Competition Law and Attention Allocation Systems

1. Introduction

Attention allocation systems are digital mechanisms that determine which products, services, advertisements, sellers, creators, applications, search results, news items, or other content receive users' attention, and in what order or prominence.

They include:

search-ranking algorithms;

social-media recommendation systems;

app-store rankings;

e-commerce product rankings;

online advertising auctions;

video recommendations;

news-feed algorithms;

travel and hotel rankings;

AI assistants and recommendation engines;

marketplace featured-placement systems.

Attention is economically valuable because businesses compete not only for customers but also for visibility before customers.

Consequently, a platform controlling a large share of user attention can become an important competitive bottleneck.

The central competition-law question is:

When does control over the allocation of user attention become a means of restricting competition rather than merely an ordinary product-design decision?

2. Economic Meaning of Attention Allocation

Traditional markets often assume competition occurs through:

Price → consumer choice → purchase.

Digital markets frequently operate differently:

Algorithm → visibility → attention → consumer choice → purchase.

If a platform controls the algorithm determining visibility, it may influence which competing firms consumers discover.

For example:

1,000 competing sellers
↓
Marketplace ranking algorithm
↓
Top 10 results
↓
Majority of consumer attention

The algorithm therefore acts as an attention gatekeeper.

3. Attention as an Economic Resource

Attention has several characteristics of an economically valuable resource:

it is scarce;

users have limited time;

visibility affects demand;

ranking affects discovery;

prominent placement can increase sales;

advertisers compete for attention;

platforms monetize attention through advertising or transactions.

A platform can therefore possess substantial economic power even when consumers pay zero monetary price for the service.

4. Attention Allocation Systems

An attention allocation system can be based upon:

Search ranking

Determines which websites or products appear first.

Recommendation systems

Determine what users see next.

Marketplace ranking

Determines which sellers receive visibility.

Advertising auctions

Determine which advertisements receive impressions.

App-store rankings

Influence application discovery.

Social-media feeds

Determine which content receives prominence.

AI assistants

Determine which products, websites or services are recommended in response to user queries.

5. Why Competition Law Is Concerned

Attention allocation becomes competition-sensitive where the platform itself competes with the businesses whose visibility it controls.

For example:

Platform controls ranking

and

Platform owns competing product

This creates a potential conflict of incentives.

The platform may have the ability to:

favour its own products;

demote rivals;

manipulate search visibility;

condition prominence on using its other services;

impose discriminatory access conditions;

exploit competitor data.

Such conduct can potentially fall within abuse-of-dominance rules where the relevant legal requirements are satisfied.

6. Relevant Competition-Law Theories

Attention allocation systems can raise several theories.

1. Self-preferencing

A platform favours its own products or services.

2. Discrimination

Competing businesses receive unequal visibility or ranking.

3. Foreclosure

Rivals cannot obtain sufficient customer attention to compete effectively.

4. Tying

High visibility is conditioned upon purchasing another service.

5. Leveraging

Market power in one market is used to gain or protect power in another.

6. Exclusionary ranking

Algorithmic decisions systematically disadvantage rivals.

7. Exploitation of data

Platform data concerning competitors is used to improve the platform's own competing services.

7. Article 102 TFEU

Under EU competition law, Article 102 TFEU prohibits abuse of a dominant position.

Attention allocation can potentially fall within Article 102 where a dominant platform uses control over visibility to:

disadvantage competitors;

discriminate;

foreclose rivals;

favour its own services;

impose exclusionary conditions.

However, algorithmic ranking is not automatically an abuse.

The assessment depends on:

market power;

conduct;

effects;

competitive process;

objective justification;

efficiencies;

proportionality where relevant.

8. Indian Competition-Law Perspective

In India, attention allocation systems can potentially engage Section 4 of the Competition Act, 2002, particularly where a dominant digital enterprise uses its position to exclude or disadvantage competitors.

Potential issues include:

discriminatory ranking;

self-preferencing;

denial of market access;

leveraging;

tying;

unfair conditions;

exploitation of platform data.

Section 3 may also become relevant where agreements or arrangements involving platforms and business users restrict competition.

9. Case Law

1. Google Search (Shopping) v European Commission

Case T-612/17, General Court

This is one of the most important authorities concerning attention allocation.

The European Commission found that Google had favoured its own comparison-shopping service in search results and had demoted competing comparison-shopping services.

The General Court largely upheld the Commission's decision.

