The Search Monopoly Case, the Remedies, and the Local Business Playbook

By Matt Griffin · Published 2026-07-16

United States v. Google LLC is the largest search antitrust case since the Microsoft litigation of the late 1990s, and it produced the strangest reaction we have seen in this industry: half the commentary declared that nothing changed, the other half declared that Google was finished. Neither reading survives contact with the actual court documents. This piece does two things, in order. First, it lays out the verified record of the case with dates and sources: what the court found, what it ordered, what it refused to order, and where the appeal sits as of mid-July 2026. Second, clearly separated from that record, it gives our analysis of what the outcome means for a local business in Ontario deciding where to spend its next visibility dollar. The short version of the analysis: the remedies loosen the pipes that kept discovery flowing through one engine, at exactly the moment AI assistants are multiplying the number of places a customer can ask for a recommendation. The businesses that win in that world are not the ones that reacted fastest to a headline. They are the ones whose facts read cleanly on every engine.

The case in one timeline, 2020 to mid-2026

The whole litigation compresses into six dated events. The Department of Justice and a coalition of state attorneys general filed suit in October 2020, alleging Google maintained its search monopoly through exclusionary distribution contracts, chiefly the multi-billion-dollar payments that made Google the default engine on Apple devices, Android phones and major browsers. The liability trial ran through 2023 and into 2024. On August 5, 2024, Judge Amit Mehta of the U.S. District Court for the District of Columbia ruled that Google had illegally maintained monopolies in general search services and general search text advertising. A separate remedies trial followed in 2025, and Judge Mehta issued his remedies opinion on September 2, 2025. The court entered the final judgment on December 5, 2025. Google filed its notice of appeal in January 2026, the government cross-appealed on February 3, 2026, and the case now sits before the D.C. Circuit Court of Appeals. Every claim in the sections below traces to one of those documents or to contemporaneous reporting on them, all listed in the sources at the end of this page.

What the court actually found in August 2024

The August 5, 2024 liability opinion found that Google violated Section 2 of the Sherman Act by maintaining its monopoly through exclusive distribution agreements. Judge Mehta's line, quoted in nearly every report of the decision, was blunt: "Google is a monopolist, and it has acted as one to maintain its monopoly." The mechanism the court identified matters more than the label. Google was not found liable for building a better search engine; the opinion accepts that Google's product quality is real. The violation was the web of default agreements, worth more than $26 billion in 2021 payments according to trial evidence, that locked competitors out of the placements where search habits form. Defaults are powerful because most users never change them, so paying to be the pre-set answer on every phone and browser effectively rented the front door of the internet.

That distinction, product quality versus distribution lock-up, is the thread that runs through everything the court did afterward. The remedies were never going to punish Google for being good at search. They were designed to unlock distribution, and that is exactly what they do.

What Judge Mehta ordered in September 2025

The September 2, 2025 remedies opinion, later formalized in the December 5, 2025 final judgment, ordered behavioural relief in four main parts. These are the ruling facts, stated before any interpretation.

The one-year cap on default agreements is the quiet centrepiece. It does not end the Google-Apple relationship; the two companies can and almost certainly will keep renewing. What it ends is the multi-year exclusive lock. Every year, the default slot on the devices where search habits form becomes contestable again, and any AI company with a competitive assistant and a chequebook can bid for placement that was structurally unavailable before.

What the court rejected, and why that matters

Reporting what a court refused to do is as much a part of the record as reporting what it ordered, and here the refusals were substantial. Judge Mehta rejected the forced divestiture of the Chrome browser, the remedy that drew the most headlines when the DOJ proposed it. He declined structural breakup generally, declined to ban default payments outright, and excluded advertising data from the sharing requirement. Markets read the September 2, 2025 decision as a Google win: Alphabet shares rose roughly 8 percent, per CNBC's contemporaneous reporting.

Why does the rejection list matter to a business owner in Brantford or Hamilton? Because it defines the tempo of change. A Chrome divestiture would have been an earthquake with immediate, unpredictable effects on how billions of people reach the web. The behavioural package the court actually ordered is a slow-release mechanism: contestable defaults, data flowing to rivals, syndication access. The competitive effects arrive over quarters and years, not overnight. Anyone selling you an urgent strategy overhaul on the back of this case has not read the judgment.

Where the appeal stands as of July 2026

Nothing in this case is beyond revision yet, and honest analysis has to say so. Google appealed both the liability finding and the remedies. The Department of Justice and 38 states cross-appealed on February 3, 2026, still pushing for the structural relief the district court declined. Google filed its opening brief at the D.C. Circuit on May 22, 2026, in case No. 26-5023, arguing among other things that the district court failed to distinguish harm to competitors from harm to the competitive process, and leaning on the D.C. Circuit's own 2001 Microsoft precedent, where that court reversed a district-court breakup in favour of narrower behavioural relief. As of this writing, July 16, 2026, oral argument has not been scheduled; observers expect argument in late 2026 or early 2027 and a decision in 2027, with Supreme Court review possible after that.

