Skip to main content
All
April 11, 2025

The CJEU’s Preliminary Ruling on Art. 102 TFEU in Alphabet and Google v. AGCM

Not a Straight Win for the AGCM

Advisory

On February 25, 2025, the Court of Justice of the European Union (CJEU), sitting in the Grand Chamber, rendered a preliminary ruling in which it answered several questions raised by the Consiglio di Stato, Italy’s highest administrative court (the Referring Court), concerning Art. 102 of the Treaty on the Functioning of the European Union (TFEU) which prohibits dominant companies from abusing their market power.1

The Referring Court had raised these questions in the context of proceedings in which Alphabet and Google (jointly referred to as Google) were seeking the annulment of a decision by which the Italian competition authority, the Autorità Garante della Concorrenza e del Mercato (AGCM), had imposed a fine on Google for infringing Art. 102 of the TFEU.

In order to understand the issues at stake, we first need to summarize the key facts that led to the AGCM’s decision (infra A). We will then review the CJEU’s answers to the Referring Court’s questions and assess to what extent these answers might have an impact on the guidelines for the application of Art. 102 of the TFEU to abusive exclusionary conduct by dominant undertakings (Guidelines) that the European Commission (Commission) intends to publish later this year (infra B).2

A. Key Facts of the Dispute Between the AGCM and Google

In 2015, Google launched Android Auto, a digital platform that enables Android smartphone users to access apps directly on the screen of their motor vehicles’ infotainment system. By late 2018, Google was offering third-party suppliers of multimedia and messaging apps templates in order to ensure interoperability between their apps and Android Auto. For navigation apps, there was no such template (yet). As a consequence, only Google’s own apps (Google Maps and Waze) could be connected to the screen of a motor vehicle’s infotainment system.

In May 2018, Enel X, a subsidiary of the Enel electricity group that manages more than 60% of the charging stations available for electric vehicles in Italy, launched the JuicePass app. Users of Android smartphones could download that app for free via Google Play and then use it on the screen of their smartphone to search for charging stations (and also to book and pay charging time at these stations) before or while driving their car.

In September 2018, Enel X asked Google to ensure interoperability between the Android Auto platform and its JuicePass app, so as to enable users of its JuicePass app to consult it on the screen of the infotainment system of their car while driving. Google refused to do so. It initially told Enel X that multimedia and messaging apps were the only categories of third-party apps for which it had developed an interoperability solution. In December 2018, when Enel X insisted, Google still refused. It invoked security grounds and also pointed to the cost of allocating the necessary resources internally to develop a new template.

In February 2019, Enel X brought the matter before the AGCM. In October 2020, Google published a template for the design of experimental versions of electric vehicle charging apps interoperable with Android Auto. The AGCM was uncertain whether that template would be sufficient to allow the full integration of JuicePass into Android Auto. In April 2021, it therefore adopted a decision concluding that Google had abusively obstructed and delayed JuicePass’ availability on Android Auto, imposing a fine of around € 102 million, and ordering Google to publish a definitive version of this template that would include any features indicated by Enel X as essential and so far missing in the then-available template.

The Regional Administrative Court in Lazio upheld the AGCM’s decision but Google brought an appeal to the Consiglio di Stato which stayed its proceedings in April 2023 in order to obtain clarifications from the CJEU on how to interpret Art. 102 of the TFEU in this case.

B. The CJEU’s Clarifications on Art. 102 of the TFEU

The Referring Court raised five specific questions regarding the interpretation of Art. 102 of the TFEU.

In order to evaluate the potential impact of the CJEU’s preliminary ruling on the Commission’s future enforcement policy, it is helpful to first consider the Commission’s current approach, as set out in its above-mentioned 2024 draft Guidelines.

