Glossary
AI search glossary
Plain-English definitions of the terms behind AI discoverability, measurement and search, and why they matter to ecommerce and retail teams.
22 terms
Core concepts
- AI discoverability
Core concepts
The extent to which a brand, product or service can be found, understood, cited or recommended when people ask relevant questions through AI assistants and generative search.
Why it matters: Customers increasingly use AI systems while researching, comparing and choosing brands.
- AI search optimisation
Core concepts
An umbrella term for improving how a brand and its information are discovered, understood, cited and represented across AI assistants and generative search experiences.
Why it matters: It brings together work across content, technical accessibility, source authority, AEO, GEO and traditional SEO.
- Answer engine
Core concepts
A search or AI system that responds to a question with a direct answer, rather than only presenting a list of links.
Why it matters: Customers can now receive recommendations and comparisons without visiting a traditional search results page.
- Answer engine optimisationAbbreviation: AEO
Core concepts
Improving the likelihood that useful, accurate brand information is selected, summarised or cited by systems that answer questions directly.
Why it matters: AEO helps a brand become part of the answer when customers ask commercially relevant questions.
- Generative engine optimisationAbbreviation: GEO
Core concepts
Improving how a brand and its content are understood, represented and cited within responses generated by AI systems.
Why it matters: GEO helps ensure that AI-generated answers contain accurate, useful and distinctive information about the brand. In practice, GEO overlaps strongly with AEO.
- Google AI Overviews
Core concepts
AI-generated summaries shown within some Google search results, often accompanied by links to supporting sources.
Why it matters: They can affect which information customers see first and which websites receive attention or visits from search.
Measurement
- AI referral traffic
Measurement
Website visits generated when someone follows a link from an AI assistant or AI-generated search result.
Why it matters: It connects visibility in AI answers with measurable visits to the website. Not every AI-referred visit can be identified, so the tracking method should always be clear.
- AI visibility
Measurement
How often a brand appears across a defined set of relevant questions and AI-generated answers.
Why it matters: It helps teams understand whether their brand is present during the customer research and comparison process.
- AI visibility score
Measurement
The percentage of sampled AI answers in which a brand appears, based on a defined set of questions, systems and measurement rules.
Why it matters: It creates a repeatable way to track visibility over time. The score is only meaningful when its methodology is included.
- Brand mention
Measurement
An appearance of a brand’s name in an AI-generated answer. A mention does not necessarily include a link to the brand’s website.
Why it matters: Mentions indicate awareness and visibility, but should be measured separately from citations and referral traffic.
- Citation
Measurement
A link or source reference included in an AI-generated answer to show where supporting information came from. A brand can be mentioned without its own website being cited.
Why it matters: Citations reveal which sources are influencing AI answers and whether those answers direct people towards the brand’s website or another source.
- SentimentAlso called: AI brand sentiment
Measurement
The positive, neutral, mixed or negative tone in which an AI system describes a brand within a sampled answer.
Why it matters: Sentiment helps teams identify favourable positioning, inaccurate descriptions or language that could affect how customers perceive the brand. It reflects the sampled answers, not overall customer satisfaction.
AI answers and evidence
- Branded and unbranded questionsAlso called: Branded and unbranded prompts
AI answers and evidence
A branded question names the company or product being investigated. An unbranded question describes a need, problem or category without naming the brand.
Why it matters: Unbranded questions show whether a brand is discovered independently, while branded questions help assess how accurately the brand is understood and described.
- Grounding
AI answers and evidence
Connecting an AI-generated answer to specific source material or retrieved information, rather than relying only on the model’s existing internal knowledge.
Why it matters: Grounded answers are more likely to be current, traceable and supported by evidence.
- Retrieval-augmented generationAbbreviation: RAG
AI answers and evidence
A method in which an AI system retrieves relevant information from external sources before generating its answer.
Why it matters: It allows answers to use current or specialised evidence that may not be contained in the model’s original training data.
Search and content
- Search engine optimisationAbbreviation: SEO
Search and content
Improving a website’s technical accessibility, relevance, content and authority so traditional search engines can crawl, understand and rank its pages.
Why it matters: SEO remains essential for website discovery and overlaps with AEO and GEO, but ranking in traditional search does not automatically guarantee visibility in AI-generated answers.
- SeedingAlso called: Content seeding
Search and content
In Lindle, seeding means a neutral follow-up to a retained AI answer, using sourced facts about the business and asking whether it is relevant to the buyer’s question.
Why it matters: The original answer and assisted response are kept separate. A follow-up can reveal factual gaps and useful content questions; it does not prove model retraining or guarantee later independent recommendations. Publishing accurate website content is a related but separate activity.
Technical and web
- AI crawlerAlso called: AI bot
Technical and web
An automated bot that visits and reads website content for purposes such as AI search retrieval, indexing or model development. Different crawlers have different names, purposes and controls.
Why it matters: Understanding which crawlers can access a website helps teams manage how their public information is discovered and used.
- llms.txt
Technical and web
A proposed website file that gives AI systems a concise guide to useful public content and important pages on a domain.
Why it matters: It can make key content easier for compatible AI systems to locate, but it is not universally adopted and does not guarantee inclusion or ranking.
- robots.txt
Technical and web
A standard text file at the root of a website that communicates which parts of the site different automated crawlers are permitted to access.
Why it matters: It helps manage crawling, but it is not a security control and does not make private content inaccessible.
See where your brand appears
Understand how customers discover and compare your brand across AI assistants, search and other decision environments.