How We Rank

Our Rankings Start With the Marketing Problem — Not the Company

AI Marketing Consensus Index does not begin by choosing a group of companies and deciding which ones we prefer. We begin with a specific marketing situation. For example:

A mid-market B2B company wants an agency that can improve its visibility and recommendation frequency across ChatGPT, Gemini, Perplexity, and other AI systems. It needs competitive analysis, citation strategy, measurement, and implementation support. We then submit substantially the same core question independently across multiple leading AI platforms. The resulting recommendations become the research dataset.

The ranking comes from that dataset.

The Short Version

Our ranking methodology can be summarized in five steps:

1. Define a specific AI marketing need
2. Ask multiple AI platforms the same core recommendation question
3. Record which companies are meaningfully recommended
4. Measure cross-platform recommendation coverage and recommendation position
5. Verify material company information and add editorial context without changing the underlying votes

The strongest ranking signal is:

Cross-Platform Recommendation Coverage

In simple terms:

How many of the AI platforms independently recommended this company for this specific situation?

Recommendation Coverage Is the Primary Ranking Signal

Suppose seven AI platforms produce usable responses. The results are: Company A — recommended by 6 of 7 platforms Company B — recommended by 5 of 7 Company C — recommended by 3 of 7 Company D — recommended by 2 of 7 Company A demonstrated the strongest recommendation coverage. Its recommendation rate would be: 6 ÷ 7 = 85.7% That is our primary consensus signal.

What a High Recommendation Rate Means

A high recommendation rate means multiple AI systems independently surfaced the same company for the marketing situation being studied. It may indicate strong:

• AI brand visibility
• Category association
• Recommendation presence
• Market recognition
• Source visibility
• Product or service positioning

It does not automatically mean the company has:

• The best software
• The best agency
• The highest customer satisfaction
• The strongest ROI
• The best implementation team
• The best pricing
• The best fit for every company

AI Marketing Consensus Index measures: AI recommendation consensus not: objective universal superiority.

One Platform Gets One Vote

Each AI platform included in a standard study contributes one vote. Our standard research universe is designed around platforms such as:

• ChatGPT
• Gemini
• Claude
• Perplexity
• Grok
• DeepSeek
• Kimi

We do not give additional weighting because one platform:

• Has more users
• Has greater market share
• Is owned by a larger company
• Performs better on a benchmark
• Is more commercially important

For standard recommendation coverage: one platform = one vote.

The Platform Is the Primary Research Unit

AI companies may offer multiple:

• Models
• Search modes
• Reasoning modes
• Interfaces
• Consumer products
• APIs

Our public methodology generally treats the platform as the stable unit of analysis. Where practical, we separately record:

• Model
• Mode
• Search state
• Interface
• Research date
• Other relevant environment information

Different models from the same company do not automatically receive separate votes in a standard consensus study.

Usable Responses Determine the Denominator

Not every platform necessarily produces a usable response in every study. A response may be unusable because of:

• Technical failure
• Refusal
• Incomplete output
• Access problems
• A response that fails to address the question meaningfully

If six of seven platforms produce usable responses, recommendation coverage uses six as the denominator. Example: 5 recommendations from 6 usable platforms = 83.3% We do not treat a technical failure as though the platform voted against every company.

What Counts as a Recommendation?

A company counts as recommended when an AI platform meaningfully presents it as an appropriate option for the specific marketing need. Examples: “For enterprise AI visibility monitoring, I would consider Company A, Company B, and Company C.” or: “Company A is one of the strongest GEO agencies for B2B SaaS organizations.” Those are recommendations.

A Mention Is Not Automatically a Recommendation

An AI system may mention a company without recommending it. For example: “Company A focuses primarily on agencies, while Company B is designed for enterprise teams.” Depending on the surrounding response, Company A may simply be part of a comparison. Our research attempts to distinguish:

• Positive recommendation
• Conditional recommendation
• Incidental mention
• Comparison reference
• Negative reference

Only meaningful recommendations count toward recommendation coverage.

Citations Are Not Recommendations

This distinction is especially important on AI Marketing Consensus Index. Suppose an AI response: cites an article from Company A but: recommends Company B's platform. Company A received a: citation Company B received a: recommendation Those are different research events. We should not give Company A a recommendation vote simply because its website was cited.

