Let’s stop pretending this is transformation and call it what it is: a full-blown industry identity crisis with better branding. On one side, you’ve got executives cycling through jobs like it’s speed dating with NDAs. On the other, you’ve got every company alive screaming “agentic AI!” like it’s sage they’re burning to cleanse the mess they made.

The real story is uglier and way more interesting. The system is broken—fragmented tools, fake metrics, duct-taped workflows—and instead of fixing it, everyone’s layering AI on top like frosting on a collapsed cake. The result? More automation, less clarity, and a growing suspicion that nobody—not the buyers, not the platforms, not the algorithms—actually knows what’s going on anymore.

This is the part of the year where advertising pretends it has evolved. The venues get more exclusive, the language gets more clinical, and suddenly everyone is talking about “outcomes” like they just discovered gravity.

What’s actually happening is simpler. The same companies that spent years building bloated, incompatible systems are now rolling out “new” products that quietly fix old problems. They are selling you clarity after monetizing confusion. They are pitching control after years of shrugging at chaos. And they are doing it over very nice dinners.

Let’s get into it.

Paramount Global skipped the stage show and went straight for the steakhouse, which honestly feels more on brand for an industry built on closed-door deals. But don’t mistake the softer lighting for humility. The pitch is louder than ever.

The big reveal is a unified ad stack across Paramount+ and Pluto TV. Translation. They finally connected systems that should have been connected years ago. This is not innovation. This is maintenance with a press release. Yes, it will improve targeting and measurement. It also confirms how fragmented things were before.

Then comes Precision+, because no product gets approved without a plus sign anymore. It promises outcomes by blending first-party data, viewing signals and an algorithm that apparently went to business school. Add in fixed ad placements and live sports DAI, and Paramount is suddenly selling control and transparency like it invented the concepts. Buyers have been begging for both for years. Now they get it, wrapped in a new name and an upfront commitment form.

Amazon did something impressive. It made its own sellers mad enough to organize.

The company tried to change ad billing so costs come straight out of seller proceeds before they even touch the money. Sellers called it what it is. Cash extraction.

The backlash was immediate. A coordinated boycott from a group that actually matters. Amazon hit pause and pushed the rollout to August, but the signal is clear. When your ad platform starts behaving like a financial middleman with attitude, people notice. This is the risk of building an ecosystem where you control the storefront, the data and the ad pipe. Eventually someone asks why you also control the wallet.

Reddit is teaming up with HubSpot to “close the loop” between messy human conversation and neat little marketing workflows.

In practice, this means brands can post, monitor and respond to Reddit activity directly inside their CRM.

Here is the problem. Reddit works because it is chaotic, anonymous and brutally honest. Turning that into a structured signal inside a dashboard risks stripping out the very thing marketers are trying to understand. You do not capture authenticity by routing it through enterprise software. You just make it look organized while missing the point.

Maine lawmakers had a chance to pass a real privacy law. Limits on data collection. A ban on selling sensitive information. Actual constraints on how targeting works.

The ad industry pushed back hard. Too restrictive. Too disruptive. Too different.

And just like that, the bill is dead. Which means the current system lives on. Maximum data, minimum friction, lots of talk about consumer benefit. The industry will keep arguing that relevance requires surveillance, right up until regulation finally forces a different answer.

Viant Technology picked up TVision to build an attention-based pricing model. The idea is simple. Not all impressions are equal, so let’s price the ones people actually look at.

The execution is where it gets slippery. Attention sounds objective until every platform defines it differently. Now you have another metric layered on top of an already complicated system, each one claiming to be the truth. This is the adtech playbook. When measurement gets messy, invent a new one and call it progress.

This is the part of the year where advertising pretends it has evolved. The venues get more exclusive, the language gets more clinical, and suddenly everyone is talking about "outcomes" like they just discovered gravity.

What's actually happening is simpler. The same companies that spent years building bloated, incompatible systems are now rolling out "new" products that quietly fix old problems. They are selling you clarity after monetizing confusion. They are pitching control after years of shrugging at chaos. And they are doing it over very nice dinners.

Let's get into it.

Linear TV commitments are shrinking. Streaming upfront commitments crossed thirteen billion dollars last cycle and are still climbing as a share of every TV deal getting written. Ad-supported viewing is now consolidating into a handful of giant platforms that, between them, control most of the ad-supported eyeballs in the eighteen-to-forty-nine demo that buyers actually pay for.

