What are Apple’s SKAdNetwork privacy thresholds?

Apple’s privacy thresholds are a SKAdNetwork (SKAN) privacy-preserving feature that aim to limit the amount of data advertisers receive per postback if a specific threshold of campaign/install volume is not met. The idea is that if campaign volume is too low, it could be possible to match specific users to installs and events, even in the absence of the identifier for advertisers (IDFA).

Since SKAN 4, privacy thresholds have been replaced by crowd anonymity, which serves the same purpose, but also provides marketers with a little bit more attribution insight on low-volume campaigns. This is largely enabled by the introduction of coarse conversion values.

For any campaign run on SKAN 3, you will need to meet the privacy threshold minimum to receive conversion value data in your postback. Unfortunately, the specifics around volume are ambiguous, but in general, the higher the campaign/install volume, the more likely you’ll be to get the conversion value you worked so hard to map.

With no conversion value, however, you’ll have no information to work with, other than the fact that an install took place. This missing data can lead to miscalculations, incorrect assumptions about post-install behavior, and poor allocation of budgets. This is one of the reasons we recommend switching to SKAN 4, or at least starting to run SKAN 3 and SKAN 4 campaigns simultaneously where possible.

The good news is that CV=null rates on SKAN 3 have, overall, decreased over time and currently sits somewhere between 10% - 15%, making it possible to fill data gaps with smart analytics and solutions that predict behavior both within the first 24 hours and later in the journey. This should be part of your overall SKAN and iOS strategy in general, meaning factoring in privacy threshold and crowd anonymity limitations will be part of your bigger, next-gen picture from the ground up.

Whether SKAN 3, SKAN 4 working within the limitations of privacy threshold and crowd anonymity should be a part of your iOS and SKAN strategy and measurement framework. At Adjust, our next-generation, machine-learning models work to keep you in control of your campaigns while plugging data gaps and maintaining the ability to scale with confidence. If you meet the privacy thresholds or crowd anonymity requires and receive the most possible data, that’s fantastic. If certain campaigns don’t, it’s still possible to perform measurement that gives you the information you need to make smart, impactful decisions.