Competition Law and Attention Hierarchy Competition Concerns

1. Introduction

Attention hierarchy refers to the way a digital platform determines which information, products, services, advertisements, sellers, applications, creators, or recommendations receive greater visibility than others.

In modern digital markets, visibility itself can be a competitive resource.

A platform may determine:

which seller appears first;

which search result receives prominence;

which application is recommended;

which news item appears in a feed;

which restaurant is ranked higher;

which AI-generated answer is displayed;

which advertisement receives attention;

which content is recommended automatically.

Consequently, competition may no longer depend only upon price and product quality. It may depend upon control over the user's limited attention.

The competition-law question is therefore:

Can a platform with substantial market power use control over attention allocation to advantage itself or exclude competing businesses?

This issue connects closely with self-preferencing, search neutrality, ranking discrimination, platform bias, tying, leveraging, exclusionary conduct and digital-market gatekeeping.

2. Meaning of Attention Hierarchy

An attention hierarchy exists where a platform places different participants at different levels of visibility.

For example:

Position 1 → Position 2 → Position 3 → Page 2 → hidden results

Although all competitors may technically have access to the platform, their actual commercial opportunities can be radically different.

A platform can create this hierarchy through:

search ranking;

recommendation algorithms;

default settings;

featured listings;

advertisements;

personalised feeds;

AI-generated summaries;

app-store rankings;

marketplace placement.

3. Attention as a Scarce Economic Resource

Human attention is limited.

Users generally do not inspect every available:

product;

search result;

application;

advertisement;

seller;

article.

Instead, they concentrate attention on a relatively small number of results.

This gives the platform controlling the ranking system significant economic influence.

A change from:

first position → tenth position

may substantially reduce a business's ability to attract customers even when the underlying product remains unchanged.

4. Attention Hierarchy and Market Power

Control over attention does not automatically establish dominance.

The competition-law analysis must consider:

relevant market;

market share;

alternatives;

switching costs;

network effects;

multi-homing;

barriers to entry;

data advantages;

user behaviour.

However, where a platform is a major gateway between businesses and consumers, control over ranking and visibility can reinforce existing market power.

5. Search Engines

Search engines provide one of the clearest examples.

A search engine determines:

which results users see first.

Suppose a dominant search platform systematically places its own specialised service above rival services.

Possible concerns include:

self-preferencing;

discriminatory ranking;

leveraging;

foreclosure.

This issue was central to the European Commission's Google Shopping case.

6. Online Marketplaces

Online marketplaces can similarly determine which sellers receive attention.

Ranking may depend upon:

price;

seller reputation;

delivery speed;

commissions;

advertising payments;

platform preferences.

A platform could potentially manipulate ranking to favour its own products or affiliated sellers.

7. App Stores

App stores control another attention hierarchy.

They decide:

featured applications;

search rankings;

recommendations;

default applications;

editorial placement.

Where the platform operator competes with application developers, preferential ranking could potentially disadvantage competing developers.

This makes app-store ranking closely connected to self-preferencing and platform neutrality.

8. Social-Media Feeds

Social networks allocate attention through recommendation algorithms.

The algorithm may determine:

which creator appears;

which advertisement is displayed;

which post receives additional distribution.

Competition concerns could emerge if a dominant platform uses recommendation control to disadvantage competing services or competing content businesses.

However, not every editorial or algorithmic choice is a competition-law violation.

The central issue is whether the conduct involves an undertaking exercising market power in a manner capable of harming competition.

9. AI-Generated Attention Hierarchy

Artificial intelligence creates a newer dimension.

Instead of showing ten links, an AI assistant may provide one answer.

Traditional search:

Result 1
Result 2
Result 3
Result 4

AI-mediated search:

One recommended answer

This gives the AI intermediary potentially greater influence over which businesses receive consumer attention.

A dominant AI platform could potentially:

favour its own services;

exclude competing providers from recommendations;

manipulate citations or referrals;

privilege affiliated businesses;

restrict access to AI-mediated distribution.

This could create a particularly strong form of attention control.

10. Self-Preferencing

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

For example:

Platform operates a marketplace + sells its own products.

