Competition Law and Ecosystem Network Mapping for Antitrust Analysis

1. Introduction

Ecosystem network mapping is a method of analysing competition by visually and analytically identifying the relationships among:

platforms;

consumers;

suppliers;

developers;

distributors;

competitors;

complementary services;

data providers;

payment systems;

infrastructure;

algorithms;

standards;

APIs;

regulators.

It is particularly useful in digital markets because competition may not occur only between two products. A powerful undertaking may occupy a central position in a network of connected markets.

The basic idea is:

MAP THE ECOSYSTEM → IDENTIFY THE GATEWAYS → LOCATE MARKET POWER → TRACE CONDUCT → ANALYSE FORECLOSURE → ASSESS COMPETITIVE EFFECTS.

2. Meaning of Ecosystem Network Mapping

An ecosystem network map represents the economic and technological relationships between different participants in an ecosystem.

For example:

                    USERS                      │                      ▼              DIGITAL PLATFORM             /        │        \            /         │         \       APP STORE    PAYMENTS    ADVERTISING          │            │             │          ▼            ▼             ▼     DEVELOPERS     MERCHANTS     ADVERTISERS          │            │             │          └────────────┼─────────────┘                       ▼                     DATA                       │                       ▼                PLATFORM POWER

The map allows an antitrust authority to ask:

Who controls the central node?

Who depends upon that node?

Which markets are connected?

Where does data flow?

Where are the contractual restrictions?

Which participants can switch?

Which participants can multi-home?

Which competitors can access customers?

Where are the bottlenecks?

Can power in one market be leveraged into another?

3. Why Network Mapping Is Important

Traditional competition analysis may focus heavily on:

Market A → Competitor B → Market Share

Ecosystem analysis requires:

Market A ↔ Platform ↔ Market B ↔ Market C ↔ Data ↔ Users ↔ Complementors

A firm may have relatively modest power in one market but exercise substantial influence because it controls a gateway connecting several markets.

For example:

Operating System → App Store → Payment System → Developers → Users → Data → Advertising

A restriction imposed at the operating-system level can therefore affect competition in downstream markets.

4. Components of an Ecosystem Network Map

A. Nodes

Nodes are participants or assets within the ecosystem.

Examples:

platform;

consumer;

developer;

supplier;

advertiser;

merchant;

payment provider;

cloud provider;

competitor;

regulator.

B. Edges

Edges represent relationships between nodes.

Examples:

contractual relationship;

technical connection;

payment relationship;

data exchange;

API connection;

distribution relationship;

ownership relationship;

licensing relationship.

C. Direction

The direction of the relationship matters.

Example:

Seller → Data → Platform

The platform receives information from sellers.

But:

Platform → Ranking → Consumer

The platform controls what consumers see.

D. Weight

The strength of each relationship can be assessed.

For example:

number of customers;

transaction volume;

percentage of revenue;

data volume;

switching cost;

dependency;

commission;

exclusivity duration.

A heavily weighted connection may represent an important competitive dependency.

5. Basic Ecosystem Mapping Formula

NODES + CONNECTIONS + DEPENDENCIES + GATEWAYS + DATA FLOWS + MARKET POWER = NETWORK MAP

The map is an analytical tool, not itself a legal test.

6. Types of Network Mapping

1. Structural Mapping

Shows who is connected to whom.

Example:

Platform → Sellers → Consumers

2. Ownership Mapping

Shows:

parent companies;

subsidiaries;

acquisitions;

joint ventures;

investments.

This is especially useful for merger control.

3. Contractual Mapping

Shows:

exclusivity;

tying;

minimum commitments;

parity clauses;

access conditions;

commissions.

4. Data Mapping

Shows:

Who collects → Who controls → Who receives → Who can use → Who cannot access

This is increasingly important in digital ecosystems.

5. Technical Mapping

Shows:

APIs;

operating systems;

protocols;

cloud infrastructure;

interoperability;

standards.

6. Customer-Dependency Mapping

Shows how strongly customers or business users depend upon each ecosystem participant.

