Precise words for precise decisions.
Each term is defined once and used the same way across the site, in reports and in contracts. Indicators point to the formula published in the methodology; former names point to the current term.
What is GEO?
Generative Engine Optimization makes a brand’s content easier for AI answer engines such as ChatGPT, Gemini, Perplexity or Claude to understand, verify and reuse. It relies on useful, sourced, structured and accessible content. Aive approaches it with video as the source. GEO guarantees neither ranking nor citation: it is measured on observed answers.
- GEO
- Generative Engine Optimization: optimising content for the way generative AI selects, rephrases and cites its sources.
- AEO
- Answer Engine Optimization: optimising for the direct answers of answer engines.
- Video GEO
- The Aive GEO module that starts from video as the source: answer-first pages, FAQs, chapters and a Content Graph, keeping the moment → statement → source → article → publication → observed answer chain intact.
- Answer-first
- An editorial format: question as a heading, a self-contained answer of 40 to 80 words, definitions, scope, proof, steps and FAQ.
How does Aive read a video?
MGT, Aive’s Multimodal Generative Technology, combines 25 models to read a video’s image, sound and text and turn it into structured data. The Playbook sets the rules before any generation, MGT structures the master, then Aive Studio and Aive GEO help teams analyse, adapt, optimise and activate versions while they keep the final decision. Outputs depend on the analyses activated.
- MGT
- Multimodal Generative Technology: Aive’s technology, combining 25 models to read a video’s image, sound and text and turn it into structured data for scoring, versioning and distribution.
- Master
- The source video the brand holds the rights to, from which versions are produced.
- Workspace
- An isolated working space for a customer or brand: corpus, roles, rules, integrations and analyses, depending on plan and configuration.
What do the Aive Studio modules cover?
Aive Studio industrialises video from the masters you already own. The Playbook encodes your brand rules; Storyboard lays out every scene; Intelligent Versioning turns a master into every format, duration, language and market variant; Creative Score assesses each version before publishing; Intelligent Distribution publishes, measures and improves the next versions. Every output stays reviewable and subject to your approval.
- Playbook
- Your brand rules encoded once (guidelines, markets, formats, rights, roles and approvers), applied before any generation, on every video.
- Storyboard
- Reading a video scene by scene: scenes, speakers and moments laid out, to prepare, select and brief without scrubbing through the whole timeline.
- Intelligent Versioning
- The module that turns a master into every format, duration, language and market variant, keeping the subject in frame.
- Creative Score
- A brand, creative quality and platform rules score, computed before publishing, with the fixes to apply.
- Intelligent Distribution
- A publish, measure, improve loop: versions go live on your channels, performance comes back, the next ones get better.
- Voice-over
- An audio track in another language, prepared for review before release.
What do the Aive GEO modules cover?
Aive GEO starts from video to make a brand readable by AI answer engines. GEO Measurement shows how the brand appears in observed answers; Video GEO turns videos into pages, FAQs, chapters and a Content Graph; Media Factory publishes owned media on your domain; EDGE deploys page fixes; Automation runs these operations with the Aive GEO agent.
- GEO Measurement
- The module that brings together the free check, GEO Audit, AI share of voice, cited sources and brand perception: a brand’s visibility in AI answers, measured and dated.
- Media Factory
- The module that turns an authorised catalogue into owned media on your domain: articles, clips and pages generated from your videos, with the Player and Aive Expert.
- EDGE
- The module that deploys page fixes (schema, structure, GEO optimisations): every change is previewed, rolled out in stages and reversible.
- Automation
- The module where the Aive GEO agent prepares and executes GEO operations, with reusable workflows, SSO and API for enterprise teams.
- 360 product page
- A unified view of a product (attributes, content, videos, data completeness and visibility signals), generated from your product videos for AI shopping.
- GEA
- Generative Engine Advertising: Aive’s advertising in the AI answers of its media (Aive Expert, answer pages), always labelled as sponsored, approved and measured.
What do the Content Graph and the evidence chain connect?
The Content Graph organises what a video or corpus contains: timestamped moments, speakers, entities, chapters, sourced statements and derived content. The evidence chain links every published statement to its proof: source moment, document, version, approver and confidence level. Together, they show where a published piece of information comes from and trace it back to the exact moment in the video.
- Content Graph
- A graph built from what a video says and shows: moments, speakers, entities, chapters, statements and links to derived content. As a module name, it now sits within Video GEO.
