Competition Law and Intelligent Reputation Ecosystems and Market Power

1. Introduction

An intelligent reputation ecosystem is a digital environment in which a platform collects, processes, scores, ranks, and distributes information about the reputation of businesses, sellers, products, professionals, creators, applications, or users.

Examples include:

• seller ratings and reviews on e-commerce platforms;
• hotel and restaurant ratings;
• app-store ratings;
• professional reputation scores;
• consumer-review platforms;
• marketplace seller-performance scores;
• trust and safety scores;
• algorithmic rankings;
• recommendation systems based on historical user behaviour;
• AI-generated reputation summaries;
• platform-generated “trusted seller” or “top-rated” labels.

Competition-law concerns arise when the operator of such an ecosystem possesses substantial market power and its reputation infrastructure becomes an essential competitive gateway. A platform may then influence which firms are visible, trusted, recommended, searched, or commercially successful.

The important competition question is therefore not simply whether a platform has a high market share. It is whether control over reputation data, ratings, ranking algorithms and consumer trust can be used to exclude or disadvantage competitors.

The EU's Google Shopping litigation demonstrates the broader principle particularly clearly: a dominant platform's control over ranking and visibility can become relevant under abuse-of-dominance law when the platform uses that control to favour its own services and disadvantage rivals. The Court of Justice confirmed the infringement in 2024.

2. Meaning of an Intelligent Reputation Ecosystem

A conventional reputation system merely records consumer opinions.

An intelligent reputation ecosystem goes considerably further.

It may:

1. collect reviews;
2. verify or authenticate reviews;
3. assign reputation scores;
4. identify allegedly fraudulent reviews;
5. rank sellers according to reputation;
6. determine search visibility;
7. recommend sellers to consumers;
8. determine access to advertising;
9. determine eligibility for premium programmes;
10. influence pricing or commissions;
11. allocate customers through algorithms;
12. generate AI summaries of reviews;
13. predict consumer trust;
14. use reputation data to determine commercial privileges.

Consequently, reputation becomes not merely information, but potentially a form of competitive infrastructure.

3. Competition-Law Structure

The issue can be analysed through five stages:

Stage 1 — Relevant market

Possible markets include:

• online marketplace services;
• search services;
• hotel-booking intermediation;
• app distribution;
• consumer-review services;
• professional reputation services;
• digital advertising;
• reputation-management software;
• data analytics;
• recommendation services.

The relevant market must be defined according to substitutability and competitive constraints.

Stage 2 — Market power

Indicators include:

• market share;
• network effects;
• scale of user data;
• access to consumer reviews;
• switching costs;
• multi-homing;
• technological advantages;
• control over ranking algorithms;
• control over reputation databases;
• dependence of business users;
• barriers to entry.

China's 2026 platform antitrust compliance guidance expressly identifies factors such as market share, ability to control the market, financial and technological conditions, dependence of platform operators, barriers to entry, and characteristics of the platform economy when assessing dominance.

Stage 3 — Reputation infrastructure

The authority should examine whether the platform controls a commercially important reputation mechanism.

For example:

Seller A has thousands of positive reviews but receives substantially less visibility because the platform's algorithm changes its reputation score.

The competitive significance of the algorithm may be much greater than the numerical rating itself.

Stage 4 — Conduct

Potentially problematic conduct includes:

• manipulation of reviews;
• discriminatory review verification;
• suppression of rival ratings;
• self-preferencing;
• preferential ranking;
• exclusion from recommendation systems;
• tying reputation scores to unrelated services;
• discriminatory access to reputation data;
• exploitative use of seller data;
• retaliation against sellers;
• algorithmic demotion;
• discriminatory “trusted seller” labels;
• manipulation of AI-generated reputation summaries.

Stage 5 — Effects

Possible competitive effects include:

• foreclosure of rivals;
• reduction of consumer choice;
• increased barriers to entry;
• weakening of competing platforms;
• raising competitors' costs;
• preventing multi-homing;
• degradation of innovation;
• exploitation of business users;
• reinforcement of network effects.

4. Reputation as a Source of Market Power

Reputation has several special economic characteristics.

A. Network effects

More consumers generate more reviews.

More reviews make the platform more useful.

Greater usefulness attracts more consumers.

More consumers attract more sellers.

More sellers generate additional data.

This produces a reinforcing cycle:

Users → Reviews → Reputation data → Better rankings → More users → More sellers → More data

A sufficiently powerful platform can therefore develop a reputation-data network effect.

5. Data Advantage

A dominant platform may possess:

• historical reviews;
• transaction data;
• cancellation data;
• return data;
• customer complaints;
• seller-response data;
• click-through data;
• conversion data;
• behavioural data;
• fraud indicators.

