Apple smartphone users are a big driver of mobile revenue, and marketers know it. From Q1 2023 to Q1 2024, 50% of apps increased their iOS ad spend, with a 28% year-over-year increase in such spending, according to analytics company AppsFlyer.
But to unlock the full potential of your app campaign, a robust strategy to measure advertising outcomes is critical.
To help you drive better performance on iOS, Google strengthens its AI models with key privacy-centric signals such as your first-party data and aggregated data from SKAdNetwork (SKAN). Here are three winning app marketing strategies.
Harness the power of first-party data
Under the App Tracking Transparency (ATT) framework, introduced with iOS 14.5 in 2021, users get a prompt to allow or deny an app’s ability to track their activity across other apps.
Once they provide their consent by opting in, advertisers gain access to more granular user-level data and attribution. This is particularly important for ad-supported and subscription-based apps, where personalised ad experiences and user engagement are key revenue drivers. The result: more observed conversions, improved conversion modelling, and better campaign performance overall.
Remember that timing is crucial. To maximise opt-in, apps should be strategic about when they show users the consent prompt. For example, it may be better to show the prompt several days after the app has been installed, giving users time to engage with the app and understand its value. Showing education cards before the prompt can also drive up the opt-in rate because they help users understand how giving consent can personalise the user experience.
But every app is different and businesses need to evaluate if implementing the ATT prompt is right for them. For example, some advertisers might avoid showing the prompt due to potential drop-off rates.
In addition to setting up your ATT prompt, consider also implementing an on-device conversion measurement solution. For apps with sign-in experiences, this solution uses consented first-party email and phone number data to increase the number of observable conversions available for your iOS campaign optimisation in a privacy-centric way.
A majority of advertisers who implemented on-device conversion measurement have seen a significant improvement in cost per action (CPA), with a median CPA reduction of 16% on Google’s owned inventory.1
Set up your app with SKAdNetwork
Next, tap additional tools to optimise your app’s performance. Apple’s SKAN, introduced in 2018, offers aggregated attribution data which does not require user consent and is a key privacy-centric signal that enables AI models to improve advertising effectiveness.
Advertisers should set up their app with SKAN for install and post-install event measurement. For post-install measurement, they can set up their SKAN conversion value schema through Google Analytics 4, the Google Ads API, or a Google-approved App Attribution Partner.
In doing so, they reap several advantages: more detailed information about what happens after someone has installed an app, a clearer understanding of campaign performance, and an optimised campaign.
Currently in beta, advertisers can also integrate their SKAN conversion value schema with Google Ads to enhance the performance of target CPA and target return on ad spend bidding in their iOS App campaigns. The results have been promising. Online manga platform LINE Manga, for example, achieved a 28% lower CPA with SKAN event integration.
“We have seen a significant improvement in our App campaign performance, improving our user acquisition and retention rate,” added LINE Manga marketing manager Tatsuro Horiuchi.
Pair SKAN with conversion modelling for measurement
Our most effective advertisers typically go a step further, finding ways to use both SKAN and conversion modelling measurement to help them better make business decisions. That move draws on the respective strengths.
SKAN provides a big picture view of how campaigns perform across different ad networks, but reports can take time to process given timers and other variables. Google's conversion modelling, on the other hand, provides a more granular, real-time understanding of campaign performance – allowing for faster optimisation decisions, such as identifying top-performing ad creatives.
As a result, many advertisers favour SKAN for long-term planning and conversion modelling for daily optimisations.
"Combining SKAN insights with conversion modelling and cohort-based analytics allows advertisers to fill in the gaps and gain a more comprehensive understanding of the user journey,” says AppsFlyer chief product officer Barak Witkowski. “This hybrid approach is essential in addressing SKAN's limitations while maintaining privacy compliance, and delivers to advertisers what they need most in today's fragmented landscape: a single source of truth.”
Pairing SKAN with Google’s conversion modelling for measurement helped South Korean entertainment company Kakao Entertainment fairly review the performance of its iOS App campaigns. A four-week test campaign confirmed that Google's App campaigns unlock high value iOS user acquisitions in a more efficient way. Cost per conversion fell by 28% while cost per install dropped by 36% based on SKAN, compared to their internal benchmark.
Google continues to improve its conversion modelling. We recently launched a series of back-end AI-driven model quality enhancements to further support the performance of your iOS campaigns. These enhancements have decreased Google iOS App campaign CPAs by a median of 17% and increased Google Ads reported in-app conversions by a median of 37% for target CPA campaigns.2
As the saying goes: If you can’t measure it, you can’t manage it. Help is at hand for marketers looking to navigate the complexities of SKAN and make more informed decisions. AppsFlyer and Google have launched a playbook with winning strategies to help you measure and improve your iOS campaigns with SKAN. By taking these simple steps, advertisers can maximise their return on ad spend and set themselves up for success.
Read the full playbook by AppsFlyer and Google.
