Every once in a while, fate throws you a bone, or perhaps a live grenade. In this case, it’s in the form of Apple’s significant changes to their App Transparency Tracking Policy and the rollout of SKAdNetwork. Apple’s most recent guidance is to prepare for a forced transition to the new paradigm by Spring of 2021 (for those of you keeping track at home that’s any day now).
As of the time of writing, all major MMPs are still lacking key functionality for usable attribution through the SKAdNetwork framework. Functionally this means integrations between the majority of ad networks and top historical attribution providers remain tacide, non-functional and incomplete.
Despite this deeply worrying fact we expect that the mobile marketing ecosystem will eventually reintegrate around Apple’s new framework and that MMP’s will remain the arbiters of marketing attribution by combining traditionally attribution practices with Apple’s new privacy first framework. In adding additional complexity with iOS 14, Apple is inadvertently entrenching existing MMP’s as a translation layer between the ad networks, Apple and app developers.
This troubling truth aside, networks are now scrambling to reintegrate with their existing measurement partners with SKAN as the new intermediary protocol. What remains unclear is how robust and actionable SKAN based data will be for optimization. Given the 64 bit, rolling 24 hour window constraint of SKAN conversion data, in the best case we will be forced to optimize at a campaign level against early funnel events (either revenue, lifecycle or retention). This will drastically upend traditional precisely targeted ROAS based optimization which is the mainstay of modern sophisticated user acquisition.
How to adapt to iOS 14.5 and SKAN
For flexible marketers these changes present an opportunity. Agile UA teams who can adapt quickly, determine new criterias for success and understand how to pivot their buying to the new ecosystem will find massive opportunities.
Identifying early optimization criteria
The biggest challenge for UA teams in the post iOS 14.5 world is understanding how profitable your campaigns and networks are without the 1:1 attribution and post install data historically used for optimization.
Analysts will no longer be able to measure each cohort purely based on cost in and revenue out. Instead analysts will need to study the early behavior of high monetizing players, find proxy events (or early revenue / retention levels) which can be tracked within the SKAN framework, correlate a dollar value to those events and then optimize your campaigns towards a cost per event metric.
Functionally, this will require a more closely aligned relationship between product and user acquisition teams as well as more in-depth analysis on game specific user behavior to identify key events and lifecycle triggers which represent a positive (or negative) optimization signal.
Adjusting to (lack of) efficiency for new ad products
This new world will also mean a functionally very different advertising ecosystem. SANs (such as Facebook and Google) have already begun to explicitly or implicitly prime advertisers to expect overall degradation in performance on their platforms.
Conversely, ad networks who already operate successful Limit Ad Tracking products (rewarded video networks come to mind here) seem primed to capture an increased market share as ad dollars seek profitable places to spend.
In both cases marketers will fundamentally need to reevaluate their underlying assumptions about the effectiveness of tried and true channels and tactics.
At Upptic, we are taking a bifurcated approach to post-14.5 optimization tactics. In cases where non-owned algorithms (like UAC and AAA) perform much of the optimization heavy lifting we intend to extensively retest ad product suites and evaluate them on a case-by-case basis. On other networks where we use first party optimization technology to do granular bidding (such as ironSource or Unity), we intend to lean heavily on historical data as we’ve already developed in-depth understanding of the effectiveness of specific publishers (even if forward looking performance will be much more obfuscated).
Adapting buying models, and the importance of organic uplift
The core issue which remains for advertisers (and ad networks) is continuing to justify the impact of marketing on incremental revenue. With the direct correlation from ad channel to performance being sharply diminished under the new framework the need for alternative methods for understanding UA impact become increasingly important.
To accurately measure marketing impact in the post-SKAN world, user acquisition teams need to correctly model correlated organic uplift (marketing impact on unattributed revenue) with more accuracy than ever before.
Building a nuanced and accurate model for organic uplift is a longstanding problem for app marketers. Many marketers have historically taken an “all or nothing” approach. Either looking at “blended” ROAS (all organic + paid revenue / ad spend) or expecting paid to recoup it’s dollars in attributed revenue alone. Neither of these approaches are optimal, but many publishers don’t have the tools to capture the nuance of differentiating baseline organic (organic installs that happen regardless of UA spend) from organics which are correlated with (and should be credited to) paid advertising.
Given SKAN’s massive obfuscation of attributed revenue from paid marketing, solving this problem becomes much more critical.
Over the last two years we’ve spent a lot of time on this problem and developed an automated tool which ingests historical attribution data ARPU curves and organic uplift and produces buying targets across country, OS and channel. This results in highly informed ROAS and CPA targets that can be backtested for accuracy.
Agile UA is the answer
Taken together, the only certainty going into the next few months is uncertainty. The teams who are able to nimbly adapt, create new criterias for success and execute on new buying strategies will be able to extract outsized results from the ecosystem.
Upptic Growth Services
Upptic Growth Services team provides strategic consulting and full-stack operational marketing to your growing mobile app or business. Our team consists of three core pillars: User Acquisition, Analytics, and Creative Development. We work with some of the biggest names in mobile including Candywriter (acquired by StillFront), GenJoy (Acquired by Scopely), GSN, Wizards of the Cost, Exploding Kittens, and many more.
Upptic is currently seeking high potential apps and companies that can support profitable scaling. If you believe partnership with Upptic might benefit your business, please reach out to our CGO Warren Woodward, who is always happy to talk.