iOS14, IDFA and Appsumer

iOS14, IDFA and Appsumer
July 30, 2020 Simon Whittick

You’ve probably seen that Apple announced iOS 14 at WWDC, which came with a number of announcements. The biggest impact for mobile advertisers was that IDFA will be limited to only be accessed by an app when consent is explicitly given by a user. IDFA is a device identifier and is used to identify the same devices across different apps. It’s what underpins attribution in iOS as well as for ad practices such as retargeting, frequency capping and machine-learning based campaign optimization within media channels such as Facebook’s value-optimised bid strategy.

I’m sure you’ve been following the news and the shockwaves in the mobile advertising industry. Whilst many vendors have been impacted and sharing a lot of thoughts on it, this change doesn’t directly impact Appsumer. This is because we’ve never needed to access IDFAs or any personal identifiers to deliver the service we offer our customers. As such, we’ve largely not talked publicly about it yet but a few of you have asked about our stance on it, so I just wanted to give you a few key viewpoints on what it means for the future from our end.

Move towards aggregate data

Firstly, we’ve always believed that strong aggregate-level data can provide user acquisition experts the insights needed to make effective optimisation decisions. For example, bid, budget & status can’t be adjusted at a user level. 

That’s why we designed Appsumer to focus only on providing an aggregated view of data since day one, and see it continue to be enough to deliver value to 100’s of UA experts. That’s not to say user-level data is not valuable (especially to feed machine learning based optimisations for other vendors), more that UA teams can be effective without it, if the aggregate data is strong delivering relevant KPIs and granular metrics.

SKAdNetwork is the long-term solution

SKAdNetwork is essentially Apple’s backup solution when users decline to be tracked. Simply put, this new service will allow advertisers to track which of their advertising campaigns led to new users installing their apps or making purchases. However, it won’t disclose much information about what specific adverts a user has seen or clicked on. It delivers just enough information about the performance of their ads to continue investing in advertising, without exposing any personally identifiable information. 

SKAdNetwork is an early solution that has the potential to be good in the long-term. However, there are a number of limitations in the short-term. In particular, the level of granularity in tracking and effort to set it up will make life hard for user acquisition experts to get the data they need to make decisions straight out-of-the-box. However, long-term we’d expect to see these initial issues ironed out as Apple sees this as the ultimate solution for vendors to build on.

Where do we see MMPs fitting in this “New world”?

With SKAdNetwork seen as the long-term attribution solution, many are asking what our view is on the role of an MMP moving forward? There are many declaring the end of MMPs, however, we see a slightly different future in the short-term and long-term for them. 

In the short-term, we still think they add a value for iOS campaigns: 1) to facilitate the implementation of SKAdNetwork which is cumbersome to say the least, and 2) to still offer a fallback attribution methods when traditional IDFA matching isn’t available, (whilst these methods are still effective and not explicitly breaching any rules). 

In the long-term – when SKAdNetwork can offer more granularity, cohorted metrics and more – the MMPs focus could switch more to media mix modelling and probabilistic methods to tie user level data to aggregate data.

What’s Appsumer’s Role?

For Appsumer, we already collect data directly from ad networks and iTunes, making us SKAdNetwork friendly from the outset, whilst still being deeply integrated with each MMP. Our job has always been to help UA experts make perfect sense of messy data from many sources. Our product by its nature has the most robust data mapping tools in the industry, and puts us in a great position to continue perfectly automating user acquisition insights in this “new world”.

We’ll be sharing more advice and guidance to our customers over the coming weeks on how advertisers could prepare for iOS14 as well as what other UA teams are doing. So stay tuned and let us know if you have any questions.