Google lets the users do the dirty work

Google lets the users do the dirty work

If you like posts like this, please subscribe to the free Marketecture newsletter.


By now, surely you've heard about the news that Google is no longer going to be removing third-party cookies unilaterally, but it will instead present end users with some sort of privacy dialog. We had a great conversation about this on a "emergency podcast" featuring myself along with Andrew Casale from Index and Paul Bannister from Raptive. Let's go through what we know and what we don't know, and some smart takes from “friends of the pod.”

Conventional wisdom

By now most pundits have come to more or less it consensus, that this news changes very little. Google is essentially still deprecating cookies just blaming it on the end users instead of taking responsibility itself.

We have zero insight into the nature of the Chrome implementation or interaction with consumers that's relevant to third-party cookies. But we know that Google’s intent is to get rid of these cookies because of their continued battering at the hand of rival apple (see meme below). We also know that Google believes that third-party cookies represents a real privacy issue as they've been saying that for 4+ years.


The regulatory deck gets shuffled

In hindsight, the Sandbox initiative was hugely ambitious. It required not just completely reworking the infrastructure of the large and complex ad tech stack, but also gaining wide adoption from parties with divergent interests.

It turns out that one group most affected by the plan struck back by enlisting regulators. Thus, CMA oversight over the cookie and Sandbox replacements became quite focused on the likely effect on publisher revenue, which, as we've been following, is certainly going to be negative (latest estimates below).

The most recent CMA report (pdf) brought up numerous fundamental issues about the Sandbox as impediments to roll out, and, as we reported at the time, these problems were not new and had been unsolved for the four year lifetime of the project. For example, the need for the Sandbox to be governed by a third-party was something that Google and others had been saying for some time, but which has not made any progress as far as we can tell. If the future of this initiative is dependent on changing quite fundamental things about the initiative that have not been changeable to date that's not a great prospect for actually releasing this at any time in the near future.

The CMA has decided to withhold publication of its upcoming report and instead ask the public for commentary. No one (other than Google, presumably) has seen this draft report, and now no one ever will. There have been rumors swirling that it was going to be quite damaging to the future of the project.

So the billion dollar question is whether Google can get out of it regulatory bind by shifting the control over cookies from their engineers to the end users. This may seem a little ridiculous, but there is a direct comparison to the way Apple rolled out ATT, which did not have any regulatory input. I’ll leave you with this tweet from Shiv:


Impact on advertising

I hate to be wishy-washy, but this announcement is both positive and negative for the overall advertising business. Here are some points of view (all mine):

Nothing changes: We don't know what Google’s approach will be, but we do know that there will be far fewer third-party cookies a year from now. Which means that we should continue the evolution of our stacks to a cookieless future even if it's not really cookieless.

More signal is good: Whether 10% of users or 30% of users still have third-party cookies a year for now, those users will represent an important signal that can be used to train our cookieless stacks. Use cases ranging from retargeting to attribution to media mix modeling will have a data rich stream to use for AI training and validation.

Lazy marketers continue to coast: Those in the trenches have seen how marketers and agencies have failed to date to step up to the challenges of a cookieless future. Numerous surveys have shown that folks are really not prepared for this eventuality. The worst effect of this news may be that it gives a further excuse to over-target the minority of Chrome users with cookies.

Other smart takes

Eric Seufert writes about ($$) the fate of Android’s GAID, which was left outside of this week’s announcement. In 2022, Google said that GAID would be deprecated in the same or similar fashion as the Chrome cookies. To me, this seems pretty different because it is unlikely to have as much opposition from publishers and there's a clear precedent from Apple.

It’s important to recognize that the GAID will be eliminated either way. But the text of an opt-in prompt likely has a meaningful effect on opt-in rates, and I don’t think that Google is as incentivized to influence low opt-in rates with Android as it is in Chrome. —Eric Seufert, MobileDevMemo

Another shoe to drop could be Google’s pre-existing consent agreements with the regulators which were largely tied to the framework around deprecation. For example Google has agreed not to use certain media from within Chrome to benefits advertising business. One wonders if these may also be to renegotiation.

Appendix: The latest Sandbox data

Less noticed within Google’s announcement was their data on Sandbox effectiveness (in bold). We can add their data to our ongoing tracking:


If you like posts like this, please subscribe to the free Marketecture newsletter.

Joe Root

Co-founder at Permutive

5mo

"More signal is good"—I think this is the most understated point in everyone's analysis, but is HUGE. The biggest risk to the Open Internet was most Publishers were struggling to amass >5% authenticated traffic in order to provide modelled activation and measurement. Even with ATT scale opt-out rates, 10% with cookies would provide enough signal to support modelled activation and measurement without the need for email collection. Assuming a publisher has invested in the tech to support these things cookie deprecation is a huge opportunity, and this change eliminates the biggest existential risk—amassing enough emails. Great article.

Alan Chapell

Privacy, Competition, AI & Music

5mo

Why do so many in adtech take everything Google tells them at face value?

Nathan Thomas

AdTech Veteran | Yield & Data Strategy, Ad Operations & Programmatic Solutions Expert

5mo

I agree this new trajectory is both positive and negative. I hope marketers (and the industry) will not keep kicking the can down the road to enact change. What I am most curious about is how the user choice will be handled, how it will be presented to the user and at what point in time.

Matt Wurst

CMO, Creator, Consultant, Catalyst for Connection & Community. (Also: Pioneer, Pragmatist & Provocateur.)

5mo

This is a cookie depreciation post

To view or add a comment, sign in

More articles by Ari Paparo

  • Two views of CTV

    Two views of CTV

    This article appeared in the Marketecture newsletter. If you like it, you can subscribe for free.

    17 Comments
  • We found the deck DoubleClick used to sell itself to Google

    We found the deck DoubleClick used to sell itself to Google

    In our first Monopoly Report newsletter we published our history of Google’s DoubleClick acquisition and made the case…

    17 Comments
  • The war on header bidding

    The war on header bidding

    This is the latest post of my new Marketecture newsletter: The Monopoly Report. Each week we will cover everything…

    22 Comments
  • Auction manipulation: what is fair?

    Auction manipulation: what is fair?

    This is the latest post of my new Marketecture newsletter: The Monopoly Report. Each week we will cover everything…

    5 Comments
  • The Monopoly Report: Everything about Google's antitrust woes

    The Monopoly Report: Everything about Google's antitrust woes

    This is the first issue of a new Marketecture newsletter: The Monopoly Report. Each week we will cover everything…

    4 Comments
  • The Big Apple (ad business)

    The Big Apple (ad business)

    This post originally appeared in the Marketecture newsletter. Subscribe for free.

    10 Comments
  • The S&P 500 — no, not that one!

    The S&P 500 — no, not that one!

    This article appeared in the free weekly Marketecture newsletter. Subscribe for free.

    27 Comments
  • State of the Sandbox, July 4th Edition

    State of the Sandbox, July 4th Edition

    This first appeared in the Marketecture Newsletter. Please subscribe for free for more of this type of analysis.

    20 Comments
  • ARPU Kidding Me?

    ARPU Kidding Me?

    This is from my weekly newsletter. If you like it, please subscribe for free.

    37 Comments
  • Introducing the SLoP model measuring publisher power

    Introducing the SLoP model measuring publisher power

    If you like this article, please subscribe to my free weekly newsletter. In last week’s newsletter I proposed The…

    7 Comments

Insights from the community

Others also viewed

Explore topics