It could rank its own products more prominently than competing products.

The competition-law issue is whether such conduct by a dominant platform has exclusionary effects.

The Google Shopping litigation is particularly important here.

11. Google Shopping

Case T-612/17, Google and Alphabet v Commission

The case concerned Google's treatment of its own comparison-shopping service within general search results.

The case is highly relevant to attention hierarchy because search ranking determines the visibility of competing services.

Competition significance

A dominant search engine can potentially influence competition not merely through prices but through:

ranking;

prominence;

traffic allocation;

visibility.

The case illustrates why control over attention can become a competition parameter.

12. Google Android

Case T-604/18, Google LLC and Alphabet Inc. v Commission

The case concerned various practices involving the Android ecosystem.

Relevance

A dominant platform may control several interconnected layers and use contractual or technical arrangements to influence how users access competing services.

Attention hierarchy can therefore operate together with:

defaults;

distribution restrictions;

tying;

ecosystem effects.

13. Microsoft

Case T-201/04, Microsoft Corp v Commission

Microsoft concerned interoperability and tying involving the Windows operating-system ecosystem.

Relevance to attention hierarchy

An operating system can influence which applications users encounter and how easily competing products can reach users.

This provides a broader lesson:

Control over a technological gateway can confer competitive advantages in downstream markets.

Modern platforms may exercise similar control through ranking and recommendation rather than operating-system defaults.

14. United Brands

Case 27/76, United Brands Company and United Brands Continentaal BV v Commission

The Court considered the concept of dominance and the ability of an undertaking to act independently of competitive constraints.

Relevance

Attention control may contribute to market power when a platform becomes an unavoidable or highly significant gateway between businesses and consumers.

However, attention control must be assessed together with other evidence of market power.

15. Hoffmann-La Roche

Case 85/76, Hoffmann-La Roche & Co AG v Commission

The Court developed the foundational concept of a dominant position.

Relevance

The case supports the broader principle that dominance concerns economic power that allows an undertaking to behave independently of effective competitive constraints.

For digital platforms, such power can potentially arise from:

network effects;

data;

user lock-in;

control over distribution;

control over attention.

16. Intel

Case C-413/14 P, Intel Corporation Inc. v Commission

The Court of Justice examined the assessment of exclusionary effects associated with rebates by dominant undertakings.

Relevance

Attention hierarchy can involve economic incentives as well as ranking.

For example, a platform could offer:

preferential placement;

discounted advertising;

visibility incentives.

Intel demonstrates the importance of examining the actual or potential exclusionary effects of conduct in the circumstances covered by the case.

17. Bronner

Case C-7/97, Oscar Bronner GmbH & Co. KG v Mediaprint

The case concerned access to a distribution system controlled by another undertaking.

Relevance

Digital attention platforms can resemble distribution infrastructure.

Search engines, marketplaces and app stores provide access to consumers.

However, Bronner also demonstrates that competition law does not automatically require a dominant undertaking to provide competitors with access to every facility it controls.

18. IMS Health

Case C-418/01, IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG

The case concerned access to protected information structures and the relationship between intellectual property and competition law.

Relevance

Where a platform's ranking system, database or information infrastructure becomes commercially indispensable, competition law may have to balance:

proprietary rights;

investment incentives;

competitive access.

19. Servizio Elettrico Nazionale

Case C-377/20, Servizio Elettrico Nazionale SpA v AGCM

The Court examined exclusionary abuse and competition on the merits.

Relevance

The case is useful for distinguishing legitimate competitive advantages from conduct involving the use of dominance to exclude competitors.

A platform's superior ranking algorithm is not inherently unlawful.

The concern arises where the ranking mechanism is used as an instrument of exclusionary market power.

20. Attention Hierarchy and Foreclosure

Foreclosure may occur where competitors technically remain on a platform but are deprived of meaningful access to consumers.

This can happen through:

demotion;

removal from recommendations;

reduced search visibility;

exclusion from default positions;

lower AI recommendations;

reduced distribution.

Thus:

formal access ≠ effective access.

This is a particularly important issue in digital competition.