Attention-allocation significance

Search ranking determines which businesses receive consumer attention.

The case demonstrates that competition analysis may extend beyond:

price → output

to:

ranking → visibility → traffic → customers.

The search-results page can therefore function as an important competitive gateway.

10. Google Android

Case T-604/18

The Google Android litigation concerned Google's practices relating to Android, search and mobile applications.

Among the issues considered were contractual arrangements affecting access to Google's ecosystem and the competitive position of rival services.

Attention-allocation significance

Mobile ecosystems determine:

which search services users encounter;

which applications are pre-installed;

which services obtain default status;

how consumers access digital services.

Defaults and ecosystem design can therefore significantly influence the allocation of user attention.

11. Google Android Auto

Case C-233/23, Alphabet and Others v Autorità Garante della Concorrenza e del Mercato

The case concerned Google's refusal to make Android Auto compatible with a third-party application.

The CJEU considered the circumstances under which refusal of interoperability may constitute abusive conduct.

Attention-allocation significance

Interoperability determines whether competing services can appear within an established digital ecosystem.

If a dominant platform controls the interface through which users discover or interact with applications, technical access can become an important attention-allocation mechanism.

12. Microsoft Corp. v Commission

Case T-201/04

Microsoft was found to have abused its dominant position through conduct concerning interoperability information and the integration of its media-player technology.

Relevance

Although not an attention-ranking case, the decision is important because it demonstrates how control over a technological ecosystem can affect the ability of competitors to reach users effectively.

Digital attention depends not merely on ranking but also on:

compatibility;

interoperability;

default settings;

technical access.

13. Bronner v Mediaprint

Case C-7/97

The case concerned access to a newspaper home-delivery system.

The CJEU established stringent conditions relevant to refusal-to-supply claims.

Attention-allocation relevance

Newspapers historically competed for consumer attention through distribution networks.

A dominant distribution system can therefore act as an intermediary between:

content producer → consumer attention.

Modern recommendation systems perform a similar intermediary function digitally.

The case is relevant to determining when control over a distribution mechanism should give rise to competition-law obligations.

14. United Brands v Commission

Case 27/76

United Brands concerned dominance and abusive conduct in the banana market.

The case remains important for broader principles concerning the conduct of dominant firms toward customers and trading partners.

Attention-system relevance

The case provides foundational guidance on how a dominant undertaking's commercial conduct can affect downstream competitive conditions.

Its principles can be adapted cautiously to digital systems where a dominant platform controls commercially important access channels.

15. Intel Corp. v Commission

Case C-413/14 P and subsequent proceedings

The Intel litigation concerned rebates and exclusionary effects in the market for computer processors.

The CJEU emphasized the importance of assessing the actual or potential exclusionary effects of the conduct in circumstances where appropriate.

Attention-allocation relevance

The principle is useful for digital-ranking cases because a competition authority may need to demonstrate how the challenged mechanism affects competitors.

For example:

ranking disadvantage → reduced visibility → reduced traffic → reduced sales → foreclosure

rather than simply asserting that a competitor received a lower ranking.

16. AstraZeneca v Commission

Case C-457/10 P

The case concerned exclusionary conduct involving regulatory procedures and intellectual-property mechanisms.

Relevance

AstraZeneca demonstrates that competition law can address conduct that indirectly restricts rivals' ability to compete.

Attention allocation similarly may involve indirect exclusion:

ranking mechanism → reduced discovery → reduced customer acquisition.

17. Self-Preferencing

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

Example:

Marketplace

→ Platform's own product: Position 1

→ Rival A: Position 15

→ Rival B: Position 20

If the platform is dominant and the ranking mechanism is capable of substantially affecting competitive conditions, competition authorities may investigate whether the conduct constitutes unlawful exclusion.

The crucial issue is not simply that the platform's product ranks first, but why it ranks first and what competitive consequences follow.

18. Algorithmic Discrimination

Attention allocation systems can discriminate through:

ranking;

recommendation;

pricing;

search visibility;

advertising access;

platform commissions.

For example, two sellers offering comparable products could receive substantially different visibility.

The competition analysis may ask:

Is the platform dominant?

Are the sellers similarly situated?

Is the difference objectively justified?

Does the discrimination disadvantage competitors?

Does it affect effective competition?

19. Ranking Bias and Foreclosure

Ranking bias becomes particularly significant where consumers rarely move beyond the first page or first few recommendations.