A new entrant may technically be able to establish a competing review system but lack the historical dataset necessary to replicate the incumbent's reputation infrastructure.

This creates an important competition question:

Can the incumbent use accumulated reputation data to make entry or expansion by rival platforms commercially impracticable?

6. Algorithmic Reputation and Market Power

Artificial intelligence can transform reputation systems.

A platform may calculate:

Reputation Score = reviews + transaction history + complaints + returns + response time + fraud probability + behavioural data

The algorithm may then determine:

• search position;
• recommendation probability;
• advertising eligibility;
• commission rates;
• consumer visibility;
• access to premium programmes.

Consequently, an apparently neutral algorithm can become a market-allocation mechanism.

The UK's competition authorities have specifically recognised that algorithmic systems can be used to manipulate rankings and favour a platform's own products. Google Shopping was identified as a major example.

7. Self-Preferencing

Self-preferencing occurs when a platform gives preferential treatment to its own products or services.

In a reputation ecosystem, examples include:

• giving the platform's own sellers better reputation visibility;
• ranking affiliated businesses above rivals;
• applying stricter review standards to competitors;
• using seller reputation data to improve the platform's own competing products;
• giving affiliated sellers preferred “trusted” status.

The Google Shopping judgment is particularly important because the Court of Justice dealt directly with a dominant platform favouring its own specialised service through ranking and display mechanisms.

8. Six Major Case Laws

1. Google Shopping — Google and Alphabet v European Commission

Case: Google and Alphabet v Commission, C-48/22 P, judgment of 10 September 2024.

Facts

Google operated a dominant general-search service while also operating a comparison-shopping service.

The European Commission found that Google gave its own shopping results favourable positioning and display while competing comparison-shopping services were subject to less favourable treatment.

Legal significance

The Court of Justice upheld the Commission's infringement finding and the €2.4 billion fine.

Relevance to reputation ecosystems

The case demonstrates that algorithmic visibility can constitute an important competitive parameter.

The analogy to reputation systems is strong:

Search ranking → consumer visibility
Reputation ranking → consumer trust and visibility

A dominant platform cannot necessarily treat its ranking infrastructure as competitively irrelevant merely because the underlying mechanism is algorithmic.

2. Amazon Marketplace

Facts

The European Commission investigated Amazon's dual role as:

1. marketplace operator; and
2. retailer competing with marketplace sellers.

Concerns included Amazon's use of non-public seller information and the operation of the Buy Box and Prime systems.

Amazon offered commitments concerning use of seller data, Buy Box selection and logistics.

Competition significance

The case illustrates the danger created when a platform:

• observes competitors;
• controls their visibility;
• controls marketplace reputation;
• and competes against them simultaneously.

Reputation relevance

Suppose Amazon—or another marketplace—uses seller-performance data to determine which products receive the most prominent placement.

The issue becomes:

Does control over seller reputation become a competitive advantage capable of reinforcing the platform's own downstream position?

This is closely related to the broader problem of platform self-preferencing.

3. Booking.com — Booking.com and Booking.com (Deutschland)

Case: C-264/23, judgment of 19 September 2024.

The case concerned price-parity clauses between Booking.com and hotels. The Court held that such clauses could not, in principle, be classified as ancillary restraints under EU competition law.

Relevance to reputation ecosystems

Booking platforms demonstrate how reputation, ranking and commercial intermediation can interact.

Hotels may depend upon:

• reviews;
• ratings;
• ranking position;
• visibility;
• customer traffic;
• platform commissions.

Therefore, contractual restrictions imposed by a powerful platform may become more significant where businesses are highly dependent upon the platform's reputation and visibility infrastructure.

The case is not itself a “reputation-score” case, but it is highly relevant to the platform-dependence component of reputation ecosystems.

4. Google Android — Competition Commission of India

Case: Umar Javeed & Others v Google LLC & Another, Case No. 39/2018, CCI order, 20 October 2022.

The CCI found Google dominant in several relevant markets associated with Android and Google's mobile ecosystem. It considered Google's contractual arrangements and the competitive advantages created by pre-installation and prominent placement.

Relevance

The case demonstrates the importance of:

• default status;
• pre-installation;
• prominent placement;
• ecosystem effects;
• network effects;
• status-quo bias.

These principles are applicable to intelligent reputation systems.

A seller may technically remain available on a platform, but if the algorithm places it below preferred competitors, formal access may not equal effective market access.

5. Epic Games v Google

The U.S. litigation concerning Google's Android app-distribution ecosystem provides another important example of ecosystem power.

The jury found Google liable for unlawfully monopolizing the market for Android app distribution. The DOJ subsequently identified the case as an important contemporary monopolization decision.

