Data Concentration in Digital Mergers

Introduction

Data concentration in digital mergers refers to the accumulation or combination of large, valuable, or strategically important datasets as a result of a merger or acquisition. In digital markets, data can function as an important competitive asset alongside price, technology, intellectual property, infrastructure, and network effects.

A merger may therefore raise competition concerns even where the parties have relatively low monetary revenues or where the transaction does not immediately increase prices. The combination of datasets may enable the merged firm to improve algorithms, target advertising more precisely, develop AI models, personalize services, strengthen network effects, reduce rivals' access to information, or create barriers to entry.

Data concentration is particularly relevant in:

• online advertising;
• search engines;
• social-media platforms;
• e-commerce;
• digital payments;
• cloud computing;
• artificial intelligence;
• ad-tech;
• digital health;
• online marketplaces;
• financial technology; and
• Internet-of-Things ecosystems.

I. Meaning of Data Concentration

Data concentration occurs when a merger causes one undertaking to control substantially more data than it controlled before the transaction.

The relevant data may include:

1. Personal data – names, identifiers, location, preferences and behavioural information.
2. Transactional data – purchases, payments and browsing histories.
3. Behavioural data – searches, clicks, viewing patterns and interactions.
4. Technical data – device information, telemetry and performance information.
5. Business data – supplier, customer and pricing information.
6. Location data – GPS and mobility information.
7. Social-graph data – connections, interactions and communications.
8. Training data – datasets used to train machine-learning and AI systems.

The competition-law question is not simply how much data exists, but whether the merged data asset materially changes the competitive conditions of the market.

II. Why Data Matters in Digital Competition

Data can produce several forms of competitive advantage.

1. Economies of scale in data

A larger dataset may improve the accuracy of algorithms.

For example:

more users → more data → better algorithms → better service → more users → still more data.

This can create a data-driven feedback loop.

2. Economies of scope in data

Different datasets can be combined to generate insights unavailable from either dataset individually.

For example:

• search data;
• location data; and
• purchasing data

may collectively provide a much more detailed picture of consumer behaviour.

3. Targeted advertising

Data concentration may enable a merged firm to offer advertisers superior targeting and measurement capabilities.

This is particularly significant in:

• ad exchanges;
• demand-side platforms;
• social-media advertising; and
• retail-media networks.

4. Personalisation

More data may allow a platform to provide more sophisticated recommendations and personalization.

5. AI development

Large datasets can constitute an important input into:

• training;
• fine-tuning;
• evaluation;
• reinforcement learning; and
• model improvement.

Consequently, a merger between two companies possessing complementary datasets can potentially create a significant AI-related competitive advantage.

III. Data Concentration as a Merger-Control Issue

Traditional merger analysis focuses on:

• market shares;
• concentration;
• prices;
• output;
• barriers to entry; and
• unilateral/coordinated effects.

Digital merger analysis additionally considers:

• data assets;
• data access;
• data portability;
• interoperability;
• network effects;
• switching costs;
• algorithmic advantages;
• ecosystem effects; and
• privacy as a dimension of competition.

Thus, data can be both an input and a competitive advantage.

IV. Relevant Market Analysis

The first issue is defining the relevant market.

A transaction may affect several interconnected markets.

Example

Suppose Platform A operates a social-media service and Platform B operates an advertising-data business.

The transaction could affect:

1. social networking;
2. digital advertising;
3. advertising technology;
4. data analytics; and
5. potentially AI/data services.

A narrow market definition may reveal substantial concentration that is hidden by examining the broader digital economy.

V. Data as a Non-Price Competitive Parameter

One of the most important developments in digital merger analysis is recognition that competition does not occur only through monetary prices.

A service may be offered to consumers for zero monetary consideration.

Consumers may instead provide:

• personal information;
• behavioural information;
• attention; and
• engagement.

Accordingly, deterioration in privacy or data protection can potentially constitute a non-price competitive effect, depending on the legal framework and evidence.

The competition authority may ask:

Would the merged undertaking have an increased ability or incentive to reduce privacy protections, increase data collection, or impose more extensive data-use conditions?

