AI Marketing Consensus Index / Publication standards

Research Definitions & Terminology

AI marketing terminology is changing quickly. Terms such as:

GEO
AEO
AI SEO
AI Search Optimization
LLM Optimization
AI Visibility
Citation Visibility
Recommendation Share

are often used differently by different companies, platforms, agencies, and researchers. Some terms overlap. Some describe different things. Others are used primarily as marketing labels rather than standardized technical definitions. AI Marketing Consensus Index therefore maintains this glossary to explain how we use important terms within our own research.

These definitions are intended to make our methodology easier to understand and our datasets easier to interpret. They should not be read as claims that the entire AI marketing industry has adopted one universal vocabulary.

Core Research Terms

AI Recommendation

An AI recommendation occurs when an AI platform meaningfully presents a company, agency, product, platform, or service as an option that addresses the user's stated need. Examples may include: “For enterprise AI visibility tracking, I would consider Company A.” or: “Company B is one of the stronger options for a SaaS company looking for GEO support.” A recommendation does not always need to be phrased as:

“I recommend Company A.” The full context of the response matters.

Recommendation vs. Mention

A mention is not automatically a recommendation. For example: “Company A competes with Company B.” mentions both companies. It does not necessarily recommend either one. Likewise: “Unlike Company A, Company B focuses primarily on enterprise customers.” may simply provide comparative context. AI Marketing Consensus Index attempts to distinguish:

Recommendations
Conditional recommendations
Mentions
Comparisons
Citations
Negative references

rather than treating every brand appearance as equivalent.

Conditional Recommendation

A conditional recommendation occurs when an AI platform recommends a company only under a specific circumstance. For example: “Company A is a good option if your main priority is enterprise reporting.” This may still qualify as a recommendation when the condition is relevant to the research scenario. Context matters.

Negative Reference

A negative reference occurs when a company appears in an AI response primarily as something the user should avoid, reconsider, or treat cautiously. For example: “I would not choose Company A if historical citation tracking is essential.” That should not automatically count as a positive recommendation.

Recommendation Coverage

Recommendation Coverage measures how many usable AI platform responses recommended a company. Formula: Platforms Recommending the Company ÷ Usable Platform Responses Example: If 5 of 7 usable platforms recommend Company A: Recommendation Coverage = 71.4% This is one of the most important metrics in AI Marketing Consensus Index research.

Recommendation Rate

Recommendation Rate is generally used synonymously with Recommendation Coverage within a single study. For example: 5 recommendations ÷ 7 usable platforms = 71.4% Recommendation Rate Where a page uses a different denominator, the denominator should be explained.

Cross-Platform Consensus

Cross-platform consensus refers to the degree to which multiple AI platforms independently surface the same company for the same research scenario. For example: 6 of 7 platforms recommend Company A represents stronger cross-platform consensus than: 1 of 7 platforms recommends Company B. Consensus measures agreement among the AI outputs studied. It does not prove that the company is objectively superior.

Consensus Is a Recommendation Signal

A high consensus result may indicate:

Strong AI visibility
Broad category recognition
Recurring recommendation presence
Strong positioning for a particular use case

It does not automatically establish:

Higher customer satisfaction
Better service quality
Better technology
Greater ROI
Superior implementation
Universal suitability

Consensus is one signal.

Consensus Score

A Consensus Score is a standardized metric used by AI Marketing Consensus Index to summarize recommendation strength. Its primary foundation is: cross-platform recommendation coverage. Additional ranking context may include:

Average recommendation position
Top recommendation frequency
Top-three recommendation frequency
Breadth across related studies

The exact calculation should be documented in: How We Rank and should not be changed simply to favor a particular company.

