The General Court of the European Union on September 9, 2026 dismissed Booking Holdings' action against the European Commission's 2023 prohibition of its planned takeover of Etraveli Group, ruling that the flight booking business would have entrenched Booking's dominance over hotel sales made through online travel agencies - even though the judges accepted that the extra slice of that market the deal would deliver might be "limited to a few tenths of a per cent".

In Short

A European court agreed that Booking, which already handles most hotel stays sold through travel websites in Europe, was rightly stopped from buying Etraveli, one of the region's big online sellers of plane tickets. The judges accepted that owning Etraveli would let Booking turn your flight search into a hotel sales pitch, which makes it harder for smaller travel sites to grow. The prohibition stays in place, and regulators now hold a ruling that says even a tiny increase in a dominant platform's position can be enough to stop a deal.

A ruling built around a small number

Case T-1139/23 was decided by the Tenth Chamber of the General Court, sitting in an extended formation of five judges: President M. van der Woude, M. Jaeger, L. Madise, P. Nihoul and S. Verschuur, who acted as rapporteur. The hearing took place on July 8 and 9, 2025, and the language of the case was English. The operative part is brief. The action is dismissed, Booking Holdings bears its own costs and those of the Commission, and Etraveli, which intervened in support of Booking, bears its own.

What makes the 81-page judgment unusual is how much of the Commission's quantitative work it takes apart before arriving at that result. The Court found that the Commission's estimate of how much Booking's hotel market share would rise after the deal was "vitiated by a number of errors", that the Commission had failed to examine one of Booking's arguments about how hotels manage multiple sales channels, and that it had not shown hotels would shift inventory or end relationships with rival agencies. None of that changed the outcome. The judges held that in a market this concentrated, the deal's effect on contestability was enough on its own.

The contested measure is Commission Decision C(2023) 6376 final of September 25, 2023, in Case M.10615. Under it, Booking Holdings, based in Norwalk, Connecticut, was barred from acquiring sole control of Flugo Group Holdings AB, the Uppsala-based company trading as Etraveli Group. The transaction excluded Etraveli's subsidiary Flightmate AB, which runs the Flygresor metasearch business mainly in Sweden. The deal was valued at about 1.63 billion euros, according to Reuters. Booking told the news agency on the day of the ruling that it disagreed with the outcome on the facts and the law and was reviewing the judgment and a possible appeal.

How two complementary agencies ended up in Luxembourg

Both companies are online travel agencies, or OTAs, which the judgment describes as online retailers that resell flights, accommodation and car rental supplied by travel service providers, earning most of their revenue from commissions charged to those providers. Booking runs several such agencies, including Rentalcars, Priceline and Agoda, but is mainly active in accommodation under the Booking.com brand. It also supplies its hotel content to rival agencies that lack their own. Etraveli sells flights through brands including Gotogate, Mytrip, Flightnetwork, travelstart, Seat24 and SuperSaver.

The relationship predates the deal. Cooperation began on January 10, 2019 under a simplified contract that redirected flight searches from Booking's platform to Etraveli's. A "Phase 1" agreement from August 27, 2019 let Booking integrate Etraveli's flight content through an application programming interface and sell it under its own brand. The "Phase 2" agreement, signed on June 9, 2021, allowed Booking for the first time to sell Etraveli's content through metasearch services such as Google Flights and Skyscanner, subject to a limit on volumes that the judgment calls the "meta cap". According to the parties, Etraveli cannot unilaterally terminate or renegotiate that agreement until 2028.

The judgment is not internally consistent on the date of a later amendment. Paragraph 7 states that Booking and Etraveli agreed the amendment on August 10, 2023, while paragraph 14 says Booking informed the Commission on August 15, 2023 that the parties had agreed it, and paragraph 37 refers to an amendment "dated 15 August 2023". A similar gap appears in the procedural history: paragraph 12 records Booking's response to the statement of objections on June 24, 2023, while paragraph 340 cites a "response dated 23 June 2023".

The route to Brussels was, in a sense, of Booking's own making. The transaction did not meet the EU Merger Regulation's turnover thresholds and could have been reviewed under national law in Germany, Cyprus and Austria. The parties asked the Commission to take the case on February 14, 2022, and the referral was accepted on March 9, 2022. The deal was notified on October 10, 2022, an in-depth investigation opened on November 16, 2022, and a statement of objections followed on June 9, 2023. Booking offered commitments on July 31, 2023 and a revised set on August 25, 2023. The Advisory Committee met on September 13, 2023, the Hearing Officer reported on September 15, and the Commission adopted its prohibition ten days later.