Referring to settled case law, the Commission indicates that the analysis of allegedly abusive conduct should proceed in three steps:

1. Does the dominant company’s conduct constitute competition on the merits with its rival competitors?

2. If not, does its conduct actually restrict competition or does it have at least the capacity to do so?

3. If so, can that conduct nevertheless be justified on “objective necessity” or “efficiency” grounds?3

The Commission acknowledges that, while step 1 is “conceptually different” from step 2, “certain factual elements may be relevant to the assessment of both.”4 It also accepts that there might be a link between step 1 and step 3 “if a dominant undertaking argues that its conduct amounts to competition on the merits because … the actual or potential exclusionary effects produced by the conduct are counterbalanced or outweighed by advantages in terms of efficiencies that benefit consumers, this argument is evaluated as part of the assessment of the objective justifications.”5

In our view, the CJEU’s clarifications of Art. 102 of the TFEU in the present case confirm that these three steps may indeed be intertwined.

Let us now take a closer look at each of the five questions and the CJEU’s responses. For each question and response, we will also share our thoughts on the potential impact of the CJEU’s ruling on the future enforcement of Art. 102 of the TFEU by the Commission or the national competition authorities in the EU Member States.

a) Competition on the Merits (Question 1)

Question

The Referring Court’s first question was based on a legal premise as well as a factual premise. The CJEU accepts both premises as a basis for the requested guidance on Art. 102 of the TFEU.

The legal premise was that Google’s refusal to ensure interoperability between its Android Auto platform and third-party apps (like JuicePass) implies a refusal to give third parties (like Enel X) access to the dominant company’s own (digital) infrastructure and that it can therefore be compared to the refusal at issue in the Bronner case.6

In Bronner, a dominant publisher of a newspaper had refused to give a rival competitor access to its nationwide home-delivery distribution network. The CJEU had ruled that, for that refusal to be found abusive, access to the infrastructure must inter alia “in itself be indispensable to carrying on that person’s business, inasmuch as there is no actual or potential substitute in existence for that home-delivery scheme” (emphasis added).7

The factual premise behind the referring court’s question was that access to Google’s digital infrastructure was not indispensable within the meaning of Bronner (because users of Android smartphones can always consult an app like JuicePass on the screen of their phones before driving or while driving their car. Displaying the app on the infotainment screen in their car would merely make it more attractive to consumers as it would be more convenient to use while driving.

On that basis, the Referring Court asked the CJEU whether “the specific characteristics of the functioning of digital markets justify departing from the conditions laid out in [Bronner] or, at the very least, interpreting them in a flexible manner” and, more specifically, whether “it [is] sufficient that access be indispensable for a more convenient use of the product or service offered by the undertaking requesting access, especially where the essential function of the product that is the subject of the refusal to supply is to make it easier and more convenient to use existing products or services” (emphasis added).8

In our view, this question essentially relates to the first step in the above-mentioned methodological analysis (i.e., competition on the merits): why should Google be obliged to assist its competitors by opening up the Android Auto platform to them if there is already an accessible market for their products without the access to the Android Auto platform?

The CJEU Response

The CJEU explains that its answer ultimately depends on the question of whether Google “has developed [the] infrastructure not solely for the needs of its own business but with a view to enabling third-party undertakings to use that infrastructure.”9

In the latter case, “the fact of requiring the undertaking in a dominant position to provide access to that infrastructure to a third-party undertaking does not fundamentally alter the economic model which applied to the development of that infrastructure.” “[I]n such a situation, neither the preservation of the freedom of contract and the right to property of the undertaking in a dominant position nor the need for that undertaking to continue to have an incentive to invest in developing high-quality products or services justify limiting a refusal to provide access to the infrastructure in question to a third-party undertaking being classified as abusive.”10

In the present case, the CJEU opines — subject to the factual verification by the Referring Court — that “Android Auto was not developed by Google solely for the needs of its own business, since access to that digital platform is open to third-party undertakings” and that, as a consequence, Google’s refusal was “capable of constituting an abuse of a dominant position even though that digital platform is not indispensable for the commercial operation of the app concerned on a downstream market.”11

