Algorithmic Influence Scoring and Ranking Control

Introduction

Algorithmic influence scoring and ranking control refers to the use of automated systems to assign scores, rankings, visibility levels, priority positions, recommendations, or access probabilities to products, sellers, users, advertisers, content, applications, or services.

In competition law, the central concern is not merely that an algorithm ranks alternatives. Ranking is normally a legitimate function of a digital platform. The competition issue arises where a firm with substantial market power controls the ranking mechanism and uses that control to disadvantage rivals, favour its own products, manipulate access to customers, or reinforce its market position.

The legal significance of algorithmic ranking is particularly clear in the Google Shopping litigation, where the EU courts examined Google's preferential positioning of its own comparison-shopping service and the simultaneous demotion of competing services. The Court of Justice ultimately upheld the €2.4 billion fine in 2024.

1. Meaning of Algorithmic Influence Scoring

An influence score is an algorithmically generated assessment of how prominently an item, business, person, product or service should appear.

Examples include:

• seller-quality scores;
• search-result rankings;
• recommendation scores;
• advertising-quality scores;
• app-store rankings;
• marketplace seller rankings;
• news-feed visibility;
• credit or reputation scores;
• delivery-platform priority scores;
• hotel and travel rankings;
• advertising-auction scores;
• consumer-review rankings;
• platform trust scores.

A simplified model could be represented as:

Ranking Score = f(relevance + quality + engagement + price + conversion + platform data + strategic variables)

The legal difficulty is that the platform may control both:

1. the market in which participants compete, and
2. the algorithm that determines how effectively those participants reach customers.

This creates a potential dual-role conflict.

2. Algorithmic Ranking as a Competitive Bottleneck

Digital markets often have strong visibility effects.

For example:

Higher ranking → greater visibility → more clicks → more sales → more data → better algorithmic performance → still higher ranking

This produces a feedback loop:

Ranking → User attention → Transactions → Data → Algorithmic advantage → Ranking

A platform with substantial market power may therefore influence competition without formally excluding a rival.

The rival may technically remain on the platform but become commercially invisible.

This is sometimes described as algorithmic foreclosure.

3. Algorithmic Influence Versus Traditional Exclusion

Traditional exclusion may involve:

• exclusive contracts;
• refusal to supply;
• tying;
• loyalty rebates;
• predatory pricing.

Algorithmic exclusion can operate differently.

The platform may simply:

• lower a rival's ranking;
• change recommendation parameters;
• reduce search visibility;
• alter the scoring formula;
• impose an opaque quality score;
• give its own product preferential treatment;
• change the default recommendation;
• increase the ranking weight assigned to a variable controlled by the platform;
• make rival products appear only after several pages.

The economic effect can nevertheless be similar to exclusion.

4. Why Ranking Control Creates Competition-Law Concerns

A. Self-preferencing

A platform may give its own products a higher ranking than competing products.

Example:

Platform's product → Rank 1

Independent competitor → Rank 17

If the difference cannot reasonably be explained by quality, relevance, price or another legitimate ranking criterion, competition authorities may investigate whether the ranking mechanism has been manipulated.

B. Discriminatory ranking

A platform may apply different algorithmic rules to:

• its own products;
• affiliated businesses;
• independent sellers;
• competing platforms.

The crucial question becomes whether similarly situated competitors receive equivalent competitive opportunities.

C. Ranking as an essential route to customers

For many digital platforms, appearing on the first page is economically important.

Consequently:

Control over ranking can become control over market access.

The legal analysis therefore needs to consider not merely the technical algorithm but its effect on:

• traffic;
• conversion;
• sales;
• customer acquisition;
• advertising costs;
• entry;
• expansion;
• innovation.

5. Algorithmic Opacity

Another problem is black-box ranking.

A platform may tell businesses that ranking is based on "relevance" or "quality" without disclosing:

• the variables;
• weighting;
• thresholds;
• penalty mechanisms;
• adjustment rules;
• exceptions;
• treatment of affiliated products.

This creates an evidentiary problem.

A competition authority may need to reconstruct the algorithm through:

• source-code evidence;
• historical versions;
• internal documents;
• A/B testing records;
• ranking experiments;
• database records;
• audit logs;
• consumer data;
• econometric analysis.

6. Algorithmic Influence and Market Power

Algorithmic ranking becomes particularly important where the platform has substantial market power.

Relevant questions include:

Market definition

What is the relevant market?

Dominance

Does the platform possess substantial market power?