The market the judges accepted

The Commission defined two markets, hotel OTAs and flight OTAs, each covering the entire EEA. In 2022, Booking held an estimated 60-70% of the hotel OTA market in both its business-to-business and business-to-consumer segments. Expedia followed with 10-20% on the business side and 5-10% on the consumer side. HRS, Airbnb, OYO, Travelminit, Weekendesk, Trip.com, Lastminute, TUI, eDreams Odigeo, Kiwi and Tix held what the Commission called modest shares.

Flights looked different. eDreams Odigeo led the 2022 flight OTA market with 20-30%. Etraveli held 10-20% and Booking 5-10%, which together made them the second-largest player, alongside Trip.com at 10-20%, Lastminute, Kiwi and Travelgenio at 5-10% each and Expedia at 0-5%. Those figures sit awkwardly with the Court's own press release, which describes Etraveli as Europe's leading online flight booking operator. The judgment itself is more careful: it calls Etraveli "a leading flight OTA" and accepts the Commission's projection that the deal would make the combined business the market leader by 2026.

Booking attacked its hotel market share on methodological grounds. The Commission had reconstructed the market from Booking's own estimates of March 11, 2023, which were in turn derived from an August 2022 Euromonitor forecast. In July 2023, Euromonitor revised its 2022 figure to roughly one and a half times the original size, which would have cut Booking's share substantially. The Court agreed it was "inconsistent" for the Commission to reject Euromonitor's figures while relying on them indirectly, but still sided with the Commission, noting that Euromonitor's broader "lodging" category inflated the market, that Euromonitor's numbers would put Booking's share above 100% in some member states, and that Booking could not identify the agencies supposedly making up an implausibly large "Others" category - despite being the main supplier of hotel content to smaller agencies. Industry experts and rival OTAs had independently put Booking's share at 60-70% or more.

Other findings survived intact. Booking charges hotels a higher average effective commission than its rivals, based on revenue and transaction data from companies accounting for about 70% to 80% of the market. In one survey of accommodation partners cited by the Commission, the number of respondents describing Booking as an expensive platform was more than ten times the number saying the same of Expedia. Booking had argued that hotels' own websites, boosted by metasearch services such as Google Hotels, exert real pressure. The Court was unpersuaded: Booking's share of bookings grew even as direct sales and Google Hotels gained ground, and hotels' own websites stayed consistently cheaper without Booking feeling compelled to match them.

The Court also accepted the Commission's reading of customer behaviour. Booking surveys run in 2020 and 2021, each with 500 respondents in Germany, Spain, France, Italy, the United Kingdom and the United States, showed a substantial share of users going straight to their usual booking site without comparing prices. For its definition of multi-homing - using several competing digital services in parallel - the Court cited its July 2024 judgment dismissing ByteDance's challenge to its gatekeeper designation under the Digital Markets Act, a case it relied on three times in the Booking judgment.

Reverse leveraging and the Google Shopping precedent

Booking's first line of attack was doctrinal. The Commission's 2008 Guidelines on non-horizontal mergers focus on foreclosure through tying, bundling and similar conduct, in which a firm uses a strong position in one market to gain ground in another. The Commission's theory ran the other way: Booking would use a non-dominant position in flights to reinforce an already dominant position in hotels. Booking called that a "novel and unsubstantiated theory".

The Court disagreed, and it reached for the Google Shopping case to do so. Citing its own November 2021 judgment in that case, which the Court of Justice confirmed in September 2024 when it upheld the 2.4 billion euro fine against Google, the judges wrote that "leveraging is a generic term in relation to the impact which a practice identified on one market may have on another market", and that the potential harm "is essentially the same, irrespective of the direction of the leveraging effect". It would be "contradictory", they added, to let the Commission act only where leveraging strengthens a non-dominant position and not where it strengthens a dominant one. The guidelines, the Court noted, were adopted before the Commission had experience of such concerns and leave room for theories of harm beyond those they list, which matters in digital markets with features the 2008 text did not anticipate.

Booking's second doctrinal argument was that cross-selling hotel rooms to flight customers is simply competition on the merits: a convenient one-stop shop that customers are free to ignore. The Court conceded the point of fact. Such cross-selling "does not involve any commercial pressure on customers", it said, and differs from tying or bundling. But the concept of competition on the merits, developed under Articles 101 and 102 of the Treaty, "does not play a similar role in the context of merger control", because "the implementation of a concentration cannot be regarded as a form of competition on the merits". Merger control looks at the structure of competition, not at whether the merged entity will later misbehave - a point the Court again supported with a paragraph from Google Shopping. The practical effect is to separate the question of whether conduct is lawful from whether a structural change is: a dominant platform can be barred from buying a business whose integration would itself raise no self-preferencing or tying concern.

