Google keeps its ad business: what the ruling changes
A judge let Google keep its ad tech and ordered behaviour changes instead. Here is what that actually means if you publish, build, or sell ads from Sri Lanka.
The Google ad business breakup ruling landed on Wednesday, 2 September 2026, and the headline is that there is no breakup. Judge Leonie M. Brinkema of the Eastern District of Virginia rejected the Justice Department's proposal to split Google's ad-tech tools apart, and instead ordered Google to adjust how it runs the business so competitors can get a look in.
I read the TechCrunch report on it. My honest first reaction: this ruling changes almost nothing for anyone reading this, and understanding why is more useful than the verdict itself.
🔍 What the court actually decided
Here is everything we know for certain, separated from everything we don't.
| Item | Status |
|---|---|
| DOJ's proposal to break apart Google's ad-tech tools | Rejected |
| Google keeps its advertising business | Yes |
| Remedy ordered | Google must "adjust its business practices to favor competitors" |
| Specifics of that remedy | Not provided in what has been made public |
| Full written ruling | Sealed for 14 days for redactions |
| Underlying finding | Ad tech ruled an illegal monopoly back in April 2025 |
Key takeaway: A court has now confirmed twice that Google broke the law and twice declined to change Google's structure. The finding of illegality and the consequence of illegality are two very different things, and only the second one affects your business.
Google's VP for regulatory affairs, Lee-Anne Mulholland, said the company was "very pleased the Court rejected the DOJ's proposal to break apart tools that help small businesses reach new customers and grow." Note the framing there. Small businesses are the shield in this argument, which is worth remembering the next time you are told a platform's scale exists for your benefit.
⚖️ This is the second identical outcome, not a surprise
Put the two big Google remedy decisions side by side and the pattern is hard to miss.
| Search case (Judge Amit Mehta) | Ad-tech case (Judge Brinkema) | |
|---|---|---|
| Ruling on liability | Illegal monopoly, 2024 | Illegal monopoly, April 2025 |
| Remedy decided | September 2025 | September 2026 |
| Structural breakup | Rejected — kept Chrome and Android | Rejected — keeps ad business |
| What was ordered instead | End exclusive default deals, share search data with rivals | Adjust practices to favour competitors |
| Current state | Under appeal | Ruling sealed 14 days |
Two courts, two years apart, both reaching for behavioural remedies over structural ones. The Mehta remedy is already being appealed, which tells you the realistic timeline for the Brinkema one: years, not months, before anything is settled and enforced.
For a small team, the practical translation is simple. Any plan that depends on regulators reshaping your market is not a plan. It is a hope with a docket number.
🌐 Why this matters differently from Sri Lanka
If you run a site, a newsletter, or a small ad-funded product from Colombo or Kandy, you are at the far end of this supply chain and you have the least leverage in it.
- You are on the sell side with no negotiating power. Ad-tech remedies are fought over by publishers and exchanges large enough to hire antitrust counsel. Nobody in that courtroom is arguing on behalf of a 40,000-pageview Sri Lankan site.
- Your revenue is priced in USD, your costs are in LKR. A change in ad-stack take rates hits you through two conversions before you feel it. If you have never modelled that, our USD to LKR earnings calculator shows what the remittance and spread actually cost you.
- Payment rails, not ad policy, are your real constraint. Many small SL publishers cannot easily receive certain ad-network payouts anyway. A ruling that reshuffles auction mechanics in Virginia does not touch that.
- Traffic risk outranks monetisation risk. For most SL sites, losing Google Search referrals would hurt far more than any change in ad yield. That is the search case, not this one, and it is still under appeal.
If Google's ad take rate dropped 10% tomorrow, most small publishers would not notice it in a month of earnings. If organic search traffic dropped 10%, everyone would notice within a week.
🛠️ What I would actually spend this month on instead
Regulatory news is a bad input for engineering decisions. Concentration risk is a good one, and you can act on it today without waiting for a sealed ruling.
- Measure your single-source dependency honestly. Pull the last 90 days of traffic by source. If one referrer is above 70%, that is your actual risk register, ranked.
- Own a channel nobody can deprecate. Email list, RSS, a WhatsApp broadcast. It is unglamorous and it survives algorithm changes.
- Fix how your pages are represented before you chase more traffic. Bad titles and descriptions waste the visibility you already have. Our meta tag preview tool shows the exact snippet Google and social cards will render.
- Decide your AI-crawler policy deliberately. Search referrals increasingly arrive through AI answers rather than blue links. Blocking or allowing those crawlers is a revenue decision now. The AI crawler robots.txt generator lets you set it per bot without breaking normal search indexing.
- Diversify demand, not just supply. One direct sponsor at a fair rate can outperform a month of programmatic, and it is a relationship no court has jurisdiction over.
None of that requires knowing what is in the sealed 14-day ruling.
💡 What this means for you
The short version: a US court found Google's ad business illegal and then decided not to change its shape. The remedy is behavioural, unspecified so far, sealed for two weeks, and following a precedent that is currently being appealed. Anyone telling you today what this means for ad rates in 2027 is guessing.
What is not a guess is the structural lesson. Platform concentration is now something courts describe rather than dismantle. That leaves the fix where it has always been for small builders: reduce how much of your business runs through any single company, and do it before you are forced to.
I would treat this ruling as a prompt, not an event. Open your analytics, find your largest single dependency, and spend one afternoon reducing it. That is worth more than the full text when it is unsealed.