Reputation relevance

App stores combine:

• ratings;
• reviews;
• downloads;
• rankings;
• recommendation systems;
• search results;
• developer reputation.

Consequently, control over app-store distribution can affect not merely whether an application is available but whether consumers discover and trust it.

This makes reputation and distribution potentially complementary sources of market power.

6. United States v Google — General Search

The U.S. search monopolization litigation provides an additional example of the relationship between data, defaults, distribution and market power.

The U.S. District Court found Google liable for maintaining monopolies in general search and general-search text advertising through exclusionary distribution agreements. The subsequent remedy proceedings addressed access to search index and user-interaction data and other measures designed to facilitate rival competition.

Relevance to reputation ecosystems

Reputation ecosystems can exhibit the same structural characteristics:

More users → more behavioural data → better algorithm → better consumer experience → more users.

If a dominant platform restricts rivals' ability to obtain the data necessary to develop competing reputation systems, data access may become an important competitive issue.

9. Chinese Competition-Law Dimension

China is particularly relevant because its platform competition framework increasingly addresses algorithmic conduct.

China's 2026 SAMR materials identify platform practices involving:

• search rankings;
• user evaluations;
• algorithmic control;
• traffic restrictions;
• false rankings;
• false online evaluations;

as areas requiring regulatory attention.

SAMR also reports enforcement against practices involving manipulation of online reviews and automated engagement mechanisms.

This is highly relevant to intelligent reputation ecosystems because reputation manipulation can simultaneously raise competition-law, unfair-competition and consumer-protection concerns.

10. Abuse of Dominance Through Reputation Manipulation

A dominant platform could theoretically engage in several forms of abuse.

A. Discriminatory ranking

The platform ranks competing businesses differently without objective justification.

B. Self-preferencing

Its own affiliated businesses receive superior reputation visibility.

C. Selective review enforcement

Negative reviews concerning the platform's own businesses are removed more readily than negative reviews concerning competitors.

D. Artificial reputation inflation

The platform artificially improves the reputation of preferred businesses.

E. Reputation-based exclusion

Businesses falling below an algorithmic score are effectively denied access to customers.

F. Retaliatory demotion

A seller challenging the platform's commercial policies experiences unexplained deterioration in ranking.

G. Data leveraging

The platform uses competitors' reputation data to enter their downstream market.

11. False Reviews and Competition

False reviews present a distinctive problem.

There are two separate questions:

Consumer-protection question

Is the review false or misleading?

Competition-law question

Does the manipulation distort competitive conditions between undertakings?

For example:

Platform A operates a dominant marketplace and systematically promotes favourable reviews for its own affiliated sellers while suppressing comparable reviews for independent sellers.

The conduct may affect:

• consumer choice;
• competitor visibility;
• entry;
• seller reputation;
• market shares.

Thus, a conduct initially appearing to be a consumer-protection issue may acquire a competition-law dimension when conducted by a dominant undertaking.

12. Reputation as an Essential Competitive Input

An emerging theory is that reputation information can sometimes function as an essential competitive input.

This requires careful analysis.

Not every review database is an essential facility.

The stronger case arises where:

1. the platform has substantial market power;
2. the reputation database is difficult to replicate;
3. consumers rely heavily upon the platform;
4. sellers depend upon the reputation system;
5. alternative reputation mechanisms are ineffective;
6. access can realistically be provided;
7. exclusion materially harms competition.

This brings reputation ecosystems into dialogue with traditional essential-facility and refusal-to-deal principles.

13. Market Definition Problems

Reputation platforms create difficult market-definition questions.

One possible approach is:

Market A

Online marketplace services.

Market B

Consumer-review services.

Market C

Online reputation-management services.

Market D

Digital recommendation services.

Market E

Data-driven ranking services.

The correct market cannot be assumed simply because the platform calls its product a “review service.”

The analysis must examine actual consumer substitutability and competitive constraints.

14. Multi-Sided Market Effects

Reputation platforms are normally multi-sided.

They connect:

Consumers ↔ Platform ↔ Sellers

The platform may therefore have different relationships with each side.

Consumers provide:

• reviews;
• ratings;
• behavioural information.

Sellers provide:

• products;
• commissions;
• transaction data.

The platform supplies:

• discovery;
• reputation;
• ranking;
• recommendations;
• trust infrastructure.

Competition analysis must therefore examine the interaction between all sides rather than treating the platform as a conventional single-product undertaking.

15. Network Effects and Tipping

Reputation markets can be especially vulnerable to tipping.

Suppose Platform A has:

• 80 million users;
• 100 million reviews.

Platform B has:

• 5 million users;
• 3 million reviews.

Even if Platform B has a technically superior algorithm, it may struggle to compete because consumers perceive Platform A as the place containing the most reliable reputation information.

