The SKAdNetwork framework is not new, Apple introduced it back in March of 2018. At the time it was positioned as an alternative, more privacy-oriented approach to campaign measurement.
SKAdNetwork is an aggregate method for measuring attribution of mobile ad campaigns for iOS apps. SK stands for StoreKit, Apple’s framework for developers to interact with the app store and in-app purchases.
The framework involves three main entities:
- The ad network signing the ad
- The publishing app (where the ad is displayed)
- The app being advertised
With SKAdNetwork, the attribution data is sent to the attributed network only when all three players “do their part.” It’s important to note that the entire flow is unrelated to and completely independent of the device’s IDFA, Apple’s device identifier for advertisers.
Since independent and unbiased attribution solutions were already very robust and widely used by advertisers even back in 2018, there was no major incentive for advertisers to adopt SKAdNetwork at the time. Especially since these solutions could deliver attribution for all networks and operating systems, not just for iOS alone.
Apple’s SKAdNetwork remained on the sidelines of the mobile attribution and marketing measurement playing field for over two years.
So What Has Changed?
At WWDC this past June, Apple announced that the upcoming iOS 14 would limit and dramatically reduce the ubiquitous access to IDFA enjoyed by all mobile app developers today.
The big change? Each app “will be required to receive user permission to track users across apps or websites owned by other companies, or to access the device’s advertising identifier.”
A new framework, AppTrackingTransparency (ATT), will require apps to ask new users to opt-in to IDFA collection by advertisers. This must be done separately for each iOS app.
With low opt-in rates anticipated by industry analysts, IDFA access is about to become severely limited to advertisers. All at once, the SKAdNetwok framework became an important measurement tool, one of the main ways advertisers will receive the attribution data for advertising campaigns on iOS.
How Will It Work?
To enjoy SKAdNetwork attribution, three things need to happen:
- Ad networks need to register and integrate with Apple and then sign each ad appropriately.
- The publisher app needs to indicate to the OS on each relevant ad click (passing the signature to the OS).
- The advertised app needs to indicate it was opened, and ‘grade’ each user’s quality based on their activity (to enable measurements of quality of users per campaign).
What Will This Mean for Publishers?
If most advertisers choose to adopt SKAdNetwork in compliance with Apple’s tracking guidelines and to attribute programmatic advertising on iOS, DSPs could potentially only bid on SKAdnetwork compatible inventory. Therefore, it would become imperative for pubs and SSPs to support SKAdnetwork, making inventory available to meet advertiser demand and not miss out on any revenue.
What Are the Limitations of SKAdNetwork?
SKAdNetwork presents new functional challenges for advertisers, compared to the data they’re accustomed to getting from their attribution providers today:
- Minimal ROI/LTV – SKAdNetwork post-install measurements are severely lacking, both in the granularity of events and window time of events that can be measured.
- Granularity – Data is presented at the campaign-level only, and is limited to 100 campaigns per network, per app.
- Postback delay – There will be a delay of at least 24 hours between when installs occur and when they are reported, making it very challenging for advertisers to optimize on-the-go.
- Mistrust from advertisers – Data is owned and reported by ad networks, who are essentially “grading their own homework”. Currently, advertisers are accustomed to receiving their data from unbiased entities acting on their behalf.
- Ad fraud risk – Data can be manipulated, potentially increasing the risk of fraud.
- No re-engagement attribution support — SKAdNetwork does not currently support deep linking or view-through attribution and does not consider anything but the act of downloading as attributable.
|SKAdNetwork also introduces structural challenges; the postbacks are sent only to the attributed ad network; therefore the advertisers, or anyone processing data on their behalf, are blind to it.
What Does Not Having First-hand Access to This Data Mean for Advertisers?
Advertisers are asking themselves four main questions:
- How will they be able to collect SKAdnetwork data from all of the different networks they work with? The list of networks can be very long, leading to extensive R&D, integration, and support cycles for each of them.
- How will they be able to view results in one centralized place, for actionable marketing insights?
- How will data be validated? The data could potentially be manipulated by many players along the way. Dirty data means bad marketing decisions.
- How will networks decipher the actual meaning of the conversion value per app?
What Is the Path Forward?
Since June, leading attribution providers have been designing rich solutions that tackle these functional and structural challenges, including:
- Data aggregation: Collecting and centralizing all SKAdNetwork information from each ad network, on behalf of the advertiser
- Data validation: Ensuring all postbacks are signed by Apple and aren’t manipulated in transit
- Data enrichment: Matching SKAdNetwork information with other data points, such as impressions, clicks, cost, organic traffic and more, for complete ROI analysis
- Data enablement: Facilitating SKAdNetwork data for convenient consumption by the advertiser, through dedicated dashboards and APIs
- Seamless integration: Full encapsulation, requiring close to zero effort from the advertiser, including for future changes in the SKAdNetwork protocol
- Conversion events: Server-side, dynamic, and flexible in-app event to conversion value configuration
So Is SKAdNetwork the Only Show in Town?
Attribution providers will likely shift to aggregated attribution solutions that combine differential privacy, deterministic (where allowed) and non-deterministic methods, probabilistic modeling, and, of course, signals from Apple’s SKAdNetwork.
Admittedly, maximizing SKAdNetwork data will be an important part of aggregated attribution solutions, as these new signals make the full attribution solution more accurate.