What Is a Postback? Where to Look When Installs Read Zero

In one line

A postback is the server-to-server signal saying "this install happened." When it does not fire, the network's report shows zero.

SAN and S2S

Networks receive conversions in one of two ways. A SAN (Self-Attributing Network) like Google or Meta decides attribution itself and reconciles with your MMP; S2S means the MMP pushes directly to the network under agreed server-to-server rules.

The distinction matters because it changes where you look when something breaks. For S2S, check the endpoint, authentication, and payload format; for a SAN, check the account link and attribution settings.

Let's be honest

A postback arriving does not mean the numbers agree. If the network and your MMP use different attribution windows and conversion definitions, the counts still diverge. Fixing an integration and reconciling numbers are two separate jobs.

Go deeper

Per-integration debugging order is covered in postback error diagnosis.

Frequently asked questions

Installs read zero — what do I check first?

Confirm whether the network receives installs via SAN or S2S. The integration type changes which checkpoints matter, and for S2S the server-to-server transfer rules and authentication come first.

Is a SKAN postback the same thing?

No. A regular postback is your MMP telling a network about a user-level conversion; a SKAN postback is Apple sending an anonymous aggregate value on a delay. They differ in timing and resolution, so they should not be compared in the same table.