This does not mean every privacy issue automatically constitutes an antitrust violation. There must be a demonstrated connection with competitive conditions.

VI. Horizontal Data Concentration

Horizontal data concentration occurs where competing firms possess similar or overlapping datasets.

Example

Two competing digital advertising platforms merge.

Before the merger:

• Firm A possesses Dataset A;
• Firm B possesses Dataset B.

After the merger:

Dataset A + Dataset B = substantially larger combined dataset.

Potential effects include:

• increased targeting accuracy;
• stronger advertising performance;
• reduced ability of rivals to compete;
• increased entry barriers; and
• greater control over advertisers.

VII. Vertical Data Concentration

Vertical transactions may also produce data-related concerns.

For example:

marketplace → payment provider → advertising platform.

A vertically integrated firm may obtain access to information generated at several levels of the supply chain.

This could allow it to:

• identify competitors' customers;
• monitor rival suppliers;
• optimize its own products;
• discriminate against competing services; or
• leverage information into an adjacent market.

VIII. Data Combination and Cross-Use

A particularly important concern is cross-use.

A firm may independently possess two datasets that are not competitively problematic when held separately.

After the merger, however, it may combine them.

For example:

Dataset A: search behaviour
Dataset B: purchase behaviour

Combined:

consumer-intent dataset

The competitive value of the combined information may therefore exceed the value of either dataset individually.

IX. Data Barriers to Entry

Data concentration can increase barriers to entry where new firms cannot reproduce the incumbent's dataset.

An entrant may have:

• capital;
• software;
• skilled employees; and
• technological expertise,

but still lack historical behavioural data.

This can be especially significant where data is:

• difficult to collect;
• accumulated over many years;
• generated automatically through network activity;
• proprietary;
• highly granular; or
• subject to network effects.

X. Data Network Effects

Data can reinforce traditional network effects.

A simplified model is:

More users
↓
More data
↓
Better product/algorithm
↓
More attractive platform
↓
More users

A merger that strengthens this feedback loop may increase the incumbent's competitive advantage.

XI. Data-Driven Self-Preferencing

A merged platform may potentially use concentrated data to favour its own products.

For example, a marketplace possessing detailed seller and consumer data could theoretically use that information to:

• identify successful products;
• develop competing private-label products;
• adjust ranking algorithms;
• target consumers;
• optimize prices; or
• disadvantage independent sellers.

Whether such conduct violates competition law depends on the applicable dominance, abuse, vertical-restraint, or merger-control rules.

XII. Data Concentration and AI

Data concentration has become particularly important with the development of generative AI.

An acquisition can combine:

• proprietary datasets;
• user interaction data;
• search data;
• content libraries;
• cloud infrastructure;
• AI models; and
• distribution channels.

The resulting entity may obtain advantages in:

1. model training;
2. model evaluation;
3. personalization;
4. AI safety testing;
5. recommendation systems;
6. advertising;
7. inference;
8. product development.

Thus, merger analysis increasingly asks whether a transaction creates control over critical data inputs for future digital competition.

XIII. Important Case Laws

1. Facebook/WhatsApp – European Commission, Case COMP/M.7217 (2014)

This is one of the foundational digital merger cases involving data.

The transaction concerned Facebook's acquisition of WhatsApp.

The European Commission examined, among other things:

• social-networking services;
• consumer communications;
• online advertising; and
• the role of data.

The Commission considered whether Facebook and WhatsApp's data resources could create competitive advantages, particularly in relation to online advertising.

The merger was ultimately cleared subject to the applicable merger-control analysis.

Significance

The case established an important principle for digital merger analysis:

A merger involving large quantities of user data can raise competition questions even where the transaction does not involve a conventional price increase.

It also demonstrated the importance of distinguishing data availability from actual data-related market power.

2. Google/DoubleClick – European Commission, Case COMP/M.4731 (2008)

Google's acquisition of DoubleClick involved important digital advertising assets.

The Commission examined the relationship between:

• search advertising;
• display advertising;
• ad-serving technology; and
• advertising intermediation.

