The document that matters most here is the offer Google made in 2024, in a separate European proceeding, to divest AdX outright [11], because the same company then told the Virginia court that unwinding the exchange would be an intricate technical undertaking causing a prolonged and damaging transition for customers [10]. Google's answer to the tension is that the DOJ's proposal bore no resemblance to that earlier offer [11], and the answer is probably right in the way that matters, since a negotiated carve-out with a buyer you pick and a perimeter you draw yourself is a different asset from one a court draws for you and then supervises. What survives is the awkward fact that the seller calling a sale technically forbidding had already drafted one.
What the government was reaching for was smaller in cash than the rhetoric implied. CNBC, drawing on Wedbush research and court filings, put Ad Manager at 4.1% of Google's revenue and 1.5% of its operating profit in 2020, with later figures redacted [12], and 1.5 divided by 4.1 is 0.37, meaning the unit was earning roughly a third of the company's blended operating margin [1]. The case turned on the tie the court itself found, not the earnings line, running from the publisher ad server into the exchange [5], with a 20% clip taken at that junction, which leaves publishers eighty cents of every dollar they clear [3][5]. Brinkema said in April that the conduct had substantially harmed publisher customers, the competitive process and consumers of information on the open web [7]; the remedy she chose leaves the clip where it was.
The allocation consequence runs the other way from the headline. Whatever engineering a separation would have consumed, and Google's own estimate of that difficulty is the only one on the record [10], now stays pointed at product rather than at disentanglement, and Cryptopolitan's read is that Google emerges largely unchanged and free to build new AI positions against a DOJ showing little appetite to obstruct it [22]. Google is meanwhile spending on appeal rather than on a carve-out, having moved separately to challenge the underlying liability finding [14], while its regulatory affairs VP 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 [15].
The readings that stay live until the order is unsealed break down into three. The conduct terms could be thin, in which case the ad stack publishers and buyers plan around is the one they already have; they could carry interoperability or auction obligations firm enough to break the tie without a sale, which is structural relief with a compliance department attached; or the appeal could vacate liability and take the remedies with it [14]. This is probably wrong, but I lean to the thin reading, because the judge accepted most of the behavioral remedies the parties themselves proposed [2] and Google was one of those parties, and because this is the third straight refusal by a US judge to break up a large technology company [16]. The falsifier is specific: an unsealed text containing a divestiture trigger on non-compliance, or a mandate to open the ad server to rival exchanges, and my read of this as a price event rather than a value event fails. Europe has already collected 5.2 billion dollars on ad tech alone [2] within more than 10 billion of Google fines [20]. Alexandria has collected a behavior change that nobody outside the case can size yet.
