When Competition Meets Consumer Protection: Europe’s Expanding Playbook on Algorithmic Pricing

In recent years, the integration of algorithms into core commercial strategies has attracted increased scrutiny from competition authorities. Pricing algorithms are now rarely out of the headlines. From surges during live entertainment ticketing sales to inquiries into corporate “surveillance pricing” by the US Federal Trade Commission, the automated determination of prices has become a point of intense legal and political debate. More recently, mainstream retail markets saw a federal antitrust lawsuit challenging McDonald’s use of an internal AI platform to pool transaction data across franchises, alleging that automated menu optimization amounted to algorithmic price-fixing.

While we previously explored the US enforcement landscape on algorithmic pricing, European enforcers are moving just as rapidly. Officials at the European Commission (Commission) have confirmed that investigations into algorithmic pricing are underway. The UK Competition and Markets Authority (CMA) has already started looking into alleged algorithm-enabled information sharing in the hotel sector. The CMA has also leveraged its consumer protection powers to secure transparency commitments from Ticketmaster regarding ticket pricing.

European enforcers are deploying the full spectrum of tools against pricing algorithms: from Article 101 TFEU (and its analytical limits) to possible forms of abuse under Article 102 TFEU, through to emerging regulatory obligations. We trace how the debate has shifted beyond traditional consumer welfare standards toward an enforcement agenda which is focused on fairness and consumer protection. This is largely driven by Europe’s expanding digital regulatory playbook, including the forthcoming Digital Fairness Act (DFA). We outline practical steps to help businesses navigate this converging regulatory environment.

Clarifying the terms: what is under the regulatory microscope?

Much of the friction in public debate stems from conflating three distinct commercial mechanisms:

• Algorithmic pricing is the broad umbrella term encompassing the use of any automated software, mathematical rule, or machine-learning model to calculate, adjust, or recommend prices.
• Dynamic pricing refers to the pricing practice of adjusting prices in real time or near-real time in response to constantly shifting external market parameters (such as aggregate demand, competitor stock levels, capacity constraints, or weather patterns). Dynamic pricing responds to overall market conditions, not individual consumer identity.
• Personalized pricing (frequently termed "surveillance pricing" by enforcers) uses consumer data (such as browsing history, geolocation, purchasing habits, and device type) to estimate an individual’s willingness to pay and serve a customized price tailored to that specific user or demographic cohort.

While these types of pricing tools are drawing increased attention, they are not new. They have been used for decades across aviation, hospitality, and retail, only that now they are more granular, operate with more autonomy, and integrate generative and agentic AI models.

The competitive implications of this sophisticated approach to pricing are still evolving. As a note by the EU to the OECD emphasizes, algorithmic pricing delivers real economic gains. Real-time matching of supply and demand cuts transaction and search costs and resolves bottlenecks that human teams cannot react to quickly enough. Dynamic discounting also plays a role in capacity management and waste reduction: surplus food platforms, for instance, rely on dynamic price reductions as products near expiration. Predictive software can also lower barriers to entry by allowing smaller entrants to monitor market movements and compete aggressively without maintaining large commercial pricing departments. 

Nonetheless, legal risks can emerge when algorithms become vehicles for anti-competitive coordination, facilitate unilateral market exclusion or exploitation, or breach consumer protection or other regulatory standards. We examine these distinct categories of risk below.

Algorithmic Collusion under Article 101 TFEU

Under Article 101 TFEU, authorities could evaluate algorithmic collusion across four main categories, although the latter two push the boundaries of antitrust law as evidence of human consensus disappears. 

The most obvious way in which liability could arise is where algorithms implement or monitor an underlying human agreement. Such software-assisted cartels have been sanctioned in the past. In the UK, the CMA secured criminal convictions and fines in Online Posters and Frames (2016), where marketplace sellers used repricing software to enforce agreed price floors on Amazon. In the EU, the Commission’s 2023 Horizontal Guidelines (Commission Guidelines) emphasizes that firms remain liable for pricing decisions implemented through algorithms. 

Additionally, hub-and-spoke coordination may arise where competitors use a common software provider to align pricing decisions or exchange commercially sensitive information. Even without direct contact between competitors, Article 101 may still apply where they knowingly participate in such coordination. In Eturas (Case C-74/14), the CJEU held travel agencies liable for automated discount caps applied via an online booking platform because they knew of the restriction and failed to publicly distance themselves.