- Evidence chain
- The thread linking every published statement to its source, version, approver and confidence level.
- Sourced statement
- A published statement linked to a verifiable source (video moment, document or page), with status, scope, owner and verification date.
- Timestamped moment
- The exact point in a video, to the second, tied to its speaker and its master; a statement, chapter or answer points to it.
- Evidence Graph
- Former name for the evidence chain: see Evidence chain.
- Claim, sourced claim
- Former name for the sourced statement: see Sourced statement.
What are semantic internal linking, the Player and Aive Expert?
Semantic internal linking connects your pages by what they prove and by what your readers look for: statements, entities, intents, articles and video moments, with links approved by your team. Within Media Factory, the Player streams your videos on your authorised domains and Aive Expert answers from your catalogue: the user asks a question, gets a sourced answer and returns to the relevant video moment.
- Semantic internal linking
- Links between content organised by proof and intent: suggested links, orphan pages detected, depth measured, every link approved by a person.
- Player
- The Media Factory video player, embedded on your authorised domains, with chapters, transcript and events.
- Aive Expert
- The Media Factory on-site assistant, which answers from your authorised corpus and points back to the source video moment.
- VideoObject
- A Schema.org type describing a video: title, description, thumbnail, duration, upload date and real playback URL.
- Semantic content network
- Former name for semantic internal linking: see Semantic internal linking.
How do you read AI visibility indicators?
Every Aive indicator is a ratio between a number of answers and an explicit denominator, calculated per engine and per period. The mention rate counts answers that name the brand, the citation rate those that cite its source, and AI share of voice compares the brand with its competitive set. None of these rates is universal: each holds for the panel observed.
- Mention rate
- Answers mentioning the brand ÷ observed answers.
- Citation rate
- Answers citing a brand URL or source ÷ observed answers.
- AI share of voice
- Brand mentions ÷ mentions across the competitive set in the observed sample.
- Prompt coverage
- Active prompts with a valid observation ÷ defined prompts.
Why query an engine more than once?
A generative AI does not always give the same answer to the same question. Aive therefore asks each question several times, at a set frequency, and measures how much the results vary. That variance always accompanies the score: it shows whether a gap between two periods exceeds the normal noise of the answers. Referred sessions are measured separately.
- Repeated sampling
- Several samples of the same question on the same engine within the same period, to account for answer variability.
- Variance
- The spread of an indicator across samples and days, always displayed with the score.
- Observed answer
- An answer actually collected, dated and archived with its question, engine and access method.
- AI referral
- A session from an identifiable AI referrer, kept separate from crawling and citations; a referral does not establish causality.
- Multi-sampling
- Former term for repeated sampling: see Repeated sampling.
Free check, GEO Audit, demo or Studio pilot: which to choose?
The free check estimates in moments how readable a domain is for answer engines, with no account. The GEO Audit is a scoped engagement on agreed questions, engines and markets, with a dated report. The demo shows Aive on your own stakes, live. The Studio pilot, which is paid, evaluates Aive Studio on one use case and is deducted from the annual plan.
- Free check
- An instant AI visibility estimate from a domain, with no account and no real query to the engines; distinct from the GEO Audit.
- GEO Audit
- A scoped engagement on agreed questions, engines and markets: dated, archived raw answers, indicators with formula and variance, action plan.
- Baseline
- A dated measurement taken before any action, within the same scope as the following measurements.
- Studio pilot
- A paid Aive Studio pilot, scoped on one use case and deducted from the annual plan.
What is the difference between GEO and AEO?
AEO targets engines’ direct answers; GEO more broadly covers how generative AI selects, rephrases and cites its sources. Both rely on clear, sourced and accessible content. Aive covers both, starting from video.
Are a mention and a citation the same thing?
No. A mention names the brand in the answer. A citation points to a brand URL or source. Both are measured separately, each with its own rate and denominator.
Why do some former names still appear here?
Claim, multi-sampling, semantic content network and Evidence Graph have been replaced by sourced statement, repeated sampling, semantic internal linking and evidence chain. They stay in this glossary so you can find the current term; the rest of the site uses the new names.
Do these definitions apply to GEO Audit reports?
Yes. Aive reports use the same terms and formulas as this glossary and the methodology. Any variant specific to a GEO Audit is described in the report itself.