Episode 9: IDFA and the future of the Mobile Gaming Ad Industry, featuring AdColony

Mobile GameDev Playbook podcast by GameRefinery

Apple shocked the mobile advertising world this summer when it gave a two months’ notice of a seismic change in how mobile advertising campaigns would be targeted, measured and evaluated with the arrival of iOS14 and IDFA changes. Luckily, with these changes now likely taking place in early 2021 instead, mobile game developers and advertisers have some more time to plan accordingly and explore new ways to build their acquisition and attribution strategies. But what is IDFA? What do the upcoming changes really mean? And most importantly, what marketing and tracking technologies will be available for the mobile games industry instead?

In this episode of the Mobile GameDev Playbook, we focus on Apple’s iOS14 ad feature updates and how to best prepare for the changes to the IDFA. Host Jon Jordan is joined by GameRefinery’s Head of Product, Lauri Heikkinen, as well as special guest Alasdair Pressney, Product Manager at AdColony – one of the world’s largest mobile advertising platforms, with a reach of 1.5billion users globally. 

We’re available on all major podcast platforms – if you enjoy the episode remember to hit subscribe!

• BuzzSprout
• Spotify
• TuneInRadio
• iHeartRadio
• Apple
• Google Podcasts

Topics we will cover in this episode:

1. What the IDFA actually is

2. What these changes mean for advertising and why the future is contextual

3. What we know and what we don’t know about IDFA’s impact

4. The importance of testing while we can during the IDFA grace period


Jon Jordan: Hello, and welcome to the Mobile GameDev Playbook. It’s the podcast that provides insights into what makes a great mobile game, what is and is not working for mobile game designers, and all the latest trends. I’m your host, Jon Jordan. Today we are going to be discussing the upcoming changes in iOS 14, notably those concerning user privacy and the IDFA, the Identifier For Advertisers. It sounds pretty dry, but it’s going to impact game developers and publishers potentially quite substantially.

We’re going to be looking at that and also what solutions exist. Recently, last week, we just heard Apple was going to push this into early 2021. So originally it was going to be mid-September, so we’ve got a bit more time than we thought, but it’s still important for developers and publishers to be thinking about this now because it’s going to happen. Joining me to discuss this is Lauri Heikkinen, who’s the head of product at GameRefinery. How’re things going?

Lauri Heikkinen: Hi, Jon. Things are great. I’m happy to be here. This is actually my first podcast, so really looking forward to it.

Jon: Good. Good. Excellent. Do you want to give us a bit of an explanation about what you do as head of product?

Lauri: Yes, sure. Well, I think my main responsibility is to ensure that we deliver the best product possible to our customers, and of course, understand the market as well as possible. I’ve been responsible for our SaaS product for a little bit over four years now, and more recently, I’ve been working on the data enrichment product as well, that we just recently launched.

Jon: Yes. We can talk about that further in the podcast, pretty important in this kind of context. Our expert guests, not that Lauri is not an expert, but our particular expert guest in terms of advertising, is Alasdair Pressney. Who is a senior product manager at AdColony, one of the largest mobile ad platforms in the world with more than 1.5 billion monthly users, how’re things going, Alasdair?

Alasdair Pressney: It’s going well, thanks. It’s good to have an extra three months to work through these things.

What IDFA actually is an why it’s happening

Jon: Yes, absolutely. To set the scene, we’ve been throwing around some acronyms already, which is always a bit dangerous, I think. If you could kind of give us a non-technical overview into what the IDFA actually is, and what is happening to it in iOS 14, so we can have a level playing field for the discussion, which is going to get a bit more technical but let’s start simple.

Alasdair: Sure. Let’s start with why. Apple cares a lot about privacy because the users of Apple devices care a lot about privacy. What they’ve done is they’ve actually made a lot of privacy changes in iOS 14, and there are two or three changes that are really significant for the advertising industry. The first one is that there’s this thing called a device identifier and think of it as the serial number that’s associated with your device and it goes everywhere you go.

“Apple cares a lot about privacy because the users of Apple devices care a lot about privacy.”

Alasdair explaining Apple’s privacy approach & IDFA

The first thing that Apple’s done is they’ve made it so that users have to opt-in to sharing this device ID. If the device IDs aren’t shared, then it’s really difficult for advertisers to track these users across apps and websites, and it makes it a lot harder to do things like analytics or measurement or to target with advertising or even things like advertising measurement. That’s the first thing that Apple did.