Recommendations Are Not Citations

The opposite can also happen. An AI system may recommend a company without citing:

• The company's website
• A review of the company
• Any visible source supporting the recommendation

That still counts as a recommendation if the response meaningfully recommends the company. Citation data can be analyzed separately.

Mentions, Citations, and Recommendations Are Different Metrics

AI Marketing Consensus Index should preserve these separately.

Mention

The company appears in the response.

Recommendation

The AI system presents the company as an appropriate option.

Citation

A source associated with the company or another publisher is cited in support of the answer.

Recommendation Position

Where the company appears within an ordered recommendation set. These measurements answer different questions.

Conditional Recommendations Can Count

AI marketing recommendations frequently depend on use case. For example: “Company A is a strong option for enterprise teams, while Company B may be better for smaller agencies.” Both companies may count as recommendations. The condition should remain attached to the recommendation. We should not rewrite that as: “Company A and Company B are equally best for everyone.”

Recommendation Position Is the Secondary Signal

Two companies can receive the same recommendation coverage but appear at very different positions. Example:

Company A

Recommended by 5 of 7 platformsAverage recommendation position: #1.8

Company B

Recommended by 5 of 7 platformsAverage recommendation position: #3.7 Company A demonstrated stronger recommendation placement. Recommendation position is therefore an important secondary signal.

Coverage Comes Before Position

Suppose:

Company A

Recommended by 6 of 7 platformsAverage position: #3.0

Company B

Recommended by 3 of 7 platformsAverage position: #1.3 Company A will generally rank higher.

Why?

Because the primary purpose of the Index is to measure: cross-platform agreement. Company B performs strongly when it appears, but fewer AI systems recommend it.

Our General Ranking Hierarchy

Study-level rankings generally prioritize:

1. Cross-Platform Recommendation Coverage

The dominant signal.

2. Average Recommendation Position

Primary tie-breaker.

3. Top Recommendation Frequency

How often did the company appear first?

4. Top-Three Frequency

Useful additional context.

5. Situation-Specific Recommendation Support

How strongly did the AI systems explain that the company actually fit the scenario?

Material verification findings and editorial review may affect how we interpret the result. They do not manufacture additional AI votes.

Genuine Ties Are Allowed

We do not need to invent precision. If two companies perform essentially identically, the published result may show:

• A tie
• Shared rank
• Near-tied consensus

A real tie is more useful than forcing one company into first place without a defensible reason.

Qualification for Full Analysis

A company should generally receive recommendations from at least: 2 independent AI platforms before receiving a full company-fit analysis within a study. Single-platform recommendations may still appear in:

• Results tables
• Research data
• Historical records

They simply may not receive the same editorial depth as companies with stronger cross-platform support.

Study-Level Rankings Matter Most

Every study addresses a specific marketing need. For example:

Best AI Visibility Platforms for Enterprise Companies

A study might produce: Rank Company AI Platforms Recommending Coverage Avg. Position 1 Company A 6 of 7 85.7% 1.8 2 Company B 5 of 7 71.4% 2.2 3 Company C 4 of 7 57.1% 2.7 For an enterprise buyer, that situation-specific ranking is generally more useful than a broad site-wide leaderboard.

Category Rankings

Individual studies roll into category-wide rankings. Our five primary categories are:

• AI Search & GEO Agencies
• AI Visibility & LLM Monitoring Platforms
• AI Search Audits & Market Intelligence
• AI Citation & Authority Building
• AI SEO & Content Optimization Tools

How Category Rankings Work

Category rankings measure how consistently a company performs across multiple related use cases. For example, an AI visibility platform may appear in studies such as:

• Best AI Visibility Platforms Overall
• Best AI Visibility Platforms for Enterprise
• Best AI Visibility Platforms for Agencies
• Best ChatGPT Visibility Tracking Tools
• Best AI Citation Tracking Platforms
• Best AI Competitive Benchmarking Tools
• Best AI Search Dashboards

A company appearing strongly across many studies demonstrates broader category-level recommendation strength.

Category Metrics

Category leaderboards may include:

• Rank
• Company
• Consensus Score
• Studies Appeared In
• Total AI Recommendations
• AI Recommendation Rate
• Average Recommendation Position
• Best Position

These metrics provide different views of the underlying research.

What Is the Consensus Score?

