What Chrome's 1% cookie test does and does not tell advertisers
Chrome began restricting third-party cookies for 1% of users on January 4. Here is what that test can show your measurement, and what it cannot.
On January 4, Chrome began restricting third-party cookies by default for 1% of its users globally, on both desktop and Android. Google announced the test in December under the name Tracking Protection and said the goal was to let developers test their readiness for a web without third-party cookies. Google also said it plans to phase out third-party cookies for everyone in the second half of 2024, subject to addressing remaining competition concerns raised by the UK's Competition and Markets Authority.
If you buy mobile app installs, most of your paid traffic never touches a web cookie in the install path. But your web layer still matters: landing pages, web-to-app links, retargeting pools built from site visitors, and the analytics you use to judge all of it. This is where the test is worth reading carefully.
Treat the test as a readiness drill, not a forecast
Per Google, participants are chosen at random and are notified when they open Chrome. That means you cannot pick out the affected users in your analytics. They are about one in a hundred Chrome visitors, mixed in with everyone else. A change that hits one percent of one browser will not move a campaign-level number you can trust.
So do not wait for a clean before-and-after readout. It will not arrive. What the test can do is surface breakage: a login embedded from another domain, a payment widget, a consent tool, a redirect that relies on a cookie set elsewhere. Those problems show up as support tickets and odd error logs well before they show up in a dashboard.
What the test cannot tell you
- The size of the measurement loss. The effect on 1% of users says little about the effect at 100%, because the restriction at full scale would also change what ad platforms and partners can match.
- The final policy. The timeline Google gave is conditional. The phase-out date depends on a regulatory process that Google does not control, so plan around the possibility that it moves.
- Anything about app-side attribution. Install attribution on iOS runs through Apple's frameworks and on Android through other mechanisms. A Chrome setting does not change those paths directly.
Audit where cookies still carry weight in your funnel
We would spend January on an inventory rather than on prediction. List every place where a third-party cookie does work for you:
- Retargeting and remarketing lists that are built from tags firing on your website. Note which platforms you use and how large each list is.
- View-through and cross-site conversion credit in any report that relies on a cookie to connect an ad exposure to a later web visit.
- Embedded third-party elements on pages that matter, such as a chat widget, a video player, a checkout, an identity provider.
- Affiliate and partner tracking that sets cookies on a domain you do not own.
For each item, write down what it would cost you if it stopped working tomorrow, and what the first-party alternative is. A first-party alternative usually means server-side events, a first-party click ID stored on your own domain, or customer lists you upload yourself.
Decide on a threshold now
The most useful thing you can do before the phase-out is to agree in advance on what you would change. A decision rule is easier to follow than a panic. For example: if more than a fifth of your retargeting audience depends on third-party cookies, start building first-party audience feeds this quarter. If your reporting leans on view-through conversions for more than a modest share of credited results, set a separate, lower-confidence column for them so a drop does not look like a campaign failure.
Also check your web-to-app flow in a browser with third-party cookies blocked. Chrome's own test gives you a reason to do it, and it takes an afternoon. If the path from ad click to store listing works, you are in better shape than most.
Practical takeaway
Run your key pages through a cookie-blocked browser profile, build the inventory above, and set your thresholds before the second half of the year. Keep reading Google's updates, since the date depends on its discussions with the CMA, and do not treat any single test as a verdict.