Dependency

Do sellers or users have realistic alternatives?

Visibility

How important is ranking to customer acquisition?

Discrimination

Are rivals subjected to different ranking rules?

Foreclosure

Does the ranking reduce rivals' ability to compete?

Effects

Are there measurable effects on traffic, sales, prices or innovation?

Justification

Can the platform demonstrate legitimate technical or consumer benefits?

7. Six Major Case Laws

1. Google and Alphabet v Commission — Google Shopping

Case T-612/17; Case C-48/22 P

This is the most directly relevant authority.

The European Commission found that Google favoured its own comparison-shopping service in general search results while competing comparison-shopping services were subject to Google's adjustment algorithms and could be demoted.

The General Court upheld the Commission's decision in 2021, and the Court of Justice dismissed Google's appeal in 2024.

The Court of Justice specifically considered:

• competition on the merits;
• leveraging;
• potential foreclosure;
• causal connection;
• counterfactual analysis;
• the role of Google's ranking mechanisms.

Principle

A dominant platform's manipulation of visibility through differentiated ranking treatment can constitute an abuse of dominance where the conduct departs from competition on the merits and is capable of producing anticompetitive effects.

Importance

This case establishes the fundamental proposition that algorithmic ranking can itself become the instrument of exclusionary conduct.

2. Google Android — Google and Alphabet v Commission

Case T-604/18; appeal C-738/22 P

The Google Android litigation concerned Google's contractual arrangements surrounding Android devices, including the Play Store, Google Search and Chrome.

The General Court examined:

• tying;
• exclusivity payments;
• anti-fragmentation obligations;
• mobile operating-system ecosystems;
• exclusionary effects;
• barriers to rival development.

The Court of Justice subsequently delivered judgment in the appeal in July 2026, addressing contractual restrictions, tying, exclusivity and exclusionary effects in Android-related markets.

Principle

Algorithmic ranking cannot be analysed in isolation where ranking control is combined with contractual or ecosystem mechanisms that influence which competing services reach users.

Relevance

This is particularly important for:

• app stores;
• search engines;
• operating systems;
• recommendation systems;
• default settings;
• platform ecosystems.

3. United States v Google LLC — Search

U.S. District Court for the District of Columbia

The U.S. search-monopoly litigation concerns Google's conduct in maintaining its position in general search and search advertising.

The court found Google liable for unlawful monopolization, and subsequent remedies have addressed distribution agreements, search data and access to search-related infrastructure. The DOJ reports that the remedies include restrictions on exclusive distribution arrangements and requirements concerning certain search index and user-interaction data.

Principle

Competition in algorithmically mediated markets can be affected by mechanisms that determine:

• default access;
• distribution;
• data availability;
• search visibility;
• rival access to users.

Relevance to influence scoring

The case illustrates that ranking control cannot always be separated from the broader ecosystem that determines which competing search services can obtain users, data and scale.

4. Amazon Marketplace Investigation — UK Competition and Markets Authority

The CMA investigated Amazon's marketplace concerning:

• third-party seller data;
• selection of the product placed in the Buy Box;
• delivery arrangements.

Amazon offered commitments addressing the CMA's concerns, and the investigation was closed in 2023.

Although this was an administrative competition investigation rather than a reported judicial precedent, it is highly relevant to algorithmic ranking control.

Principle

The platform's ability to determine which seller receives the most commercially valuable position can materially affect competition.

The Buy Box illustrates a form of algorithmically mediated market access:

Seller eligibility → algorithmic selection → prominent placement → consumer attention → transaction.

5. Booking.com / Hotel Online-Booking Investigations

European competition authorities examined Booking.com's parity clauses and related mechanisms in online hotel-booking markets.

The European Competition Network reported investigations involving HRS and Booking.com, including concerns surrounding parity clauses and their competitive effects.

Importantly, the European Commission's monitoring work specifically noted that Booking.com's commitments addressed measures that could link display ranking to compliance with wide parity obligations.

Principle

Contractual restrictions can become particularly significant where they interact with ranking and visibility mechanisms.

Relevance

A platform could theoretically combine:

contractual restriction + ranking preference + customer visibility

to produce a stronger exclusionary effect than either mechanism alone.

6. United States v Google LLC — Advertising Technology

The U.S. ad-tech litigation provides another important example of algorithmic control.

The DOJ alleged that Google used mechanisms affecting advertising auctions, including preferential treatment and access to competitors' bidding information. In 2026, the Eastern District of Virginia ordered substantial behavioral relief, including interoperability, data-sharing and anti-discrimination measures.