This creates a potential data-network-effect barrier.

16. AI-Generated Reputation

Generative AI introduces a new category of competition concern.

A platform may automatically create:

“AI summary: This seller is highly reliable and delivers quickly.”

The summary might be based on:

• verified reviews;
• unverified reviews;
• transaction history;
• complaints;
• returns;
• platform-generated behavioural scores.

Competition questions include:

1. Is the methodology transparent?
2. Are rival sellers treated equally?
3. Can sellers correct erroneous data?
4. Does the platform favour its own businesses?
5. Is the AI model trained on competitors' proprietary data?
6. Can competitors access comparable data?
7. Does the AI summary systematically affect ranking?

Thus, AI reputation generation can become an extension of algorithmic ranking power.

17. Foreclosure Theory

The central foreclosure theory can be represented as:

Dominant platform

↓

Controls reputation data

↓

Controls algorithmic ranking

↓

Controls consumer visibility

↓

Competitors receive less traffic

↓

Competitors generate less transaction data

↓

Their reputation becomes weaker

↓

Platform's reputation ecosystem becomes even stronger

This creates a potentially self-reinforcing reputation foreclosure loop.

18. Consumer-Welfare Effects

Potential adverse effects include:

Price

Reduced competition may eventually permit higher prices or commissions.

Quality

Competitors may have fewer incentives to innovate.

Choice

Consumers may see fewer alternatives.

Accuracy

Manipulated rankings can reduce the reliability of consumer information.

Privacy

Reputation systems may depend upon extensive behavioural data.

Innovation

New entrants may be unable to establish sufficient reputation to reach consumers.

19. Objective Justifications

Not every algorithmic ranking is unlawful.

A platform may legitimately argue that:

• fraud prevention requires sophisticated scoring;
• fake reviews must be removed;
• customer safety requires reputation screening;
• poor-performing sellers should be demoted;
• ranking improves consumer experience;
• recommendation algorithms increase relevance;
• reputation scores reward genuine quality.

Therefore, competition analysis should distinguish between:

legitimate quality control

and

strategic exclusion of competitors.

The existence of an algorithm or AI system is not itself evidence of anticompetitive conduct.

20. Possible Competition Remedies

Authorities may consider:

1. Non-discrimination

Require comparable sellers to receive comparable treatment.

2. Ranking transparency

Require disclosure of material ranking factors.

3. Algorithmic auditing

Independent auditing of ranking and reputation systems.

4. Data-access remedies

Permit appropriate access to reputation data.

5. Data portability

Allow businesses to transfer relevant reputation information.

6. Interoperability

Permit reputation information to function across platforms where legally and technically appropriate.

7. Separation

In particularly serious cases, structural separation between marketplace operation and competing retail activity may be considered.

8. Prohibition of retaliation

Prevent platforms from lowering reputation or visibility because businesses challenge platform policies.

21. Key Doctrinal Principles

22. Six-Case-Law Synthesis

The principal lessons from the cases are:

1. Google Shopping — ranking and visibility can become competition-law relevant when controlled by a dominant platform. 
2. Amazon Marketplace — a platform acting simultaneously as intermediary and competitor creates risks involving data and preferential treatment. 
3. Booking.com — contractual restrictions can reinforce platform dependence and affect competitive conditions. 
4. Google Android/CCI — defaults, pre-installation, ecosystem effects and status-quo bias can reinforce market power. 
5. Epic Games v Google — control of a digital ecosystem can materially affect competitors' access to consumers. 
6. U.S. v Google Search — exclusionary control over distribution and data can reinforce a digital platform's market power. 

Importantly, not all six cases are direct reputation-score cases. Their value lies in the competition-law principles they establish concerning ranking, visibility, data, platform dependence, ecosystem power and foreclosure—the same mechanisms through which an intelligent reputation ecosystem can acquire or reinforce market power.

23. Conclusion

Intelligent reputation ecosystems can transform reputation from a passive consumer-information mechanism into a competitive infrastructure.

Where a powerful platform controls:

reviews + reputation data + AI scoring + rankings + recommendations + consumer access,

it may acquire a significant ability to influence the competitive position of businesses operating on that platform.

The central competition-law inquiry should therefore be:

Is the platform using objectively justified reputation mechanisms to improve competition and consumer choice, or is it using control over reputation, data and algorithmic visibility to reinforce market power and foreclose competitors?

Modern competition law increasingly examines precisely these forms of algorithmic ranking, self-preferencing, platform dependence, data advantage and ecosystem foreclosure. China's recent platform guidance and enforcement materials are particularly explicit about the competition significance of search rankings, user evaluations, algorithms, traffic controls and false rankings.