Recommendation Position

Recommendation Position refers to where a company appears within an AI platform's ordered recommendation set when a meaningful order is present. For example:

Company A
Company B
Company C

Company A has recommendation position: 1 Company B: 2 Company C: 3

Average Recommendation Position

Average Recommendation Position is the average position at which a company appears across the AI responses that recommend it. Example: A company appears at:

Position 1
Position 2
Position 3

Its average recommendation position is: 2.0 Lower values generally indicate stronger placement.

Best Recommendation Position

Best Recommendation Position is the highest position a company achieved in any usable platform response. If a company appears:

#4 in ChatGPT
#2 in Gemini
#1 in Claude

its best recommendation position is: #1

Top Recommendation

A Top Recommendation generally means the company appeared in the first recommendation position within a platform response. This should not be confused with overall Consensus Index rank. A company could receive several #1 placements but still have weaker overall recommendation coverage than a company appearing consistently across more platforms.

Top-Three Recommendation

A Top-Three Recommendation means the company appeared among the first three meaningful recommendations in a platform response. This can provide useful secondary context about recommendation prominence.

AI Visibility

AI Visibility describes how often, where, and in what context a brand, company, website, product, or service appears within AI-generated responses. AI visibility can include several different forms of presence:

Recommendations
Mentions
Citations
Comparisons
Source appearances
Category associations

Because these are different behaviors, AI Marketing Consensus Index generally avoids treating “AI visibility” as one single measurement unless the metric is clearly defined.

Recommendation Visibility

Recommendation Visibility refers specifically to how often a company is actively recommended by AI platforms. This is different from: Citation Visibility and: Mention Visibility. A company can have strong recommendation visibility without being frequently cited.

Citation Visibility

Citation Visibility refers to how frequently a domain, page, publisher, or other source is cited or linked by AI systems. A company can have strong citation visibility even when its product or service is rarely recommended.

Mention Visibility

Mention Visibility measures how frequently a company appears by name within relevant AI responses, regardless of whether the appearance is:

Positive
Neutral
Comparative
Negative
Recommendatory

For this reason, mention visibility should not automatically be interpreted as recommendation strength.

Recommendation Share

Recommendation Share describes a company's share of recorded recommendation activity within a defined research universe. The exact denominator must be stated. For example, Recommendation Share could measure: Company A's recommendations as a percentage of all company recommendations recorded across 100 prompts. That is a different metric from Recommendation Coverage in a single study.

AI Share of Voice

AI Share of Voice is a broader market metric that may measure how frequently a brand appears relative to competitors across a defined set of AI prompts. Depending on the methodology, it may include:

Mentions
Recommendations
Citations
Weighted positions

Because different providers calculate AI Share of Voice differently, AI Marketing Consensus Index should state the methodology whenever using the term.

Citation

A citation is a source reference supplied by an AI platform in support of or alongside its answer. A citation may take the form of:

A linked webpage
A source card
A footnote
A referenced publication
Another visible source attribution

Citation formatting varies considerably across AI platforms.

Citation Occurrence

A Citation Occurrence is one recorded instance in which an AI response cites a source. If the same domain is cited three times across three responses, that may represent: three citation occurrences.

Unique Citing Response

A Unique Citing Response is an AI response that contains at least one citation to a particular source or domain. Multiple links to the same domain within the same answer may therefore represent:

Multiple citation occurrences
But only one unique citing response

depending on the metric being reported.

Citation Coverage

Citation Coverage describes the percentage of eligible AI responses that cite a particular source or domain. Example: If 4 of 7 responses cite Domain A: Citation Coverage = 57.1% Citation Coverage does not equal Recommendation Coverage.

Source

A source is a webpage, publication, document, database, company page, or other information resource referenced or relied upon by an AI platform. Where an AI platform exposes visible citations, we can often observe at least part of its source selection. Where citations are not visible, the complete underlying source set may be unknown.

Source Domain

A Source Domain is the internet domain associated with a cited source. For example: A citation to: example.com/research/article would normally be attributed to the source domain: example.com

Source URL

A Source URL is the specific page cited by the AI platform. This allows research to distinguish: which domains are cited from: which individual pages are cited.

