Skip to content
  • There are no suggestions because the search field is empty.

Attribution methods: SANs and platform frameworks

The way attribution data is shared with an MMP depends on how each ad network is designed. Ad networks generally fall into two models: Self-Attributing Networks (SANs) and platforms that rely on privacy-preserving attribution frameworks. Understanding which model applies to each of your networks helps explain why attribution data can look different from one network to the next.

Self-Attributing Networks (SANs)

Some ad networks act as Self-Attributing Networks. A SAN performs attribution within its own systems, decides whether an install belongs to it, and shares the final attribution result with the MMP.

Common SAN examples:

  • Google Ads
  • Meta Ads

How it works

  1. The user interacts with an ad on Google or Meta.
  2. The network tracks the interaction using its own identifiers and logic.
  3. When the app is installed and opened, the network determines whether it should claim the install.
  4. The network sends the attribution decision to the MMP.
  5. The MMP records the install as attributed to that network.

 

Note: In the SAN model, the MMP does not independently decide attribution. It trusts the network's decision.

Platform-specific attribution frameworks

Other platforms do not directly share raw attribution data. Instead, they rely on privacy-preserving attribution frameworks provided by the platform owner. Apple Ads, for example, uses AdAttributionKit.

How it works

  1. The user views or taps an Apple Ads ad.
  2. Apple records the interaction internally.
  3. When the app is installed and opened, Apple evaluates the interaction using AdAttributionKit.
  4. Apple sends a privacy-safe attribution signal to the MMP.
  5. The MMP uses this signal to attribute the install to the correct Apple Ads campaign.

In this model:

  • No user-level identifiers are shared
  • Attribution is delayed and aggregated
  • Data is limited to protect user privacy

How the two models compare

Aspect

SANs / Platform frameworks

Who decides attribution

The network itself / The platform owner

Data granularity

Varies by network / Aggregated and privacy-safe

User-level identifiers

Sometimes available / Not shared

Timing of attribution data

Typically faster / Often delayed

Examples

Google Ads, Meta Ads, TikTok / Apple Ads via AdAttributionKit

Related links

Need more help?

If you have further questions on the process, contact your dedicated Customer Success Manager or contact the support team via live chat.