The EU Fines Google €890M: What the First DMA Fine Changes

The EU's first DMA fine against Google landed on July 23, 2026 — €890 million total, split across two separate infringement decisions: €460 million for self-preferencing Google's own shopping, hotel, transport, and sports results in Search, and €430 million for anti-steering restrictions on Google Play. Most coverage blends the two into one headline number. They are two distinct compliance problems, and they demand two distinct responses.

The fine itself is not the story — €890 million is small next to the roughly €8.25 billion in pre-DMA EU antitrust penalties Google has already accumulated. The story is the mechanism. Google has 60 days, to roughly September 21, 2026, to end both practices. Past that deadline, the Commission can impose periodic penalty payments of up to 5% of Alphabet's average daily worldwide turnover — a per-day, revenue-scaled lever, not a one-time cheque. And an appeal does not pause the clock.

This guide separates the two infringements precisely, explains the penalty mechanics most coverage muddles, places the fine on Google's decade-long EU timeline, maps which SERP verticals are likely to shift vertical by vertical, revisits what the 2024 compliance round actually did to traffic, and closes with a 60-day playbook for teams competing in the affected verticals.

The European Commission's announcement on July 23, 2026 covers two separate infringement decisions under the Digital Markets Act — the first fines the DMA has produced since the regulation's obligations took effect. Some outlets report the total as roughly $1 billion; the $525 million figure circulating in travel-trade coverage is simply the USD conversion of the €460M Search decision, not a separate penalty. Precision matters here because the two decisions target different products, different conduct, and different competitor sets.

Google's response came the same day from Kent Walker, President of Global Affairs, arguing the decisions force it to "strip away real-time Search features Europeans love — like instant pricing and direct availability for hotels, flights, and restaurants — and dismantle safety protections on Google Play." That is Google's characterization of the remedy, not a neutral description — but it is a useful tell about where Google expects the changes to land: the rich travel and local modules that currently dominate EU commercial SERPs. As of publication, Google says it is evaluating whether to appeal, with no commitment to litigate.

The fine amounts are backward-looking. The forward-looking mechanism is what should concern anyone planning around EU search: Google has 60 days from July 23 — roughly September 21, 2026 — to end both practices. If either continues past that date, the Commission can impose periodic penalty payments of up to 5% of Alphabet's average daily worldwide turnover, per day, until compliance. Two structural details make this sharper than any previous EU action against Google.

First, an appeal to the EU General Court does not automatically suspend the compliance obligation or the fine — Google must comply first and litigate afterward. A General Court ruling earlier in July 2026 reinforced the point, confirming that designated DMA gatekeepers cannot obtain pre-decision injunctions to delay enforcement. Second, this fine is not a first warning: per 9to5Google's timeline, it caps an enforcement track running since a March 2025 preliminary finding on Search self-preferencing, followed by a May 2026 rejection of Google's proposed remedy as unacceptable.

This is Google's first DMA fine, but its fourth major EU competition penalty. The pre-DMA record, documented across Bloomberg's antitrust timeline and elsewhere: €2.42 billion for Google Shopping self-preferencing in 2017, €4.34 billion for Android in 2018, and €1.49 billion for AdSense advertising restrictions in 2019 — roughly €8.25 billion combined. Notably, the Android fine was upheld by the EU's top court on July 2, 2026 at approximately €4.1 billion (about $4.7 billion, after a modest reduction during earlier appeals) — three weeks before this DMA decision, and a reminder that Google's EU appeals have largely failed.

Put on one timeline, the pattern is striking: the fines are getting smaller while the enforcement mechanism gets structurally more dangerous. Classic antitrust cases took years and ended in one-time penalties Google could absorb and appeal. The DMA compresses that into a 60-day compliance window backed by per-day penalties that scale with revenue and cannot be paused by litigation.

One more piece of context: Google was designated a DMA gatekeeper on September 6, 2023 — alongside Amazon, Apple, ByteDance, Meta, and Microsoft — with Search, Play, Shopping, Maps, YouTube, Ads, Chrome, and Android all listed as core platform services. Gatekeepers had six months to reach full compliance, meaning Google has operated under active DMA search and Play obligations since March 2024 — more than two years before this first fine landed. The DMA track now runs alongside the EU's other major 2026 tech-regulation front, the AI Act — two regulatory programs any team operating in European markets needs to track together.