Suppose:

Top 5 results = 80% of user clicks.

A platform that systematically moves a rival from position 2 to position 50 may substantially affect that rival's ability to compete.

Thus, the relevant competitive variable is not merely:

“Was the competitor listed?”

but:

“Was the competitor sufficiently visible to compete effectively?”

20. Attention and Consumer Choice

Consumers generally have limited time.

Therefore, platforms can influence competition through:

default settings;

rankings;

recommendation;

personalization;

notifications;

featured placements;

autocomplete;

AI-generated answers.

These mechanisms may influence the choice architecture within which consumers make decisions.

21. AI and Attention Allocation

Generative AI creates a new attention-allocation problem.

Traditional search:

User → search results → multiple websites.

Generative AI:

User → AI answer → potentially one or a few recommendations.

This can dramatically change competitive visibility.

If an AI assistant recommends:

“Use Service X”

instead of presenting numerous alternatives, the AI provider may become an important attention gatekeeper.

Potential concerns include:

preferential recommendations;

suppression of competitors;

exclusive data access;

self-preferencing;

biased training;

discriminatory API access.

22. AI Assistants as Competitive Gatekeepers

AI systems can control:

product discovery;

information discovery;

service recommendations;

purchasing decisions;

travel planning;

software selection.

The economic structure can therefore become:

Competitors → AI intermediary → consumer

instead of:

Competitors → consumer directly.

This gives the AI intermediary potentially significant influence over attention allocation.

23. Advertising Auctions

Attention allocation is particularly important in digital advertising.

Platforms may determine:

which advertiser receives an impression;

position of advertisements;

auction winners;

targeting;

frequency;

recommendation.

Competition concerns may arise where a platform:

favours its own advertising service;

discriminates against rival ad exchanges;

uses advertiser data to compete against advertisers;

restricts access to advertising inventory.

The Google advertising-related cases provide an important broader competition-law context for such concerns.

24. Marketplace Rankings

E-commerce platforms frequently rank products using:

price;

reviews;

conversion rates;

seller reputation;

advertising payments;

inventory;

delivery speed.

The competitive concern becomes stronger where the ranking system is opaque and the platform competes with the sellers it ranks.

For example:

Seller pays platform → higher visibility.

If commercial payment determines ranking without appropriate disclosure or competitive safeguards, questions may arise regarding discrimination and market access.

25. Attention Allocation and Network Effects

Attention systems can create powerful feedback loops:

Visibility → sales → reviews → higher ranking → more visibility.

This creates a potential self-reinforcing cycle.

A new entrant may have difficulty obtaining sufficient attention to generate the data and sales needed to improve its ranking.

This can produce an attention-based entry barrier.

26. Data Feedback Loops

The cycle can become:

More users

↓

More behavioural data

↓

Better recommendation algorithm

↓

Better targeting

↓

More users

This can strengthen incumbency.

Competition analysis may therefore examine whether competitors can realistically reproduce the same data advantage.

27. Attention Bottlenecks

An attention bottleneck exists where one platform controls a disproportionately important gateway to consumer attention.

Examples include:

dominant search engines;

major app stores;

large marketplaces;

social networks;

digital advertising exchanges;

AI assistants.

The bottleneck becomes competition-sensitive when control over attention is used to disadvantage rivals.

28. Objective Justification

Ranking differences can have legitimate reasons.

For example:

product relevance;

quality;

delivery reliability;

consumer preferences;

security;

fraud prevention;

technical performance.

Therefore, every ranking difference cannot be treated as discriminatory.

A platform should be able to demonstrate legitimate and consistently applied criteria.

29. Transparency

Transparency can help businesses understand:

why they are ranked;

what factors affect visibility;

how advertising affects placement;

how recommendations are generated.

But complete disclosure of algorithms may create:

manipulation risks;

gaming;

security problems;

intellectual-property concerns.

Competition law therefore does not necessarily require full disclosure of source code.

30. Competition Between Platforms

Attention allocation also affects competition between platforms.

A dominant platform may have incentives to:

prevent multi-homing;

restrict interoperability;

impose exclusivity;

make switching difficult;

control user data portability.

These practices can reduce contestability even where the platform does not directly discriminate against individual sellers.

31. Attention Allocation and Tying

Suppose a dominant operating system makes its own AI assistant the default and restricts alternative assistants.