“It makes it a lot harder to do things like analytics or measurement or to target with advertising or even things like advertising measurement.”

Alaisdair explaining Apple’s IDFA changes

The second thing that Apple did was they introduced what’s an attribution system, an ad measurement system called SKAdNetwork, which allows ad measurements to take place when IDFA is not available. Apple is saying, “Hey, guys, we’re not going to let you track users if they don’t want to be tracked, but we’re still going to let you measure the success of your advertising campaigns if the users are not trackable.

Jon: Good, excellent. That’s a very good overview. One thing just to point out, so obviously, this is a big change that Apple has made, but this is part of– Certainly, as you mentioned, a long term plan for Apple to be a privacy-based platform and more generally, we see across all nationalities and across most companies in this space, we see more privacy anyway, so it’s not just Apple that’s pulled this one out of the hat, this is Google and similar companies are moving in the same direction. Is that broadly correct?

Alasdair: I think the best way to express that or see that change which is happening is by looking at the privacy legislation that’s come out in various countries. In Europe, we have GDPR, in California we have CCPA, other states are considering their own privacy laws. There’s one that just came out in Thailand, Australia is also looking at one too, so it’s becoming much more of a global thing.

Jon: Correct me if I’m wrong, which could be the case, but as far as I understand it, at the moment, in iOS 13, if you went into the settings, you could opt-out of tracking across all apps as a universal statement so you couldn’t be tracked. With the iOS 14, the interesting switches, when you download an app now, you’ll be asked- there will be a pop up on a per-app basis whether you want to be tracked or not.

In some ways, it’s a more granulated decision, but I guess we assume that most users when they get a pop up saying, “Do you want to be tracked?” Are going to say no. That’s why we are so interested in this particular change. Is that correct?

Alasdair: That’s right. Today, there’s about 30% of users who don’t let their device identifier be shared. What’s going to happen is this option is going to be made a lot more obvious and a lot more available to any user that downloads an app and installs an app for the first time. As opposed to users navigating to an option that’s a little more hidden, you have something that’s going to pop up right in front of you. We expect that the number of users or the percentage of users who are going to say, yes, I agree to tracking is going to be more. Right now it’s 30%. It could be more than 30% in the future.

“We expect that the number of users or the percentage of users who are going to say, ‘Yes, I agree to tracking’ is going to increase. Right now it’s 30%. It could be more than 30% in the future.”

Alasdair commenting on IDFA predictions in how the public will respond

Jon: Yes, we have some interesting stats that could be. I don’t know how well organized these stats are or whether they’re gut instincts, but people seem to be saying like 70%, 80%, 90% of people are going to opt-out now. It seems like even if it’s a smaller number because you just lose the granularity across what’s going on over the entire ecosystem, the whole thing falls apart. Lauri, are you someone who opts in? How focused you are on privacy at the moment? Or maybe you’re using Android, so it doesn’t affect you?

Lauri: No, I’m on iPhone, but so far, I really haven’t cared, but I think I like to get targeted if that helps me to get better games, yes. I’m the person who likes to get ads delivered as sharp as possible.

Jon: It is an interesting sort of thing, isn’t it? Because I guess a lot of this privacy stuff gets confused with maybe hacking or all that sort of stuff, where privacy is actually that, as far as I understand, the IDFA isn’t tracking you as an individual, it doesn’t know your name and where you live. It just gives an anonymous tracker to you so they can tell what that device is doing.

I’m with you, but maybe we’re losing breed. It’s nice to get adverts related to things you’re interested in, rather than adverts for things that you’re not interested in. I guess things are changing, and I guess we will have the chance to opt-in now on a per-app basis. There’s some apps we might not want to be tracked in and some apps we might be very happy with. Okay, so let’s move on.

You’ve mentioned a little bit about the SKAdNetwork, which is Apple’s proprietary ad solution or effectively will now be replacing IDFA. Alasdair, can you say a little without getting too technical how is this different and what kind of functionality are we losing in this new system that Apple is effectively asking everyone to use?