A more challenging case arises where competitors independently deploy algorithms programmed to react predictably to each other’s prices, for example through instantaneous price-matching. Economic research shows that real-time algorithmic conditioning can prevent price wars, sustaining supra-competitive prices without any inter-firm communication. However, because Article 101 requires a “meeting of minds”, purely unilateral parallel conduct remains exceptionally difficult to penalize.

The most contentious scenario involves self-learning AI, where regulators fear that algorithms could independently learn to sustain supra-competitive prices without any human instruction to collude. In its March 2026 paper, AI and Collusion, the CMA highlighted how self-learning agents could theoretically learn tacit signaling or punishment strategies. To date, there has been no formal infringement decision sanctioning purely autonomous algorithmic collusion, and it is doubtful how a “digital eye” could meet the agreement or concerted practice criterion. Nonetheless, as we discuss in our blog post here, the Commission Guidelines explicitly refer to “collusion by code”. 

Algorithms in abuse of dominance cases (Article 102 TFEU)

Digital platforms considered dominant face antitrust exposure under Article 102 TFEU when algorithmic pricing produces exclusionary or exploitative effects. Depending on the design and impact of the pricing tool, enforcers can categorize these unilateral practices under several theories of harm.

On the exploitative front, authorities could challenge automated price surges as excessive pricing under Article 102(a) TFEU (under EU law, this test is satisfied if the price charged is both excessive and unfair). Conversely, on the exclusionary side, real-time data could be used to execute predatory pricing, using algorithms to detect rivals and apply targeted below-cost price cuts.

Dynamic or personalized pricing could also take the form of price discrimination, which is not unlawful per se under EU competition law. On the contrary, by allowing sellers to segment customers according to their demand elasticity, algorithms can expand overall output and serve price-sensitive buyers who would otherwise be priced out of the market. In line with Ramsey pricing principles, this mechanism can broaden access and enhance total welfare. For price discrimination by a dominant firm to be abusive under Article 102(c) TFEU, EU law requires that applying dissimilar conditions to equivalent transactions must place commercial trading parties at a competitive disadvantage and be capable of distorting competition.

Finally, competition authorities may take unlawful personalized pricing behavior under the General Data Protection Regulation (GDPR) into account when establishing an abuse of dominance under Article 102 TFEU.

Beyond competition law: the EU’s regulatory toolbox on pricing

Beyond classic antitrust enforcement, algorithmic pricing sits at the centre of a converging web of European digital, consumer, and data protection rules. Enforcers are no longer looking only at collusion or market power; they are policing the data inputs that fuel pricing engines and the consumer-facing tactics used to display those prices.

Under EU consumer protection rules, traders face direct transparency obligations regarding price-setting. Under the Consumer Rights Directive (CRD) (as amended by the Modernisation Directive), traders must inform consumers in a clear and prominent manner whenever a price has been personalized through automated decision-making. In parallel, the Unfair Commercial Practices Directive (UCPD) prohibits deceptive pricing mechanisms. The Commission guidance on consumer rights makes clear that while dynamic pricing is inherently lawful, it breaches the UCPD if algorithms engineer false scarcity countdowns or advertise attractive “starting” prices that automated systems rapidly render unavailable during checkout (drip pricing). We also see enforcement in this area. In 2024, the Consumer Protection Cooperation (CPC) Network sanctioned dating platform Tinder for deploying algorithmic, personalized discounts without transparent disclosure. While the CPC Network solely coordinates national consumer protection enforcement, the CPC Regulation is now undergoing reform, envisaging the possibility of granting the Commission direct enforcement powers in certain cross-border consumer cases. Meanwhile, private litigation from consumer bodies plays a prominent role in some Member States, such as Germany. Algorithmic pricing could therefore face scrutiny on multiple fronts: from national consumer protection authorities and potentially the Commission, to individual national courts.

At the same time, digital and data protection frameworks increasingly restrict the data inputs that fuel these pricing engines. Under the GDPR, profiling consumers to set personalized rates requires a robust lawful basis, with Article 22 granting individuals specific rights against solely automated decision-making that produces legal or similarly significant economic effects. For designated gatekeepers, Article 5(2) of the Digital Markets Act (DMA) may restrict the cross-service data pooling that gatekeepers could rely on to feed predictive pricing models without explicit consent. Similarly, the Digital Services Act (DSA) sharply limits commercial profiling inputs by barring the use of special categories of sensitive personal data (such as health or political beliefs) under Article 26(2), and prohibiting profiling-based commercial targeting of minors under Article 28(2).