Recommendation and Citation Are Different Events

Suppose an AI platform recommends: Company A while citing an article published by: Company B Then: Company A receives a recommendation event. Company B receives a citation event. We do not automatically give Company B a recommendation because its content influenced or supported the answer.

Citation Does Not Prove Causation

If an AI platform cites a particular webpage, it is reasonable to report that the page appeared as a citation. It is generally not appropriate to claim without additional evidence: “This page caused the AI platform to recommend Company A.” AI systems may rely on multiple information sources and internal processes that are not fully visible.

AI Search

AI Search refers broadly to search or discovery experiences in which generative artificial intelligence plays a major role in producing, synthesizing, ranking, or presenting answers. Examples may include:

Search-enabled AI assistants
Generative search results
AI answer engines
AI-generated recommendation experiences

AI Search Optimization

AI Search Optimization is the practice of improving the likelihood that a brand, company, product, service, website, or content asset will be:

Found
Understood
Cited
Mentioned
Recommended

within AI-powered search and answer environments. The term is broader than any one platform.

GEO

Generative Engine Optimization

GEO commonly stands for: Generative Engine Optimization We use GEO to describe strategies intended to improve visibility within generative AI and AI-powered answer environments. Depending on the context, GEO may involve:

Content structure
Entity clarity
Source authority
Third-party references
Citation acquisition
Brand positioning
Technical accessibility
Information consistency
Original data
Digital PR
Traditional SEO foundations

GEO does not have one universally accepted industry definition.

AEO

Answer Engine Optimization

AEO generally stands for: Answer Engine Optimization. The term predates the current generative-AI boom and has historically included optimization for:

Featured answers
Voice search
Direct-answer systems
Knowledge panels
Answer engines

Today, some marketers also use AEO to describe optimization for AI-generated answers. We therefore evaluate the context rather than assuming AEO and GEO always mean exactly the same thing.

AI SEO

AI SEO is an informal term commonly used to describe search optimization adapted for AI-generated search experiences. Depending on the speaker, it may refer to:

Traditional SEO using AI tools
SEO for AI search
GEO
AI content optimization
AI visibility measurement

Because the term can be ambiguous, we prefer more specific terminology where practical.

LLM Optimization

LLM Optimization generally describes efforts intended to improve how a company, brand, or source is represented within responses generated by large language models. The term overlaps substantially with:

GEO
AI Search Optimization
AI Visibility Optimization

but may focus more specifically on LLM-generated answers rather than the broader AI search ecosystem.

LLM Visibility

LLM Visibility refers to a brand or source's presence within outputs from large language model-based systems. This can include:

Recommendations
Mentions
Citations
Comparisons
Brand descriptions

In many commercial contexts, LLM Visibility and AI Visibility are used almost interchangeably. AI Marketing Consensus Index uses AI Visibility as the broader term.

Generative Search

Generative Search refers to search experiences in which generative AI produces synthesized responses rather than merely presenting a conventional list of links. Generative search can combine:

Retrieval
Search indexes
Language models
Knowledge systems
Citation systems
Recommendation logic

Answer Engine

An Answer Engine is a system designed to respond directly to a user's question rather than requiring the user to review a conventional list of search results. Modern AI platforms may function as:

Chatbots
Search engines
Answer engines
Research assistants

simultaneously.

AI Marketing

For this site, AI Marketing refers primarily to the emerging marketing discipline concerned with how brands are discovered, represented, cited, mentioned, and recommended within generative AI and AI search systems. It does not primarily refer to: using generative AI to write advertisements or automate marketing tasks. That is a separate field of AI-enabled marketing technology.

AI Marketing Agency

An AI Marketing Agency, in the context of AI Marketing Consensus Index, generally means an agency offering services materially related to:

GEO
AI Search Optimization
AI Visibility
AI citation strategy
Recommendation visibility
AI market intelligence
AEO

We do not assume every agency using the phrase “AI marketing” provides these services.