Translation: fewer sellers, more leverage, less inventory to spread around. Which is, conveniently, the perfect environment to invent "exclusive" showcases, "intimate" dinners, and curated salons where the lucky few get to hear about your "performance ecosystem" between courses.

This isn't intimacy. It's scarcity theater. When you have less to sell, you make people feel special for being allowed to buy it. The candles are doing a lot of work.

If you went to enough of these things this year, you noticed the same word ricocheting off every wall: performance. Performance TV. Performance CTV. Performance pause ads. Performance everything except the actual presentations, which mostly performed adequately.

Buyers were told, with straight faces, that platforms can now not only track outcomes — sales, signups, store visits — but explain them. The pitch is that television, the medium that for seventy years has measured itself in vibes and Nielsen households, is now going to behave like Google Search. Attribution. Optimization. Real-time loops. Just like the digital channels everyone is currently disillusioned with.

Which raises the obvious question nobody on stage wanted to answer: if all of this is finally measurable and accountable and precise, what exactly were you selling me for the last fifteen years?

Here is the part that should make any honest person in the room uncomfortable. Programmatic and connected TV got rich on opacity. Multiple DSPs, SSPs, exchanges, identity graphs, data clean rooms, measurement vendors — every layer takes a cut, every layer fragments the data, and nobody can tell you with a straight face where the dollar actually went.

That confusion was the business model. Take rates love a fog bank.

Now, with budgets tightening and CFOs asking inconvenient questions, the same ecosystem has discovered a new product: clarity. Unified measurement. Single-source panels. Cross-platform attribution. The supply chain optimization tool that finally tells you which intermediary in the supply chain you are paying. Sold, of course, by the people who built the supply chain.

This is the equivalent of a contractor demolishing your kitchen, walking away for a decade, and then coming back to charge you to renovate the rubble. Except the contractor is also the building inspector. And also, somehow, your insurance company.

The word "precision" got thrown around so much this year it should be unionized. Precision targeting. Precision outcomes. Precision attribution. Precision audiences.

Here is what precision actually looks like inside a walled garden: each platform measures very accurately, within its own walls, the things it owns. Each one will hand you a beautiful dashboard showing how brilliantly your campaign performed — on their inventory, by their definition, attributed by their model, deduplicated against nobody.

Run the same campaign across CTV, social, and retail media, and you will get three different precise stories that don't add up. Each platform will claim the conversion. Your media team will be asked to reconcile it. They will fail, because reconciliation is mathematically impossible when nobody shares the same identity spine. So they will pick the story that flatters the spend, and everyone will go to lunch.

That is not precision. That is three people in a dark room with flashlights, each describing a different elephant, and a fourth person in the corner sending you the invoice.

The dirty secret of every "outcomes" pitch this season is that cross-platform measurement still doesn't work. Not in a hand-wavy way. In a literal-numbers-don't-line-up way. One system measures impressions, another measures lift, a third can't deduplicate at all. Marketers, faced with this, do the rational thing: they optimize toward whatever is easiest to measure, not whatever actually drives the business.

The "fix" being sold to them is, predictably, more vendors. More frameworks. More clean rooms. More "single source of truth" tools, each of which is a single source, none of which agree. The unified view is always one acquisition away.

Meanwhile the underlying incentive — keep dollars inside my garden, make it expensive to leave, make it impossible to compare me to the garden next door — has not budged. The vocabulary updated. The incentives are exactly where you left them.

A brand spends across CTV, social, and a retail media network. The campaign produces, let's say, ten thousand attributable conversions. The CTV platform claims six thousand of them. The social platform claims seven thousand. The retail media network claims five thousand. That is eighteen thousand conversions claimed against ten thousand actual conversions, and the brand is being asked to write the check for all of them.

The platforms are not lying, exactly. They are each correctly measuring what they can see. The problem is that nobody can see the whole journey, and nobody is incentivized to.

This is the math that doesn't work. It is not a metaphor. It is arithmetic.

Watch the language. When a platform tells you they have "discovered consumers prefer integrated ecosystems," what they mean is they have discovered they prefer keeping your dollars inside their ecosystem. When they tell you they are "committed to outcomes," ask them whose outcomes, measured by whom, audited by whom, and paid to whom.

When the dinner is very nice, ask why the dinner is very nice. The answer is usually that the inventory got harder to sell.

The upfronts didn't get more sophisticated this year. They got more performative. The rooms got smaller. The promises got bigger. And the business model — sell you the chaos, sell you the cleanup, sell you the dashboard that proves the cleanup worked — quietly added another product line.

Pass the bread.