Skip to Content
Platform.Ontology

SKAN postback windows and timers

Postbacks arrive days after install, in three time windows with random delays.

Works without askingAds and install trackingApple attributionAndroid and iPhoneBefore you installIn the background

In one line

Can only get install reports days later, in three windows with random delays

Postbacks arrive days after install, in three time windows with random delays.

Delays break the link between a user's behavior and the postback, so an ad network cannot tell which person it was.

What it can and can't do

What the user does: Nothing. Apple sends the reports after you install and use the app

It can

  • Send the first report after days 0 to 2, plus a random delay of 24 to 48 hours
  • Send later reports for days 3 to 7 and 8 to 35, with delays of 24 to 144 hours
  • Link a report to a campaign, not to you

It cannot

  • Send a report the moment you install
  • Tell which person did what
  • Change the delays, since Apple sets them

How it differs between platforms: iPhone only (iOS 16.1 and later). Android has no such timers.

On Android and iPhone

AndroidiPhone
StatusNot supportedSupported
First versionNot availableiOS 16.1
Permission or entitlementNot listedNot listed
Protection levelnonenone
How access is grantednot-exposednone
What the user seesNot listedNot listed
How long it lastsNot listedNot listed
How to take it backNot listedNot listed
Store policyNone listedNone listed
APINot listedNot listed
NotesNot applicable.Window 1 covers days 0 to 2, window 2 days 3 to 7, window 3 days 8 to 35. A randomized timer of 24 to 48 hours after the first window, and 24 to 144 hours for the later windows, delays each send.

Data you get

What comes back: 3 postbacks per install
Shape: event
How often: windows 0 to 2, 3 to 7, 8 to 35 days

FieldTypeUnitRangePrecision
window 1eventdays0 to 2Not listed
window 2eventdays3 to 7Not listed
window 3eventdays8 to 35Not listed
delay timerfloathours24 to 144Not listed

Technical limits

  • Randomized delay
  • Windows 2 and 3 only report if a coarse value is available

The story

What it is

Apple delays and randomizes ad reports so no one can tie them to a person's actions.

For you

You never see it. You install an app today and the ad network hears about it days later, mixed with other installs.

For makers

The three windows are 0 to 2 days, 3 to 7 days and 8 to 35 days. Random delays break the link between your actions and the report, so marketers must wait and plan around slow data.

Where you see it

TypeExampleApps
gamesA studio waits days to learn if a campaign paid back, so it spends cautiously at first.Zynga, Supercell
adsAn ad network gets a first postback after two days and a second after a week.AppLovin
shoppingA retailer launches a sale and cannot see its ad results until the next week.Target
analyticsA team builds a model to predict full results from the first postback.Singular, Adjust
securityRandom delays stop an attacker from matching an install to a person.None named
fitnessA wellness app waits for the second window to see who bought a subscription.Calm

Risks

  • Slow feedback for advertisers
  • Data may arrive partial

Android and iPhone in short

  • Android: Android had no delayed postback system, since it uses the advertising ID.
  • iPhone: SKAN 4 has three windows with random delays.

What the platform says

Good uses and abuse

Legitimate uses

  • Privacy-preserving measurement

Known abuse

None listed.

Scores

ScoreValueMeaning
Sensitivity1 of 5How revealing the data is
Utility3 of 5How useful it is to honest apps
Friction0 of 5How much the user has to do
Fingerprint bitsNot ratedRough identifying entropy, when known

History

  • 2022-10, iPhone iOS 16.1: Three windows and timers (SKAN 4); SKAN 3 had one 24 to 48 hour timer reset on each update

Sources

How well checked: Based on official documentation.

On this page