GEO Agency

A GEO Agency is an agency that provides Generative Engine Optimization or closely related AI search optimization services. Because there is no licensing body or standardized certification for GEO agencies, we may independently verify the actual services offered by companies appearing in our research.

AI Visibility Platform

An AI Visibility Platform is software primarily designed to help organizations measure how brands, products, services, competitors, or sources appear across AI systems. Features may include:

Prompt monitoring
Brand mention tracking
Recommendation tracking
Citation tracking
Competitive benchmarking
Historical trends
Source analysis
Share-of-voice metrics

Capabilities differ considerably by provider.

LLM Monitoring Platform

An LLM Monitoring Platform is software that tracks outputs from large language model-based systems. Within marketing, the term often refers to monitoring:

Brand mentions
Recommendations
Citations
Competitor appearances
Prompt-level visibility

It should not be confused with technical LLM observability software used by developers to monitor their own AI applications.

AI Citation Tracking Platform

An AI Citation Tracking Platform monitors when websites, domains, or pages are cited by AI systems. Citation tracking and recommendation tracking are related but different capabilities.

AI Search Audit

An AI Search Audit evaluates how a brand or company currently appears across relevant AI search and generative systems. An audit may examine:

Recommendation visibility
Mention visibility
Citation visibility
Competitor performance
Source patterns
Prompt coverage
Content gaps
Entity consistency

The exact scope should be defined by the provider.

AI Market Intelligence

AI Market Intelligence, in the context of this site, means research about how AI systems represent a market, its competitors, and the sources influencing those representations. It may include:

Recommendation share
Competitor visibility
Citation patterns
Market leaders
Prompt-level differences
Historical movement
Platform disagreement
Source ecosystems

Competitive AI Visibility

Competitive AI Visibility compares one brand's presence in AI outputs with the presence of competing brands. It may measure:

Recommendation rate
Mentions
Citations
Ranking position
Prompt coverage
Share of voice

The metric should always identify what type of visibility is being compared.

Citation Architecture

Citation Architecture refers to the strategic development of sources, references, information structures, and third-party authority signals that may improve the likelihood that AI systems can discover, understand, and cite useful information about a brand or subject. It can involve both:

A company's own content
Third-party sources

We do not claim that any particular citation-building activity guarantees an AI citation or recommendation.

Authority Building

Authority Building refers broadly to efforts designed to strengthen the credibility, discoverability, and external validation of a company or subject. In an AI-search context, this may include:

Original research
Third-party coverage
Expert commentary
Citations
Industry references
Consistent entity information
High-quality informational resources

Authority is not represented by one universal numeric score.

AI Content Optimization

AI Content Optimization refers to improving content so it is more useful and understandable within AI-powered discovery environments. This may involve:

Clear structure
Direct answers
Strong sourcing
Entity clarity
Supporting evidence
Original data
Appropriate metadata
Logical sectioning

Content optimization alone does not guarantee AI visibility.

Prompt

A Prompt is the question, request, scenario, or instruction submitted to an AI platform. Prompts are a fundamental research unit within AI Marketing Consensus Index.

Research Prompt

A Research Prompt is the standardized core question used for a Consensus Index study. Where practical, substantially the same core research prompt is submitted across all included AI platforms.

Open Recommendation Prompt

An Open Recommendation Prompt asks the AI platform to identify suitable companies without supplying a predetermined candidate list. Example:

Which AI visibility platforms would you recommend for a mid-market B2B company?

This is different from a named-company comparison.

Named-Company Comparison

A Named-Company Comparison explicitly asks an AI system to compare specific companies. Example: Compare Company A, Company B, and Company C for enterprise AI visibility monitoring. Named-company comparisons can be useful but answer a different research question from open recommendation research. They should be labeled accordingly.

