person or place. AI systems reason about entities rather than keyword strings, so consistent naming, descriptions and identifiers across the web materially affect whether a model can resolve your brand.
- Entity consistency
- Describing your brand identically — same name, same description, same contact details, same category — across your site, directories, review platforms and structured data, so models resolve every mention to one entity instead of several ambiguous ones.
- Structured data / Schema.org
- Machine-readable markup that states explicitly what a page is about. Well-formed schema removes guesswork for both crawlers and language models, and is the cheapest reliable way to make key facts like services and service area unambiguous.
- JSON-LD @graph
- A single structured-data document containing multiple linked nodes with cross-referenced @id values. It lets a crawler or model traverse an entire entity model in one pass instead of stitching together disconnected fragments.
- llms.txt
- A plain-text file at the root of a site that gives AI agents a curated map of the most important pages with short descriptions. It is a routing convention rather than a ranking factor, and its value depends entirely on being kept current.
- llms-full.txt
- A companion to llms.txt containing the full context an agent might need — complete features, positioning and FAQ content — in one plain-text document optimised for ingestion rather than browsing.
- Retrieval Augmented Generation (RAG)
- An architecture where a model retrieves source documents at query time and generates its answer from them. It is why content that is well-chunked and self-contained gets cited: retrieval pulls fragments, not whole pages.
- Semantic chunking
- Structuring content so each section stands alone as a complete answer to one question. A well-chunked section survives being extracted out of context, which is precisely what retrieval systems do to it.
- BLUF (Bottom Line Up Front)
- A writing pattern where the direct answer appears in the first sentence and supporting detail follows. It suits answer engines because they lift the opening statement, and it suits humans because it respects their time.
- Prompt set
- The fixed collection of buyer questions used to measure AI visibility. Holding the prompt set constant across runs is what makes month-over-month comparison meaningful rather than noise.
- AI crawler
- A bot operated by an AI company to fetch web content, either for live retrieval during a user conversation or for model training. Retrieval crawlers such as OAI-SearchBot and PerplexityBot can drive real referral traffic.
- AI receptionist
- A voice agent that answers inbound business calls, converses naturally with the caller, answers business-specific questions and books appointments — replacing both voicemail and, in most cases, a traditional answering service.
- Missed call cost
- The revenue lost to calls that go unanswered, calculated as missed calls multiplied by average job value and close rate. It is usually the largest unmeasured leak in a service business.
- Speed to lead
- How fast a business responds to an inbound enquiry. It is one of the strongest predictors of whether the lead converts, because the first business to respond typically wins the customer.
Know the terms. Now check your score.
An AEO report tells you how often AI assistants actually name your business across 100+ real buyer searches, run through all seven engines, with every answer shown word for word.