User acquisition on iOS is broken

We have to face the harsh reality.

It’s been a year since Apple has fundamentally changed user acquisition with iOS 14.5. While the best mobile marketers are using SKAN effectively, many others are not. And even for the best, performance is well below IDFA-based user acquisition. The result: ad spend on iOS, which dropped significantly immediately after the launch of iOS 14.5, is still down:

1. ROI is down 38%
2. Spend is down 25%

The conclusion is obvious: the basic framework that SKAdNetwork provides does not enable sufficient growth on iOS. If it did, more brands would have cracked the code. They simply have not, and the inescapable conclusion is simple.

The industry needs help. We can wait for another big change in iOS and SKAN, or we can invest in technology to close the gap.

IDFA >> SKAN: what changed?

Before iOS 14.5, Apple’s identifier for advertisers (IDFA) connected all the components of the mobile performance marketer’s toolkit.

An ad network targeted a device ID that looked appropriate. The owner of that device clicked on an ad which was connected to a campaign. That device ID installed an app. And then that device engaged with the app and monetized by watching ads or buying in-app purchases, and marketers could connect it all back to an ad, a partner, and a campaign.

IDFA didn’t just enable complete data: it offered instant data: as installs happened, marketers got data. Ad networks got data too, which enabled near real time optimization of campaigns.

Plus as people abandoned apps, marketers still had IDFAs so they could create new audiences for an ad partner to retarget them in win-back campaigns.

Very little of that works the same way now, and much of it doesn’t work at all.

Now we have SKAdNetwork, or SKAN. With SKAN, marketers also need to decide which events to optimize by.

But with SKAN, the number of possible events is tiny, the window to get them is brief, and the postbacks that deliver data are randomly delayed and partially censored for privacy. Plus marketers need to set up new types of integrations with ad networks, separate integrations with SANs who only share SKAN data over APIs, run conversion model management, and pull the few available fragments of SKAN data into their analytics to compare and dedupe from ad partners’ numbers.

All of this is really hard.

Furthermore, there are 7 massive problems that severely impact SKAN-based user acquisition.

Broken: 7 massive problems with SKAN

ROI for mobile user acquisition is down 44% under SKAN. Campaign performance is getting crushed. Spend is down overall on iOS as marketers lose confidence in their ability to profitably acquire new users.

Why?

Because there are 7 massive problems with user acquisition in iOS with SKAdNetwork.

Problem	Details	Impact
1. Missing data	Apple’s privacy thresholds censor roughly 20% of SKAN data (and even more in lower volume campaigns)	Marketers have incomplete data, making optimization less reliable and SKAN-based KPIs inaccurate.
2. Randomized timing of SKAN postbacks	Apple waits up to 24 hours before sending install postbacks	Marketers have unknown skews in their data at any given point in time Ad networks receive postbacks with delays
3. Limited conversion value reporting	Apple offers only 6 bits of data for post-install app conversions	Marketers struggle with estimating the value of their new users SKAN-based reporting is limited to a single KPI
4. Limited time for conversion value updates	Post-install conversions can be encoded into install  after installs, but in practical terms when marketers need to optimize campaigns, that commonly means 1 or at most 3 days	Marketers have very limited long-term data on the value of their users by source Apps with longer conversion cycles struggle to use SKAN effectively
5. Limited campaign attributes	Apple limits campaign IDs to 100	Marketers lack granularity on campaign and creative performance Ad networks struggle with scaling campaigns
6. Complex conversion value encoding	Encoding for conversion values is extremely challenging	Marketers are forced to spend a huge amount of time deciding on and implementing conversion models, which are very susceptible to human error resulting in suboptimal performance, delays, missed opportunities, confusion, and frustration
7. No cohort data	Because of missing data and delayed data, traditional cohorts are unavailable	Marketers lack data on LTV and ROAS that is essential for strategic planning and tactical optimization

Fixing UA on iOS: Making SKAN work

The solution is not fingerprinting. It doesn’t work on the majority of the adtech ecosystem, is getting increasingly less viable as ATT, ITP, and Private Relay converge, and is starting to cause app rejections as Apple enforces its rules.