Although the decision predated the modern data-driven merger debate, the transaction became an important reference point for understanding the competitive significance of combining digital capabilities and information.

Significance

The case is frequently discussed in relation to the development of Google's position across digital advertising markets.

It illustrates why merger authorities must consider not only conventional market shares but also:

• technological integration;
• access to information;
• advertising infrastructure; and
• future competitive development.

3. Microsoft/LinkedIn – European Commission, Case M.8124 (2016)

Microsoft acquired LinkedIn in 2016.

The transaction raised questions concerning:

• professional social networking;
• customer relationship management software;
• online advertising;
• data;
• productivity software; and
• Microsoft's ecosystem.

The Commission examined whether Microsoft's ability to combine LinkedIn's professional-network data with Microsoft's software and services could affect competition.

The transaction was cleared subject to commitments.

Significance

The case demonstrates the importance of data as an ecosystem asset.

The competitive concern was not merely the number of LinkedIn users. The analysis involved the potential interaction between:

professional-network data + enterprise software + Microsoft's ecosystem.

4. Google/Fitbit – European Commission, Case M.9660 (2020)

Google's proposed acquisition of Fitbit generated significant concerns concerning health and fitness data.

Fitbit collected substantial information concerning:

• physical activity;
• health-related measurements;
• fitness behaviour; and
• user activity.

The European Commission examined whether Google could use Fitbit's data to strengthen its position in online advertising and related markets.

The transaction was ultimately cleared subject to commitments concerning the use of Fitbit data and related competitive conditions.

Significance

The case is particularly important because it demonstrates that the strategic value of data can extend beyond the market in which the data was originally generated.

Fitness data could potentially become valuable in:

• advertising;
• health services;
• AI;
• personalization; and
• digital ecosystems.

5. Meta/Giphy – UK Competition and Markets Authority

Meta's acquisition of Giphy became a major digital merger case in the United Kingdom.

The CMA examined the transaction in relation to:

• display advertising;
• digital platforms;
• GIF services; and
• access to digital content.

The CMA ultimately required the transaction to be unwound.

Significance

Although the transaction was not purely a data-concentration case, it illustrates the broader principle that a seemingly small digital acquisition can have implications for a major platform's:

• ecosystem;
• advertising position;
• user engagement;
• access to digital content; and
• competitive advantages.

It is particularly relevant to the concept of ecosystem-based merger control, in which the competitive value of a digital asset cannot be assessed solely by its standalone revenue.

6. Microsoft/Activision Blizzard – European Commission, Case M.10646 (2023)

Microsoft's acquisition of Activision Blizzard involved a major combination of digital assets, including:

• gaming content;
• user ecosystems;
• cloud gaming;
• distribution;
• software;
• platform infrastructure; and
• consumer information.

The European Commission's analysis focused heavily on vertical and ecosystem effects.

Significance for data concentration

Although data was not the sole focus, the case demonstrates the increasing importance of examining ecosystem-level advantages in digital mergers.

A digital merger can create competitive advantages through the combination of:

content + users + platform + infrastructure + data.

7. Illumina/GRAIL – European Commission, Case M.10188

The Illumina/GRAIL transaction is important for understanding modern merger-control treatment of innovation and strategically important assets.

GRAIL developed technology for early cancer detection, while Illumina supplied sequencing technology.

The transaction raised concerns about:

• access to essential technology;
• innovation;
• foreclosure;
• emerging markets; and
• control over strategically important technological inputs.

Significance for digital/data analysis

Although this is not a conventional digital-platform merger, it demonstrates an important principle relevant to data-driven transactions:

competition authorities may consider the future competitive significance of technological and informational assets even where existing revenue or market shares do not fully capture their strategic importance.

8. Meta/Kustomer – European Commission, Case M.10262

Meta's acquisition of Kustomer involved customer relationship management technology.

The transaction was examined against the background of Meta's existing digital ecosystem and its position in online communications and business messaging.

The case illustrates the importance of considering whether an acquisition gives a major platform greater control over:

• business-user information;
• customer interaction data;
• messaging infrastructure; and
• adjacent digital services.