Potential competition concerns can involve:

tying;

default bias;

exclusion;

interoperability;

leveraging.

The relevant analysis would depend upon the applicable jurisdiction and evidence concerning competitive effects.

32. Attention Allocation and Exclusive Dealing

A platform may contractually require businesses to:

use its advertising system;

use its payment system;

avoid competing platforms;

purchase preferred placement;

use its analytics tools.

Such contractual architecture may restrict the ability of rivals to compete for attention.

33. Attention Allocation and Market Definition

Several relevant markets may coexist.

For example:

Market 1

Search services.

Market 2

Online advertising.

Market 3

Comparison-shopping services.

Market 4

Marketplace services.

Market 5

AI recommendation services.

The same attention-allocation mechanism may affect multiple markets.

Competition authorities therefore need to distinguish:

where the platform is dominant

from

where the competitive harm occurs.

34. Evidence in Attention-Allocation Cases

Relevant evidence may include:

ranking algorithms;

A/B testing;

search logs;

click-through rates;

recommendation data;

internal strategy documents;

changes in traffic;

consumer behaviour;

competitor complaints;

ranking criteria;

advertising records;

platform communications.

The strongest analysis normally connects the algorithmic mechanism to measurable competitive effects.

35. Remedies

Possible competition-law remedies include:

Non-discrimination

Require equivalent treatment of similarly situated competitors.

Ranking neutrality

Prevent unjustified preferential treatment.

Transparency

Require explanation of major ranking factors.

Data access

Permit appropriate access to relevant platform data.

Interoperability

Enable competing services to interact with the platform.

Choice screens

Give users meaningful choices among competing services.

Structural remedies

In exceptional circumstances, separation of platform and downstream operations may be considered.

36. Indian Competition-Law Application

Under Indian law, attention-allocation systems can potentially be examined under Section 4 of the Competition Act, 2002 when a dominant enterprise uses its position to:

discriminate;

deny market access;

leverage dominance;

impose unfair conditions;

favour its own services;

restrict competing enterprises.

The Competition Commission of India has increasingly had to consider the economics of digital platforms, where ranking and visibility are important competitive variables.

37. Important Distinction: Algorithmic Error vs Anticompetitive Conduct

An algorithm may produce an unfair or inaccurate ranking without violating competition law.

For example:

faulty data → incorrect ranking.

That may raise consumer-protection, data-governance or platform-governance issues.

Competition law requires a separate inquiry into:

market power;

conduct;

competitive effects;

exclusion;

legal classification.

38. Case-Law Summary

39. Key Legal Principles

Principle 1

Control over user attention can constitute an economically significant form of market power.

Principle 2

Algorithmic ranking is not inherently anticompetitive.

Principle 3

Self-preferencing becomes competition-sensitive particularly where a dominant platform controls a critical route to consumers.

Principle 4

Interoperability can determine whether competing services can obtain meaningful user access.

Principle 5

Ranking discrimination requires examination of both differential treatment and competitive effects.

Principle 6

A ranking system should not be assessed solely by its algorithmic design; its market position and actual operation matter.

Principle 7

AI recommendation systems may create a new form of attention bottleneck because they can replace lists of competing options with a small number of machine-generated recommendations.

40. Conclusion

Attention allocation systems are becoming an important frontier of competition law.

In traditional markets, competition is largely expressed through price, quality and output. In digital markets, competition increasingly depends upon who gets seen, recommended, ranked, displayed, or presented to consumers.

The basic competitive chain is:

Algorithmic ranking → visibility → attention → traffic → sales → market share.

When a dominant platform controls this chain, its ranking and recommendation architecture can become a significant competitive gateway.

The most important legal issues therefore include:

self-preferencing;

discriminatory ranking;

foreclosure;

interoperability;

default bias;

tying;

data advantages;

advertising allocation;

marketplace visibility;

AI recommendations.

The authorities in Google Shopping, Google Android, Google Android Auto, Microsoft, Bronner, United Brands, Intel, and AstraZeneca demonstrate different aspects of the broader principle: competition law can examine not merely the price charged to consumers, but also the mechanisms through which dominant undertakings control access to customers and competitive opportunities.

At the same time, effective competition analysis requires care. A platform's ranking system may legitimately prioritize relevance, quality, safety or consumer preferences. The critical issue is whether the attention-allocation mechanism, in the circumstances of the particular market, is being used in a manner that satisfies the applicable legal test for anticompetitive conduct.