iOS14 and IDFA Changes: What It Means for Mobile Developers

We recently held a webinar to discuss the changes we can expect from iOS14 and rule changes to IDFA that are due in early 2021. It was an interesting, lively discussion with GameAnalytics and Tenjin, answering some of the developers’ and publishers’ most pressing questions.

We’ve had some time to reflect on all the topics we discussed. Here are our takeaways:

So what is IDFA?

IDFA stands for ‘identifier for advertisers. It allows advertisers to follow users through the experience – from seeing the ad in another app and then actually ending up in your app.

Apple first introduced it and it’s exclusive to iOS Devices. But there’s a similar system on Android. Users do have the option to disable IDFA, but it’s always on by default.

And what’s changing?

From some point in early 2021, the default for IDFA will change from ‘opt-out’ to ‘opt-in’. So the first time you go into an app, it’ll ask you if you want to be tracked in this way.

Apple originally planned to roll out this change with the release of iOS14, but it’s been delayed to an unconfirmed date. It should happen within the next few months. At the moment, roughly 30% of users actively go into the settings and disable IDFA. Once the default changes, most developers expect around 90% of players to opt out.

This will make for a fundamental change for the industry – one that most people aren’t on top of. We asked our audience about it in a small survey, and 75% say they’re ‘not prepared’ for the IDFA change.

What does this mean in practice?

IDFA has mainly been used for very granular, targeted advertising purposes – targeting a specific type of highly engaged, high spending user.

But soon, advertisers won’t be able to submit IDFA to Facebook to target specific people. Apple even went a step further by saying in their terms of service that you can’t even do this with emails anymore. If they catch you doing it with emails, they’ll ban you.

So the market for those types of ads will go away.

You can still use tracking links that don’t have IDFA and still work with non-SAN ad networks. People will be shocked by the SAN ones where all of those direct connections and MMP exclusivities for those APIs will break.

Apple is clamping down on advertising to specific users. And if you try to work around it, Apple will catch you and ban you. We think it’s a much better idea to pay the toll now and update your systems, so you don’t have to change the way you do things.

How will this affect the mobile gaming industry?

This depends on who you ask. Some people are taking a blind approach to the change – either ignoring it completely or assuming opt-in rates will be high enough for them to carry on as normal.

Others have come up with specific methodologies to work around the changes, but haven’t had any confirmation that they’ll actually be allowed by Apple. Some people think they’ll rely on fingerprinting technology. But Apple has said they’re against this.

We think the IDFA changes will cause a major shift in how marketers should be measuring their ROI – or even verifying that they’re getting return on investment. It’ll mostly affect companies that help developers figure out their attribution methodologies.

Mediation and analytics won’t be highly affected, but attribution will

It’ll soon be better to have an attribution tool than not to have one. This being said, not everybody is ready to pay for attribution. Some won’t have enough player volume to get accurate attribution enough to be worth the money.

In attribution, we’ll all be starting from zero again

When attribution was first introduced on mobile, it took at least four years to become standardized. The IDFA changes will likely start this cycle over again. Third-party IDs are out, in our opinion – we expect Apple to ban anyone trying to use third-party IDs to track across different publishers. So we really will be starting over again.

There’ll be a scramble to solve the problem

Whenever there’s a problem to be solved, money is made from the solution. So we expect many people to race to create the first product that fits the niche. This won’t be easy though, because it all depends on Apple’s accepted solution. And then we’ll see how Google reacts and so on.

Data science will become more important

We’ll be moving from a very deterministic world using IDFA to a non-deterministic world, using other kinds of signals you can get from an app. Computing and interpreting these signals for attribution purposes will need data science expertise.

SAN networks will be severely impacted

Half of the users have unlimited ad tracking. And in those cases, we see a drop in, for example, Facebook spend, or spending on a SAN network where integrations completely rely on IDFA. So if you’re using an attribution partner that integrates with a self-attributing network, like Facebook, Twitter, or Snapchat, all those connections will be severed when IDFA is opt-in.

How will this affect studios?

It won’t make things easier for any studios. But it will affect certain kinds of studios more than others.

It might have an equalizing effect

Not everyone has attribution. But the information that attribution provides is very valuable at scales. So if you’re a smaller or mid-sized studio, the playing field for you probably looks a little better than bigger studios. The gap in knowledge between studios will also get a bit smaller. So if you use the information you have, you should become more competitive after this change.

