How to optimize performance app marketing in a post iOS 14 world
We met with Laura Nys, Gaming & App Measurement Lead of Northern Europe, from Google who focuses on measurement and she answered some of our top burning questions around driving app campaign performance in Northern Europe post iOS 14+.
Here’s what she had to say.
“All right. Good. Let’s get started.
Hi everyone. My name is Laura from Google, I work for the Northern European Gaming & Apps team. My focus is measurement, and I’m here to talk you guys through a few burning questions on the use of SKAdNetwork and more generally the trends we’ve seen as a consequence of iOS 14.
How to optimize your conversion values and measurement windows. Should you extend beyond the 24-hour window?
The short answer to this great question is it depends, but I will give you guys the long answer.
It’s true that the faster you can send a value, the faster you’re able to respond to changes in cohort behavior, but according to the garbage in, garbage out principle, sending inaccurate values to a campaign will not improve the quality of your UA efforts. Essentially you’re striking a balance between speed and accuracy. If a user’s behavior in those first 24 hours is very meaningful to predict their value to your business over time, say if your app has high usage frequency and low conversion hurdles, then stick to the 24-hour window. However, if you feel you do need more time to capture some of those valuable signals, don’t hesitate to extend to a longer window. I think ideally that would be three days.
How important are concepts like LTV and prediction modeling and how do you go about it with PredictSK?
Availing of a simple and powerful way to compute an approximation of LTV beyond direct attribution has never been as important as it is today. It will become even more important in a privacy-centric future. Analyzing LTV is an essential part of competitiveness and long-term profitability. PredictSK is a simple and powerful way to do that. Using an existing AppsFlyer integration and in-app measurement that’s already embedded in the SDK.
PredictSK can provide you with an understanding of how your channels are performing and where to adjust strategies by giving you an LTV bucket between 1 and 9 within three days after the install. It will be a complement to the strategy you’ve chosen for your SKAdNetwork conversion values, and it will result in a more holistic view of your performance over time.
Finally, how do you think the performance marketing world in Northern Europe changed both iOS 14 and potentially in comparison to some of the other markets?
Northern Europe has an especially strong developer footprint, Helsinki being a hotbed for mobile gaming globally, so I believe we will find smart new ways to go about marketing beyond the narrow scope of unique identification and rely more on first party data and predictive analytics.
The bottom line is that heavily performance-focused marketing is under a lot of pressure, and it’s being forced to shift away from individual identification. We need to get creative about how we look at profitability of marketing efforts in this new world, which is turning out to be very challenging, but you know what they say, never waste a good crisis.”