Prompt Seeding

Prompt Seeding occurs when specific companies or brands are inserted into a prompt before the AI responds. For standard open recommendation studies, AI Marketing Consensus Index generally avoids seeding:

Related companies
Affiliates
Advertisers
Competitors

unless the research is specifically designed as a named-company comparison.

Research Scenario

A Research Scenario describes the type of person, company, need, budget, or use case represented by the prompt. Examples:

Enterprise CMO
B2B SaaS company
Digital marketing agency
Small business
Ecommerce brand

Specific scenarios help produce recommendations that are more useful than asking only for the “best” company overall.

AI Platform

An AI Platform is the consumer-facing AI product or service treated as one voting unit in our standard research. Our current standard research universe includes:

ChatGPT
Gemini
Claude
Perplexity
Grok
DeepSeek
Kimi

One Platform = One Vote

For standard Consensus Index research: one platform receives one vote. We do not normally give several votes to one provider because it offers multiple:

Models
Modes
Subscription tiers
Interfaces

AI Model

An AI Model is the underlying machine-learning model or model family generating or helping generate a platform's response. The specific model may change even while the public-facing platform remains the same. For that reason, our public consensus framework generally uses the platform as the voting unit while preserving model metadata where practical.

Search-Enabled Response

A Search-Enabled Response is generated in an environment where the AI platform can retrieve current information from the web or another search system. This can materially affect recommendations.

Non-Search Response

A Non-Search Response is generated without active web retrieval in the research environment. Such a response may rely more heavily on:

Existing model knowledge
Training data
Internal knowledge structures

Usable Response

A Usable Response is an AI response that meaningfully addresses the research question and contains enough information to be analyzed under the study methodology.

Unusable Response

An Unusable Response may include:

Technical failure
Timeout
Refusal
Empty result
Irrelevant answer
Severely incomplete response

An unusable response should not automatically count as a negative vote against every company.

Usable Platform Denominator

The Usable Platform Denominator is the actual number of AI platforms that returned usable responses for a study. Example: If seven platforms were scheduled but only six produce usable responses: Usable Platform Denominator = 6 A company recommended by five receives: 5 of 6 = 83.3% Recommendation Coverage

Research Universe

The Research Universe is the complete set of:

Platforms
Prompts
Companies
Time period
Categories

included in a particular analysis. Metrics should be interpreted within the relevant research universe.

Research Date

The Research Date is the date on which the underlying AI responses were collected. This is one of the most important dates on a Consensus Index study.

Publication Date

The Publication Date is the date the research was first published. It may differ from the research date.

Verification Date

The Verification Date records when material company, product, service, or platform facts were checked. Research collection and factual verification are separate processes.

Reviewed Date

The Reviewed Date indicates when a named human reviewer completed meaningful editorial review. A formatting change or minor typo correction should not automatically create a new human review date.

Date Modified

Date Modified records a meaningful change to the published page. It should not be artificially refreshed merely to make old research appear new.

Research Snapshot

A Research Snapshot is a dated set of AI responses and associated analysis representing the recommendation environment at that time. A future rerun creates: a new snapshot rather than rewriting the old one.

Rerun

A Rerun is a new execution of the research using the same or substantially similar methodology at a later time. A rerun can reveal:

Recommendation gains
Recommendation losses
New competitors
Platform changes
Citation shifts

Historical Research

Historical Research refers to prior research snapshots preserved for comparison. Historical results should not be deleted merely because they become commercially inconvenient.

Research Update

A Research Update occurs when new information or a new research cycle materially changes the page. This may include:

New AI responses
Changed company features
New platforms
Updated pricing
New verification information

Correction

A Correction occurs when previously published information was wrong. Examples:

Recommendation count incorrectly calculated
Company misidentified
Product feature reported incorrectly
Data-processing error

A correction is different from a normal research rerun.

Company

A Company is the commercial or organizational entity being analyzed. Depending on the study, this may include:

Agency
Software company
Research firm
Consultancy
Platform provider

Platform vs. Company

The word platform can refer to two different things in AI marketing.