Principle

Algorithmic competition concerns can extend beyond ordinary search ranking to auction ranking and allocation.

An advertising algorithm may determine:

• which advertisement wins;
• which advertiser receives exposure;
• what price is paid;
• which inventory is available;
• how rival exchanges compete.

Thus, auction algorithms themselves can become competitive infrastructure.

8. Comparative Case-Law Table

The first two are judicial authorities; several of the others are competition-enforcement proceedings or administrative investigations rather than court judgments, so they should be cited accordingly in an examination answer.

9. Algorithmic Ranking and Self-Preferencing

Self-preferencing becomes particularly problematic where a platform operates simultaneously as:

1. platform operator, and
2. competitor to businesses using the platform.

For example:

Marketplace owns Brand A
Marketplace ranks Brand A first
Independent Brand B ranks tenth
Consumers predominantly click the first results
Brand A receives more sales
Marketplace obtains additional transaction data
Ranking advantage becomes self-reinforcing.

The Google Shopping litigation demonstrates how this type of conduct can be analysed under abuse-of-dominance principles.

10. Algorithmic Ranking and Data Advantage

Ranking systems depend heavily upon data.

A dominant platform may possess:

• click-through data;
• conversion data;
• customer searches;
• purchase histories;
• seller performance;
• pricing information;
• inventory data;
• advertising data.

The platform can use these data to improve its own ranking algorithms while competitors cannot access equivalent information.

This can generate:

Data advantage → better prediction → better ranking → greater traffic → more data

This is a data–ranking feedback loop.

11. Algorithmic Ranking and Network Effects

Ranking systems can reinforce network effects.

Suppose:

More users → more transactions

More transactions → more data

More data → better algorithm

Better algorithm → better user experience

Better user experience → more users

This can create a self-reinforcing competitive structure.

The competition-law question is not whether network effects are unlawful. They are ordinarily legitimate economic phenomena.

The question is whether the dominant firm artificially reinforces the network effect through exclusionary conduct.

12. Algorithmic Demotion

Algorithmic demotion is potentially as important as preferential promotion.

For example:

Competitor initially ranked #3

↓

Algorithm modification

↓

Competitor ranked #35

↓

Traffic falls

↓

Sales decline

↓

Lower conversion data

↓

Algorithm assigns still lower score

This creates a demotion spiral.

The Google Shopping litigation is especially important because the Court examined Google's treatment of competing comparison-shopping services alongside its preferential treatment of Google's own service.

13. Ranking Manipulation Through Hidden Variables

A platform may alter ranking indirectly.

Instead of explicitly saying:

"Rank our product first."

it may modify the algorithm so that variables disproportionately favour its own product.

For example:

• higher weighting for fulfilment controlled by the platform;
• higher weighting for advertising expenditure;
• higher weighting for platform-generated reviews;
• lower weighting for external sellers;
• preferential treatment for platform logistics;
• different quality thresholds for affiliated products.

The legal inquiry should therefore examine the functional effect of the algorithm, rather than merely its formal wording.

14. Competition on the Merits

An important distinction is:

Legitimate competition

A product ranks highly because it has:

• better quality;
• lower price;
• higher consumer satisfaction;
• better delivery;
• stronger relevance.

Potentially problematic ranking

A product ranks highly because:

• the platform owns it;
• the platform receives higher margins from it;
• rivals are subjected to additional algorithmic penalties;
• the ranking criteria are selectively applied;
• affiliated products are exempted from ranking adjustments.

The Google Shopping judgment is particularly significant because the Court addressed the distinction between competition on the merits and potentially exclusionary conduct.

15. Evidentiary Issues

Algorithmic cases present unusual evidentiary problems.

Competition authorities may need to establish:

A. Algorithmic design

What variables does the algorithm use?

B. Differential treatment

Does the algorithm treat affiliated and independent products differently?

C. Historical changes

When was the ranking algorithm changed?

D. Causation

Did the algorithmic change cause the competitor's loss of visibility?

E. Counterfactual

What would rankings have looked like without the disputed mechanism?

F. Effects

Did the ranking change affect:

• traffic;
• sales;
• prices;
• market share;
• entry;
• innovation?

The Court of Justice's Google Shopping judgment expressly addressed issues involving causal links, counterfactual scenarios, foreclosure capability and the burden of proof.

16. Remedies for Algorithmic Ranking Abuse

Potential remedies include:

1. Non-discrimination obligations

The platform must apply equivalent ranking principles to comparable products.