The Consensus Score is a normalized category-level measurement designed to summarize recurring recommendation strength. It may incorporate factors such as:

• Breadth across studies
• Recommendation coverage
• Recommendation position

The exact formula should remain consistent within a methodology version. The Consensus Score is a summary. It does not replace the underlying transparent metrics.

Breadth Across Studies Matters

Consider:

Company A

Appears in 16 of 20 studies.

Company B

Appears in 5 of 20 studies. Company B may occasionally rank first. Company A has demonstrated much broader recommendation presence across the category. That breadth matters when calculating category-level strength.

Category Rank Does Not Mean Universal Superiority

A platform ranking #1 in: AI Visibility & LLM Monitoring Platforms does not automatically mean it is best for:

• Every company size
• Every budget
• Every AI platform
• Every reporting need
• Every agency
• Every enterprise use case

It means it demonstrated the strongest recurring AI recommendation performance across the studies included in that category.

Site-Wide Rankings

AI Marketing Consensus Index may aggregate research across all five categories. A site-wide ranking can reveal companies with broad recommendation presence across the AI marketing ecosystem. However, it must be interpreted carefully. An agency and a software platform may solve fundamentally different problems. A broad ranking should not erase those distinctions.

Site-Wide #1 Does Not Mean Best for Everyone

A company could rank highly site-wide because it appears across many:

• Agency
• Software
• Audit
• Citation
• Intelligence

studies. Another company may dominate a single narrow niche. The site-wide leaderboard measures: broad recurring AI recommendation presence not: universal superiority.

Company Type Matters

AI Marketing Consensus Index may compare entities including:

• Agencies
• Software platforms
• Research firms
• Market-intelligence providers
• Consulting firms
• Content tools
• Citation tools

We preserve those distinctions. A recommendation for an agency should not automatically become a recommendation for software. Likewise, a software recommendation should not be interpreted as an endorsement of consulting services the company does not provide.

Entity Normalization

AI systems may refer to the same company in different ways. We normalize obvious equivalent references where appropriate. For example: CompanyName and: Company Name may refer to the same entity. But normalization should not merge:

• Different companies with similar names
• Separate products that materially differ
• Parent and subsidiary companies without justification
• Agencies and related software products that operate separately

The original response should remain preserved.

Verification Is Separate From Ranking

Suppose six AI systems recommend Company A because they believe it monitors seven AI platforms. Current verification shows it monitors four. The research result remains: 6 of 7 AI systems recommended Company A. The factual verification layer should separately explain: Current company documentation does not support the seven-platform claim repeated in several AI responses.

We Do Not Rewrite AI History When the AI Is Wrong

If an AI platform makes a factual mistake, we preserve the recommendation as part of the research record. We separately report the discrepancy. This distinction is essential to the integrity of the dataset.

Verification Can Affect Our Editorial Interpretation

Although verification does not change the AI vote count, it can affect our analysis. For example: Company A had the strongest recommendation consensus, but several AI systems attributed capabilities to the platform that we could not verify in its current product documentation. That can be highly relevant to a buyer.

Related Business Relationships Require Additional Safeguards

AI Marketing Consensus Index has relationships with: LLM Authority Index and: CiteWorks Studio Mark B. Huntley, J.D., our AI Search & Visibility Research Reviewer, also has financial and operational interests connected to those companies. Because those companies may appear in research published on this site, we apply additional transparency rules.

LLM Authority Index Provides Research Infrastructure

LLM Authority Index provides AI research data and measurement infrastructure supporting AI Marketing Consensus Index. That may include support for:

• AI response collection
• Recommendation data
• Citation data
• Competitive analysis
• Historical visibility research
• Measurement infrastructure

This contribution does not give LLM Authority Index additional recommendation credit. If it appears in a ranking, its recommendation count must come from the same underlying AI research used for competing companies.

CiteWorks Studio Provides Strategy and Subject-Matter Support

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

• GEO/AEO concepts
• Research taxonomy
• AI-search strategy
• Citation strategy
• Research-question development input
• Practical marketing context

This relationship does not give CiteWorks Studio additional recommendation credit.

Related Companies Receive No Bonus Points

Neither CiteWorks Studio nor LLM Authority Index receives:

• Extra AI votes
• Additional recommendation coverage
• Better recommendation position
• Bonus Consensus Score
• Automatic inclusion
• Guaranteed qualification
• Preferential tie-breaking

Their ranking treatment must follow the same underlying methodology applied to competitors.