Research AI Platform

Examples:

ChatGPT
Gemini
Claude

These are the AI systems providing recommendations.

Commercial AI Visibility Platform

Examples would be companies selling software for tracking AI visibility. Our articles should make the context clear.

Entity Normalization

Entity Normalization is the process of identifying different references that represent the same company or product and recording them consistently. For example:

“Example AI”
“ExampleAI”
“exampleai.com”

might all represent the same company. Normalization prevents artificial fragmentation of recommendation counts.

Entity Disambiguation

Entity Disambiguation distinguishes similarly named but genuinely different:

Companies
Products
Divisions
Platforms
Agencies

We do not collapse materially different organizations merely because their names are similar.

Study

A Study is a specific research question analyzed across multiple AI platforms. Example: Best AI Visibility Platforms for Enterprise Companies Each study has its own:

Prompt
Research date
Platform responses
Recommendation data
Rankings
Verification
Analysis

Category Ranking

A Category Ranking aggregates performance across multiple related studies. It measures breadth and recurring recommendation strength across a category rather than performance on only one prompt.

Site-Wide Ranking

A Site-Wide Ranking summarizes broad recommendation presence across multiple categories or studies. It should not be interpreted as: “This is objectively the best AI marketing company.” It measures performance within the AI Marketing Consensus Index dataset.

Ranking

A Ranking is the ordered presentation produced by the published methodology. For individual studies, Recommendation Coverage is the primary signal. Average Recommendation Position and other defined metrics may provide secondary ranking context.

Tie

A Tie occurs when two or more companies produce equivalent ranking results under the defined methodology. We allow genuine ties. We do not force a winner merely because an ordered list is visually convenient.

Affiliate

An Affiliate is a company with which AI Marketing Consensus Index may have a performance-based commercial relationship. We may receive compensation if a user:

Clicks
Registers
Requests information
Purchases
Completes another qualifying action

Affiliate status does not create recommendation credit.

Advertiser

An Advertiser is a company that pays for promotional exposure. Advertising does not create or alter earned AI recommendation results.

Sponsor

A Sponsor financially supports specific content, research, or another project. A sponsor may help define a legitimate research topic. It cannot purchase the resulting AI recommendations or Consensus ranking.

Related Company

A Related Company is a business with a material ownership, financial, operational, or contributor relationship with AI Marketing Consensus Index or a person materially involved in its editorial process. For this project, particularly important related companies include: LLM Authority Index and: CiteWorks Studio.

LLM Authority Index

LLM Authority Index provides AI research data and measurement infrastructure supporting AI Marketing Consensus Index. Its contribution may include:

AI response collection
Recommendation data
Citation data
Competitive benchmarking
Historical AI visibility data
Research infrastructure

LLM Authority Index may itself appear in relevant research. Its relationship with this project does not create additional recommendation votes or ranking weight.

CiteWorks Studio

CiteWorks Studio provides AI search strategy and subject-matter support to AI Marketing Consensus Index. Its contribution may include:

GEO/AEO strategy context
Research taxonomy
Research-question input
AI citation concepts
AI authority concepts
Practical marketing interpretation

CiteWorks Studio may itself appear in relevant research. Its relationship does not create recommendation votes or ranking preference.

Research Reviewer

A Research Reviewer is a named person who reviews applicable research for:

Terminology
Interpretation
Methodological context
Unsupported conclusions
Relevant subject-matter issues

A research reviewer does not automatically determine the underlying recommendation data.

Mark B. Huntley, J.D.

Mark B. Huntley, J.D. serves as: AI Search & Visibility Research Reviewer for applicable AI Marketing Consensus Index research. Mark also has financial and operational interests associated with:

LLM Authority Index
CiteWorks Studio

Those relationships are disclosed because both companies may appear in research published by this site.