Finally, the EU AI Act seeks to address the manipulative potential of these systems directly. Under Article 5, it prohibits AI systems that deploy subliminal, manipulative, or deceptive techniques that exploit consumer vulnerabilities (including age, disability, or specific socio-economic conditions) to distort economic decision-making and cause significant harm.

Spotlight: The DFA & the fairness debate

Algorithmic pricing is also set to feature prominently in the forthcoming DFA. Prompted by the Commission’s “Fitness Check” that found existing EU consumer law too fragmented to police commercial personalisation, the DFA aims to bridge these perceived regulatory silos and place consumer fairness at the heart of the digital agenda. 

A draft legislative proposal is expected on 11 November. So far, the DFA consultation has focused on several pricing practices. On drip pricing, the EU is considering a prohibition akin to the UK’s outright ban under the Digital Markets, Competition and Consumers Act 2024. Dynamic pricing is also under scrutiny where traders advertise attractive “starting prices” that dynamic repricing software renders unattainable for many buyers, placing consumers under artificial pressure. The Commission also takes aim at misleading price reductions, including discounts calculated in breach of the Price Indication Directive, as well as misleading comparisons with unsubstantiated recommended retail prices (RRPs). A sweep (screening) on such misleading prices has already been concluded by the CPC Network in March 2026.

Apart from these obligations, this upcoming regulatory instrument introduces a wider fairness consideration for businesses that must navigate both antitrust and regulatory regimes. In EU competition law, “fairness” has historically occupied an ambiguous role, often considered as one of the discipline’s objectives. Empirical research has shown, however, that fairness has functioned more as a “buzzword”, featuring prominently under policy discourse (especially of former Commissioner Vestager) and very rarely in CJEU judgements or in Commission decisions. However, fairness has increasingly moved to the centre of EU enforcement, first with the DMA, which places fairness as one of its core objectives, and now through the renewed focus on consumer protection regulation. While competition law’s consumer welfare framework has traditionally focused on economic effects, including prices, output and efficiency, consumer protection addresses issues such as transparency and behavioural pressure on purchasing decisions. It is this dimension that the Commission is now seeking to strengthen, meaning that businesses must increasingly assess their practices through both a competition law and a consumer fairness lens. 

It is also helpful to think about what algorithmic pricing fairness actually means to consumers. As a pricing strategist has observed, consumers are often less troubled by whether a price is objectively high than by whether it is higher than what somebody else paid, or higher than it was only moments earlier. The public outcry surrounding Ticketmaster’s dynamic pricing for the Oasis Live ’25 tour illustrated this breakdown: fans waited in hours-long virtual queues only to find tickets priced at more than double their advertised cost. Academic research, however, shows that consumer perceptions of fairness are highly nuanced. A 2024 study found that consumers actually perceive demographic price discrimination as fairer when prices are set by algorithms rather than humans, because they feel less personally judged. However, this acceptance can disappear where price discrimination involves volatile, contentious one-off transactions or lacks a clear pro-social justification. Insofar as algorithmic price-setting may challenge more fundamental social and economic values, engagement by public stakeholders is likely to be high requiring a disciplined regulatory approach that identifies the proper legal tool backed by a sound analytical framework.

Outlook and managing exposure

As companies develop increasingly sophisticated technologies, Europe is casting a wider regulatory net to address the risks of algorithmic pricing. Competition law is no longer the only tool in the enforcement toolbox. Instead, a dual-track approach is emerging: competition law, traditionally focused on consumer welfare and competitive effects, alongside regulation targeting consumer protection and fairness. The distinction was recently underscored by the UK Competition Appeal Tribunal, which observed that “competition law is not a general law of consumer protection” and that not every consumer transparency issue is an abuse of dominance. For businesses, this means that algorithmic pricing must now be assessed through both lenses. This starts with understanding where and how pricing algorithms are deployed, what data they rely on and how much autonomy they have. Particular attention should be paid not only to competition risks, such as information sharing and coordination, but also to consumer-facing concerns, including personalised pricing, transparency and potential consumer manipulation

At Freshfields, we advise businesses across the full spectrum of competition, data protection and digital consumer regulation. Our cross-disciplinary approach brings together antitrust, privacy and consumer expertise to identify the implications for your business. If you have questions about this evolving regulatory landscape or would like integrated advice across these areas, please reach out to the authors or your usual Freshfields contact.