IDFA - 6 Things Your Company Needs to do Yesterday
Yesterday
All my troubles seemed so far away
Now it looks as though they're here to stay
Oh, I believe in yesterday…- the Beatles
Through changes announced at WWDC, Apple has flipped the table and reset the mobile advertising landscape. While the full implications are still unclear, and Apple may make adjustments before iOS 14 ships, it is safe to say that the changes will be monumental and that companies that fail to prepare for these changes will be at risk of being left behind.
The MMPs (mobile measurement partners), ad networks, Facebook, Google and UA (user acquisition) teams are all bracing for the impact to come. And while Apple has created the SKAdNetwork its current implementation leaves a lot to be desired.
In the end, no matter how freaking viral your app is, you need advertising. Just look at TikTok. As viral as it can get. Yet according to ByteDance insiders the company spent a Billion dollars on advertising the app.
Bottom line: IDFA allows effective targeting. Without being able to target (and retarget) specific users and track them all the way through to purchase, measuring and optimizing the effectiveness of ad spend will change dramatically.
To better understand the future I sat down with three fantastic experts to discuss the implication of the changes. This post is based on our discussion, but I suggest you listen to the full Deconstructor of Fun podcast episode below 👇
This post wouldn’t have been possible without the help of Tatyana Bogatyreva, Head of UA at Gram Games, Nebojsa Radovic, Growth Lead at N3TWORK and Yevgeny Peres, VP of Growth at ironSource. You can hear the full in-depth conversation from the link above. I’ve also relied on fantastic MobileDevMemo and Revenue Cat’s blog for generating questions.
#1 Get Agile
The methods and tools currently used for paid user acquisition on iOS will change dramatically as iOS14 goes live. Some developers likely feel that spending the summer ramping up with tools that are destined to change seems like a waste of time. At the same time, trying to build new tooling now while so much is still up in the air can lead to simply redoing everything later.
Most of the teams have accepted new reality and are preparing to make the most out of uncertain terrain. The fact that most MMPs have various approaches is definitely worrying. Maintain level of analysis and optimization with lacking inputs.
Key questions for your company to ask:
How agile are our product, UA and analytics teams
How quickly can SDK updates be made
How quickly can our BI tools, both internal and external, adapt to a new reality
What is the minimum viable platform from UA and BI perspective? You have to be prepared to intake data from multiple sources and maintain the basic capabilities.
How can you run ios14 simulation campaigns to understand targeting capabilities of post IDFA-depreciation
How can you most effectively ingest data from Android users
In the end it comes down to learning as much as possible in a very short time. In practice this means working with your partners to run as many tests as possible as long as there’s time. Companies are engineering a solution without knowing what the problem actually is. But based on everything that we know, it's safe to say that advertisers should prepare for the worst case scenario.
#2 Figure the Best Messaging
image: Adjust blog
Apple is providing companies with an opportunity to ask users to allow tracking. The giant issue is that you can’t even change the message or where the request is generated in the funnel without re-submitting the app. The clear implication is that companies should be testing their messaging already. This limitation also speaks volumes to the importance of agility within product, UA and BI teams.
At the same time a recent survey from TapResearch suggests that the opt-in rate will be well below 30%, which would make the IDFA essentially useless for tracking users at scale. Even if 50% of users opt in when prompted, users with Limit Ad Tracking (LAT) turned on in iOS 13 (roughly 30% of users) won’t even see the prompt in iOS 14. Which means that a very optimistic 50% opt-in rate amounts to just 35% of all users. To make matters even worse, LAT users on iOS are some of the biggest spenders on the platform.
Even if you are getting only a small percentage of users to opt in, it is still valuable data that you can leverage in your marketing campaigns going forward.
#3 Redefine your performance marketing methodology
Measuring return on ad spend is severely limited by Apple’s in iOS14. In the post-IDFA era when an ad is displayed, there’s no way to reliably track revenue generated from that ad, even at the campaign level. In other words, you will be able to measure at the campaign level, but some data might be missing, especially for smaller campaigns.
Apple has severely limited the timing of post-back events and the data they contain. This means that developers could get a relatively clear picture of app installs from SKAdNetwork but will get a much fuzzier sense of ROAS.
At the same time campaigns are being hard limited by SKAadNetwork to 100 max per network per app. Lots of smaller developers would argue that 100 campaigns is a lot. Yet keep in mind that ad networks will now be forced to do all targeting and optimization within these 100 buckets. There are no ad groups and no creative sets. It’s a question how this affects and testing of creatives.
Postback events post IDFA.
image: Singular
The main challenge is the missing information. Optimization will be on campaign levels and focus on conversion value events. Advertisers will have to trust the algorithms of the partners but there's always a question on how accurate they will be. The fact that marketing partners fail to provide a cohesive clarity is worrisome. And given the unknowns and the data flow limitations advertisers may focus even more on the biggest networks (Facebook and Google) forgoing multiple channel approaches due to added complexity.
