If you believe the story I was told, two years ago a bunch of adtech bros got together in NYC and tried to solve an issue: more and more of the industry was questioning what they were selling, and pointing out that the media never hit any actual metrics, never helped with sales, never had clicks, no one seemed to be looking at it.

If you believe the story, they then decided to start a myth, a lie, a fraud.

They called it the "Age of Outcomes."

I thought it was a joke when the bruhs started talking about it.

Turns out it wasn't a joke. It was a tent revival for people who measure salvation in basis points.

Every company in adtech that couldn't actually provide proof decided they'd start claiming "outcomes" and just lying.

The smart money already knows. They're not talking about it at conferences. They're running holdouts. They're measuring incrementality. They're watching their competitors burn budgets on attribution theater while they figure out what actually works.

You were lied to. By PR folks who smile with their eyebrows. By adtech bros selling vaporware wrapped in destiny. They lied the way prophets lie because they believe so hard they forget to check if it's true.

Just an expensive game of pretend where everyone agrees not to ask what happens when you turn the machine off.

Erez Levin spent thirteen years at Google.

Product guy. Strategy guy. Account Executive for Google Ad Manager. Top 2% of salespeople in their Global Business Organization. He wasn't watching this happen. He was making it happen. Then one day he was gone. No one knows why, but many say the ethics got to him.

Now he runs Emet Advisory. Emet means truth in Hebrew.

The truth he tells now is about the thing he sold for thirteen years. Or at least that’s the marketing.

"Media is taking attribution credit for business outcomes it did not contribute to," Levin says, "and the propagation of easily gamed metrics masquerading as outcomes."

Read that again.

He worked at Google. He sold this. He benefited from it. He's telling you now it was a lie.

I'm amazed Google's lawyers let him say it. I'm more amazed he said it at all.

Google benefited more than most from the outcomes driven status quo. Levin knows. He was there. Selling it. Making award winner numbers doing it.

"Enough buyers have moved on from attribution," Levin says now. "They know it's highly inaccurate and likely misleading."

While you're celebrating dashboard wins, they're measuring what actually moved.

This is mea culpa as public service. The tent preacher admitting the tongues were fake.

Except Levin's right. And he knows where the bodies are buried because he helped dig the holes.

Outcomes platforms optimize to whatever acronym calms the CFO. Those numbers come from attribution models. Attribution answers "who gets credit," not "what caused the sale."

But dashboards are framed like lab results. Like someone saw the exact photon from your display ad enter the customer's brainstem.

Correlation in a lab coat.

When Google, Meta, or Amazon grade their own homework, surprise, straight A's.

Avinash Kaushik spent sixteen years at Google. Analytics Evangelist. Now he's at Croud and he's been trying to explain the con for years. Again, Google.

"Attribution is simply the science, or, too often, the art, of distributing credit for conversions across your various marketing channels," Kaushik wrote. "The question incrementality asks is 'How many of those conversions would have happened anyway, without any advertising spend?'"

"Did you make it happen?" Kaushik asks. "Or were you just there, like a rooster thinking his crowing makes the dawn?"

Someone tells Kaushik: Yes, we're measuring incrementality. We have robust attribution measurement.

"No," Kaushik says. "I'm sorry, no."

For larger companies, he says, true incrementality from marketing is between 5% and 25%.

Not 80%. Not 100%. Not the numbers the platforms show you.

Between 5% and 25%.

The rest? Would have happened anyway.

They're reallocating budget while you're celebrating vanity metrics.

Tell an algorithm to maximize ROAS and watch it become a preacher who only visits rich people.

Retarget people already buying. Show up for the baptism, claim you saved the soul.

Inflate view through windows until Tuesday counts as "exposure."

Hoover up branded search like you invented desire itself.

You harvested demand you didn't create and billed it like a miracle. You're not a marketer. You're a crop thief with a keynote slot.

Dr. Augustine Fou has a PhD from MIT. He worked at McKinsey. He's been studying ad fraud for fifteen years. He’s been kicked out of more conferences than ADOTAT.

"The platforms and the fraudsters are claiming credit for sales that would have happened anyway, or that had already happened," Fou says.

"Every form of digital ad fraud involves the falsification of digital metrics, analytics, and attribution, from creating fake impressions, traffic, and clicks with bots, to falsely claiming credit for organic sales."

Here's his example:

"If you wanted Uber, and you heard about it from your friend, you probably go to the Google play store and just download it, install it yourself. That's called an organic install. The human just installed it without clicking an ad. But what the bad guys are doing, just like the eBay super affiliates, they were stealing credit for the organic install and claiming that it was a paid install."

Stealing credit for what was going to happen anyway.

That's the game. That's the whole game.

And it's not just fraudsters. It's the platforms.

Google built an enormous business on click based tracking that "proves" that lower funnel paid search ads drive most digital sales. The platforms tell you marketing drives 80% to 100% of sales. The real number? 30%. Maybe.

The math is simple and brutal. You got played.

True outcomes require counterfactuals. What would have happened without the ad?

Platforms can't show that without exposing their pipes.

The pipes are where the magic happens. The magic is an algorithm that mistakes proximity for causation.