Alasdair: I think it’s important to note that I don’t think it’s necessarily different from what we’ve seen in the past. Before the IDFA existed, there was a different identifier that was used called a UUID. Apple replaced that with the IDFA a few years ago. The main reason they replaced the UUID with the IDFA is that the IDFA gave the users the ability to turn on or turn off whether or not it was going to be exposed.

Now what’s happened is Apple’s just taking that option to turn on and turn off whether or not that identifier has been exposed, and they made it significantly more obvious. We’re talking about a gradual shift over a long period of time that’s manifested in a change today, where there’s a pop up that exists for the IDFA. This is something that’s been a long time coming.

Jon: Although we’ve pushed it out a few months, we’ve seen some quite kind of apocalyptic headlines written around this. Obviously, we don’t know exactly how the rollout was and how the potential solutions that are going to come into place are, but do you think, potentially it will completely disrupt this– Will it completely tank kind of UA, user acquisition for mobile game developers? Or is there more nuance in there, it’ll be like a journalist slightly making up headlines. We don’t do that, really. 

Alasdair: Performance advertising is about data-driven decision making when you’re doing your optimizations, and so when you’re tracking individual users, and the ability to track individual users goes away, then the data that you have at your disposal for making those campaign optimizations is no longer available as well, so advertising if you’re a performance advertiser, someone who tries to get their app installed on a user’s device, for example, they’re going to have to make an adjustment.

“Performance advertisers will still be able to do advertising. It’s just they are not going to be able to target those individual users that they know are going to provide a certain value to their app.”

Alasdair describing the limitations that performance advertisers will have after the IDFA come into effect

With the absence of specific user-level targeting and advertisers no longer being able to snipe specific users that they’re going after, they will, however, still be able to target specific users in a contextual way or a more broad basis. They’ll still be able to do advertising. It’s just they are not going to be able to target those individual users that they know are going to provide a certain value to their app.

What this means for advertising and why the future is contextual

Jon: They’re targeting and can they do, will they be able to look at other apps or games that I’m playing or how will that work or will there just be- it’ll just be everyone will get Pepsi ads sort of thing with that level of advertising?

Alasdair: Contextual targeting is anything that doesn’t involve an individual user or device. It could be things like which country the user’s in, what game they’re playing, what operating system they’re running, or what device they’re on in general. As opposed to a specific device they can say things like or they can identify things like, “Oh, I’m on an iPhone 10, but I don’t know which iPhone 10 this is.” Those sorts of contextual targeting options have always been made available and those will continue to be made available into the future.

Jon: Lauri, you mentioned in the introduction that one of the things you’ve been working on recently is the data enrichment product. This actually is interesting in a sense that GameRefinery has been very focused on a specific way of looking at quantifying mobile games and who likes them. That’s been more a game design sort of tool. Game developers have used your data to see what features and what things they should be doing in their games or maybe not what they should be doing, but looking at what their competitors are doing sort of things.

It’s a game design tool. This new data enrichment is more in the advertising sphere. What was the vision behind that and how do you think it plays into this new advertising ecosystem, which is less focused or less targeted on the individual idea of users, but maybe more interested in broadly what games people are playing?

Lauri: The main idea is still the same as we’ve done previously. We’ve been deconstructing games and trying to get different types of features out of there such as sub-genre defining features and we’ve also expanded to motivations and demographics. What’s the visual style, try to paint as good of a picture of the game as possible. Previously, we’ve been only doing that manually, but more recently, we’ve been developing machine learning models that can automate this data.

We’re now able to cover basically every game in App Store and Google Play and how the contextual targeting ties to this. First of all, contextual targeting per se isn’t new, but it hasn’t been used that much in mobile because of the performance targeting ability to target individual users. What we are now trying to chip in is that we’re able to tell you more about the game and publisher ID is still past even in SKAdNetwork.

We can inject some data into that. The idea here is that we can match advertisers with the most suitable publishers. The hypothesis here is that, if I’m, for example, shown a game that is similar to a game that I’m playing, it could be in terms of sub-genre, or it may be the motivations are the same, or maybe the average demographics is the same or who is playing the game. I’m more likely to, first of all, maybe install that game, but also play that game more, maybe even monetize, look at, and so forth.

I’m just trying to narrow the gap and give transparency to the apps where you’re advertising and also give the tools to both the game developers and to ad networks, give them the data about the advertisers and publishers for themselves to do the analysis and optimization when it comes to contextual targeting.