7. Network Centrality

One of the most useful concepts is centrality.

A firm is central when many important relationships pass through it.

For example:

Developer ──┐ Merchant ───┤ Advertiser ─┼── PLATFORM ── Consumers Cloud ──────┤ Payment ────┘

The platform becomes a central node.

Centrality does not automatically mean dominance.

But high centrality can be evidence that the firm has an important gateway or bottleneck position.

8. Gateway Analysis

A gateway is a point through which competitors or business users must pass to reach customers.

Examples:

app store;

operating system;

search engine;

marketplace;

payment network;

cloud infrastructure;

digital identity system.

Antitrust question

Can a rival realistically bypass the gateway?

If the answer is no, the gateway may have significant competitive importance.

9. Bottleneck Analysis

A bottleneck is a node or infrastructure component where access is constrained.

Example:

Many Developers      ↓   APP STORE      ↓    USERS

If developers cannot effectively reach users without the App Store, the App Store becomes an important bottleneck.

The legal analysis must still determine whether the relevant access conditions satisfy the applicable abuse or regulatory standard.

10. Dependency Mapping

Network mapping should identify who depends upon whom.

Example

Small Seller     │     │ 90% of online sales     ▼ Marketplace     │     ▼ Consumers

The seller has high dependence on the marketplace.

Now add:

Marketplace + Search Ranking + Payments + Logistics

The dependency becomes even stronger.

Possible indicators

revenue dependence;

customer dependence;

technical dependence;

data dependence;

contractual dependence;

switching costs;

single-homing.

11. Network Effects Mapping

Network effects should be shown as feedback loops.

Example

More Users    ↓ More Sellers    ↓ More Products    ↓ More Consumer Choice    ↓ More Users

This can generate legitimate efficiencies.

But it can also create:

Scale → Network Effects → Entry Barriers → Greater Scale

This is sometimes described as a self-reinforcing ecosystem.

12. Data Network Mapping

Data may create another feedback loop:

More Users     ↓ More Data     ↓ Better Algorithms     ↓ Better Service     ↓ More Users

An antitrust investigation may therefore examine:

whether competitors can obtain comparable data;

whether data is portable;

whether data is interoperable;

whether the incumbent receives privileged information;

whether the incumbent uses third-party data to compete against those third parties.

13. Multi-Homing Mapping

Network mapping should identify whether participants use multiple ecosystems.

Example

Seller ├── Marketplace A ├── Marketplace B └── Own Website

This is multi-homing.

If instead:

Seller   │   ▼ Marketplace A only   │   ▼ Consumers

the seller is single-homing.

Single-homing may increase the platform's bargaining and competitive power.

14. Switching-Cost Mapping

The map should also identify what happens if a participant leaves.

Switching costs can include:

loss of data;

loss of customers;

technical migration;

retraining;

contractual penalties;

loss of reputation;

incompatibility;

loss of applications;

loss of transaction history.

Formula

Dependency = Switching Cost + Network Effects + Data Lock-in + Technical Dependence

This is an analytical framework rather than a statutory test.

15. Ecosystem Network Mapping and Relevant Market

Network mapping does not replace market definition.

The authority must still determine:

product market;

geographic market;

demand-side substitution;

supply-side substitution;

competitive constraints.

However, network mapping helps reveal that several relevant markets may be economically connected.

Example

Market 1 Operating System      │      ▼ Market 2 App Distribution      │      ▼ Market 3 Search      │      ▼ Market 4 Digital Advertising

The authority can then examine whether power in Market 1 affects competition in Markets 2–4.

16. Network Mapping and Leveraging

Leveraging occurs when a firm uses power in one market to influence another market.

Basic structure

Market A Power → Gateway → Conduct → Market B → Rival Foreclosure

Examples:

operating system → search;

search → shopping;

marketplace → logistics;

app store → payments;

cloud → software;

payment network → digital wallet.

17. Case Law 1 — Google Android

Google and Alphabet v Commission, C-738/22 P

This is one of the clearest ecosystem cases.

