Seventeen billion dollars sounds like the kind of number that ends a business practice. I do not think it does, and I think marketing leaders who read this week’s Meta settlement as the moment engagement-optimized design got shut down are misreading it. What it actually did was set a price for that design, payable over nine years, by a company whose ad business alone generates far more than that in a single year. A price is not a prohibition.
The strongest case against me
The counter-argument deserves stating plainly, because it is a good one. This is not a token fine. Forty-seven states plus four territories signed on. Colorado alone gets 626.4 million dollars. The operational mandates, a two-hour daily cap with mandatory pauses, an overnight lockout, a school-day notification blackout, age verification, dialed-back comparison mechanics, are specific, enforceable, and backed by an independent auditor for at least five years. Colorado Attorney General Phil Weiser said the relief is “very meaningful and well beyond what any court has ordered or is likely to order,” and on the narrow question of what a court would have awarded, he is almost certainly right. This is the strongest state consumer-protection outcome of its kind on record. That is a real accomplishment for the coalition that built it.
Why I still think it prices the habit instead of ending it
My disagreement is not with the size of the number. It is with what the number is attached to. The settlement changes Instagram and Facebook’s rules for teen accounts. It does not change the underlying business model, which is still built on maximizing time and attention across the rest of the user base, still monetized the same way, still measured by the same engagement metrics internally. A five-year design mandate on one age cohort, on two products, at one company, is a patch applied to the part of the system that generated the worst headlines, not a rewrite of the incentive structure that produced it in the first place.
There is also no admission of wrongdoing attached to any of it, which matters more than it sounds like it should. A settlement without an admission is a company agreeing to change specific features under specific conditions, not a company agreeing that the underlying approach was wrong. That distinction is exactly why I expect the same playbook, engagement loops tuned for time-on-platform, comparison mechanics tuned for return visits, to keep shipping everywhere this particular settlement does not reach: adult accounts, other Meta products, and every competitor that has not yet been sued. It also is not happening in isolation: this outlet covered a separate 400 million dollar DOJ settlement with TikTok just yesterday, over children’s data collection, a different legal theory aimed at the same underlying pattern.
What this means for how you plan, not just what you feel about it
Whatever my read of the deterrent value, the operational reality is not optional. Teen-account reach and frequency on Meta platforms are about to shrink, on a fixed schedule, for years. Any brand whose media plan quietly depends on late-night or school-hours engagement from that cohort needs a new plan, not a wait-and-see posture. The same goes for creative built around visible likes and beauty filters, both are explicitly named in the reforms, so assume the surfaces they run on will change under you.
The bigger planning question is whether your own first-party product, not the platforms you buy media on, has engagement loops that would read the same way to a regulator applying this theory. This settlement proved that “how the product is designed to hold attention” is now a legitimate target independent of data handling. If your loyalty program, your app, or your owned engagement mechanics were built on the same growth playbook Meta was, the fact that you are not Meta-sized is not the same as being out of scope, it is only a matter of when the theory reaches your category.
I would love to be wrong about the deterrent question. Seventeen billion dollars and a five-year audited mandate is a genuinely hard number to shrug off, and if it changes how the next platform designs its defaults before getting sued rather than after, that would be the best outcome available. I just would not build next year’s plan on the assumption that it already has.
Source: Colorado Attorney General