Jon: I guess at the highest level it would be something like I’m playing some sort of match-3 puzzle game and obviously, to give you an advert for another match-3 puzzle game. I guess that’s so obvious that you actually- that would be something within your framework, but it’s so obvious that doesn’t make any sense. You’re not really enriching any data to do something like that.

How surprising could be the adverts that are being supplied to someone playing a game because I guess you’re talking about when it comes to game and motivations, that is potentially very different to genre or subgenre. Could it be the case that someone’s playing a match-3 game and for some reason, they get an advert for a hardcore strategy game, and you somehow found this like an audience affinity between those things? Is this possible? Is that likely?

Lauri: If you look at the individual users, then it’s unlikely, but when you look at the aggregates of users, then it’s actually likely that we are able to give you recommendations for the apps that would work better as a publisher as for your own game.

Jon: I guess you’ve built these- how many data points per game is it now that you’re collecting?

Lauri: Currently, we have a little bit over 700 data points and, of course, as you mentioned, if you’re looking only at match-3 games, that’s your obvious answer, and you know that the top match-3 games that you would like to go for, but usually, they are blacklisting the biggest competitors if they are even showing ads. The publishers might be elsewhere. The thing is that there are tens of thousands of games that are potential publishers.

What we try to do is we try to use different signals in the data set, whether it’s a sub-genre, motivations, demographics, or visual style, and come up with a list of games that you could advertise in. As you said, they can vary a lot. It could be so that a match-3 player is playing an RPG game, but it could be that there are actually motivations that are shared among the games.

“With Ad Enrichment, we use different signals in the data set, whether it’s a sub-genre, motivations, demographics, or visual style, and come up with a list of games that you could advertise in”

Lauri explaining how Ad Enrichment works

What we’ve been trying to do is that we actually use either the game developers or the ad network’s actual performance data to use that as a predictor. We’ve been modelling against that. We then know what types of games have historically worked and then use the whole of metadata to then optimize towards that.

Jon: You have your theoretical view on what should work, and then you fine-tune that with the feedback you get from running all these ads. Then the combination of the two things is what gives you the secret sauce, the finesse to maximize the performance advertising hopefully.

Lauri: Yes, exactly. There’s, of course, that second part of the equation, which is the creatives. We also have models for that. We’re able to pull metadata out of the creatives, which completes the trio. Finding the fit between the advertiser, creative, and then the publisher game. We’ve seen that there’s quite a big correlation between the fit and then the quality of traffic.

“It is about finding the fit between the advertiser, creative, and then the game publisher. We’ve seen that there’s quite a big correlation between the fit and the quality of traffic of a game.”

Lauri talking about ad success in games 

Jon: Alasdair, this contextual targeting, is this something you think has promise, particularly in this situation where we’re focusing less on individuals?

Alasdair: Absolutely. One of the first things that happened when Apple announced the IDFA changes were a lot of partners that AdColony works with, came to us and said, “Hey, what other contextual features can you provide us so that we can do different sorts of targeting?” Having these sorts of options available, especially the unique app analysis when it’s done manually, like the way that GameRefinery was doing to initially be able to scale that, it’s pretty great.

Having those options available for contextual targeting in tomorrow’s world of advertising is going to be very useful. At the least, advertisers are going to test it out and see if there’s any strong correlation. Then if there’s a correlation, advertisers will define bidding strategies around it. Right now, just having the ability to go explore new contextual targeting features is something I would encourage any advertiser to do in the absence of user-level targeting.

“Having the ability to explore new contextual targeting features is something I would encourage any advertiser to do in the absence of user-level targeting.”

Alasdair giving advice to advertisers in response to IDFA

What we know and what we don’t know about IDFA’s impact

Jon: More generally, what other approaches are people experimenting with to try and get back this granularity that they had previously?

Alasdair: Removing the IDFA caused some fundamental features in ad targeting to go away. For example, without an IDFA, it’s very difficult to do frequency capping. Just limiting the number of ads you would show a user because if you don’t know which device you show an app to, then you can’t keep a record of the history of ads that you’ve shown to that user and so, advertising is basically doing more or less a full restart.