So you get "modeled lift," "probabilistic incrementality," phrases that sound scientific until you ask to audit them.

Then: silence.

The kind that fills a megachurch after someone asks where the money goes.

"The only way this could possibly work," Levin says, "is if the system adopts, and buyers accept, an egregious level of unexplainable and unauditable probabilistic modeling, which is simply untenable."

Translation: More lying, but with math.

He would know. He was inside Google selling this exact lie for thirteen years.

Top 2%. Award winner. Beneficiary.

Now he's out here telling you the emperor's dick is hanging out and always has been.

"Even the big guys that have benefited the most from these legacy systems know that they're not long for this world," Levin says, "so they're trying to build alternatives."

Google's releasing open source MMM tools. Meta too. Backing away from attribution quietly, like a televangelist slipping out the side door before the IRS shows up.

If outcomes worked, brands would reduce spend while growing revenue.

Instead, budgets rise. Finance teams ask why growth feels mushy.

The only consistent outcome? Higher platform take rates.

It's not a business model. It's a tithe.

The platforms give you the numbers you want. You give them the money they want. Everybody's happy until somebody runs the math.

Moving from platforms that tell you marketing drives 80% to 100% of sales to actually measuring what works means admitting marketing drives 30% of sales.

Daunting because you have to admit you got played. Or that you played yourself. Or that you knew all along and didn't care because the dashboard looked good in the deck.

You've heard every measure called an outcome. Sales, store visits, reach, impressions, clicks.

If the word means everything, it means nothing.

Levin's blunt: "MFA sites, outstream video misdeclared as instream and even nonviewable impressions have notoriously been given attribution credit for outcomes justifying many billions of dollars."

Billions. For ads no one saw. For credit no one earned.

"Does anybody seriously think this industry is in a good place with its obsessive focus on proving efficacy with outcomes?" Levin asks. "This obsession often results in media taking attribution credit for business outcomes it did not contribute to and the propagation of easily gamed metrics masquerading as outcomes."

The man spent thirteen years helping Google perfect this exact hustle. Now he's calling it what it is.

Either he found Moses or he's got a new hustle.

Either way, he's not wrong.

Last click failed. Multi touch attribution failed. MMM got ignored.

So they rewrapped attribution, sprinkled AI on top, AI being the new way to say "trust me," and called it Outcomes.

Same math. New fonts. Different prophets, same book.

Dr. Fou watched it happen in real time.

"Too many marketers have taken easy to measure digital metrics like impressions, traffic, and clicks to mean performance," he says. "They have lost sight of and haven't put in the hard work necessary to measure incremental sales, sales that would not have occurred in the absence of said marketing. These marketers mistake coincidence for causation."

Coincidence for causation.

"Digital marketing was being done at the same time as the occurrence of the sales. Coincidence. But the digital ads did not cause the sales."

That's the con. Ad runs, sale happens, dashboard says you did it.

You didn't. You were just there. Like a rooster at dawn.

Incrementality asks one question: What changed because the ad ran?

Not who gets credit. Not which dashboard glowed green. What actually moved.

Everything else is storytelling.

Most adtech testing asks: Can I assign success to myself?

Attribution assumes the conversion was happening, then argues over custody.

Incrementality asks: Would this have happened if we did nothing?

That's the question that empties the tent.

If you don't compare exposed vs. not exposed, you're not testing. You're narrating.

Platforms optimize inside closed systems that only see what happens with ads. So they infer. Inference feels smart.

Still guessing. Just in a nicer suit.

Incrementality forces a holdout. Reality gets a vote.

Incrementality is uncomfortable. Makes sales decks shorter. Removes the theatre.

Removes plausible deniability.

Forces you to ask: What if I didn't matter as much as I said I did?

That's a 3 AM question. The kind you have alone after the conference, wondering if any of it meant anything.

Kaushik again: "Companies stay stuck on last click attribution because it's been the default for such a long time."

The default. The lie everyone agreed to believe because the alternative was admitting we'd been burning money in a barrel.

The Age of Outcomes is dead.

Because eventually, someone runs the holdout.

The tent goes quiet.

You realize the Age of Outcomes wasn't an age.

It was just a tour of aging adtech bros.

And we're all standing in an empty parking lot, holding our lanyards, wondering what we just paid for. Wondering if anyone else feels like they just got saved by someone who needed the money more than we needed salvation.

Levin left Google. Got out. Started telling the truth.

Maybe he feels bad about the thirteen years. Maybe he just saw a new market opportunity in honesty.

Either way, the smart money moved on six months ago. They're not waiting for you to figure it out.

The age is dead. Long live the holdout.

The people who matter already know this. They're having different conversations now. If you're still defending attribution models in 2026, you're the last one in the tent.

Behind the paywall is the part everyone actually cares about: the receipts, the whispered-over-latte power plays, the charts that make CMOs sweat, and the stuff PR teams wish you’d stop noticing. If the free feed is the trailer, ADOTAT+ is the director’s cut where the knives come out and the plot twists finally make sense.

Stay Bold, Stay Curious, and Know More than You Did Yesterday.