The General Court expressly described the case in terms of a multi-sided platform and ecosystem, involving:

Android operating system;

Play Store;

Google Search;

Chrome;

device manufacturers;

mobile network operators.

The case concerned product bundles, exclusivity payments and anti-fragmentation obligations. (Infocuria)

The Court of Justice dismissed Google's appeal in July 2026 and upheld the major infringement findings and a fine of approximately €4.1 billion. (curia)

Network-map significance

Android   │   ├── Play Store   │   ├── Google Search   │   ├── Chrome   │   ├── Device Manufacturers   │   └── Users

Principle

The case demonstrates that an antitrust authority can examine interconnected contractual arrangements as part of an overall ecosystem strategy.

18. Case Law 2 — Google Shopping

Google and Alphabet v Commission, C-48/22 P

The Court of Justice upheld the €2.4 billion fine concerning Google's favouring of its own comparison-shopping service in general search results. (Infocuria)

Network map

General Search      │      ▼ Search Results      │      ├── Rival Shopping Services      │      └── Google Shopping

Google occupied the central search gateway.

Principle

The case is important for analysing how control over a gateway can influence competition in a connected downstream market.

Network lesson

Gateway node + dominant position + preferential conduct = potential leveraging concern.

19. Case Law 3 — Microsoft v Commission

Case T-201/04

Microsoft involved:

client PC operating systems;

work-group server operating systems;

media players;

interoperability information.

The General Court upheld major elements of the Commission's finding that Microsoft abused its dominant position by refusing to provide interoperability information and by tying Windows with Windows Media Player. (curia)

Network map

Windows   │   ├── Server Interoperability   │   └── Media Player          │          ▼     Consumers/Developers

Network significance

The operating system functioned as a central technological node.

Principle

Control over a core technological interface can affect competition in connected markets.

20. Case Law 4 — Intel v Commission

Case C-413/14 P

Intel concerned conditional rebates involving computer manufacturers and a retailer.

The Court of Justice required proper consideration of the circumstances relevant to whether the rebates were capable of foreclosing an equally efficient competitor. (Infocuria)

Network map

Intel │ ├── Computer Manufacturer A ├── Computer Manufacturer B └── Retailer        │        ▼     Consumers

Network significance

The case illustrates how contractual incentives can be mapped across several commercial relationships.

Principle

Network mapping can help identify:

coverage of agreements;

duration;

customer dependence;

foreclosure opportunities;

rival access.

But the map must be followed by an effects analysis.

21. Case Law 5 — IMS Health

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

IMS Health concerned access to a pharmaceutical sales-data structure.

Network map

Pharmaceutical Companies          │          ▼      Data Structure          │          ▼       IMS Health          │          ▼   Competing Information       Providers

Principle

Dominance over an important data structure does not automatically create an unlimited obligation to license it.

The strict conditions for compulsory access remain relevant.

Network significance

It demonstrates that centrality + dependence ≠ automatic access right.

The authority must examine:

indispensability;

elimination of effective competition;

objective justification;

whether realistic alternatives exist.

22. Case Law 6 — Bronner

Oscar Bronner v Mediaprint, C-7/97

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

Network map

Newspaper Publisher       │       ▼ Delivery Network       │       ▼ Consumers

A competitor wanted access to the network.

Principle

A refusal to provide access to infrastructure becomes abusive only under demanding conditions, including indispensability and the absence of realistic alternatives.

Network significance

Network mapping helps identify:

whether the infrastructure is genuinely indispensable;

alternative routes;

duplication costs;

customer reach;

dependence.

23. Case Law 7 — Google Android Auto

Android Auto, C-233/23

This case is particularly relevant to network mapping because it concerns interoperability within a digital ecosystem.

The Court considered the circumstances in which a dominant platform's refusal to make its platform interoperable with third-party applications may constitute an abuse.

Network map

Android   │   ▼ Android Auto   │   ├── Google Applications   │   └── Third-Party Applications                 │                 ▼               Users

Principle

The characteristics of an open digital platform can matter when analysing access and interoperability.

Network significance

It demonstrates that technical architecture itself can become a competitive node.