Two practical facts follow. First, the behavioural remedies, including the data-sharing obligations, are in effect while the appeal proceeds. Second, the appeal is two-sided: the D.C. Circuit could soften the remedies, but the cross-appeal means it could also strengthen them. Planning a business strategy around either outcome is speculation. Planning around the trend both outcomes share, which is more competition for the discovery moment, is not.

Our reading: the ruling meets the AI assistant wave

Everything above is the court record. What follows is Formative Digital's analysis, and we flag the boundary deliberately, because a research page that blurs ruling facts into agency opinion is not worth your trust.

These remedies would have mattered far less in 2015, when the only realistic beneficiary of a contestable default was Bing. They land in 2026, when OpenAI, Perplexity, Anthropic and Microsoft all operate assistants that answer the questions people used to type into a search box, and when Google's own AI Overviews appear on roughly half of queries. The one-year contract cap gives those AI companies a recurring shot at distribution. The data-sharing provisions lower the cost of building a competitive index. The court did not create the multi-engine world; usage trends were already building it. The judgment removes the contractual friction that was slowing it down.

We have been measuring this fragmentation directly. Our AI engine consensus gap study ran identical local-intent questions through four engines and found they frequently disagree about which businesses to name, because each engine retrieves from different sources and weighs different evidence. That disagreement is the practical face of the post-remedies world: there is no single gatekeeper whose favour settles whether you get found. The mechanics of how each surface selects its sources differ too, which we cover in AI search versus traditional SEO and in our comparison of AI search and Google written for business owners.

What this means for an Ontario local business

A necessary caveat first: this is an American judgment. Its remedies govern Google's U.S. distribution contracts, and Canada's Competition Bureau runs its own files on its own schedule. But the products are global, the default deals the judgment restructures cover the same iPhones and browsers sold in Ontario, and the AI assistants gaining ground on the discovery moment do not check passports. The competitive weather changes here too, just without a Canadian court order behind it.

So the sensible posture is neither panic nor dismissal. Keep the Google fundamentals that already work: an accurate Business Profile, reviews, local pages, structured data. Then add the engine-neutral layer that AI systems retrieve from: consistent name, address and phone data everywhere the engines look, schema markup machines parse with confidence, citations from sources the models already trust, and pages that genuinely answer the questions your customers ask. This is Vector 2, Anchor, and Vector 5, Cite, in our 12 Vectors methodology, and the reason those vectors exist is that the same work raises your odds on every engine simultaneously. None of it is wasted if the D.C. Circuit rewrites the remedies next year.

What not to do: the panic moves that waste money

Court decisions of this size generate a predictable crop of bad advice, and part of a research page's job is to name it. Do not abandon Google work; the judgment restructures Google's contracts, not its market share, which remains dominant in Canada. Do not buy a standalone "Bing optimization package"; when your entity data and content are sound, the secondary engines largely inherit the benefit. Do not pay anyone claiming special access to the court-ordered shared data; the sharing runs between Google and qualified competitors under Technical Committee supervision, not through agencies. And do not treat the case as settled; the cross-appeals mean the remedy set could move in either direction in 2027. The pattern behind every one of these mistakes is identical: converting a slow structural shift into a fake emergency, because emergencies sell retainers.

The honest framing for budget decisions is this: the case changes the probability distribution of where your next customer's question gets asked. It does not change what makes an answer engine, any answer engine, choose to name your business. Evidence does that. Structure does that. Consistency does that.

The work that compounds across every engine

Here is the operational conclusion we build client roadmaps around. In a single-gatekeeper world, an optimization strategy could live parasitically on one company's algorithm, and much of the SEO industry did exactly that. In the world this judgment accelerates, the only strategy that does not require predicting court outcomes is to become the kind of source every retrieval system prefers: a well-documented entity with verifiable claims, consistent data, structured markup and genuine topical depth. Princeton's GEO research (Aggarwal et al., KDD 2024) quantified the direction: content carrying statistics, quotations and source citations gained 30 to 40 percent visibility in generated answers. That finding is engine-mechanical rather than Google-specific, which is exactly why it survives the antitrust turbulence.

The standard qualifier belongs here: visibility outcomes depend on your industry, competition and existing digital presence, and no agency can promise how any engine will answer a given query. What we can describe is where the odds live. The full workstream, from entity validation through citation building to per-engine measurement, is laid out on our GEO services page, and the rest of our study library sits in the research hub. To talk through how the case applies to your specific market, our contact page is the direct route.