Human Review

Human Review is the editorial process in which a qualified reviewer examines applicable research after the underlying AI recommendation data has been collected. Human review may:

Correct interpretation
Add context
Flag limitations
Improve terminology
Identify unsupported claims

Human review should not manufacture additional AI votes.

Editorial Independence

Editorial Independence, as used by AI Marketing Consensus Index, means that commercial relationships do not purchase or alter the underlying research result. It does not mean the publication has:

No affiliates
No advertising
No related businesses
No commercial incentives

Those relationships exist and should be disclosed.

Research Independence

Research Independence refers to safeguards intended to prevent commercial interests from changing valid AI recommendation data after collection. For example: CiteWorks Studio cannot receive an extra AI vote merely because it is related to the project.

Raw Response

A Raw Response is the original AI output collected during research before recommendation extraction or editorial interpretation. Where practical, raw responses should be preserved as part of the research record.

Structured Dataset

The Structured Dataset is the organized record created from the underlying research. It may include:

Platform
Prompt
Research date
Companies recommended
Recommendation position
Citations
Source URLs
Model metadata
Search state
Verification fields

The structured dataset serves as the source of truth for published ranking metrics.

Verification

Verification is the process of checking material factual claims about a company, service, product, or platform using appropriate current sources. Verification is separate from the AI recommendation event.

Verified Fact

A Verified Fact is a factual claim that has been checked against a reasonably authoritative current source. Examples might include:

Current product availability
Publicly listed pricing
Supported AI platforms
Service descriptions
Feature availability

AI-Reported Claim

An AI-Reported Claim is a statement made by an AI system that has not necessarily been independently verified. AI Marketing Consensus Index should not automatically present every AI-generated statement as fact.

AI Hallucination

An AI Hallucination is a generated statement that is unsupported, fabricated, materially inaccurate, or incorrectly presented as fact. Recommendation data can still be preserved even when an AI platform's stated reasoning includes an inaccurate factual claim.

Verification Does Not Rewrite the AI Vote

Suppose five AI systems recommend Company A because they believe it offers Feature X. Current verification shows Feature X has been discontinued. The historical result remains: 5 of 7 AI platforms recommended Company A. We separately disclose: Current verification does not support the feature claim contained in the AI responses. This preserves the distinction between: what AI recommended and:

what is currently factually true.

Shared Source Ecosystem

A Shared Source Ecosystem refers to the possibility that several AI platforms may rely upon overlapping:

Websites
Publishers
Company pages
Research
Reviews
Datasets

Cross-platform agreement therefore does not prove that each AI system independently reached the conclusion from completely different evidence.

AI Recommendation Consensus Is Not Independent Factual Consensus

If six AI platforms recommend the same company: that means: six platform responses recommended the company. It does not necessarily mean: six independent factual investigations concluded the company is objectively best. This distinction is central to interpreting our research responsibly.

Buyer Intent

Buyer Intent refers to the degree to which a research question reflects a user who may be approaching a purchasing or vendor-selection decision. For example: Best AI visibility platform for an enterprise marketing team typically reflects stronger commercial intent than:

What is AI visibility?

AI Marketing Consensus Index may intentionally research commercially relevant questions. Commercial relevance can influence: which questions we study. It does not determine: which company wins.

High-Intent Prompt

A High-Intent Prompt is a question indicating that a user is actively evaluating:

Companies
Vendors
Agencies
Software
Services
Solutions

These prompts are especially relevant for recommendation research.

Informational Prompt

An Informational Prompt primarily seeks education rather than a recommendation. Example:

What is generative engine optimization?

Informational prompts can be useful for citation research but may not belong in recommendation rankings.

Navigational Prompt

A Navigational Prompt primarily seeks a particular company, platform, page, or resource. Example:

What is the pricing for Company A?

Navigational prompts should not generally be treated as open recommendation research.