24. Case Law 8 — Google Android: Ecosystem and Exclusion

The Android litigation is especially important because the General Court explicitly referred to the concept of an ecosystem and examined how:

barriers;

network effects;

pre-installation;

exclusivity;

anti-fragmentation

could interact.

The Court of Justice's 2026 judgment also emphasised that exclusionary effects must be demonstrated through specific analysis and evidence, taking account of the relevant conduct, markets and functioning of competition. (Curia)

Key lesson

A network map is evidence architecture, not the final legal conclusion.

25. Case-Law Comparison

26. Network Mapping of Self-Preferencing

Consider:

                PLATFORM                    │          ┌─────────┴─────────┐          ▼                   ▼     Third Parties        Own Service          │                   │          ▼                   ▼      Ranking A            Ranking B          │                   │          └─────────┬─────────┘                    ▼                 USERS

Antitrust questions:

Is the platform dominant?

Does it control user visibility?

Does it favour its own service?

Are rivals dependent upon the platform?

Does the conduct materially alter traffic?

Is there foreclosure?

Are there objective justifications?

27. Network Mapping of Tying

Dominant Product A       │       │ mandatory/conditional       ▼ Product B       │       ▼ Competitors in B       │       ▼ Potential foreclosure

The map allows the investigator to identify:

tying product;

tied product;

customer groups;

distribution channels;

rival access;

foreclosure mechanism.

The Microsoft and Google Android cases illustrate this type of ecosystem analysis. (Infocuria)

28. Network Mapping of Data Advantage

Consumers   │   ▼ Platform   │   ├── Consumer Data   ├── Seller Data   ├── Transaction Data   └── Behavioural Data          │          ▼       Algorithm          │          ▼     Own Products

Potential competition question:

Does the platform use information obtained through its intermediary role to compete against the businesses that depend on it?

This is especially relevant in marketplace ecosystems.

29. Network Mapping of Vertical Integration

              Platform             /        \            /          \      Infrastructure   Marketplace           │               │           ▼               ▼       Competitors       Sellers           │               │           └───────┬───────┘                   ▼                Consumers

A competition authority can examine whether the platform:

raises rivals' costs;

discriminates;

self-preferences;

restricts access;

uses competitor data;

forecloses alternative suppliers.

30. Network Mapping and Merger Control

Network mapping is also useful for mergers.

Suppose:

Platform A acquires Platform B

The map should identify:

horizontal overlaps;

vertical relationships;

complementary services;

data combinations;

potential competitors;

future competitors;

interoperability effects;

network effects.

Acquisition structure

Platform A    │    ├── Users    ├── Data    └── Infrastructure            + Platform B    │    ├── Users    ├── Data    └── Technology            ↓      Combined Ecosystem

The question is whether the merger eliminates an important competitive constraint or creates additional ecosystem power.

31. Network Mapping and Killer Acquisitions

A small target may appear insignificant if analysed only by current revenue.

Network mapping can reveal:

future technology;

innovation pipeline;

user base;

data assets;

complementary technology;

strategic position.

Therefore:

Current market share may underestimate future ecosystem significance.

32. Network Mapping and Essential Facilities

Network mapping helps determine whether a facility is truly essential.

Questions

Can competitors build an alternative?

Is duplication technically possible?

Is duplication economically viable?

Is access necessary to compete?

Is the facility controlled by a dominant firm?

Does refusal eliminate effective competition?

Bronner and IMS Health demonstrate why these questions must be examined carefully.

33. Network Mapping and Interoperability

Interoperability mapping asks:

Platform A    │    X  ← blocked interface    │ Platform B

versus:

Platform A    │    ↕  ← interoperable interface    │ Platform B

The authority may examine:

technical necessity;

standards;

APIs;

licensing;

access terms;

security;

compatibility;

rival viability.

Microsoft and Android Auto provide useful authorities for this analysis.

34. Network Mapping and Algorithmic Competition

Algorithms create invisible network relationships.