So making SKAN work is non-optional for serious mobile growth marketers.

Here are the 7 problems … their impact … and how to solve each one of them with Singular’s solution.

Problem	Singular solution
1. Missing data	SKAN Modeled Metrics Singular models missing SKAN data so that marketers can view their complete data set, calculate KPIs, and make truly informed decisions about ad spend.
2. Randomized timing of SKAN postbacks	SKAN Estimated Install Date Singular estimates actual install dates to reconstruct reporting by install date for marketers, which is the standard view for marketing teams to analyze data.
3. Limited conversion value reporting	SKAN Smart Conversion Management Singular provides all the tools you need for maximizing value from SKAN’s six bits of data. That includes 7 conversion models to choose from including mixed models, ad monetization, events, and engagement, ensuring every app can utilize the best suited model. Singular also offers server-to-server functionality to support downstream and first-party events that allow for additional optimization.
4. Limited time for conversion value updates	SKAN Instant Campaign Optimization (coming soon) Singular includes near real-time predictive D7 LTV for instant campaign optimization, enabling ad networks to optimize campaigns quickly via modeled revenue baked into your conversion management model. High-volume enterprise customers can also use server-to-server modes that enables customers to achieve 90% accurate pLTV to Singular to report via the SDK and speed up campaign optimization even more.
5. Limited campaign attributes	SKAN Combined Cost & Conversion Singular enriches SKAN data with campaign dimensions and cost from your ad partners to provide bottom-funnel metrics like ROAS and critical data such as country-level reporting to offer the best source of intelligence on campaign performance possible.
6. Complex conversion value encoding	SKAN Simple Set-up Singular provides the easiest-to-use SKAN set-up process available, with optimal recommended models, a super-simple UI, and automatic set-up to ensure marketers get it right. Plus: automated encoding and decoding, and the ability to simulate multiple models with historical data before deploying.
7. No cohort data	SKAN Cohorts Singular uses data science and statistical algorithms to give marketers back visibility into cohorted metrics such as D7 Revenue and ROAS, which they need to compare campaigns over time.

 

The solution is here: Singular SKAN

First you need the basics: setting up your apps to interface with SKAdNetwork, selecting endpoints for SKAN postbacks, enabling conversion models, configuring Facebook and other SANs that have specific integration instructions, and reporting on results.

Once you have the basics up and running, you’re ready for the next step.

And that’s taking SKAdNetwork from a framework that gives you incomplete data, minimal insight, and almost non-existent campaign optimization to a fully-functioning, high-performance, scalable user acquisition engine.

Smart Conversion Management

Singular provides more models than any other marketing measurement platform, plus allows you to change them on the fly from our dashboard: no app code changes required. They include:

1. Revenue
   1. IAP
   2. Ad monetization
2. Conversion events
3. Engagement
4. Funnel
5. Mixed models
   1. Revenue + Conversion Events
   2. Revenue + Engagement
   3. Revenue + Funnel

“Singular’s Smart Conversion Management reduced the time and effort needed to set up the conversion value schema, integrate it with partners, and receive postbacks data. Additionally, Singular provides fantastic customer support.”

— Frederick Ayala Gómez, Senior Data Scientist, Team Manager , Rovio

But it isn’t just about picking one: Singular helps you select the right model based on our experience of working with hundreds of apps, so you don’t end up with limited and unusable data. And Singular provides automatic event encoding plus leverages exclusive server-to-server functionality to give you and your partners the optimization data you need, when you need it.

SKAN Advanced Reporting

Data you can’t enrich, summarize, and activate is useless. SKAN Advanced Reporting enriches SKAN data by combining it with ad network and cost data, plus decodes conversion values and events, unlocking campaign ROI.