2. Algorithmic auditing

Independent experts periodically examine ranking systems.

3. Ranking transparency

Platforms may be required to disclose significant ranking parameters.

4. Data-access remedies

Rivals may obtain specified categories of data.

5. Interoperability

Competitors receive technical access necessary to compete.

6. Separation

The platform's marketplace function may be separated from its competing commercial operation in particularly serious cases.

7. Monitoring trustees

An independent monitoring mechanism can supervise compliance.

8. Prohibition of self-preferencing

The platform may be prohibited from placing its own services ahead of equivalent third-party services solely because of corporate affiliation.

The EU's Digital Markets Act illustrates a more direct regulatory approach: gatekeepers are required to apply fair and non-discriminatory conditions when ranking their own services against third-party services. In July 2026, the European Commission announced a €460 million DMA fine concerning Google's self-preferencing of its own services in Search.

17. Algorithmic Influence Scoring and the Digital Markets Act

The DMA is important because traditional Article 102 TFEU analysis generally requires examination of dominance and abuse.

The DMA establishes more specific obligations for designated gatekeepers.

For ranking systems, the regulatory concern is particularly direct:

A gatekeeper should not use its privileged position as an intermediary to give its own services more favourable ranking treatment than competing third-party services.

The Commission's 2026 Google enforcement decision illustrates the continuing importance of this principle.

18. Algorithmic Influence Scoring in Different Digital Markets

E-Commerce

Ranking determines:

• product visibility;
• seller visibility;
• Buy Box allocation;
• recommendation placement.

Search Engines

Ranking determines:

• website traffic;
• discovery;
• commercial referrals.

App Stores

Ranking affects:

• downloads;
• discoverability;
• developer revenues.

Food-Delivery Platforms

Ranking can determine:

• restaurant visibility;
• delivery allocation;
• promotional exposure.

Travel Platforms

Ranking affects:

• hotel visibility;
• booking conversion;
• consumer choice.

Advertising Exchanges

Auction scores determine:

• advertisement placement;
• winning bids;
• publisher access;
• advertiser exposure.

19. Competition-Law Test

A useful analytical framework is:

Step 1 — Identify the ranking system

What exactly does the algorithm rank?

Step 2 — Identify market power

Does the operator possess substantial market power?

Step 3 — Identify the competitive dependency

How important is the ranking position for obtaining customers?

Step 4 — Examine algorithmic criteria

Which variables determine ranking?

Step 5 — Test differential treatment

Are the platform's own products treated differently?

Step 6 — Examine exclusionary capability

Can the conduct materially disadvantage rivals?

Step 7 — Establish effects

Examine traffic, sales, prices, innovation and entry.

Step 8 — Examine objective justification

Is the difference reasonably connected to legitimate consumer or technical considerations?

Step 9 — Examine proportionality

Could the legitimate objective be achieved through a less restrictive mechanism?

Step 10 — Design the remedy

Consider transparency, non-discrimination, interoperability, data access, monitoring or structural remedies.

20. Algorithmic Ranking Control as a New Form of Market Power

The deeper competition-law significance is that market power increasingly operates through control over information architecture rather than merely through control over price.

A platform can influence:

• what consumers see;
• what consumers do not see;
• which seller receives attention;
• which competitor receives traffic;
• which advertisement wins;
• which application is discovered;
• which product becomes commercially viable.

Thus, ranking power may be understood as a form of intermediation power.

Conclusion

Algorithmic influence scoring and ranking control represents an important development in digital competition law because algorithms can determine practical access to consumers without formally excluding a competitor.

The strongest judicial foundation is Google Shopping, where the EU courts examined preferential placement of Google's own comparison-shopping service and the demotion of competing services, ultimately upholding the infringement finding and €2.4 billion fine.

The broader case law and enforcement experience involving Google Android, Google Search, Amazon Marketplace, Booking.com and Google Ad Tech demonstrate that competition authorities increasingly examine not merely whether a platform is available to rivals, but how the platform's technical architecture determines the conditions under which rivals can actually compete.

Accordingly, the central legal question in algorithmic ranking cases is:

Does the algorithm merely identify the most relevant or efficient competitive option, or is ranking control being used by a powerful intermediary to distort the competitive process itself?

That distinction connects algorithmic ranking with the established doctrines of abuse of dominance, self-preferencing, tying, discrimination, exclusionary conduct, essential facilities, leveraging, foreclosure and competition on the merits.