We Predicted a Cookieless Future in 2022. Here Is What Actually Happened.

In March 2022 we published CRM in a Cookieless World: Paid Media Solutions. It opened with a joke about someone taking all the cookies and went on to explain how marketers should prepare for Google removing third-party cookies from Chrome.

Google never did it.

The deadline moved to 2023, then 2024. In July 2024 Google said it would not deprecate third-party cookies after all, and in 2025 it confirmed the decision. Chrome supports them today.

Plenty of agencies would quietly delete a piece like that. We would rather grade it, because the scorecard is genuinely more useful than the original article was.

What we got right

First-party data was the answer. The 2022 piece argued that marketers needed to grow and organize the data they own, through a more disciplined CRM strategy, and use it to build lookalike audiences and retargeting pools. Every brand that did this is better off today. Not because cookies vanished, but because owned data is simply a better asset.

The three strategic pillars held. We recommended a diversified media mix, stronger attribution and analytics, and a testing culture built around progressive profiling. All three aged well, and all three would be the same advice today.

Second-party partnerships still matter. Data shared by platform partners remains a legitimate way to build scale without depending on cross-site tracking.

What we got wrong

We treated one browser's deadline as the event. This was the real mistake. Safari and Firefox had already blocked third-party cookies by default, so a meaningful share of every audience was effectively cookieless while the industry was still preparing for a future date. The change was not coming. Part of it had already arrived.

We overweighted Privacy Sandbox. We described Google's replacement technology as the thing that would fill the gap. In practice it never became the successor the industry expected, and the deprecation it was designed to enable never happened.

We wrote in future tense. "Advertisers will lose the ability to track users." That framing invited people to wait for a starting gun rather than to act on the signal loss already underway.

What actually changed

The measurement problems marketers complain about today are real. They simply came from somewhere else:

• App tracking permission prompts made cross-app tracking opt-in on mobile, and most people decline.
• Email privacy protection made open rates unreliable, which quietly broke a metric a lot of CRM programs were built to optimize.
• Consent requirements under GDPR, CPRA and a growing list of state laws mean a share of your own site traffic is not measurable at all.

None of those depended on Chrome. All of them are still in force. The signal got thinner exactly as predicted. The mechanism was just different.

Why the advice survived being wrong

Here is the part worth keeping.

First-party data was usually sold as a replacement. Cookies are going away, so you need a substitute. That framing made it sound like a consolation prize, and it set up exactly the disappointment marketers felt when the deadline evaporated.

It was never a substitute. It is a better asset on its own terms:

• It is yours. No browser update, platform decision or policy change can remove it.
• It is more accurate. A customer record reflects what someone actually did with your brand, rather than what a network inferred about them.
• It compounds. Every purchase and every interaction makes the next campaign smarter. Third-party segments never improved with use.
• Competitors cannot buy it. Anyone can license the same third-party audience. Nobody else has your customer file.

An argument that only works if a prediction comes true is a fragile argument. This one did not need the prediction.

What to do now

The deadline disappeared. The direction of travel did not. Regulation keeps expanding and platforms keep tightening defaults, so plan for signal to keep getting scarcer rather than for a single cliff edge.

Grow and organize first-party data. Your CRM and loyalty program are the collection mechanism. Focus on giving people real reasons to identify themselves.

Diversify the media mix. If one targeting signal carries the plan, the plan is fragile.

Measure incrementality, not last click. As signal thins, models built on complete tracking get less trustworthy. Holdouts answer whether the spend caused the outcome.

Treat consent and governance as a capability. Clean, consented, well-structured data is what activates reliably across platforms.

The lesson worth keeping

We were wrong about the deadline and right about the work.

That is a reasonable outcome, and it points at something more durable than any prediction: a marketing program built on somebody else's infrastructure inherits somebody else's roadmap. The brands that came through this in the best shape were not the ones who forecast the outcome correctly. They were the ones who reduced how much the outcome mattered.

Whatever the next platform announcement turns out to be, that is still the move.

Frequently Asked Questions

Are third-party cookies going away in Chrome?

No. Google announced the plan in 2020, delayed it repeatedly, and in July 2024 said it would not deprecate third-party cookies in Chrome. In 2025 it confirmed they would stay. Chrome still supports third-party cookies today.

Does cookieless marketing still matter if cookies survived?

Yes, because the cookie was never the whole story. Safari and Firefox have blocked third-party cookies by default for years, so a meaningful share of any audience has been effectively cookieless the entire time.

What actually caused the loss of marketing signal?

Changes that had nothing to do with Chrome. App tracking permission prompts made cross-app tracking opt-in, email privacy protection made open rates unreliable, and consent requirements mean a share of site traffic is not measurable at all. All of those are still in force.

Was investing in first-party data a waste if the deadline never came?

No. Brands that invested got better targeting, better measurement and an asset competitors cannot buy. That payoff never depended on cookies disappearing, which is why it survived the reversal intact.

What should marketers prioritize now?

Grow and organize first-party data through CRM and loyalty, diversify the media mix so no single signal carries the plan, measure with incrementality and holdouts rather than last-click, and keep consent and data governance clean.

Could Google reverse the decision again?

Possibly, which is rather the point. The lesson is not to predict the next announcement correctly but to reduce how much any single platform decision matters to your program.