Impact

The Commission will likely interpret the CJEU’s ruling as a confirmation of its view, as set out in its draft Guidelines, that “access restrictions can be liable to be abusive even if the input at stake is not indispensable, as the need to protect the undertaking’s freedom of contract and incentives to invest does not apply to the same extent as in a refusal to supply setting ….” More specifically (with reference to the General Court’s judgment in Google Shopping), an access restriction would be abusive “where the dominant undertaking develops an input for the declared purpose of sharing it widely with third parties but later does not provide access or restricts access to that input.”12

This reminds us of a similar observation by the General Court in the Microsoft case when it found no objective justification for Microsoft’s refusal to ensure interoperability between its Windows operating system and that of competitors in the work group servers market. According to the General Court, “it is normal practice for operators in the industry to disclose to third parties the information which will facilitate interoperability with their products and Microsoft itself had followed that practice until it was sufficiently established on the work group server operating systems market” and that “such disclosure allows the operators concerned to make their own products more attractive and therefore more valuable.”13

b) Restrictive Effects (Questions 2 and 5)

Questions

With its second question, the Referring Court wished to know whether the fact that Google’s refusal to grant interoperability had not stopped Enel X and other competitors from growing on the market to which their app belonged, “is such as to indicate in itself that the refusal by the undertaking in a dominant position to act on that request was incapable of having anticompetitive effects” (emphasis added).14

With its fifth question, the Referring Court was asking whether, assuming Google’s refusal was capable of having anticompetitive effects, “a competition authority is required to define the downstream market on which that refusal is capable of having anticompetitive effects, even if that market is only a potential market.”15

As will be discussed below, the CJEU’s answers to both questions seem pretty mainstream. However, its answer to the second question contains a “wrinkle” which we believe is worth noting.

The CJEU Responses

The CJEU’s short answer to the second question is that a refusal to ensure interoperability can be abusive even if the party requesting the interoperability has been able to grow its market position in the absence of such interoperability.

Referring to its judgment in SEN,16 the CJEU opines that “the maintaining of the same degree of competition on the market concerned, or even the growth of competition on that market, does not necessarily mean that the conduct in question is incapable of having anticompetitive effects …” and it concludes that the continued presence or even the increased presence of Enel X and its competitors “does not in itself mean that Google’s refusal to grant access to Android Auto was incapable of having anticompetitive effects.”17

However, the CJEU adds that the continued or increased presence “may nevertheless constitute evidence that Google’s conduct at issue in the main proceedings was incapable of having the alleged exclusionary effects” (emphasis added) and that “any evidence adduced to demonstrate the attractiveness to users of electric motor vehicles of an app such as JuicePass … may be relevant, despite the fact that that app could not be used on those vehicles’ infotainment systems via Android Auto” (emphasis added).18

With regard to the Referring Court’s fifth question (i.e., whether a precise definition of the relevant market is needed for the purpose of locating the actual or potential anticompetitive effects of the dominant company’s conduct, is required), the CJEU opines that this is not necessary because “it is sufficient that a potential or even hypothetical market can be identified,” for instance where that market “is still developing or is evolving rapidly and, therefore, its scope is not fully defined on the date on which the undertaking in a dominant position implements the allegedly abusive conduct.”19

Impact

At first sight, the CJEU’s response to the Referring Court’s second question is not groundbreaking. Well before the SEN case, the General Court had already held in the British Airways (BA) case that “the growth in the market shares of some of BA’s airline competitors, which was modest in absolute value having regard to the small size of their original market shares, does not mean that BA’s practices had no effect” and that “in the absence of those practices, it may legitimately be considered that the market shares of those competitors would have been able to grow more significantly.”20

Not surprisingly, the Commission considered this point in its draft Guidelines by noting that the fact that “a faster or more significant decline in the dominant undertaking’s market share may have been prevented” can be evidence that the company’s practice is at least capable of having exclusionary effects.21