What advertisers should do is update API integrations to make sure they’re getting the most comprehensive data flow. It is also imperative for advertisers to define what their UA and their methodology of optimizing campaigns look like. Being a front runner rather than a follower when the market dynamics are shifting allows for breakthroughs in the competitive landscape.
#4 Master the first 24 hours of user life-cycle
The assumption is that Facebook’s Value Optimization campaigns are out the window. In other words, campaigns that show your ads to people who are likely to maximize the amount of spending they generate are not on the menu in the post-IDFA era.
In the new world users will be signaled only once inside the first 24 hours. This pushes for changes in the event mapping to focus on early signals of long-term engagement and monetization. Product, BI and UA teams will have to adopt the methodology of mastering the first 24 hours, understanding users based on very early signals and converting them into marketing decisions.
In addition to advertisers implementing and testing early game events they will have to make sure to have control of the events from the server side. This allows the advertiser to push new events without the need to submit a new version.
Keep in mind that the end result is a hybrid approach. Advertisers will not be flying blind as they will have the data of conversion value events from users that have opted in as well as the data from Android users and potential similar titles in their portfolio.
#5 Learn to distinct between paid and organic users
image: Quick Marketing
On top of it all cohorting and differentiating users between paid and organic becomes nearly impossible. SKAdNetwork has no way of measuring when an ad impression leads to an install if the ad doesn’t directly lead to the install via a trackable click. User-level tracking allows MMPs and ad networks to better tie ad impressions to “organic” installs. On some networks, this tie-in known as view-through attribution (for example: watching an ad of AFK Arena on YouTube and then deciding to install the game), can account for up to 50% of traffic.
Developers will have to figure out the organic baseline for their game and then based on that triangulate the effectiveness of a paid channel per country. Assuming that there’s no IDFA, this is going to be pretty much the only way to do so. And while it sounds wildly inaccurate, it is actually similar methodology to how off-line marketing functions.
#6 Invest into CRM
Tracking limitations will make re-targeting much more difficult. This will especially affect whale-driven genres in which games have a limited but highly monetizing audience such as casino and 4X strategy games.In the future there will be retarging but it will be uncontrollable due to tracking limitations.
Having a wider portfolio of diverse user base gives an ability to do somewhat effective retargeting due to IDFV - a code assigned to all apps by one developer and is shared across all apps by that developer on your device.
It seems like Supercell is ahead of the curve with their CRM program the Supercell ID.
Now of course only few publishers have these wide portfolios and even fewer has the funds to accumulate them quickly via M&A. So the most straightforward thing for developers that have whale game(s) in their portfolio is to collect emails of their most precious users and invest into CRM. Something that most of these companies do as it is. So theoretically, it’s business as usual.
What will happen to the CPIs?
Possible outcome is that CPMs drop due to value optimization tools going away and UA teams getting more cautious and lowering their bid. There’s also the campaign limitations that can decrease IPM. These two together will lead to decrease in CPI and ad monetization.
At the same time there’s a case to be made for CPIs staying the same or even increasing on a short-term basis. With lots of unknowns advertisers will stick to campaigns and channels that work. This will lead to even more cut-throat competition for those channels. Companies that have executive vertical integration through M&A, such as AppLovin, will benefit in this scenario.
Either way, advertisers shouldn’t bank on lower CPIs. A tide that would rise all ships, like seen during the first months of lockdown. Instead advertisers should double down on finding and proven channels for their titles. Channels they can trust blindly to deliver quality traffic.
In the Post IDFA-era
Because much is still unknown it is imperative that your company invests into the agility of the UA, Product and BI teams. The focus should be on building the MVP platform and then build up rapidly as IDFA-depreciation rolls onto iOS.
Now is also the time to beef up the creative teams. If you don’t have a creative director in charge of marketing content, this is the time to consider hiring one. Where today creatives are driven by almost endless amounts of iterations in the future creatives will have to get - well, more creative. Ads that connect with the wider target audience in an engaging way. Tactile’s Lily's Garden is the pound-for-pound great when it comes to creative ads that connect with target audience.
And of course your company has to make sure that you will have all the data possible and actionable. You need to be able to create conversion value events at a rapid pace and have a way to measure channel contribution. The performance marketing landscape will perhaps simplify to some extent. But that doesn’t mean it will get easier. It just means that you’ll have less tools but you will have mastered using each and every on of them.
Our partner ironSource builds technologies that help game developers take their games to the next level. They boast industry’s largest in-app video ad network, a robust mobile ad mediation platform, and a data-driven user acquisition platform. In other words, everything you need for game growth.
Make sure to follow their Level Up blog for unmissable insights.
☝️ If you enjoyed this read and the podcast, please make sure to listen to our previous episode with Yevgeny and Eric Seufert of MobileDevMemo. We discussed everything one should know about IDFA and the impact the change bears on industry level.
You should also check out Eric’s post on Deconstructor of Fun blog insinuating that there will be a Ragnarok of M&A fueled by depreciation of Identifier for Advertisers and rising importance if IDFV.