Frequency capping has gone away but also things that are more targeting base like segments or targeting specific user segments. A list of device IDs is still going to be there, but the size of the list is going to be considerably smaller because you’ll have fewer IDFAs available to generate those lists. 

Another thing that’s not going to be feasible as much is anti-targeting using a suppression list. A suppression list is very similar to a segment where you generate a list of device IDs, and you say, I don’t want to target those devices. When you’re advertising an app campaign, it’s one of the things that a suppression list is used for is listing out the users who have already installed your app, because in that case, you don’t want to advertise to those users again.

That’s also going to be severely diminished with having less IDFA availability. Ultimately what this means is that there could be some implications for the efficiency of campaigns and so, campaigns are going to have to find new ways to be efficient to adjust for this, and that’s where contextual targeting comes in and fills the gap. It’s not a one-to-one replacement of these features. These features aren’t going away. You’ll be able to do it for the percentage of users that do opt-in but to maintain efficiency, you have to look at other areas, and so, contextual is the best place to go.

“Ultimately what this means is that there could be some implications for the efficiency of campaigns and so, campaigns are going to have to find new ways to be efficient to adjust for this, and that’s where contextual targeting comes in and fills the gap. It’s not a one-to-one replacement of these features. These features aren’t going away. You’ll be able to do it for the percentage of users that do opt-in but to maintain efficiency, you have to look at other areas, and so, contextual is the best place to go.”

Alasdair on how the IDFA changes will impact campaign efficiency

Jon: If I’ve understood you correctly there, there’s quite a lot of games that have in-app purchases and advertising and one of the things might spend $5 and if we find something we’ll switch off advertising. That kind of has gone away now, because they won’t know who I am to switch off my advertising. Is that right?

Alasdair: I’m going to relay your question back to you to make sure I understand it.

Jon: Yes.

Alasdair: The question was whether it’s going to be possible to track a user after they’ve made a purchase for an app?

Jon: Yes.

Alasdair: That would be possible for, within a specific publisher or developer, because what Apple’s done is they’ve exposed an identifier called an IDFV, and that allows publishers and developers to have a serial number that they can use to track that user within their own games. But that serial number, that identifier is going to change from one game to another so you cannot track that user’s purchase history from one developer to another developer. The IDFA could do that.

Your publisher development is still going to be able to measure things like LTV and things like that, but they’re not going to be able to do it across multiple apps and websites unless the user opts-in.

Jon: Yes. Do you think that we’ll start to see not just game developers but app companies in general, in some ways, try to incentivize people or try and explain to them in broad terms why they think they should be opting in, or would Apple allow them to do that? Are they not very keen on the incentivized actions in general?

Alasdair: It’s possible. We’ve seen something similar in the webspace already. There are cookie walls that exist on the browser. I think the big question mark right now is whether or not that’s going to be permissible, not just by Apple, but also by various legal authorities in Europe and the US who are building and enforcing these privacy laws. You have to clear two hurdles now.

You have to clear both Apple and legal authorities. If you’re able to clear both hurdles, then I can see incentivize opt-in as a way for publishers to get a higher percentage of users to opt-in, provided that they have a strong value proposition for their users.

Jon: Lauri, would you- there is something you would consider opting in, do you think, gamers, taking yourself as a mobile gamer? Do you think there’s something, if they were given just kind of an argument or the facts for why it’s good, that you could see maybe a slight reversal of consumer behaviour towards adopting an opt-in, at least for apps and products that they’re really interested in? Because obviously, we have this, the things we really like, we’re fans of, and we feel like we want to help the developer and for other things, we might have a very different attitude?

Lauri: Yes. I think at least for me, it’s about transparency. Then Alasdair mentioned the cookie walls and I just pictured a few of them that I saw today, and I usually approached through them and if they make it sound reasonable and if it’s not too vague, I usually just go for it. Of course, it depends on the site as well, so I would assume that would be the same thing with apps.

It depends from app to app, but also how they would be using it and how would they communicate it to me. I would be ready to opt-in if I feel fine with it; otherwise, I would opt out probably.

Jon: We’ve been talking about some of the nuts and bolts of what’s going away and what potentially is coming. I guess that could still seem, I think, to lots of people, very abstract. I guess that’s because it’s hard to understand how integrated mobile advertising is to the mobile games industry. It seems like obviously, in-app purchases drive most of the revenue. There’s a lot of money from ads, but it’s small compared to in-app purchases.

