SKAdNetwork 4.0 vs SKAdNetwork 3.0

In this article, we’ll dive into the latest SKAN updates introduced by Apple and what they mean for app marketers looking to grow their iOS campaigns.

Guido Crego

November 9, 2022

UPDATE - OCTOBER 2022

SKAdNetwork 4.0 was released on Oct 24, 2022 together with iOS 16.1. We’ve updated this article according to Apple’s SKAN 4.0 release notes.

Welcome, SKAdNetwork 4.0 👋

Apple updated its SKAdNetwork documentation to reflect the new changes available in the new SKAN version. If you’re anything like us, you know that while the industry has made a lot of progress in figuring out iOS 14.5+, there’s still a lot of room for improvement in the SKAN framework itself.

So is SKAdNetwork 4.0 the turning point we’d been expecting? The main changes in SKAdNetwork 4.0 include: 

  1. Hierarchical source identifiers
  2. Conversion values
  3. Multiple postbacks
  4. SKAdNetwork attributions for web

In this article, we’ll dive into each of the updates introduced by Apple and what they actually mean for app marketers looking to grow their iOS campaigns.

#1 Hierarchical source identifiers

🆕 What changes?

  • The campaign ID field is now called source identifier: the new name better suits the purpose of this field—letting advertisers identify which campaign an app install was attributed to, or any other additional attribution information. 
  • The range of the source identifier is increased from 2 to 4 digits: the easiest way to understand this is to consider this field as a single number composed of combinations of 2, 3, and 4 digit numbers.

➡️ What does this mean?

There’ll be much more flexibility for campaign structure and variable testing: instead of only having 100 campaign IDs for testing campaign variables to get deterministic attributions, there will be, potentially, 10.000 different ID combinations available.

This is super good news 👍 because the former 2-digit format was extremely limiting. For example, a campaign running five different creatives across five different cities and at two different times of day would make use of all the 100 IDs. 

HOWEVER, while 2 digits will always be present in the postback, the remaining 2 digits will be subject to Apple’s “privacy threshold”, in a similar fashion to how conversion values work.

To put it simply, if your campaign doesn’t have enough installs to meet the privacy threshold, only 2 digits will be available—and therefore only 100 campaign IDs. 

Note that the higher the budget, the higher the chances of a campaign meeting the threshold and making the most of the 4-digit identifier in SKAN 4.0. The number of digits available is tied to the number of installs: there are two different thresholds, one that allows you to get 3 digits, and another unlocking the complete 4-digit source identifier.

#2 Conversion values

🆕 What changes?

Conversion values are now classified into two types: 

  • Fine-grained value: this type refers to the conversion values as we know them today, a 6-bit value (with up to 64 variables) that can be set up to collect valuable performance insights.
  • Coarse-grained values: this new type can assume three different values, high, medium, or low.

Note, however, that while both values are captured during the measurement window, only one of these two values, if any, will be sent back to the advertiser (more on this below).

➡️ What does this mean?

The indicators of post-install activity are changing: with SKAN 4.0, whether you get a conversion value or not still depends on the number of downloads and the level of user privacy met by the campaign.

To decide what conversion value you’ll get (or whether you’ll receive a conversion value at all), Apple relies on crowd anonymity. This system has four different tiers based on counts of installs:

Tier 0️⃣

The install count is low and therefore doesn’t meet the privacy threshold. In this scenario you get no conversion values and the postback features the source identifier with two digits.

Tier 1️⃣

The install count is at the lower end but enough to meet the privacy threshold. Postbacks in this stage get a coarse-grained conversion value to have at least some indication of quality and the source identifier with two digits. 

Tier 2️⃣

An intermediary level where the count is high enough to meet the threshold and the user's uniqueness starts to blend with the crowd. Postbacks in this stage will get a fine-grained conversion value in the first postback, a coarse-grained conversion value in the second and third postbacks, and the source identifier with two, three, or four digits.

Tier 3️⃣

The count of installs generated is high enough to access detailed attribution data without compromising crowd anonymity. Postbacks in this stage will get a fine-grained conversion value in the first postback, a coarse-grained conversion value in the second and third postbacks, the source identifier with two, three, or four digits, and a source-app-id for ads that display in apps, or source-domain for attributable web ads.

From an implementation perspective, advertisers will need to map these new coarse-grained values into their SKAN implementation by working with their MMP partner of choice.

#3 Multiple postbacks

🆕 What changes?

Instead of receiving only 1 postback, advertisers can now receive up to 3, each one based on a specific conversion window:

  • Postback window 1: 0-2 days
  • Postback window 2: 3-7 days
  • Postback window 3: 8-35 days

Every conversion window can include multiple engagements. Note, however, that only the first postback will get the fine-grained conversion value mentioned in the previous point (the rest will receive coarse-grained conversion values, if any).

💡 Update October, 2022: while Apple has introduced a postback-sequence-index parameter to distinguish between postbacks 1, 2, and 3, it has also clarified that postbacks are not connected, so it will not be possible to tie postback 1 to any subsequent postbacks.

➡️ What does this mean?

Advertisers will have further visibility into how much a user engages with their app after clicking on an ad. This is an improvement compared to previous SKAN versions, where the only available postback was tied to a timer that would restart after each time the user interacted with the app—without taking into account any conversion windows.

💡 Update October, 2022: while this change provides more insights into campaign performance, we still see a delay when getting the postbacks, and it varies according to whether it’s postback number 1, 2, or 3. In SKAN 4.0, you get the first postback 24-48 hours after the first conversion window ends (or after you lock the conversion window—more on this below). Unfortunately, for postbacks 2 and 3, Apple has increased the length of that randomized timer: you will receive these postbacks between 24-144 hours after each conversion window has finalized.

Lock Window

This new method gives you the option to stop tracking activity and finalize a conversion value before the actual conversion window ends. This allows advertisers to receive the postbacks early but not immediately: the randomized timer starts when the conversion value is finalized, which means that you will get the postback 24-144 hours after the conversion is locked (or within 24-48 hours if it’s the first postback).

The lockWindow parameter can be applied to any or all conversion windows.

Note that after receiving a locked conversion value, the system stops registering any further conversion value updates for the remaining time of that conversion window.

#4 SKAdNetwork attributions for web

🆕 What changes?

SKAdNetwork will start supporting attribution for web-based ads that direct to the advertiser's App Store page.

💡 Update October, 2022: the SKAdNetwork 4.0 release notes clarify that web-to-app SKAN postbacks will only be available for Safari.

➡️ What does this mean?

Campaigns running in mobile web pages (different from in-app inventory) can now be attributed via SKAN, whereas previous versions could only track app-to-app and app-to-web campaigns.

What are the next steps for advertisers running SKAdNetwork campaigns?

The introduction of coarse-grained conversion values and the new attribution windows for postbacks means advertisers need to rethink their overall conversion value mapping, again. SKAdNetwork 4.0 is officially out, and our recommendation is to get ready for this new framework as soon as possible.

At Jampp, we’ve been relentlessly working to adapt our platform to Apple’s new privacy policies since ATT was announced, and we’ve already made all the necessary changes to make the most of the new SKAN 4.0 updates. If you’re looking to drive incremental performance in iOS campaigns, or want to take your SKAN campaign to the next level, send us a message.