One type of studio will take the hardest hit

Any studio that relies on a small group of highly profitable users will suffer from this change. They won’t be able to tell when they’re buying these players or not. They’ll have a much harder time figuring out how to get ROI on ads.

But at the same time, these are usually the studios that can afford to buy the best tools. So they’ll also find very sophisticated data teams and get the best advice in the industry.

Hyper-casual studios should be the least affected

For hyper-casual developers, the economics don’t change – because they rely on users who have very small individual value, but very high value as a big group. So being unable to target specific users shouldn’t be too much of a problem.

People have been waiting for the hyper-casual wave to die down because it’s powered by ads. But really what iOS14 is doing is making it even more necessary to do many ads per user, which suits hyper casual fine.

But we’re seeing this consolidation trend in buying hyper-casual publishers to get a bigger user base to advertise to. With advertising to specific users becoming harder, we might see this trend accelerate.

So how will the mobile gaming industry look in 2021?

These changes to IDFA come after a year of huge economic turmoil. So with all things considered, we can make some fairly strong predictions for the year ahead.

We should expect a new period of growth

It’s not like users have stopped liking games. Many businesses in the industry have been affected by the economics of 2020, and advertising now becomes a little trickier, but we should see more money coming back into the market. As this happens, it’ll push CPI up to some extent and ECPMs. We’ll continue to see an effervescent market where more studios will be making money and launching new games – and new entrants will still be coming in.

We’ll see more innovation

Whenever there’s a fundamental change to regulations, there are always opportunities for anyone who can solve the problems it creates. The ways those users seek out new games will change without the ability to advertise directly to specific players. And there needs to be new technology to support that.

There are so many unknowns about how the platform will work and how the marketing will change. There’ll be many new companies offering creative solutions – some of which will work and some won’t. So expect to learn some new acronyms within the next year.

We’re going to see a lot more focus on broader types of games

As we mentioned before, the IDFA changes are much more of a challenge if you focus very specifically on a small group of high-paying players. So the types of games that create this monetization model will suddenly seem like a riskier investment.

Games with a broader appeal that can cultivate a much larger player base will now seem like a much safer bet for developers and publishers alike.

We might finally see brand advertisers getting involved

The increasing ‘snackable’ of games is leading more and more people to integrate mobile games into their lifestyles. Brands are finally realizing that regular consumers, not just gamers, play games while on the bus or sitting in a waiting room.

So the purchase experience for brands and the gaming experience is much more intertwined than they used to be. This was hard to ignore in 2020 when many people were deprived of the leisure activities they’d usually use to fill their idle time.

If brands dip their toes into the mobile gaming market, it’ll probably be small experiments. But it’s slowly going to accelerate as economies start to recover.

General tips for developers and studios

Read up on data science

Developers will need even more data understanding because you have less granular data to work with. You have to rely much more on correlations to understand your attribution, where your players are coming from, and how profitable they are.

Ask providers the right questions

You need to know how each provider handles things with Apple’s terms of service. You don’t want to get your app rejected for a reason you don’t understand, or for the ways they calculate an attribution or the ways you will measure your ROI.

You don’t want to go into a ‘black box’ scenario where you don’t know anything about what’s under the hood. So you need your providers to show you exactly how they comply with the terms of service.

Tips for testing games post-IDFA

Expect to change how you measure conversion values and how you spend against them

We expect to see less testing in Facebook and similar platforms, and more in non-San networks – simply because they’re cheaper and there’s no real ability to tell the difference between users.

When you start thinking about testing, you test to scale. And while you can still buy ads on Facebook that target certain specificities, scaling beyond that isn’t as applicable anymore. That’s going to be a shift where you’ll start to see more testing within non-sans.

Measure your LTV a lot sooner

It used to be that, for prototypes, you only looked at engagement metrics like retention, playtime, and CPI. But now you’ll need to assume a certain LTV, based on the engagement metrics.

Try using ‘ad ops’

Ad ops are like live ops, but for hyper-casual games. You measure your ad frequency from the prototype stage and run A/B tests like you would for your core mechanic.

This is important because you’ll now have more variation in how your ads perform. So look at the increases in LTV from changing ad placements or types. You’ll need to know this early to get a predicted CPI.

Talk to ad networks about how to test

All the ad networks are doing very different things with conversion values. Some are ignoring them. And the ones that are using them are doing so in very different ways.

So this is a big unknown for developers. When you go out there and talk to ad networks about how to test, try to understand what they’re doing with conversion values.

Watch the full webinar to learn more

We went into more depth on these topics in the full webinar, which you can watch on our YouTube channel.