I think it’s worth, maybe setting out the knock-on effects, I guess, of what we think is going to happen with Apple. Alasdair, you can- I can come to you, and you can tell me if I’m wrong, but it seems to me that a lot of developers know that they like this ad tracker because they can find anonymous individuals and they can know exactly what those people have done if they’re big spenders or if they’re really into a certain type of game. Because they know that, that means that they can, when they’re doing their advertising, they’re really interested in getting those people to play their games.

They will bid higher amounts for someone who they know has been playing certain types of games and they’ve spent a certain amount of money and has been in these games for a certain period of time, because they know, they can see they can generate a return on that. They’ve built those kinds of models. That also means that the people who have games that have- are showing advertising. Obviously, they get the share of the advertising that’s being spent by the developer who wants to attract people to download his game.

Some of that goes to the developer who’s shown the ad. In a sense, advertising becomes so big because it’s win-win. After all, the advertiser is going to find potentially a quite small group of people who are really interested in their Japanese RPG who live in Norway. If they wanted to find those people, there’s probably ten of them or something, but they could find them and advertise to them and make money from them, and the people who are showing the adverts will also get that money.

Now that’s going to go away. Particularly on one end, you may see a lot of developers who are doing quite well in terms of the advertising money they made. Now, they might see a quite significant drop-off in advertising revenue. Is that right? Do we have any idea quite how big a drop they could be impacted by?

Alasdair: It’s really hard to make firm projections at this point. As you said, we do know that user-level targeting is largely going to go away. To use an analogy, essentially that the toolkit an advertiser’s going to use is going to switch from a very accurate sniper rifle to go after very specific users, to more of the advertisers on a lake somewhere fishing for users and it might be quite dark, so you don’t exactly know where those users are.

As opposed to making your targeting as accurate as possible, it’s now more about making sure you’re fishing in the lake at the right time of day, or you’re using the right kind of lure. Again, it’s much more about contextual targeting. Now, we don’t know how advertisers are going to set up their campaigns around contextual targeting because they haven’t done it yet.

We know that advertising revenue flows from the advertisers to the publisher. So, it’s really hard to make any projections because we don’t know how advertisers are going to bid using their new toolset that they’re going to have. We don’t know what their tool set’s going to be yet and therefore, we don’t know what publisher revenue is going to look like either.

Jon: You’re not getting– I guess you can’t publicly go, and I was looking to see if you’d get any kind of percentage figure because that’s what we like to see. Revenue is still going to be down by 50%, obviously. That’s not very nuanced.

Alasdair: There’s just no data that we can use to make an accurate prediction, unfortunately.

Jon: The reason that we’re doing this podcast, and it is right about this is, we don’t expect it to be fairly substantial, I think, do we? We don’t expect it to be like a 5% drop. We can’t tell, it’s a prediction, but we expect it to be- perhaps some of the smaller developers are going to be more here, I don’t know.

Alasdair: When advertisers are able to target specific users, they are able to make bids that are significantly higher for those specific users that are high value. Something like that might not be possible anymore and as a result, we might see lower bids across the board if the upper percentile bids are no longer possible. Maybe those users still opt-in. We don’t know. It’s difficult to really make guesses based off of that.

Now, basically all the users are treated in aggregate, so whatever targeting rules that you create contextually when you’re targeting these users, the price that you set for acquiring those users, the cost per install is going to be based off the aggregate of all of those users that are within that group, as opposed to having a specific bid for segments of that group or something like that, where it’s much easier to go and do price differentiation.

Because there’s less price differentiation and because we’re treating things as aggregates, we might see different pricing behaviour but if everything in aggregate results in a very similar average bid, then we may not see it drop in advertising revenue. Again, it’s very difficult for us to make these projections without actually seeing some data first, and for that, we need the advertisers to go in and build campaigns for this new world, contextual signals.

Jon: I can see it’s more complicated of course than I’m suggesting. Which is why you’re the expert and I’m not. Broadly, everyone’s happy that Apple has pushed it into next year. They’ve been a bit unclear. They just said early 2021. Do we think there’s any potential that they may be dialing back on some degree on what they’re up to, or are their lots of other things going on with Apple’s kind of world at the moment, there’s the whole Epic thing and the EU and stuff like that. Do we just think they just pushed that away for a little bit because it’s easy at the moment rather than there’s going to be any substantial rethinking of what they’re up to?

