Glossary
A list of core AI search terms and definitions.
1. AI Citations
AI citations are references included by generative engines (e.g., ChatGPT, Perplexity, Gemini) to show where an answer’s facts or phrasing originated.
Earning citations increases brand visibility and trust because users can verify claims and click through to the source. Brands improve citation likelihood by publishing clear, well-structured content with evidence, bylines, dates, and consistent terminology that matches user intent.
Technical hygiene—schema markup, canonical URLs, fast performance—also helps engines resolve entities correctly.
Related Definitions: GEO, Authority Signals, Structured Data
Related Pages: Signals of Trust; The AI Search Era
2. Generative Engine Optimisation (GEO)
GEO is the practice of making your brand’s content and data understandable, cite-worthy and reusable by generative engines. It blends strategy (positioning, proof), content design (clear information architecture), and technical enablement (schema, formats, APIs).
Unlike traditional SEO focused on ranking pages, GEO focuses on earning trustworthy answers and citations.
Related Definitions: AI Citations, Structured Data, Prompt Engineering
Related Page: GEO: An Executive Guide
3. Authority Signals
Authority signals are cues that help humans and AI determine whether to trust your brand: clear authorship, credentials, verifiable sources, consistent branding, case studies with outcomes, and third-party references.
Technically, these become machine-readable via schema (e.g., Organization, Person, Product, Review) and well-named assets.
Related Definitions: E-E-A-T, Brand Identity as Data
Related Page: Signals of Trust
4. E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness)
E-E-A-T is a framework used by search platforms to evaluate content quality. For brands, it means demonstrating real-world experience, subject expertise, recognised authority, and trustworthy practices. Publish bylines, methods, sources, dates, and update logs; align claims with evidence and outcomes.
Related Definitions: Authority Signals, Editorial Policy
Related Page: The AI Search Era
5. Structured Data (Schema Markup)
Structured data is machine-readable markup (e.g., JSON-LD) that annotates your pages with meaning—people, organisations, articles, products, events. It helps engines disambiguate entities, connect facts, and generate richer results and citations. Prioritise Organization, Website, BreadcrumbList, and Article/TechArticle, plus DefinedTerm for glossary entries.
Related Definitions: Knowledge Graph, Entity
Related Page: Positioning to Prompts; 90-Day Roadmap
6. Brand Identity as Data
Treat brand elements—name, tagline, mission, tone, colour, logo, product names—as data. Store them consistently, name files clearly, and reflect them in schema and design tokens.
This reduces ambiguity, improves entity resolution, and keeps your brand consistent across citations and generative answers.
Related Definitions: Authority Signals, Design Tokens
Related Page: Signals of Trust
7. Entity (in Search)
An entity is a distinct “thing” (person, organisation, product, concept) that engines recognise and connect across sources. Optimising for entities means being unambiguous: consistent naming, schema, sameAs links, and corroborating references.
Related Definitions: Knowledge Graph, Structured Data
Related Page: The AI Search Era
8. Knowledge Graph
A knowledge graph is a network of entities and relationships used by search and AI systems to understand the world. Your goal is to be a well-defined node: consistent identifiers, rich attributes, and trusted external references (company registers, profiles, press, standards bodies).
Related Definitions: Entity, AI Citations
Related Page: GEO: An Executive Guide
9. Prompt Engineering (for Brand Ops)
Prompt engineering is the craft of structuring inputs and reference materials so AI systems produce accurate, brand-aligned outputs. In brand ops, it means turning positioning and proof points into reusable prompt blocks and retrieval snippets, with guardrails for tone, claims and compliance.
Related Definitions: RAG (Retrieval-Augmented Generation), Design Tokens
Related Page: Positioning to Prompts
10. 90-Day Roadmap (AI Search)
A focused implementation plan to modernise your brand for AI Search: (1) Audit content, entities, schema, and citations; (2) Prioritise critical fixes and high-leverage pages; (3) Ship structured data, glossary terms, cornerstone updates, and measurement. Finish with governance and an update cadence.
Related Definitions: Structured Data, Authority Signals, GEO
Related Page: 90-Day Brand Modernisation
11. RAG (Retrieval-Augmented Generation)
Retrieval-Augmented Generation (RAG) is a method where an AI model fetches relevant, vetted source content at query time and uses it to generate a more accurate, grounded answer. Instead of relying only on what the model “knows,” RAG pulls from your approved knowledge base—playbooks, case studies, product docs—then cites or reflects that information in its output.
For brands, RAG reduces hallucinations, improves compliance, and keeps messaging consistent. To make RAG work well, curate a clean corpus (single source of truth), use consistent naming and metadata, add schema for entities, and maintain versioning and access controls. Pair with prompt patterns that specify tone, boundaries, and citation rules, and track performance with feedback loops and update cadences.
Related Definitions: Prompt Engineering, Knowledge Graph, Structured Data
Related Page: From Positioning to Prompts: Translating Strategy into Machine-Readable Assets
12. Design Tokens
Design tokens are the smallest reusable pieces of your brand’s visual system—stored as named variables (e.g., color.primary, font.size.body, radius.lg). They translate brand identity into machine-readable data that can be consumed by design tools, code, and AI systems.
