FAQs about iOS 14 changes

As of April 30, 2021, SKAdNetwork reports are available to all App campaign advertisers. SKAdNetwork reporting is one of multiple data sources to consider for campaign reporting. Compared to Google’s conversion modeling, SKAdNetwork has core differences in underlying logic and settings, which can lead to reporting discrepancies between the 2 methodologies.

To improve the quality and consistency of our modeled reporting in the Google Ads UI, we’re deepening our integrations with SKAdNetwork data. However, some level of discrepancy between the two are normal and expected. Learn more about these 2 methodologies and best practices for iOS App campaigns.

Additionally, App Attribution Partners (AAPs) have access to the same information of SKAdNetwork conversions and App campaign conversions. However, AAPs don't receive attribution claims on modeled conversions, so you’ll notice differences among App campaigns, SKAdNetwork, and AAP reporting.

Setting up your SKAdNetwork conversion value schema can help you measure in-app event performance and more effectively optimize your campaigns for post-install actions or a target return on your ad spend.

You can set up your conversion value schema through Google Analytics 4, a Google-approved App Attribution Partner, or the Google Ads API directly. It’s highly recommended that advertisers choose one place to set up their conversion value schema. Conversion value schemas set up through Google Analytics 4 can be exported with the Admin API. We also plan to introduce the ability to utilize your schema to improve the optimization of your tCPA and tROAS App campaigns. Currently, this is available in limited beta.

Learn how to Set up your SKAdNetwork conversion value schema and best practices for iOS App campaigns.

• Automatically measures key interactions, such as first opens and in-app purchases, for consented users. Learn more about Automatically collected events.
• Allows you to use Google’s on-device conversion measurement for iOS tools and improve conversion modeling and reporting for your iOS App campaigns.
• Allows you to experiment with explainer messages to communicate why your users should opt into personalized ads.
• Ensures your campaigns will be compatible with upcoming privacy-safe measurement innovations from Google.
• Helps you unify data across your app and website. Learn how to Set up Analytics for a website and/or app.

We separate ad interactions based on whether we were able to observe the link between that ad and the conversion.

Then, we identify a group of similar-behaving users in the observable group and calculate their conversion rates. We also analyze other factors, including device type, time of day, and operating system, in order to assess similarity.

Google AI is trained to determine the likelihood of conversions across all users that interacted with the ad, even those that weren’t directly viewed because of browser or OS restrictions. We apply these training techniques uniquely for each advertiser’s specific business needs.

Then, we validate that our models are representative by leveraging data from other calibration signals, such as historical conversion trends. Learn more About modeled online conversions.

Our bidding systems rely on the measurement solutions used to collect data such as the Google tag (gtag.js), Google Tag Manager (gtm.js), and Google Analytics for Firebase. As we develop new measurement solutions like conversion modeling and new URL parameters, we closely monitor their effect on our bidding systems and make adjustments accordingly.

Campaigns using Smart Bidding will continue to set bids based on the most accurate measurement available. However, we recommend that you prepare for the above changes by closely monitoring the performance and delivery of all campaigns serving on iOS traffic. If necessary, you can also make adjustments to budgets and Smart Bidding CPA or ROAS targets to help you achieve your goals. For example, if you'd like to increase spend, you can raise CPA targets or lower ROAS targets as needed.