21. Ranking Discrimination

Suppose Platform A ranks competing businesses according to one algorithm but uses a more favourable algorithm for its own affiliated business.

Potential questions include:

Is the platform dominant?

Does it compete downstream?

Are competitors treated differently?

Is the difference objectively justified?

Does the ranking significantly affect consumer traffic?

Is the conduct capable of foreclosing equally efficient competitors?

22. Paid Placement

Advertising can complicate the analysis.

A platform may legitimately sell:

sponsored search;

promoted listings;

featured products.

Paid placement is not automatically anticompetitive.

However, concerns could arise if the platform:

conceals the commercial nature of placement;

systematically disadvantages non-paying competitors;

requires competitors to purchase visibility;

uses advertising to exclude rival services.

23. Default Positions

Defaults can have powerful attention effects.

Examples include:

default search engine;

default browser;

default payment method;

default AI assistant;

default marketplace;

default recommendation service.

Users often do not change defaults.

Therefore, default arrangements can potentially reinforce network effects and make entry more difficult.

24. Tying and Attention

Tying can influence attention.

Suppose a dominant operating system requires its AI assistant to be installed and prominently displayed.

If competing AI assistants are technically available but difficult to access, the platform may effectively control the user's attention.

Competition authorities would need to determine whether the arrangement satisfies the legal elements of tying or another theory of abuse.

25. Data and Attention

Attention hierarchy can also generate data advantages.

A platform that receives more user attention obtains:

more clicks;

more interactions;

more behavioural information;

more feedback.

This produces a reinforcing cycle:

more visibility → more users → more data → better recommendations → more visibility.

Such feedback loops can contribute to durable market power.

26. Network Effects

Attention markets are particularly susceptible to indirect network effects.

For example:

More users → more sellers

More sellers → greater consumer choice

Greater consumer choice → more users

The platform then becomes more attractive to advertisers and developers.

This can make entry increasingly difficult.

27. Multi-Homing

Multi-homing can reduce attention-based market power.

If sellers can easily appear on:

Google;

Amazon;

specialised marketplaces;

social networks;

then a single platform may have less control.

However, if consumers strongly concentrate their attention on one platform, businesses may have little practical choice but to participate.

28. Switching Costs

Switching costs can reinforce attention hierarchy.

Users may remain on a platform because they have accumulated:

search history;

preferences;

social connections;

reviews;

recommendations;

subscriptions;

loyalty benefits.

The greater the switching cost, the more durable platform power may become.

29. Algorithmic Transparency

Competition law does not necessarily require platforms to disclose their complete algorithms.

Full disclosure could create:

security risks;

gaming;

intellectual-property problems.

Nevertheless, competition authorities may need sufficient access to determine whether ranking practices are discriminatory or exclusionary.

Potential investigative evidence includes:

ranking experiments;

internal documents;

algorithmic changes;

traffic data;

A/B testing;

developer complaints.

30. Objective Justifications

A platform may legitimately rank one result above another because of:

relevance;

quality;

safety;

reliability;

fraud prevention;

delivery performance;

user preferences.

Therefore, ranking differentiation is not automatically discriminatory.

Competition analysis must distinguish legitimate product design from exclusionary manipulation.

31. Attention Hierarchy in App Stores

App stores present several potential competition concerns.

A dominant app store can determine:

featured apps;

search rankings;

recommendations;

category placement;

default applications.

If the platform also owns competing applications, self-preferencing becomes a potential concern.

The issue can be particularly significant where developers cannot easily reach consumers outside the platform.

32. Attention Hierarchy in E-Commerce

E-commerce platforms can rank:

private-label products;

third-party sellers;

sponsored listings.

Potential concerns include:

self-preferencing;

discriminatory ranking;

manipulation of seller visibility;

leveraging marketplace power into private-label markets.

The platform's dual role as:

marketplace operator + competitor

can make attention allocation particularly important.

33. Attention Hierarchy and News

News platforms may determine which publishers receive visibility.

A dominant intermediary could potentially influence:

referral traffic;

advertising revenue;

audience access.

However, competition analysis must distinguish market-power issues from editorial decisions and legitimate content moderation.

