Operated by Airbridge (AB180 Inc.), a mobile measurement partner that appears in this comparison. How we compareReviewed 2026-11-10
bestappattribution.comAn Airbridge reference
How we compare

Guide

How iOS and Android attribution differ

Reviewed 2026-11-10 · By the Airbridge team, which operates this site

Answer in brief

On iOS, device-level attribution needs the user's permission, and Apple's own frameworks report delayed, campaign-level results with no user identifier. On Android, the Google Play Install Referrer API gives the app the referrer URL and the click and install-begin timestamps for an install. The same campaign therefore produces faster, more detailed data on Android than on iOS for users measured through Apple's frameworks.

The short version

PointiOSAndroid
Permission for device-level trackingRequired through App Tracking Transparency[1]Set by Google's own policies, which this page does not cover
Store-provided install signalNone that identifies the click. Apple's frameworks report at campaign level, with no user identifierPlay Install Referrer: referrer URL plus click and install-begin timestamps[2]
Privacy framework for ad measurementSKAdNetwork and AdAttributionKit[3, 4]Attribution Reporting API, which Google has decided to retire[5]
Speed of campaign dataFirst postback after a 0 to 2 day window plus a 24 to 48 hour delay[6]Available when the app first opens and reads the referrer
Post-install detailConversion values, fine in the first postback and coarse later[6]Your in-app events, tied to the attributed install

How does an install get matched on iOS?

There are two paths, and which one applies depends on the user's answer to one prompt. With permission, your attribution tool can read the advertising identifier and match the install to a click on that device.[1] Without permission, the identifier is all zeros.[1]

The second path belongs to Apple. The device itself decides which ad wins and later sends a postback to the ad network. This works with or without permission.[3] It is the reason iOS reports have a consented part and a framework part. What you can still measure without the IDFA covers what each part is good for.

How does an install get matched on Android?

For installs from Google Play, the store passes information to the app. Google describes the Install Referrer API as a way to securely retrieve referral content from Google Play.[2] That content includes the referrer URL of the installed package, the timestamp of the referrer click and the timestamp of when the installation began.[2]

An attribution SDK reads those fields on first open and matches the install to the click that carried the referrer. The mechanism has its own page: the Google Play Install Referrer. Installs from outside Google Play do not come with this signal, so ask your provider how it handles other Android stores if you use them.

What this does to your reports

  • Timing. Android campaign results fill in as users open the app. iOS framework results trail by days, because each postback waits for its window and then a random delay.[6]
  • Depth. On Android you can follow an attributed user's events for as long as you collect them. On iOS, postbacks stop after the window that ends at day 35, and later ones carry only coarse values.[6]
  • Small campaigns. iOS may limit postback data to ensure crowd anonymity, so an install from a small crowd can return less detail.[6] The Install Referrer has no such tiering in the documentation cited here.
  • Totals. Do not expect the platforms to reconcile line by line. Compare each platform's trend with itself over time.

How to run one campaign across both

  1. Split the campaign by platform in the ad channel, so each has its own budget and its own report.
  2. Pick one success event that exists on both platforms, and make it the first thing your iOS conversion values can report.
  3. Set a review rhythm per platform: Android as often as you like, iOS weekly.
  4. Keep the lookback window settings you control the same across platforms where you can, so the device-level parts are comparable.
  5. When you compare cost per result, compare iOS with iOS and Android with Android first, then look across.

Questions and answers

Why do my iOS and Android numbers for the same campaign look so different?

They are produced by different mechanisms. Android installs from Google Play can be matched through the Install Referrer, while iOS installs from users who declined tracking arrive later as SKAdNetwork postbacks with limited detail.[2, 6]

Is there an Android equivalent of SKAdNetwork?

Google built an Attribution Reporting API as part of Privacy Sandbox on Android, and has since said it decided to retire that API.[5] For installs from Google Play, the Install Referrer remains the signal an attribution SDK reads.

Continue reading

Sources

  1. Apple: https://developer.apple.com/app-store/user-privacy-and-data-use/ Checked 2026-11-10.
  2. Google: https://developer.android.com/google/play/installreferrer Checked 2026-11-10.
  3. Apple: https://developer.apple.com/documentation/storekit/skadnetwork Checked 2026-11-10.
  4. Apple: https://developer.apple.com/documentation/adattributionkit Checked 2026-11-10.
  5. Google: https://privacysandbox.google.com/blog/update-on-plans-for-privacy-sandbox-technologies Checked 2026-11-10.
  6. Apple: https://developer.apple.com/documentation/storekit/receiving-postbacks-in-multiple-conversion-windows Checked 2026-11-10.

Cite this page

Airbridge. “How iOS and Android attribution differ.” bestappattribution.com. Reviewed 2026-11-10. https://bestappattribution.com/ios-and-android-differences