Tokens ensure consistency across channels, speed up changes (update once, propagate everywhere), and make your brand easier for generative engines to recognise and reproduce. Maintain tokens in a central registry, use clear, stable names, and map them to components and schema (e.g., including references in your style guide and Organization markup).
For AI workflows, provide token sets alongside content and prompts so outputs adhere to brand colour, type, spacing, and tone rules.
Related Definitions: Brand Identity as Data, Authority Signals, Prompt Engineering
Related Page: Signals of Trust: Brand Identity as Data for AI
13. SERP
Definition: The Search Engine Results Page—the list (and now AI summaries) a search engine returns for a query.
Why it matters for GEO: Competition is shifting from ranking in the list to being cited inside AI-generated answers at the top of the SERP.
How to use it: Design pages to be answerable (clear definitions, steps, references) and machine-readable (schema, entities).
Related: AI Overview, Answer Engine, E-E-A-T, FAQPage.
14. knowsAbout (Property)
Definition: A schema.org property for Person (and some other types) indicating subjects the entity is knowledgeable about.
Why it matters for GEO: Strengthens author expertise signals and helps disambiguate topics for answer engines.
How to use it: Add concise topic phrases or entity IDs (where possible) to author pages.
JSON Snippet:
{
“@type”: “Person”,
“name”: “Alex Morgan”,
“knowsAbout”: [
“Generative Engine Optimisation”,
“schema.org structured data”,
“B2B content strategy”
]
}
Related: about, mentions, E-E-A-T, Author Page.
15. sameAs (Property)
Definition: A schema.org property linking an entity to its official profiles/IDs on other sites.
Why it matters for GEO: Creates a canonical identity trail across the web, boosting confidence and reducing ambiguity.
How to use it: Link to authoritative profiles only (e.g., Companies House, LinkedIn, Wikidata, official docs).
JSON Snippet:
{
“@type”: “Organization”,
“name”: “Brand Strategy AI”,
“url”: “https://brandstrategyai.co.uk/”,
“sameAs”: [
“https://www.linkedin.com/company/brandstrategyai”,
“https://find-and-update.company-information.service.gov.uk/company/00000000”,
“https://www.wikidata.org/entity/Q000000”
]
}
Related: identifier, url, Organization schema, Entity Disambiguation.
16. Service schema (Service Type)
Definition: The schema.org type that describes a professional service offering (what it is, who provides it, scope, audience, area served).
Why it matters for GEO: Makes offerings legible to machines, enabling clearer inclusion in AI answers comparing providers.
How to use it: Create a node per service with name, description, provider, audience, areaServed, and (optionally) pricing/offer details.
JSON Snippet:
{
“@type”: “Service”,
“name”: “GEO Readiness Audit”,
“description”: “An executive audit of entities, schema, and content to improve inclusion in AI answers.”,
“provider”: {
“@type”: “Organization”,
“name”: “Brand Strategy AI”,
“url”: “https://brandstrategyai.co.uk/”
},
“serviceType”: “SEO & GEO Strategy”,
“areaServed”: [“GB”, “IE”, “EU”],
“audience”: { “@type”: “BusinessAudience”, “audienceType”: “B2B” }
}
Related: Product, Offer, Organization, areaServed, audience.
17. areaServed (Property)
Definition: A schema.org property indicating the geographic area where a service, organisation, or offer is available.
Why it matters for GEO: Helps answer engines align you with location-qualified queries (e.g., “in the UK” / “Europe”).
How to use it: Use ISO country codes, regions, or place names. Add to Service, Organization, and Offer.
JSON Snippet:
{
“@type”: “Organization”,
“name”: “Brand Strategy AI”,
“areaServed”: [
{ “@type”: “Country”, “name”: “United Kingdom” },
{ “@type”: “Country”, “name”: “Ireland” },
“European Union”
]
}
Related: serviceArea, availableAtOrFrom, Local/Regional SEO, NAP Consistency.
18. YYML
Regulated sectors classified as YMYL (Your Money or Your Life) are those where the content can significantly impact a user’s health, financial stability, safety, or overall well-being, leading to high stakes if the information is inaccurate or misleading.
This category includes industries such as healthcare, finance, legal services, tobacco, alcoholic beverages, and beauty, where content must meet stringent standards for accuracy, trustworthiness, and compliance with both industry regulations and platform policies.
Google applies stricter quality evaluations to YMYL content, emphasizing the need for high levels of Experience, Expertise, Authoritativeness, and Trustworthiness (EEAT) to ensure users receive reliable information.
19. Governance Frameworks
A governance framework is a structured set of rules, processes, and standards that ensure brand information, content, and data remain consistent, accurate, and compliant over time.
In the context of AI Search and digital strategy, governance frameworks define how decisions are made, who maintains data, and what criteria determine quality and approval. They create accountability and prevent “data drift” — ensuring that every page, schema, and citation reflects the same verified source of truth.
A governance framework may include naming conventions, schema templates, approval workflows, review policies, and change control procedures — all documented to keep brand signals aligned as teams, tools, and platforms evolve.