Below are frequently asked questions about SKAdNetwork and ATT.
This document aims to help our clients and channel partners. The answers may be revised depending on Apple’s policies.
SKAdNetwork
Q. I haven’t updated the SDK yet, but can I set up conversion values in advance?
Yes, you can set conversion values before updating the SDK. However, the conversion values will not be applied before updating the SDK to 1.14.3 or later versions.
Q. Will there be any duplicated data between Actuals and SKAN Report?
Yes, there might be duplicated data between Actuals and SKAN Report for various reasons. It is recommended to double-check the data for each report.
e.g.) “Install” Events
The SDK collects “Install Events” to help you figure out where the app install originated from (organic inflow/advertisements/SKAN). However, app install through SKAN cannot be attributed by Airbridge (and other MMPs). So if a SKAN install occurs, the event is recorded on both “Install (APP)” and “SKAN Install (APP)”, but attributed as “unattributed” or the respective channel.
The attribution can also be duplicated depending on ATT's permissions or if SKAN events happen with tracking links.
- ATT is permitted to collect ADID -> Recorded as a channel attributed in-app event + SKAN event
- ATT is NOT permitted to collect ADID -> Recorded as an “unattributed” in-app event + SKAN event
Q. What if I skip the updates for iOS 14?
If you do not make the necessary updates before the official release of iOS 14.5 (estimated to be released before early May), you will not be able to gain insights for new app install campaigns. Please make plans for SKAdNetwork support to avoid any data loss.
Q. When I activate SKAdNetwork, is IDFA only collected if users opt-in for ATT?
SKAdNetwork and ATT have no direct relation as they are two separate systems. This means that SKAdNetwork does not require ATT consent. Also, SKAN does not prevent ATT Prompt from automatically appearing or IDFA access. Regard them as separate systems while developing them in parallel.
Q. Can we get SKAN conversion values even if users have the app installed already?
No. SKAdNetwork currently supports only app install events. So if the app has already been installed, SKAdNetwork postback is not sent.
Q. Why can’t I see any numbers for SKAN data?
You can see numbers related to SKAN only if you are running SKAdNetwork campaigns.
Q. There is no SKAN conversion value data in the report.
(When you can find SKAN install data but not conversion values)
If you are running SKAN advertisements and can view install data but not conversion values in reports, it is likely that it hasn’t met Apple’s minimum requirement threshold.
In order to prevent conversion values from being used to track a particular device, Apple sends the conversion value data through postbacks only if a certain amount of conversion values have been recorded. (The exact number has not been disclosed by Apple). If you are running a small campaign for testing purposes, you might not meet the threshold to get conversion values data.
Q. What is the Source App ID value? What does it mean if this value doesn’t exist?
Source App ID is part of the information included in SKAN postbacks. It is an ID value that can help you identify the app that the app installs in the App Store. You might not receive this value if you fail to meet Apple’s threshold explained in the previous question.
Q. What is the difference between SKAN Install and SKAN Re-install?
SKAdNetwork postbacks include the “redownload” field. According to Apple, if the same user (same Apple ID) has already installed the app before, the “redownload” parameter is set as “true”. Airbridge records every install through SKAdNetwork as “SKAN Install” or “SKAN Re-Install” accordingly.
“SKAN Install” is the number of all installs including re-installs. For example, if the number of “SKAN Install” is 10 and the number of “SKAN Re-install” is 2, it means there were eight new installs and two re-installs.
Q. The number of SKAN Installs is lower than expected.
SKAdNetwork framework has several limitations as stated below. The number of installs may be lower than expected for these reasons.
- Web-to-App installs and probabilistic attribution installs are not supported
- Re-engagement is not supported.
- SKAdNetwork install postbacks are sent from 24 hours to 48 hours after the app installs. This means that the data will show up on the report at a later time.
Q. What are the timestamps of events received through SKAN data based on?
The timestamps of these events are based on the event timestamp received from media channel postbacks. It will be reflected as such on the Airbridge dashboard.
ATT (App Tracking Transparency)
Q. When should I prompt users for ATT?
The timing for ATT prompts is completely up to you.
However, the install event is not attributed even if a user permits app tracking. This is because the SDK transmits install events as soon as the app opens, regardless of the user’s ATT consent status. Use “TrackingAuthorizeTimeout” to avoid this, which postpones the transmission of install events, to include attribution information of install events after users consent to ATT.
Please note that the longer you postpone the event transmission, the more chance of attribution results being different due to additional touchpoints. Additionally, the event itself can be omitted if users exit the app before deciding to consent to ATT.
Q. Is the Deep Linking still available?
Yes. Apple’s policies only cover paid advertisement tracking, attribution, and targeting. Deep Linking, which sends users to particular pages, is available regardless of the user’s ATT consent status. However, DeeplinkOpen events are attributed to the relevant touchpoints if the user accepts ATT. Otherwise, DeeplinkOpen events are recorded as “unattributed”.
Q. Even if ATT is denied or restricted, is the relevant data still reflected on the Active Users report?
Airbridge identifies users through DeviceID values. So even if ATT status is denied or restricted, users will be identified through IDFVs or randomly generated Airbridge IDs. Accordingly, there will be no difference in active user data due to ATT statuses.
Q. How to test ATT?
ATT - Authorize tracking
- Delete the app
- Install the app
- Allow ATT
- View install events in Real-time log
- Check if device.appTrackingTransparency of install events is 3 (authorized)
- Check if device.ifa has been collected for install events.
ATT - Reject tracking
- Delete the app
- Install the app
- Decline ATT
- View install events in Real-time log
- Check if device.appTrackingTransparency of install events is 2 (denied)
- Check if device.ifa has not been collected for install events
Q. What if a user declines ATT?
Device-level attribution will not be possible, because the collection of IDFA will be restricted and iOS 14.5 policies prohibit fingerprinting. SKAN app-to-app install events will still be counted if the media supports SKAN. Attribution will not be possible for web-to-app and re-engagement advertisements.
Q. What kinds of attribution methods are available following whether app tracking is permitted or not?
Refer to the following table.