Related Companies Are Not Automatically Added to Results

In ordinary open recommendation studies, we do not insert CiteWorks Studio or LLM Authority Index into the result after the fact. If the AI platforms do not recommend them: they receive no recommendation credit.

Related Companies Should Not Be Seeded Into Ordinary Open Prompts

Generic recommendation studies should normally ask open questions such as:

Which AI search agencies would you recommend for this situation?

They should not say: Choose between CiteWorks Studio, Agency A, Agency B, and Agency C. unless the study is intentionally designed as a named-company comparison. This rule also applies to competitors.

Related Companies Can Rank Poorly

CiteWorks Studio or LLM Authority Index may:

• Rank below competitors
• Receive few recommendations
• Fail to qualify for full analysis
• Fail to appear entirely
• Lose recommendation share in later research

Those results should remain visible.

Related Companies Can Rank First

If the underlying research independently shows: CiteWorks Studio — 6 of 7 or: LLM Authority Index — 7 of 7 the result can be published. The relationship does not invalidate the research. It requires prominent disclosure.

Mark Huntley Does Not Choose the Winner

Mark B. Huntley, J.D. serves as: AI Search & Visibility Research Reviewer His role includes reviewing issues such as:

• AI search terminology
• Citation interpretation
• Recommendation interpretation
• GEO/AEO distinctions
• Competitive analysis
• Research conclusions
• Methodological context

Mark does not manually alter:

• AI recommendation counts
• Recommendation coverage
• Recommendation position
• Consensus Score

because he prefers one company.

Mark Can Challenge the Interpretation

Suppose: Competitor A — 6 of 7 CiteWorks Studio — 4 of 7 Those numbers remain the research result. Mark may still identify:

• An incorrect product description
• A methodology problem
• Confusion between citation tracking and recommendation tracking
• An outdated feature claim
• A misleading interpretation

That context can be added without changing the votes.

Reviewer Involvement Must Be Disclosed

Where Mark reviews an article, the page should state: Reviewed by Mark B. Huntley, J.D. Where a related company appears, the page should also display an appropriate: Related Business Disclosure Reviewer involvement is not a substitute for conflict disclosure.

Commercial Relationships Do Not Affect Rankings

AI Marketing Consensus Index may generate revenue through:

• Affiliate relationships
• Referrals
• Advertising
• Sponsorship
• Data licensing
• Leads
• Related businesses

These relationships do not determine:

• Recommendation counts
• Recommendation coverage
• Recommendation position
• Study rankings
• Category rankings
• Consensus Scores

Commercial Value Can Influence Which Questions We Study

AI Marketing Consensus Index is a commercial publication. We may prioritize research topics because they have:

• Buyer demand
• Search demand
• Strategic importance
• Affiliate potential
• Lead-generation potential
• Market interest

That is different from selecting the winner. Commercial value may influence: the question but should not influence: the answer.

Advertisers Cannot Buy Consensus

An advertiser cannot pay to receive:

• An AI recommendation
• A higher recommendation rate
• A better average position
• A higher Consensus Score
• Category leadership

Paid promotional placement must remain separate from earned research rankings.

Affiliate Links Do Not Create Ranking Credit

A company may have:

• An affiliate program
• No affiliate program
• A direct commercial relationship
• No commercial relationship

None of those states should alter the research calculation. A non-commercial company can rank first.

Historical Rankings Are Preserved

AI marketing changes quickly. A company could receive: 6 of 7 recommendations in January and: 3 of 7 in July. The January result should not be erased simply because newer research differs. Where practical, we preserve historical research snapshots.

A Rerun Is Not a Correction

New research producing a different result does not necessarily mean the old research was wrong. A rerun reflects a new point in time. A correction is reserved for actual errors such as:

• Miscounted recommendations
• Incorrect company normalization
• Wrong recommendation position
• Incorrect ranking calculation
• Incorrect factual reporting

Methodology Changes Should Be Versioned

If we materially change:

• AI platform set
• Recommendation extraction rules
• Ranking formula
• Qualification threshold
• Category aggregation

the methodology should be updated. Where practical, historical research should remain tied to the methodology version used at the time.

AI Platforms Can Share Information Sources

Seven-platform consensus does not necessarily mean: seven completely independent evidence ecosystems. Different AI systems may rely on overlapping:

• Publishers
• Company websites
• Reviews
• Industry research
• Source documents

That is one reason we call this: recommendation consensus rather than: independent factual validation.