However, it would seem to us that the Commission should now reflect on how to interpret the CJEU’s observation that, while JuicePass’ continued or (even) increased its presence does not in itself prove that Google’s refusal to grant access was incapable of having anticompetitive effects, the “attractiveness” of JuicePass to Android smartphone users, even if the JuicePass app could not be used on the infotainment screens, “may be relevant.” Clearly, in the CJEU’s eyes, it is a factor that may indicate that Google’s refusal was incapable of having anticompetitive effects. Its refusal only prevented users of JuicePass from consulting that app on a larger screen when driving their vehicle. It will be interesting to see how the Consiglio di Stato will deal with this suggestion and how the Commission will interpret it in the final text of its Guidelines.

Assuming the refusal did have the capacity to restrict competition, a separate issue — not addressed by the CJEU (because it was not raised by the Referring Court) — was whether the refusal’s anticompetitive effect was appreciable enough to raise concern.

Of course, in the Post Danmark II case, the CJEU already observed that “fixing an appreciability (de minimis) threshold for the purposes of determining whether there is an abuse of a dominant position is not justified” and the Commission referred to it in its draft Guidelines.22 The CJEU also firmly rejected a de minimis test in SEN — a case ironically involving Enel and its subsidiary EE when they were “on the receiving end” in another case handled by the AGCM.23 In that case, the CJEU had observed that “the fact (on which the companies concerned rely in order to dispute the existence of abuse of a dominant position) that EE obtained, by means of the use of the SEN lists, just 478 clients, that is to say, 0.002% of the customers in the protected market, cannot be regarded as sufficient of itself to show that the practice in question was not capable of producing an exclusionary effect.”24

That said, it is our understanding that Google developed a de minimis argument in the present case. At the time of Enel’s request, only 0.04% of all Italian cars were electric. Many Italian electric car drivers used an iPhone (not an Android smartphone) and not every car driver owning an Android smartphone used the JuicePass app. Moreover, many owners of an Android smartphone who did have the JuicePass app, only used this app on their smartphone screens instead of connecting it with Android Auto, even after Google had developed the template that would have enabled them to connect their smartphone to their car’s infotainment system.25

The CJEU’s “no” answer to the fifth question of whether it is necessary to define precisely the downstream or neighboring market where the competitive impact of the dominant company’s conduct is taking place, will likely be welcomed by the Commission.

In its draft Guidelines, the Commission does not address this issue. When observing that “to assess dominance, it is generally necessary to define the relevant market,” it indeed only refers to an upstream market where the dominant power must have sufficient market power to leverage its position elsewhere.26

In contrast, in its 2024 Notice on the definition of the relevant market, the Commission has already indicated that a precise market definition for the purpose of assessing the negative impact on competition of mergers or conduct should not be required in cases where the contours of any such market(s) are still uncertain.27 With reference to the Dow/DuPont merger case, it noted that “… in some cases, an R&D process may not (yet) be closely related to any specific product but related to early stages of research, which may serve multiple purposes and, in the longer term, feed into various products” and that “although the fact that such early innovation efforts do not immediately translate into tradeable products may render it difficult to identify a relevant product market in the strict sense, it may still be relevant to identify the boundaries within which undertakings compete in such earlier innovation efforts, in order to assess whether there could be a loss of innovation competition due to a concentration or behaviour.”28

c) Objective Justification (Questions 3 and 4)

Questions

The CJEU examines the Referring Court’s third and fourth questions together. According to the CJEU, the Consiglio di State essentially wishes to know whether the absence of a template for interoperability — at the time of Enel X’s request — objectively justifies Google’s refusal to ensure interoperability with the Android Auto platform or whether Google has an obligation to develop that template and, in the latter case, whether it must offer that template within a certain timeframe.