Branded Prompt

A Branded Prompt includes a particular company or brand name. Branded prompts answer different questions from unbranded category prompts and should generally be analyzed separately.

Unbranded Prompt

An Unbranded Prompt does not name a preferred company. Example:

Which AI citation tracking tools would you recommend for an agency?

Unbranded prompts are especially useful for measuring natural recommendation visibility.

Natural AI Visibility

Natural AI Visibility, as used on this site, refers to a company appearing in an AI response without that company being intentionally supplied in the prompt. This does not mean the recommendation occurred without outside influences such as:

SEO
PR
Publishing
Citations
Brand recognition
Third-party content

It simply means the research prompt did not name the company.

Research Limitation

A Research Limitation is a condition that affects how confidently a result should be interpreted. Examples include:

AI nondeterminism
Search-state differences
Personalization
Platform changes
Geographic variation
Overlapping information sources
Small sample size
Incomplete citations

Limitations should be disclosed rather than hidden.

Our Terminology Standard

AI marketing vocabulary will continue to evolve. When terminology is ambiguous, AI Marketing Consensus Index aims to:

Define how the term is being used.
Avoid treating citations, mentions, and recommendations as interchangeable.
State the denominator behind percentages.
Separate AI output from verified facts.
Separate platform-level research from model-level research.
Distinguish open recommendations from seeded comparisons.
Disclose related-company relationships.
Preserve historical definitions where needed for consistent comparisons.

Frequently Asked Questions

Is GEO the same thing as AI SEO?

Not necessarily. The terms overlap substantially, but different companies use them differently. We generally use GEO for Generative Engine Optimization and AI Search Optimization as a broader category.

Is AEO the same as GEO?

Not exactly. AEO historically referred to Answer Engine Optimization and can include older direct-answer and voice-search environments. Modern usage increasingly overlaps with GEO.

Is a mention the same as a recommendation?

No. A company can be mentioned without being recommended.

Is a citation the same as a recommendation?

No. An AI system can recommend one company while citing another company's content.

What is the most important ranking metric?

For individual standard Consensus Index studies: Recommendation Coverage is the primary ranking signal.

What does 5 of 7 mean?

Five of seven usable AI platform responses recommended the company. That equals: 71.4% Recommendation Coverage.

What happens if only six platforms provide usable answers?

The denominator becomes six. A technical failure is not treated as a negative vote.

Does a #1 recommendation from ChatGPT count more than a recommendation from another platform?

Not in our standard cross-platform consensus methodology. Each platform receives one vote. Recommendation position may be used as a secondary metric.

Can LLM Authority Index appear in these rankings?

Yes. It may legitimately fall within the candidate universe for certain studies. Its role as a provider of research data and measurement infrastructure does not give it additional recommendation credit.

Can CiteWorks Studio appear in these rankings?

Yes. It may legitimately fall within the candidate universe for GEO, AI search agency, citation, audit, or related research. Its strategic relationship with this project does not create additional AI votes.

Does Mark Huntley decide whether those companies rank?

No. Mark may review interpretation and methodology. Valid AI recommendation counts come from the underlying research responses.

How We Rank →

Research Methodology →

Platforms We Analyze →

Data & Research Limitations →

How We Verify AI Marketing Companies, Platforms & Services →

Editorial Independence →

Human Review Policy →

Our Definitions Standard in One Sentence

AI Marketing Consensus Index defines recommendations, mentions, citations, visibility, prompts, platforms, ranking metrics, and commercial relationships separately so readers can understand exactly what each published metric measures—and what it does not.

Transparency

AI research data and measurement infrastructure provided by LLM Authority Index. AI search strategy and subject-matter support provided by CiteWorks Studio. Research reviewed by Mark B. Huntley, J.D., where indicated.

Mark has financial and operational interests associated with LLM Authority Index and CiteWorks Studio. Those relationships do not alter raw AI responses, recommendation counts, recommendation coverage, or ranking calculations. Last Updated: Not yet supplied