Significance

It is relevant to data-driven ecosystem expansion, particularly where a dominant platform acquires a company operating in a complementary data-intensive service.

XIV. Chinese Competition-Law Perspective

China's Anti-Monopoly Law (AML) is particularly relevant to data concentration because digital platforms are major participants in the Chinese economy.

Chinese merger analysis can consider:

• market power;
• network effects;
• data;
• technology;
• platform ecosystems;
• user dependence;
• entry barriers; and
• control of important resources.

The Chinese regulatory framework is especially relevant to transactions involving major platform businesses.

XV. Alibaba/Youku Tudou and Digital-Ecosystem Concerns

Transactions involving large Chinese digital platforms have demonstrated the importance of examining combinations of:

• user data;
• content;
• platform traffic;
• advertising;
• payment services; and
• ecosystem infrastructure.

The broader Chinese enforcement environment has increasingly recognized that digital platforms can derive competitive advantages from their control of data and technological resources.

XVI. Alibaba Case – Platform Data and Competition

The Alibaba enforcement action under China's AML is not itself a merger decision, but it is highly relevant to understanding the regulatory treatment of data-rich platform ecosystems.

The Chinese authorities examined Alibaba's platform conduct, including its "choose one from two" arrangements.

The case demonstrated that:

• platform ecosystems can generate significant market power;
• network effects can reinforce dominance;
• data can contribute to platform advantages; and
• exclusionary conduct may have effects beyond a single conventional market.

Importance for mergers

In merger analysis, similar considerations can arise where an acquisition gives a dominant platform control over another major source of:

• users;
• transaction information;
• supplier data; or
• behavioural data.

XVII. Data Concentration and Privacy

Competition law and privacy law are related but distinct.

Competition law asks:

Does the transaction substantially lessen or restrict competition?

Privacy law asks:

Is personal data collected, processed, transferred or used lawfully?

A single transaction may therefore require analysis under:

• competition law;
• data-protection law;
• cybersecurity law;
• consumer-protection law; and
• sector-specific regulation.

In China, this may additionally involve the:

• Personal Information Protection Law;
• Data Security Law; and
• Cybersecurity Law.

XVIII. Data Portability as a Remedy

One possible remedy for data-related competition concerns is data portability.

This may permit users to transfer their information between competing services.

Potential benefits include:

• reducing switching costs;
• encouraging multi-homing;
• lowering entry barriers;
• facilitating competition; and
• reducing lock-in.

However, portability can raise practical questions concerning:

• privacy;
• cybersecurity;
• interoperability;
• technical standards; and
• consent.

XIX. Data Silos as a Structural or Behavioural Remedy

Authorities may require the merged firm to maintain separate data environments.

For example:

Dataset A cannot automatically be combined with Dataset B.

Such remedies may prevent the merged company from immediately exploiting the informational advantage created by the transaction.

However, monitoring such remedies can be difficult because authorities must determine:

• which data may be combined;
• which employees may access it;
• how algorithms use the data;
• whether indirect combination occurs; and
• whether anonymisation is genuinely effective.

XX. Data-Access Remedies

Another possible remedy is requiring the merged company to provide competitors with access to specified datasets or interfaces.

This may involve:

• API access;
• interoperability;
• data-sharing;
• technical interfaces; or
• access to certain information.

Such remedies must be carefully designed because forced access to personal or commercially sensitive data can create privacy and security problems.

XXI. Privacy Commitments as Merger Remedies

A competition authority may also consider commitments concerning:

• data collection;
• data combination;
• data retention;
• user consent;
• privacy standards; and
• use of acquired data for advertising.

The Google/Fitbit transaction is particularly important in this respect.

XXII. Killer Acquisitions and Data

A large platform may acquire a smaller company before the latter becomes a significant competitor.

The acquisition may be attractive because the target possesses:

• innovative technology;
• valuable data;
• a rapidly growing user base;
• unique algorithms; or
• a potentially competing service.

Traditional turnover thresholds may fail to capture the significance of such a target.

Therefore, merger regimes increasingly consider whether transaction value and strategic importance, rather than turnover alone, should trigger review.