Flights as a data channel for hotel advertising

For the advertising industry, the most concrete part of the ruling is its account of how a flight booking becomes a hotel sale. The Commission described flights as the "entry point" of a trip and the travel service most likely to be booked on the same website as accommodation. A traveller searching for a flight hands over a profile and a planned journey. With a logged-in account, the agency gains an email address and booking history and can show tailored hotel offers by email, on its own site, "or elsewhere whilst the customer was surfing the internet". App users can be reached with notifications. A completed flight booking adds confirmed dates, more personal details and a reason to return to the platform to check the reservation.

Booking argued that most flight customers book hotels separately and often much later. The Court treated that as beside the point: the agency still holds data on where and when the customer will stay, which it can use to target an offer for a later booking. The cross-selling opportunity, the Commission said and the Court accepted, is "not limited to the actual moment of a sale of the flight".

The numbers behind that opportunity were contested. Booking pointed out that rival agencies selling both flights and hotels reported attach rates - the share of customers who add a second product after buying a first - below 15%, and that only a small share of traffic on the ten main hotel OTA platforms comes from flight sites or airlines. The Court found Booking's own attach rate for new customers barely above that level, but much higher for existing customers, around 20-30% of whom Booking classifies as "high value". Internal documents described the travel ecosystem as making the platform "stickier". Etraveli called itself the "#1 flight platform" with a "#1 position" in "Meta traffic"; Booking described it as "best-in-class at combining flights and ancillary content from multiple suppliers".

Paid media runs through the barriers-to-entry analysis too. Among the parameters on which the Commission found Booking outperforms every rival were advertising strength and the ability to attract online traffic. Smaller agencies and new entrants, it found, face an "inability to generate online traffic to their platform by investing in online advertising and in search engine optimisation", and Booking has "significant strength in paid advertising".

Where the Commission's arithmetic failed

The Commission calculated the deal's effect on Booking's 2025 hotel market share using two methods derived from Booking's own transaction valuation model, drawn up in November 2021 to price Etraveli and its synergies. The "modified TVM methodology" ran in eight steps: estimate flight transactions available for cross-selling, apply an attach rate, apply a cannibalisation rate, multiply by nights per booking, add loyalty effects tied to flights, strip out non-hotel stays, convert the result into euros and divide by the size of the market. The second method used a figure from a slide in a Booking presentation, referred to as Figure 123. Both produced increments in the confidential range of 0-5%.

The cannibalisation rate is the step that removes hotel bookings Booking would have won anyway, with or without a flight offering - the same logic behind incrementality testing in advertising measurement. It proved decisive for one of the two methods. The Figure 123 method applied no such rate, and when asked at the hearing, the Commission "was unable to explain" why the two methods diverged so sharply given that both drew on Booking documents prepared in May 2022. The Court disregarded the Figure 123 result.

Three more corrections followed. The Commission had attributed to flights all the extra hotel bookings Booking credits to "Value Leadership", where a discounted taxi or car rental nudges a hotel booking, and "Halo Effects", the lower cancellation and higher rebooking rates of multi-product trips. According to the judgment, the vast majority of both effects involved non-flight services, and their growth was mainly explained by the post-pandemic recovery and by more taxi and car rental bookings attached to hotel stays. The Commission had also held the 2025 hotel market at its 2022 size, even though its own decision assumed inflation of 5.4% for 2023 and 3% a year for 2024 to 2026, and Euromonitor projected the 2025 market to be 53% larger than in 2022. And the calculation started from a "zero-flights" baseline, although the Commission accepted that Booking would sell flights in every realistic scenario.

Not every correction went Booking's way. The Court upheld the Commission's upward adjustment of flight volumes based on a "Go-to-Market model" Booking drew up in May 2022 for integrating Etraveli, rejected Booking's claim that the cannibalisation rate had to mirror its market share, and accepted the use of the total average room-night value on the basis that the connected trip strategy targets high-value customers. The net result was that the only surviving estimate was an overestimate, and the true figure might be a few tenths of a per cent. For comparison, the same court annulled a 1.49 billion euro fine in the AdSense case in September 2024 after finding gaps in the Commission's evidence. Here, the arithmetic errors were not fatal.