AI Visibility Can Influence AI Recommendations

Companies with greater online visibility may have an advantage because AI systems encounter more information about them. Potential influences include:

• Brand awareness
• Search visibility
• Publisher coverage
• Company documentation
• Third-party references
• Reviews
• Citations
• Historical online presence

Our ranking measures the recommendation environment as it exists. It does not assume every company entered that environment with equal visibility.

New Companies May Be Underrepresented

A newer company may offer an excellent service yet have limited:

• Web coverage
• Historical mentions
• Third-party references
• AI training exposure
• Current AI visibility

Low recommendation frequency does not prove low product quality. It means the company was not frequently recommended in the research conducted.

Absence Is Not Rejection

If an AI platform does not mention a company, that does not necessarily mean the platform evaluated and rejected it. The system may simply not have surfaced it. This distinction should be considered when interpreting low recommendation coverage.

Rankings Are Research Snapshots

Every ranking should be understood in the context of its: research date. AI models change. Search integrations change. Companies change. The market changes. Our rankings are therefore snapshots of AI recommendation behavior rather than permanent awards.

Ranking Tables

Study pages should generally use a table such as: Rank Company AI Platforms Recommending Recommendation Coverage Avg. Position Category pages may use: Rank Company Consensus Score Studies Recommendation Rate Avg. Position Best Position Related businesses should be visibly identified where appropriate.

How to Read a Ranking

Before relying on a result, ask:

• What specific marketing situation was studied?
• How many AI platforms produced usable responses?
• How many recommended the company?
• Where did it typically rank?
• Why did the AI systems recommend it?
• Were the supporting product or service claims verified?
• Is the company an agency, software platform, research provider, or another type of solution?
• Does AI Marketing Consensus Index have a related business relationship with the company?
• Does the scenario actually resemble my organization's need?

Example

Imagine seven platforms respond: Company Recommended By Coverage Avg. Position Company A 6 of 7 85.7% 2.0 Company B 6 of 7 85.7% 3.2 Company C 5 of 7 71.4% 1.8 Company D 2 of 7 28.6% 1.5 Company A would generally rank above Company B because they have equal recommendation coverage but Company A has stronger average placement. Company C performs very well when recommended but has less cross-platform agreement.

Company D appears near the top when it appears but lacks broad consensus.

What a #1 Ranking Means

A #1 ranking means: This company demonstrated the strongest recommendation performance under the published methodology for the specific research being summarized. It does not automatically mean:

• Best company for every buyer
• Best software
• Best agency
• Highest ROI
• Best pricing
• Best service
• Guaranteed business results

Our Ranking Principle

Our ranking methodology can be summarized in one sentence:

AI Marketing Consensus Index gives the strongest ranking credit to companies recommended most consistently across multiple AI platforms for a defined marketing situation, uses recommendation position as a secondary signal, and keeps factual verification, editorial review, related-business relationships, and commercial interests separate from the underlying recommendation vote count.

Transparency About Our Related Businesses

LLM Authority Index provides AI research data and measurement infrastructure supporting AI Marketing Consensus Index. CiteWorks Studio provides AI search strategy and subject-matter support. Mark B. Huntley, J.D. serves as AI Search & Visibility Research Reviewer and has financial and operational interests associated with these businesses.

Because CiteWorks Studio and LLM Authority Index may appear in the same markets that we research, their involvement creates a potential conflict of interest. Our response is: disclosure + structural separation. Their involvement does not alter:

• Raw AI responses
• Recommendation counts
• Recommendation coverage
• Recommendation position
• Consensus Score calculations
• Competitor inclusion
• Historical research

Read Related Business & Conflict of Interest Disclosure →

Related Pages

Research Methodology →

Platforms We Analyze →

How We Verify AI Marketing Companies, Platforms & Services →

Research Definitions & Terminology →

Data & Research Limitations →

Editorial Standards →

Editorial Independence →

Human Review Policy →

Related Business & Conflict of Interest Disclosure →

How LLM Authority Index and CiteWorks Studio Contribute →

Corrections & Updates →

Affiliate Disclosure →

Related-company relationships do not influence the underlying AI recommendation data or ranking calculations.

Related business disclosure · Research methodology