The CJEU Responses

The CJEU starts with a reminder that it is up to a dominant company to provide an objective justification for its anticompetitive conduct.29

It then shifts gears by taking a closer look at the facts at hand. According to the CJEU, the absence of a template for the category of apps concerned can only be objectively justified “where to grant such interoperability by means of such a template would, in itself and in the light of the properties of the app for which interoperability is sought, compromise the integrity or security of the platform concerned, or where it would be impossible for other technical reasons to ensure that interoperability by developing such a template” (emphasis added).30

This sets a high standard of proof. If the dominant company fails to demonstrate that the security of its own platform is at risk or that it is technically impossible to develop the template, it indeed has an obligation to develop one: “… the fact that there is no template for the category of apps concerned or the difficulties involved in its development which the undertaking in a dominant position may face cannot in themselves constitute an objective justification for that undertaking’s refusal to grant access.”31

However, the CJEU acknowledges that the dominant company must be given “a reasonable period of time to that development” and it identifies three factors that should assist the Referring Court in determining when the delay would be reasonable: “(i) the degree of technical difficulty in developing the template …, (ii) constraints related to the fact that it is impossible for it to equip itself, within a short time, with some of the resources, in particular human resources, necessary to develop that template …, or even (iii) constraints external to the undertaking in a dominant position which have an impact on its ability to develop that template, such as, for example, constraints relating to the applicable regulatory framework.”32

In other words, while mere difficulties (due to technical issues, challenges in allocating human resources internally or external regulatory constraints) in developing the required template do not exempt Google from an obligation to develop the template, the dominant company can invoke these difficulties to “buy” reasonable time to comply with its obligation.

The CJEU also acknowledges that the development of such a template “is likely to represent a cost” for the dominant company and, as a consequence, that Art. 102 of the TFEU “does not preclude [it] from requiring an appropriate financial contribution from the undertaking which requested interoperability,” provided it is “fair and proportionate” and allows it “to derive an appropriate benefit from it.”33

Impact

The CJEU’s observations on the objective justification issue are rather unusual insofar as they are more closely connected to the facts at hand than in most other preliminary rulings on Art. 102 of the TFEU. In such rulings, the CJEU usually confines itself to mentioning pro memori that dominant companies have the possibility to bring forward a justification for their alleged abuse. Here the CJEU has gone a significant step further by identifying factual circumstances that may justify Google’s refusal — at least temporarily — to ensure interoperability between Enel X’s JuicePass app and Google’s Android Auto platform.

In that sense, this ruling reminds us of the CJEU’s preliminary ruling in the Lelos case. In Lelos, the CJEU indicated that a dominant pharmaceutical company could not flatly stop sales of its medicines to Greek wholesalers who wished to buy these medicines in order to export at least in part to high-price countries elsewhere in Europe. However, it also indicated that the pharmaceutical company should have the possibility of operating a sales quota system in order to limit at least parallel exports of its medicines out of Greece and thus secure sufficient availability of these medicines for Greek patients.34

However in the Lelos case, the CJEU’s guidance for the referring court strikes us as more high level. It observed that “it is for the referring court to ascertain whether the abovementioned orders are ordinary in the light of both the previous business relations between the pharmaceuticals company holding a dominant position and the wholesalers concerned and the size of the orders in relation to the requirements of the market in the Member State concerned” (emphasis added).35

It will be interesting to see how the Consiglio di Stato will follow up and apply the CJEU’s response to the facts at hand in the case before it. In this regard, it is unclear (but important to know) to what extent Google already relied — before the AGCM or later, when it appealed the AGCM’s decision — on any or all of the three factors mentioned by the CJEU as relevant factors that may justify the delay in developing the template that Enel required for the interoperability between its app and Google Auto.

Leaving aside the final outcome of the Italian case, it would seem to us that the Commission will have to clarify the scope of the “objective necessity” test in light of the CJEU’s judgment.