For example:

Seller Data     ↓ Algorithm     ↓ Ranking     ↓ Consumer Visibility     ↓ Sales     ↓ More Data     ↓ Improved Algorithm

This creates a data-ranking feedback loop.

Antitrust investigators may therefore need to map not only legal and commercial relationships but also algorithmic relationships.

35. Network Mapping and Competitive Effects

The map should ultimately answer:

A. Foreclosure

Can competitors reach customers?

B. Entry

Can new firms enter?

C. Expansion

Can existing rivals grow?

D. Innovation

Can firms innovate independently?

E. Choice

Can consumers access alternatives?

F. Contestability

Can users realistically switch?

36. Network Mapping Does Not Equal Market Power

This is a critical distinction.

A firm may be the most connected participant without being legally dominant.

Similarly:

High network centrality ≠ dominance

High dependency ≠ abuse

Large ecosystem ≠ unlawful monopoly

The network map is therefore an evidentiary and analytical instrument.

37. Network Mapping and Objective Justification

Not every restrictive connection is anticompetitive.

A platform may restrict access because of:

cybersecurity;

privacy;

technical integrity;

fraud prevention;

consumer protection;

quality control;

intellectual property;

system stability.

The investigator should map the justification as well:

Restriction    ↓ Legitimate Objective?    ↓ Necessary?    ↓ Proportionate?    ↓ Less Restrictive Alternative?

38. Complete Antitrust Network-Mapping Framework

N-E-T-W-O-R-K

N — Nodes
Identify all relevant participants.

E — Edges
Identify contractual, technical, financial and data relationships.

T — Traffic
Identify users, transactions and data flows.

W — Weak points
Identify bottlenecks, dependencies and entry barriers.

O — Ownership/control
Identify who controls infrastructure, data and gateways.

R — Rivalry effects
Analyse foreclosure, entry and innovation.

K — Key remedy
Identify the least restrictive effective remedy.

39. Ultra-Short Revision

Ecosystem Network Mapping

NODES → CONNECTIONS → DEPENDENCY → GATEWAYS → DATA → NETWORK EFFECTS → MARKET POWER → CONDUCT → FORECLOSURE

Main things to map:

Users

Competitors

Complementors

Platforms

Infrastructure

Data

Contracts

APIs

Payments

Algorithms

Switching costs

Network effects

Main legal questions:

Who controls the gateway?

Who depends upon whom?

Can users switch?

Can rivals bypass the platform?

Can competitors interoperate?

Does the leader favour itself?

Does it leverage power into another market?

Does conduct foreclose competitors?

Is there an objective justification?

40. Memory Formula

NETWORK MAP = NODES + EDGES + GATEWAYS + DEPENDENCIES + DATA + FEEDBACK LOOPS

And the antitrust formula:

NETWORK STRUCTURE → MARKET DEFINITION → CENTRALITY/POWER → CONDUCT → FORECLOSURE → EFFECTS → JUSTIFICATION → REMEDY

Final Conclusion

Ecosystem network mapping provides a structured way to understand competition where markets are interconnected rather than isolated. It enables an antitrust authority to identify central nodes, bottlenecks, dependencies, data flows, network effects, switching costs and gateways and then connect those structural features to specific conduct and competitive effects.

The most important authorities include Google Android, Google Shopping, Microsoft, Intel, IMS Health, Bronner and Android Auto. Google Android is especially significant because the EU litigation expressly analysed the multi-sided platform and ecosystem involving Android, Play Store, Search, Chrome, device manufacturers and network operators. (Infocuria) Google Shopping demonstrates gateway leverage, while Microsoft demonstrates the importance of interoperability. (Infocuria) Intel emphasises evidence-based effects analysis, while IMS Health and Bronner place limits on treating every central infrastructure or proprietary resource as an automatically accessible facility. (Infocuria)

Final exam formula:

MAP THE NETWORK → FIND THE CENTRAL NODE → IDENTIFY DEPENDENCIES → DEFINE MARKETS → MEASURE POWER → ANALYSE CONDUCT → TRACE FORECLOSURE → TEST JUSTIFICATION → DESIGN REMEDY.