Apple has released its new SKAdNetwork (SKAN) 4.0 updates, and we have taken a first look, here’s a run-down of what we think about what they mean for you as a marketer, or app owner.

Hierarchical Campaign (Source) IDs gets new 2 digits

In the previous iteration of SKAN, campaign ID supported only 2 digits. Now, SKAN 4.0 supports up to four digits for campaign IDs, which will mean that instead of the limited 100 possible values for campaigns, we will have up to 10,000 values. This expansion will mean that the identifier can be used for campaigns and narrower ad elements such as ad groups, and creatives. The last bit leads us to the decision by Apple to rename Campaign IDs to ‘source ID’ to express the new intended function.

This added granularity is immense, a two orders of magnitude (or 100x) increase means that Apple needed to counterbalance it with privacy-protecting measures which brings us to privacy thresholds. The new privacy thresholds will be tested by marketers and MMPs alike to understand how the changes will affect tracking and attribution, because if the thresholds are still contingent on unique conversion values per ID, that will limit the full use of the 10000 source IDs.

Specifically, Apple discussed three tiers of privacy thresholds or “crowd anonymity.” The update to these tiers is a natural response to the 100x increase of number of possible source IDs. The general hypothesis is that the campaigns with higher volumes of installs will get much more data as it will be impossible to reverse engineer the aggregate data reported to get device and user-specific data. This is still limiting, but it is a great improvement on the previous iteration of SKAN.

Here is Apple’s example below regarding the data a source ID will get depending on the privacy threshold: 


Overall, we think this is a welcome change from Apple! With one hundred times more available granularity, we are excited to test how this new level of detail will help add further value to campaign management and analysis.

Hierarchical Conversion Values

We get a bit excited when we see any update to SKAN that allows us to extract more info from conversion values!

Here, there were two major changes to conversion values:

  • TWO new postbacks!
  • New ‘coarse-grained’ conversion values (that will apply for two newly added SKAN postbacks)

Key Considerations

It is important to consider these factors for now:

  1. ‘Coarse’ and ‘fine’ values will not be available together for postbacks. More on this in later editions of our newsletter.
  2. The crowd anonymity thresholds will affect the kind of conversion values you get, with ‘fine’ values preserved for IDs that clear the high anonymity threshold as that is in line with Apple privacy-centric framework.

When we said we were excited about these changes, we meant it, but what we are really excited about is the updates to: Multiple Conversions (Postbacks)!

The update to multiple postbacks goes hand in hand with the ‘coarse’ conversion values, which means marketers will have much more insights with a more extended time-frame to absorb much more information from users’ journey. The three SKAN postbacks come with a great new change, the postback timer mechanism.

The three postbacks will have time windows being 0-2 days, 3-7 days, and 8-35 days respectively. These postbacks will be updated via MMPs using ‘updateConversionValue’. That method was used to delay the previously single postback beyond the original 24 hours in order to capture more information. The constraint on this improvement is that the postbacks can be delayed up until the next time frame, i.e. the first postback can be delayed up to 2 days, second up to 7 days, and the third up to 35 days.

Previously, with only 6-bit conversion values, and only increasing conversion values, we only had 64 possibilities for postbacks, now with new decreasing conversion values functionality and the ‘coarse’ conversion values, the amount of information available will be at least tripled!

There’s more to come…!

SKAdNetwork 4.0 is still in its preliminary stages, and there’s a lot more to come! So far so good! We are excited about the introductions of the new hierarchical campaign Source IDs, conversion tracking and the like: but there will be lots more to evaluate, share, test and learn! We’ll keep our eyes peeled for updates (so you don’t have to) and will be sharing them all here, soon.

Discover if it’s worth it to have Appflyer’s SSOT Feature switched on..?

Categories:

No responses yet

    Leave a Reply

    Your email address will not be published.