The €460M Search decision names four verticals: shopping, hotels, transport, and sports. In each, the Commission found Google gave its own services more prominent placement — including at the top of the results page — with enhanced visuals and filters not extended to comparable third-party services. Commission Executive Vice-President Henna Virkkunen put it directly: "We found that Google harms businesses offering similar services, such as shopping or sports, by not granting them the same level of prominence on Google Search."

The stakes are clearest in travel. Skift's coverage surfaces the number that explains why the Commission acted: Google's share of hotel price-comparison usage rose from 37% in 2013 to 80% in 2023 — a 43-point gain over that ten-year window. EU Travel Tech — the industry group whose members include Booking Holdings, Expedia, Airbnb, GetYourGuide, Tripadvisor, and Skyscanner — publicly hailed the decision as a milestone. The table below maps each named vertical to the change to expect and what to monitor during the 60-day window.

Note what the table doesn't include: flights and restaurants appear in Google's own response — Walker's statement names "instant pricing and direct availability for hotels, flights, and restaurants" — suggesting Google expects the remedy to reach adjacent travel and local surfaces beyond the four verticals the decision names. If you compete anywhere in the travel funnel, treat the whole category as in scope for monitoring purposes.

This is not the first time Google has redesigned EU results under DMA pressure, and the last round left a data trail worth studying before assuming who wins. When Google rolled out its first DMA compliance changes in January–March 2024, hotel-distribution consultancy Mirai measured the impact on European hotels — and the initial numbers were dramatic: a 30% drop in Google Hotel Ads clicks and a 36% decline in direct bookings versus non-DMA markets, as the redesigned pages shifted prominence toward large intermediaries like Booking.com and Expedia rather than hotel-owned sites. But Mirai's follow-up analysis, after effects normalized, refined the net impact on overall direct bookings across Europe to just −0.8%.

The lesson for the current cycle: EU SERP redesigns produce a sharp, headline-grabbing swing toward intermediaries in the first weeks, followed by months of normalization. Layer on the tests already underway before this fine landed — the November 2024 "ten blue links" hotel format in Germany, Belgium, and Estonia, and the February 2026 tests giving competing vertical-search services default top placement in some EU markets — and the realistic expectation is not a single dramatic flip on September 21, but an acceleration of layout experiments that have been running for over a year and a half. Teams that panic-react to week-one numbers will misread the trend; teams that instrument now and measure across the full cycle will see where the visibility actually settles.

Here is the forward-looking question that matters more than the fine itself: this decision addresses self-preferencing in classic search results — the carousels, modules, and rich units of the "ten blue links" era. It does not resolve how the same self-preferencing rules apply to AI Overviews and AI Mode, a gap Skift flagged as a key unresolved question in its same-day coverage. That gap is enormous, because Google's most valuable self-preferencing real estate today is arguably inside the generative answer surfaces, not the shopping and hotel carousels the Commission just fined.

Consider the trajectory: AI Overviews are already reshaping SERP real estate across a growing share of queries, and an AI-generated answer that synthesizes hotel prices or product comparisons is functionally a Google-owned vertical service — just without the visual carousel that made classic self-preferencing easy to identify and regulate. If Google complies with this decision by demoting its classic modules while commercial intent migrates into AI surfaces the decision doesn't cover, the economic substance of self-preferencing could survive the remedy. Regulators elsewhere are already circling the same territory — the UK's parallel search-visibility rules under the CMA regime take up AI Overviews directly, where this DMA decision does not address them.

Our projection: the AI Overviews question becomes the center of the next DMA enforcement round. The Commission's July 16 orders — requiring Google to share Search data with rival AI services from January 2027 and open Android features to competing assistants — show it is already treating AI surfaces as DMA territory. A self-preferencing case aimed squarely at generative results is the logical next step, and teams building EU search strategy should assume the rules land there within the next enforcement cycle, not treat today's decision as the final word.

The compliance window is a measurement opportunity. Whatever layout Google ships by late September, the teams that benefit will be the ones with clean baselines captured before the changes land. Four positions, four moves:

Two practical notes on instrumentation. First, set up tracking for AI Mode visibility in Search Console alongside classic rank tracking — if the AI Overviews gap closes the way we project, you'll want the historical baseline. Second, treat this as a portfolio question, not a single-market one: layout changes have historically rolled out market-by-market (Germany, Belgium, and Estonia first in the 2024 hotel tests), so a per-country view will catch shifts weeks before an EU-wide aggregate does. If your team doesn't have the tooling or capacity to run this kind of monitoring, our agentic SEO service builds exactly this instrumentation, and our analytics engagements cover the measurement layer underneath it.