In the wake of Apple’s deprecation of IDFA and the introduction of App Tracking Transparency (ATT), all of FeedMob’s mobile measurement partners (MMPs) partners have issued guidance that details three paths for mobile attribution. These rules that determine credits toward a campaign event under iOS 14.5 are as follows:

• Deterministic attribution (only available when user opts in to IDFA tracking)
• Probabilistic attribution (similar to “fingerprinting”)
• Tracking through SKAdNetwork

Deterministic attribution through an MMP uses device IDs (IDFA on iOS) to make a highly accurate event attribution. Deterministic matches are largely unavailable as iOS 14.5 adoption has increased and more people have opted out of tracking.

Probabilistic attribution uses non-deterministic features (such as IP address or user agent) to make predictive matches when a user does not consent on both apps (double opt-in). This is less accurate than deterministic, but still has wide user coverage and includes the Mobile Web. 

• Probabilistic Aggregated Mode: This methodology allows for aggregated (grouped or batched) data to be recorded. While advertisers and publishers can see grouped data, no user-level data will be provided and therefore granularity will be lost.
• Probabilistic User-Level Mode: This allows for a more granular, user-level probabilistic attribution. 

SKAdNetwork is Apple’s privacy-friendly app install measurement solution for iOS in-app traffic. While SKAdNetwork is said to be a secure attribution solution, real-time and user-level data will not be available for campaign performance metrics which may be problematic for most advertisers and publishers. FeedMob is an approved SKAdNetwork partner and can work with clients to leverage this solution if requested.

If you would like to learn more about the latest methods of mobile attribution in a post-IDFA world, subscribe to our newsletter.