XXIII. Data Concentration and Coordinated Effects

A merger can also facilitate coordination where it makes market behaviour more transparent.

For example, concentrated data may provide information concerning:

• prices;
• inventories;
• demand;
• customers;
• competitors; and
• market conditions.

If several major firms possess highly sophisticated data analytics, transparency can potentially make coordinated conduct easier.

However, the mere existence of extensive data does not establish coordinated effects. Authorities must establish the relevant mechanism and competitive impact.

XXIV. Key Factors for Merger Assessment

A competition authority examining data concentration should consider:

XXV. Difference Between Data Concentration and Traditional Market Concentration

Traditional concentration:

Firm A + Firm B = larger market share

Data concentration:

Firm A's dataset + Firm B's dataset = greater informational capability

The second may matter even when the parties' conventional market shares are relatively modest.

For digital markets, therefore:

Market share + data + network effects + technology + ecosystem control

may provide a more complete picture of competitive conditions.

XXVI. Potential Theories of Harm

Data concentration can generate several theories of harm:

1. Data foreclosure

Competitors may lose access to an important source of information.

2. Input foreclosure

The merged firm may restrict data necessary for rivals.

3. Customer foreclosure

The merged firm may use data to control access to customers.

4. Algorithmic advantage

The combined dataset may improve algorithms beyond competitors' ability to replicate.

5. Entry deterrence

Potential entrants may be unable to obtain comparable data.

6. Privacy deterioration

The merged company may have increased ability or incentive to weaken privacy protections where privacy competes as a non-price dimension.

7. Ecosystem strengthening

The transaction may reinforce an already powerful digital ecosystem.

8. Innovation foreclosure

Rivals may lose access to data necessary for developing innovative services.

XXVII. Important Legal Principles Emerging from the Cases

The case law collectively demonstrates several principles:

1. Data can constitute a strategically important competitive asset.
2. Zero-price services do not eliminate merger concerns.
3. Privacy may operate as a non-price competitive parameter.
4. Data can create economies of scale and scope.
5. Unique datasets may create barriers to entry.
6. The competitive importance of data may extend beyond the market where it was initially collected.
7. Digital mergers must sometimes be assessed at ecosystem level.
8. Behavioural remedies may be used to control data combination or access.
9. Data concentration should be assessed together with network effects and switching costs.
10. The mere possession of large quantities of data does not automatically establish market power or an antitrust violation.

XXVIII. Examination-Oriented Framework

For an examination problem involving Data Concentration in a Digital Merger, the following sequence can be used:

Step 1 — Identify the parties

↓

Step 2 — Identify the relevant digital markets

↓

Step 3 — Identify the datasets controlled by each party

↓

Step 4 — Determine whether the datasets are unique or replicable

↓

Step 5 — Examine data economies of scale and scope

↓

Step 6 — Examine network effects and switching costs

↓

Step 7 — Assess horizontal, vertical and conglomerate effects

↓

Step 8 — Examine privacy as a non-price competitive parameter

↓

Step 9 — Consider foreclosure and entry barriers

↓

Step 10 — Examine innovation and AI effects

↓

Step 11 — Consider possible remedies

↓

Step 12 — Determine whether the transaction substantially restricts competition under the applicable merger-control regime

Conclusion

Data concentration is an increasingly important dimension of digital merger control. The competitive significance of a transaction cannot always be measured through conventional market share or turnover because data may function as an essential input, an innovation asset, an advertising resource, and a source of algorithmic advantage.

The cases involving Facebook/WhatsApp, Google/DoubleClick, Microsoft/LinkedIn, Google/Fitbit, Meta/Giphy, Microsoft/Activision Blizzard, Illumina/GRAIL and Meta/Kustomer illustrate the evolution from traditional product-market analysis toward a broader examination of data, technology, ecosystems, network effects, innovation and non-price competition.

The central legal question is therefore not simply “How much data will the merged firm possess?” but rather:

“Will the combination of data materially enhance the merged firm's ability or incentive to restrict competition, foreclose rivals, raise entry barriers, reduce innovation, or otherwise alter competitive conditions?”