Clicks are not purchases

One thread runs through the judgment that measurement teams will recognise: the Court's repeated refusal to treat clicks and traffic as evidence of purchases. When Booking argued that little of the traffic reaching hotel platforms came from flight sites, the Court said traffic data "do not demonstrate the potential cross-selling of hotel rooms from flights" and are not representative of the money customers actually spend. When the Commission pointed to flight search views on Booking's platform, the Court applied the same standard in reverse, noting that "the websites which customers view or the offers on which they click when searching for a travel service do not reflect the actual purchases they make".

The same logic defeated Booking's efficiency case. Booking cited data from Etraveli's Swedish metasearch site, flygresor.se, showing that 87.3% of clicks select the lowest fare for an itinerary. The Court held that "proposals on which customers click when searching for flights on an MSS are not indicative of what they purchase", since users may click several options before deciding. Figures from Kayak, the metasearch service Booking operates, on the gap between the lowest and second-lowest fares were declared inadmissible because they had not been provided during the administrative procedure.

Why a sliver was enough

If the increment was so small, what justified the prohibition? The Court's answer rests on three linked ideas, all tied to market structure rather than to the size of the change.

The first is network effects. The more travellers an agency attracts, the more hotels list with it and the better their rates, which in turn attracts more travellers - a "self-reinforcing dynamic" the Commission described as difficult for smaller agencies and entrants to break. In a market with that dynamic and "a significant gap between the dominant leader and its main competitors", the Court held that "even a relatively small increase, in quantitative terms" in the leader's share can strengthen network effects and have "a chilling effect on the already weak competitive dynamics". It added that because cross-selling can happen before or after a flight sale, the real impact "is likely to exceed what that modest percentage suggests".

Hotel data supported the dependence. For most hotels, between 61% and 100% of bookings through online agencies come via Booking. Among independent hotels, 88% list on Booking and 61% on Expedia, but mostly on no other agency: 28% use one OTA, 39% use two, 17% use three and only 15% use more. In the Commission's market investigation, 63% of responding hotels said they would not change the inventory offered on Booking after the deal, 5% would reduce it, 4% would increase it and 27% would wait and see. Booking cited those replies as evidence that the deal would not change how hotels behave. The Court read it differently: few hotels had inventory left to move, and a 2019 Booking document had already concluded that "the ... [return on investment] of capturing the remainder appears low".

On one point, Booking won. It had argued that channel managers, software that lets hotels upload inventory once and distribute it to many agencies, make multi-homing cheap. The Court agreed with the Commission that a supporting study dated December 4, 2023 was inadmissible because it post-dated the decision, but found Booking had made the same argument during the administrative procedure and that the Commission "failed to examine" it. As a result, the Commission's evidence could not show that hotels would drop rival agencies. It was a procedural loss for the Commission that the Court then turned against Booking, observing that channel managers also let small agencies reach hotels, which makes those rivals more relevant to competition.

The second idea is residual competition. Trip.com, Lastminute, TUI and eDreams Odigeo each hold roughly 1% or less of the business side of the hotel market and rely heavily on other agencies' hotel content. But Booking itself had named them as "main rivals" in its application. The Commission's case was that these mainly flight-focused agencies could use flights to build hotel customer bases, and that Booking's expansion in flights would squeeze exactly that route. "The residual competition exerted by the few remaining competitors is of particular importance in a market with low levels of competition," the Court wrote.

The third is contestability. The Court held that the EU Merger Regulation does not require the Commission to show that a deal would reduce competition further. It can instead show that a deal would consolidate and perpetuate an already low level of competition, and that effect counts as significant. Post-deal, Booking would lead both hotels and flights, creating "a travel ecosystem which would be difficult for other OTAs to replicate", according to the judgment.

Philipp Westerhoff, counsel at Geradin Partners in Berlin, who was not among the lawyers listed in the case, wrote on LinkedIn on the day of the ruling that the judgment "is particularly interesting for its treatment of digital market dynamics in ex ante merger control" and "may also become an important reference point beyond merger control for the assessment of leveraging, network effects and contestability in digital markets". He noted that the Court "reached this conclusion even though it identified errors in the Commission's calculations". He also flagged a link the judges did not make explicit. "The recent Android judgment is not cited, but the parallel is notable," he wrote, adding that both rulings focus on how positions across complementary digital services and ecosystem effects can reinforce an existing dominant position. A search of the judgment's text confirms no reference to Android.

Harm to hotels, and efficiencies that did not count

The Commission never claimed that commission rates or room prices would rise. Its argument, which the Court accepted, was subtler: as Booking's position becomes harder to contest, its incentive to lower commissions and offer better terms declines. Because the deal would divert some bookings from rival agencies and from hotels' own websites, which are cheaper for hotels and generally show lower prices to travellers than Booking does, at least some hotels and their guests would pay more. Most hotels that expressed a view in the market investigation expected Booking to take at least 5% of sales from their direct channel.

