By Grant Simmons, VP of Client Analytics, VP of Kochava Foundry
SKAdNetwork will send anonymous and delayed postbacks, but conversion models will help advertisers maximize insights from it.
The adtech world will never be the same once Apple begins enforcing the App Tracking Transparency (ATT) framework, effectively deprecating the IDFA (Identifier for Advertisers) as we know it.
Rumour is this will happen with Apple’s iOS 14.4 update, but only time will tell. What will this mean for marketers and how can they build resilience for the changes ahead?
Measurement – but not as we know it
Since the inception of smartphones more than a decade ago, marketers have had access to device identifiers for use in their marketing efforts. Apple turned all that around when they announced iOS 14 last year along with their ATT framework. Soon users will be required to give consent to access each and every app separately in order for app activity to be tracked for advertising purposes.
That’s a game-changer. With the user opt-in rate expected to be low, marketers will be forced to get the majority of their marketing performance insights from Apple’s SKAdNetwork. SKAdNetwork is how Apple will protect user privacy. Reporting of attribution and postbacks will be anonymous and delayed, signaling the end (for the most part) of real-time, user-level feedback from iOS campaigns that marketers have been accustomed to.
The silver lining is this–SKAdNetwork conversion values and the potential insights that can be extracted from them via conversion models. By employing these, marketers can understand user quality by proxy of post-install engagement insights. For instance, our Kochava Foundry team partners with brands to help them choose and configure the best conversion model for them.
Choosing a conversion model for your SKAdNetwork campaigns
There are multiple conversion models available from adtech companies, so marketers must consider what they want to measure as their key performance indicators. Our own solutions aim to pack in as much insight as possible, aligned to a marketer’s particular KPIs. These four models glean the most insights from SKAdNetwork within a seven-day measurement window post-install.
Receiving performance feedback is more valuable the closer it is to the original campaign ad interaction, so going beyond this timeframe is counterproductive. Beyond a week and there is too much delay between the install and final SKAdNetwork postback since each update to the conversion value function resets timers that postpone the postback.
So, how should a marketer choose the right conversion model and measurement window for their app? It’s not an easy decision to make and once made, adjusting it again will disrupt your data reporting flow until the new model and configuration has populated the full measurement window worth of data.
Marketers, therefore, should strive to be data-driven as they determine the type of conversion model, the event sequence, and the measurement window they’ll choose. Here’s how we do it.
Using historical data, we recommend the best model and measurement window based on an advertiser’s first-party app event data. App data is run against all four conversion models to determine the model, events, and measurement window that will work best for the app and produce the quickest and most meaningful performance insights.
We help recommend the event sequence because marketers often haven’t the means to measure all the events run previously because of the shorter measurement window that allows postbacks to be received as soon as possible. Often, if an event is too common or rare, feedback within that window will be meaningless, so it is important to isolate the most qualitative ones according to a marketer’s goals.
Apple has changed the game of adtech with respect to mobile marketing–and much of the wider digital ecosystem–irrevocably but change brings innovation and new ways of looking at things.
With Apple’s latest initiatives, marketers are being forced to change their ways. If they do nothing, they will see installs in their iOS campaigns. If they simply map events to conversion values, they will only have extremely limited insight to user quality. And if they choose one of our conversion models on their own, they most assuredly will have questions. These are not easy decisions and that’s one of the reasons why companies such as ours exist for marketers.