Without getting into technical details, we can summarize it as an operation aimed at fighting the use of cookies by third-party services, which are intended to track user behavior.
Specifically, the ITP 2.1 greatly limits the duration of cookies, forcing them to expire after 7 days.
One would ask: is Apple trying to implement an anti-tracking maneuver to protect its users?
If this were the case, it would not be surprising considering the recent statements by Tim Cook that Apple user data is not for sale and that “each Apple product is designed to minimize the collection and use of personal data […] provide transparency and control over information “.
Direct consequences of Safari upgrade
Among the most immediate consequences, limited to the users of Safari browser, we can easily identify:
- An increase in “user” and “new user” metrics
- Changes to segments and user-based audiences
- The 7-day limit in the ADS / Facebook conversion window
- Persistence of remarketing audiences limited to 7 days
- Persistence of the variations provided in the case of A / B tests
- A limitation of cross-site tracking
Within a website, the entity of these changes is directly proportional to the percentage of users who utilize the Safari browser. The remaining metrics not linked to the user, such as sessions, source channels, events, goals will not be affected by the update.
As for the CRO, what changes with the introduction of ITP 2.1?
Technically, the A / B testing platforms are affected by the problem on two levels:
- Custom audience. Often the cookies set by the browser are used to show an experiment or a personalization to a single group of users (for example “only the new ones”). With ITP 2.1 this operation can no longer be carried out reliably. The solution will be to build audiences not based on cookies, but on information saved elsewhere (LocalStorage is an example) and then switched to the platform via JS.
- In an experiment, we use cookies to keep track of the users that we want to see a specific version of the page. Their deactivation after 7 days, therefore, leads to poor reliability of the obtained results. The solution proposed by Convert is to set the experiment cookie on the server side considering that the limitation of ITP 2.1 concerns only those created by the browser. If using the loophole it will be possible to make changes via JS and without cancellation after a week, but this requires an integration of the server.
Conclusions: why it is not worth being alarmed by ITP update 2.1
The world of web analytics and advertising has suffered a strong shake from ITP 2.1, to prove it, it is also the failure to take a position from the side of big players like Google itself.
We must not be alarmed though and, indeed, we must change our point of view: we are following the direction of greater privacy for users, a significant improvement on the internet. It’s about finding the middle ground along the way: protecting privacy on the one hand and the need for data collection – for marketers – on the other.
We will continue to keep you updated on upcoming developments – an ITP 2.2 seems to be already in the pipeline – in order to monitor its effects and minimize its impact.