Alasdair: I don’t think there’s going to be any substantial rethinking of what they’re up to because again before the idea failed there was a UUID. They implemented some privacy features and replaced the UUID with the IDFA. They’ve implemented privacy features on Safari, and now they’ve just moved on to making some changes to the IDFA. We’re talking about a long-term privacy-focused shift in the industry that’s really started with Apple. I don’t expect that to change anytime soon.

I think what’s happened in this case is developers and advertisers just simply need more time to work through what they need to do to adjust. Apple has decided to give us more time because that’s good for the overall ecosystem. The last thing Apple wants is for app developers to have difficulty updating their products for the app store because then there won’t be any apps in the app store. They’re trying to make sure that the ecosystem is able to transition in a way that maintains continuity, which I think is a good thing.

The importance of testing while we can during the IDFA grace period

Jon: Okay. Time to wrap up now. Advice for developers and other people in the ecosystem. What should they be doing in these few months of grace that they’ve been magically granted?

Alasdair: It’s very important to test. What Apple has essentially done is they said, “Hey, you still have to make these changes. You just have a little more time to make these changes, and you’re going to be able to release your app on iOS 14, and we’re not going to enforce these changes that are going to be made for a few months.” As a product manager at AdColony, I would encourage all of our developers to test what this new world looks like.

“It’s very important to test. What Apple has essentially done is they said, “Hey, you still have to make these changes. You just have a little more time to make these changes, and you’re going to be able to release your app on iOS 14, and we’re not going to enforce these changes that are going to be made for a few months.”

Alasdair on the importance of testing before the IDFA changes come into effect

Take a small percentage of your devices and request the Apple pop-ups, see how your BI systems behave, see how your ad targeting behaves and your LTV analysis behaves when you do this. Also when you’re doing advertising, set up a few campaigns that run on SKAdNetwork. There’s quite a lot of limitations with SKAdNetwork that you need to be aware of. It’s better to plan ahead for when enforcement does come so that you are prepared. 

Just to give an example of the limitations, when you’re setting up campaigns, you can only use up to 100 campaign IDs and think of a campaign idea, some distinguishable way of targeting a certain supply. For example, if I’m targeting an app as an advertiser with 100 campaign IDs I can only split my campaign up to 100 distinct ways and still be able to measure. If you’re mixing in different creatives or you have different prices by different countries or different prices by different device tiers, you’re going to get 200 campaigns very quickly.

Advertisers should start thinking about that ahead of time and start planning ahead, restructuring their campaigns. Actually now that Apple has given us a few months where there’s going to be apps that do request for pop-up and apps that don’t request the pop-up and Apple is going to let both kinds of apps exist you have the ability to test, and you have the ability to make sure that you’re prepared for 2021.

Jon: Excellent advice. Thank you very much for that. Thank you, Lauri for being our expert from GameRefinery.

Lauri: Thanks, my pleasure. Basically, from my point of view, when I’ve been talking to game companies, a lot of them seem to be all over the place. What I say to them is first of all, don’t panic. The change will come, but it’s the same for everyone. As Alasdair said, test all the things and just be open-minded about that shift in the mindset to the new world and things will work out.

Jon: That’s pretty calm advice. Don’t panic. Don’t panic and test. You’ll find your nirvana. Thank you very much for that, Lauri. Thank you very much for your time, Alasdair.

Alasdair: Thank you for having me.

Jon: Thank you very much listeners for listening. That’s our time up for today. Just a reminder that we are doing podcasts every month about what’s going on in the ever-changing world of mobile games. Please do find us on your podcast provider of choice if you’re not already subscribed, and we have a whole backlog now of episodes covering a whole range of stuff.

If you’ve come across us for the first time, please check out what we’ve been up to. If you like us give us some feedback on your podcast platform of choice, always good to get some reviews. Thank you very much for listening to this one, and come back next time.

Trusted by leading game studios and publishers

Zynga logoWooga logoG5 Games logoRovio logoUbisoft logoWargaming mobile logoFunplus logoGameInsight logoludia grey logoMag Interactive logoPixonic logo