The result is both visual insight into performance plus raw data accessibility for those who want it.

SKAN Advanced Analytics

The cherry on top is SKAN Advanced Analytics, which uses data science and ever-improving machine learning to model missing data, generate predictive LTV, enable SKAN cohorts, and provide predictive analytics for key cohort KPIs.

Using this technology, enterprise customers have achieved instant modeled insights with  accuracy to ensure ad partners can optimize campaign performance in near real time.

Results

Singular SKAN is a complete solution that gives mobile marketers back the data and insights they need to grow at scale on iOS. That includes 7-day revenue within 87% accuracy, helping you boost ROI with rapid feedback loops that return the optimization levers you used to enjoy.

SKAN will never be as good as IDFA.

That’s the cold hard truth.

But Singular   due to SKAN. Improvements in reporting, over time, translate to improvements in campaigns. And this improvement has a snowball effect: better data lets you optimize faster, which then boosts results incrementally day by day, week by week.

Singular SKAN is getting better every day

Not only is Singular continuing to add functionality every single quarter, the machine learning algorithms running predictive LTV, cohorts, and modeling in Singular SKAN continually improve as they see more data, test more premises, and incorporate campaign results.

You need this. Here’s how to get it

Singular's is growing faster than any other MMP, and there’s a reason why. Our technology works, and our customers are beating their competition.

But that’s enough talk.

Schedule some time with one of our experts for a no-BS overview of how Singular SKAN works, what makes it better, and how you can now run high-performance user acquisition campaigns at scale on iOS again.

If you would like to learn more about Singular or schedule a demo for the SKAN Advanced Analytics suite, reach out to Niels Beenen ([email protected]). 

Here's what people had to say:

Rovio

“Singular’s Smart Conversion Management reduced the time and effort needed to set up the conversion value schema, integrate it with partners, and receive postbacks data. Additionally, Singular provides fantastic customer support.”

• Frederick Ayala Gómez, Senior Data Scientist, Team Manager , Rovio

Spacesheep

“Accurately measuring and attributing ad revenue was a challenge that got even more complex with SKAdNetwork. At Space Sheep Games, we depend on a consolidated view of revenue from in-app ads and IAP to understand ROAS. Partnering with Singular has been super valuable as they are at the forefront of the changes in the market.”

• Rene Retz, CEO - Spacesheep

Qiiwi

“Collecting and analyzing data from SKAdNetwork can quickly become a time-consuming pain. Singular's SKAdNetwork suite has helped us improve significantly. We've been able to optimize data collection and BI models to match our needs and to accurately predict future revenues.”

• Marcus Dale, CTO, Qiiwi Games

 

With the fundamentals solved,

Singular’s Complete SKAN solution

• A comprehensive and innovative SKAN solution invests heavily in executing the fundamentals well and simultaneously looks to how can we get marketers the same level of data, insights, and optimization that they had before iOS14.5
• Singular first addressed the fundamentals needs of SKAdNetwork to ensure marketers and developers alike have their needs met i.e. solving for and managing the intricacies of integrating with SKAdNetwork’s framework and the ecosystem that has evolved to work with the framework (re-thinking data relationships with ad networks).
• Singular’s Smart Conversion Management and SKAN reporting provide superior fundamental support with the most comprehensive solution in the market. Every aspect of Singular’s solution (even and especially with the fundamentals) is optimized to be as flexible as possible for sophisticated use cases whilst still providing an intuitive and resource friendly solution that does all the heavy lifting - promising ease of use for even the most resource constrained organizations.
• Once the fundamentals are addressed organizations (and I say organizations because SKAdnetwork doesn’t just impact the marketer - developers, data scientists, and marketers are required to overcome SKAdnetwork challenges) need to think about what’s next as the fundamentals ensure you’re compliant and that you can run campaigns, but the data is still so limited and the true