Booking's efficiency defence failed on admissibility and evidence. The one-stop-shop benefit had not been presented as an efficiency during the administrative procedure, and an expert opinion supporting it was dated December 4, 2023, after the decision. For flight customers, Booking argued that owning Etraveli would remove a per-transaction fee and double mark-ups, lowering fares, and that rivals would follow. The Court found no evidence Booking intended to pass savings on: its November 23, 2021 board presentation, "Project Bahamas Board Materials", referred to lower prices only in general terms. A March 16, 2023 economic adviser's report assumed a rival price response without showing how that figure was derived. Finally, citing the Court of Justice's MasterCard ruling, the judges held that gains to flight customers could not offset harm to hotel customers because the two groups are not substantially the same.

What happens next

An appeal on points of law to the Court of Justice may be brought within two months and ten days of notification of the judgment. Booking has said it is considering that route. If no appeal is lodged, the judgment becomes final, leaving in place the General Court's endorsement of reverse leveraging, of qualitative evidence in merger assessments and of a theory of harm built on contestability rather than on an immediate loss of competition.

Why this matters for the marketing community

Booking has been a designated gatekeeper under the Digital Markets Act since May 13, 2024, and it already sits inside several enforcement stories PPC Land has followed. The Court of Justice ruled on September 19, 2024 that its parity clauses breached competition law, and on December 16, 2025 a Berlin court found that 1,099 hotels could claim damages from Booking.com. The Etraveli judgment adds a structural limit: a dominant platform's acquisitions of complementary services can now be blocked on the strength of data flows and ecosystem effects, even where the market share arithmetic is thin.

That matters for travel advertisers because the same market is being reshaped from the search side. Google's hotel search usage in Europe rose from 37% in 2013 to 80% in 2023, according to figures European travel technology groups put forward in 2025. Early DMA compliance changes to Google's hotel units were reported to increase hotels' reliance on intermediaries, and research from Mirai cited in a March 2026 coalition letter to the Commission found 30% fewer clicks and 36% fewer direct bookings for hotels in DMA markets. On July 23, 2026, the Commission fined Google 890 million euros, naming hotels among the verticals affected by search self-preferencing. A month later, Google added hotel checkout to AI Mode in the United States, with Booking.com among ten partners.

Against that background, the Court's acceptance that Booking's share kept growing despite the rise of Google Hotels and direct booking is significant. So is its finding that paid search and SEO capacity form part of the barrier facing smaller agencies. For programmatic buyers and retargeting vendors, the judgment's description of flight data feeding email, on-site and off-site hotel offers now sits inside a court-endorsed theory of competitive harm, not only inside a CRM playbook.

The reasoning also tracks the Court of Justice's July 2, 2026 Android judgment, which rejected the as-efficient-competitor test as a universal requirement in digital markets marked by network effects and high barriers to entry. One case is enforcement after the fact under Article 102; the other is merger review before the fact. Read together, as Westerhoff suggested, the two rulings point the same way: in digital markets, what an effect adds to an already entrenched position can weigh more than its size.

Timeline

Summary

Who: Booking Holdings, supported by Etraveli Group, against the European Commission, before the Tenth Chamber of the General Court sitting with five judges. Hotels, rival online travel agencies such as Expedia, Trip.com, Lastminute, TUI and eDreams Odigeo, and travellers in the EEA are the parties affected.

What: The General Court dismissed Booking's action for annulment of the Commission's 2023 decision prohibiting its acquisition of Etraveli, endorsing a reverse leveraging theory of harm, accepting qualitative evidence, and holding that consolidating an already low level of competition can amount to a significant impediment to effective competition, despite finding errors that may shrink the deal's effect on Booking's hotel market share to a few tenths of a per cent.

When: The judgment was delivered on September 9, 2026, following a hearing on July 8 and 9, 2025. The Commission's prohibition dates from September 25, 2023, and any appeal on points of law must be lodged within two months and ten days of notification.

Where: The General Court in Luxembourg, ruling on markets for hotel and flight online travel agencies covering the entire EEA.

Why: The Court accepted that flights are an entry point for trips and a source of customer data that Booking could use to sell hotel rooms, and that in a market with strong network effects, a large gap between Booking and its rivals, and few remaining routes to acquire hotel customers, even a small strengthening of Booking's position would make it harder to contest.