In its 2009 Guidance paper on enforcement priorities in applying Art. 102 of the TFEU to abusive exclusionary conduct by dominant undertakings, the Commission had already accepted “objective necessity” as a defense, but it had observed that this necessity must be linked to “factors external to the dominant undertaking” (emphasis added).36

In its 2024 draft Guidelines, the Commission acknowledges that the objective necessity may also stem from factors internal to the dominant company. It explicitly mentions (1) ”legitimate commercial considerations” (as accepted in the Lelos case), (2) ”technical specifications,” including those “linked to maintaining or improving the performance of the dominant undertaking’s product” (although specifications linked to performance may fit better under an efficiency defense), and (3) public health, safety, or other public interest considerations (but with the caveat — for safety-related considerations — that “it is not the dominant undertaking’s task to take steps on its own initiative to eliminate products which, rightly or wrongly, it regards as dangerous or as inferior in quality to its own products,” as the General Court held in the Hilti case and the Tetra Pak case).37

In our view, the Commission will have to elaborate on the current passage in the draft Guidelines concerning “technical specifications.” First, the CJEU has held that a dominant company’s refusal to give access to an infrastructure that it designed to be open is not abusive if giving access is technically impossible in a specific case. In practice, this will likely be hard to prove, but in principle, it creates an additional opportunity for dominant companies to justify their conduct. Second, the CJEU has identified three factors (i.e., technical difficulties, allocation of scarce human resources, and certain external constraints) that may justify a “reasonable” delay in a third party’s access to the dominant company’s infrastructure. In practice, this will likely give dominant companies more leeway.

* * *

Most press reports have presented the CJEU’s judgment as a loss for Google (or a win for the AGCM).

This may be true insofar as the CJEU has ruled that the strict indispensability requirement under Bronner does not apply when the dominant company has developed an infrastructure that is, in principle, meant to be accessible to its competitors. However, we believe that other parts of the CJEU’s judgment may give dominant companies some fresh ammunition in defending themselves in alleged abuse cases.

First, with regard to the question of whether Google’s refusal to ensure interoperability between its Android Auto’s platform and Enel X’s JuicePass App had produced exclusionary effects, the CJEU has identified the growth of Enel X’s market position notwithstanding Google’s refusal, in combination with various attractive functionalities in its JuicePass App (which might actually have generated this growth), as a relevant factor for the assessment of the alleged exclusionary effects of Google’s refusal. In our view, this significantly qualifies the contrary, and rather principled, observations about the irrelevance of rival competitors’ growth in the British Airways case (by the General Court) and in the SEN case (by the CJEU).

Second, with regard to the question of whether Google’s refusal could be objectively justified, the CJEU has identified three factors that might have justified Google’s delay in delivering the interoperability template requested by Enel X. The timeframe was an important issue, given that the AGCM had considered Google’s delay as unreasonably long. In addition, the CJEU also opined that Google was, in principle, entitled to a fair remuneration for its investment in developing such a template.

It will therefore be interesting to see how the Consiglio di Stato will pick up the CJEU’s guidance on both points when delivering its ruling in the case at hand.

© Arnold & Porter Kaye Scholer LLP 2025 All Rights Reserved. This Advisory is intended to be a general summary of the law and does not constitute legal advice. You should consult with counsel to determine applicable legal requirements in a specific fact situation.

  1. Judgment of February 25, 2025 in case C-233/23, Alphabet Inc., Google LLC and Google Italy Srl v. Autorità Garante della Concorrenza e del Mercato. On September 5, 2024, Advocate-General Medina published her opinion in this case.

  2. Draft text was published on its website on August 4, 2024, inviting written comments on that draft by October 31, 2024. A conference to collect oral comments was held on February 13, 2025.

  3. See the Commission’s draft Guidelines, paragraphs 45 and 48.

  4. Id., paragraph 46.

  5. Id., paragraph 58.

  6. Judgment of November 26, 1998 in case C-7/97, Bronner v. Mediaprint.

  7. Id., paragraph 41. In Bronner, the CJEU opined that this was “certainly not the case” (paragraph 42) and it explained why (cf. paragraphs 43-46).