34. Attention Hierarchy and Advertising

Digital advertising is itself an attention market.

Advertisers compete for:

impressions;

clicks;

engagement;

conversions.

A platform controlling both:

advertising infrastructure

and

consumer attention

may possess significant vertical advantages.

Potential concerns could include:

discriminatory ad auctions;

self-preferencing;

exclusionary access rules;

tying advertising services to other platform services.

35. Artificial Intelligence and Attention Allocation

AI introduces a major transformation.

Traditional platforms:

rank multiple options.

AI assistants:

select or synthesise an answer.

The intermediary therefore moves from being a ranking system to being a decision-making intermediary.

This may increase the economic importance of AI-generated recommendations.

For example:

"Which hotel should I book?"

An AI assistant might recommend only one or two hotels.

The ranking hierarchy becomes much more concentrated.

36. Potential Competition Concerns in AI Recommendations

A dominant AI platform could theoretically:

favour its own products;

exclude competitors from AI answers;

manipulate recommendations;

give affiliated businesses greater visibility;

impose fees for recommendation;

restrict access to recommendation infrastructure.

Such conduct could potentially implicate existing competition-law doctrines, depending upon the market and circumstances.

37. Competition Remedies

Possible remedies may include:

Non-discrimination

Comparable businesses receive comparable ranking opportunities.

Transparency

Platforms disclose relevant commercial ranking factors.

Interoperability

Competitors receive appropriate technical access.

Data portability

Users can transfer relevant data.

Restrictions on self-preferencing

Dominant platforms may face limits on preferential treatment.

Monitoring

Independent monitoring may assess whether ranking practices remain compliant.

38. Challenges in Proving Harm

Attention-based competition cases can be difficult because authorities must demonstrate:

actual or potential foreclosure;

causal connection between ranking and competitive harm;

relevance of the platform;

significance of visibility;

counterfactual ranking;

consumer impact.

A competitor becoming less visible does not automatically mean competition has been harmed.

39. Counterfactual Analysis

Authorities may ask:

What would have happened if the platform had ranked competitors neutrally?

Possible evidence includes:

historical rankings;

controlled experiments;

traffic changes;

conversion rates;

internal platform documents;

user behaviour.

The counterfactual can help establish whether ranking conduct materially affected competition.

40. Economic Significance of the First Position

The commercial value of ranking positions is usually unequal.

For example:

Position 1 > Position 2 > Position 3 > Page 2

This means that even apparently small changes in ranking can have substantial economic consequences.

The competition analysis should therefore examine actual traffic and conversion effects rather than simply asking whether competitors remained technically listed.

41. Key Legal Questions

An investigation into attention hierarchy should consider:

Who controls the platform?

Is the platform dominant?

What is the relevant market?

Does the platform compete with ranked businesses?

How is ranking determined?

Are own products treated differently?

Does the ranking significantly influence consumer behaviour?

Can competitors effectively reach consumers elsewhere?

Are users able to multi-home?

Are there objective justifications?

Does the conduct foreclose competitors?

Does it reduce innovation or consumer choice?

42. Conclusion

Attention has become an important competitive resource in digital markets. Search engines, marketplaces, app stores, social networks and AI assistants do not merely provide access to products and information; they increasingly determine which products and businesses users actually notice.

The principal competition concerns include:

self-preferencing;

ranking discrimination;

search manipulation;

preferential defaults;

recommendation bias;

exclusive visibility arrangements;

tying and bundling;

foreclosure of rival businesses;

AI-mediated recommendation bias;

control over advertising visibility.

The leading cases of Google Shopping, Google Android, Microsoft, United Brands, Hoffmann-La Roche, Intel, Bronner, IMS Health and Servizio Elettrico Nazionale provide useful legal principles for examining these issues.

The central conceptual shift is that competition in digital markets may concern not merely access to consumers, but access to consumer attention. A platform can therefore possess substantial competitive power through its ability to determine what users see, what they are recommended, and what they are unlikely to see. Competition law must distinguish legitimate ranking based on quality, relevance and user preferences from the strategic use of a dominant platform's attention-allocation mechanism to disadvantage competing undertakings.