  8. Judgment of February 25, 2025, cit. note 1 at paragraphs 23 and 24.

  9. Id., paragraph 44.

  10. Id., paragraphs 44, 46 and 45 (quoted in that order). With regard to a dominant company’s freedom of contract, right of property, and incentives to invest in efficient facilities, see the CJEU’s preliminary thoughts in paragraphs 41 and 42.

  11. Paragraphs 49-50.

  12. Draft Guidelines, cit. note 2, paragraphs 165 and 166 (d). In its judgment of November 10, 2021 in the case T-612/17, Google and Alphabet v. Commission, the General Court observed “the fact (…) that Google favours its own specialised results over third-party results, which seems to be the converse of the economic model underpinning the initial success of its search engine, cannot but involve a certain form of abnormality” (paragraph 177) and therefore required a justification.

  13. Judgment of September 17, 2007 in case T-201/04, Microsoft v. Commission, paragraph 702.

  14. Paragraph 53.

  15. Paragraph 82.

  16. Judgment of May 12, 2022 in case C-377/20, Servizio Elettrico Nazionale, ENEL SpA, Enel Energia SpA v. AGCM.

  17. Judgment of February 25, 2025, cit. note 1, paragraphs 58 and 59. In SEN, the CJEU had observed that the absence of effect (in the course of the period during which the conduct of the dominant company had occurred) did not disprove that conduct’s potentially restrictive effect because the absence of an actual effect “could stem from other causes and be due to, inter alia, changes that occurred on the relevant market since that conduct began or to the fact that the undertaking in a dominant position was unable to complete the strategy underpinning that conduct” (paragraph 54).

  18. Id., paragraphs 59 and 60.

  19. Paragraph 85.

  20. Judgment of March 15, 2007 in case T-219/99, British Airways v. Commission, paragraph 298. In SEN, the CJEU had repeated this: “it cannot be ruled out that, in the absence of that conduct, competition on that market could have grown even further” (cit. note 16, paragraph 54).

  21. Cit. note 2, paragraph 70 (g).

  22. Judgment of October 6, 2015 in case C-23/14, Post Danmark v. Konkurrencerådet, paragraph 73. See the Commission’s draft Guidelines, paragraph 75.

  23. Judgment of May 12, 2022, cit. note 16.

  24. Id., paragraph 57.

  25. Cf. a post published by Tero Louko (senior EMEA competition counsel at Google) on LinkedIn.

  26. Draft Guidelines, cit. note 2, paragraph 20.

  27. Notice on the definition of the relevant market for the purposes of European Union competition law, O.J. 22.02.24 C/2024/1645.

  28. Id., paragraph 92.

  29. Judgment of February 25, 2025, paragraphs 70-71.

  30. Id., paragraph 73.

  31. Id., paragraph 74.

  32. Id., paragraphs 74 and 75.

  33. Id., paragraph 76.

  34. Judgment of September 16, 2008 in joined cases C-468/06 to C-478/06, Lelos and others v. GSK. See, in particular, paragraph 71: “Thus, although a pharmaceuticals company in a dominant position, in a Member State where prices are relatively low, cannot be allowed to cease to honour the ordinary orders of an existing customer for the sole reason that that customer, in addition to supplying the market in that Member State, exports part of the quantities ordered to other Member States with higher prices, it is none the less permissible for that company to counter in a reasonable and proportionate way the threat to its own commercial interests potentially posed by the activities of an undertaking which wishes to be supplied in the first Member State with significant quantities of products that are essentially destined for parallel export.”

  35. Id., paragraph 73. It borrowed the notion of “ordinary” orders from its judgment in United Brands. See its judgment of February 14, 1978 in case 27/76, United Brands v. Commission, paragraph 182 where it observed that a dominant company “cannot stop supplying a long standing customer who abides by regular commercial practice, if the orders placed by that customer are in no way out of the ordinary.”

  36. Guidance Paper, O.J. C 45 of February 24, 2009, p. 7, paragraph 29.

  37. Draft Guidelines, paragraph 168.