VIEW THE SLIDES: Slides



ABOUT ROCKETSHIP HQ: Website | LinkedIn  | Twitter | YouTube

KEY HIGHLIGHTS

🤔What exactly is this Apple update?

📚The history leading up to this decision

📱What is the SKAdNetwork?

⚠️What will happen to the different elements of the mobile ad ecosystem? – MMPs, Facebook and Google, DSPs, SDK ad networks, Android, and you and me

📍Impact on ROAs and CPM without user level data

🏃🏽‍♂️The expectation that Google will likely follow Apple’s track

🔄Marketing team transition from last click focused to incrementally focused

🛠Very different tools for a post IDFA world

💭Looking forward to an era of challenges, change, and opportunity 

FULL TRANSCRIPT BELOW

Hey everyone, welcome to this edition of the Mobile User Acquisition Show. And this is an important, very crucial time for all of us as marketers because Apple has changed what has been the very basis of mobile marketing by making IDFAs to be opt-in rather than opt-out. This is going to take effect this September. 

These are challenging times, these are monumental times, or these can be times of opportunity. This is a presentation I made as a part of a YouTube Live earlier today on the 26th of June. I will walk through the presentation in this audio. To start off, I started off the presentation with what is an apocryphal Chinese curse: “May you live in interesting times. May you be recognized by people in high places. May you find what you’re looking for.”  I think all three of those elements of that curse can be applicable in some way to our situation to where we are as an industry today. 

As I had said, what has happened is that IDFA will become opt-in, which means every app, starting with iOS 14, in September 2020 will show a prompt that says, “Hey, this app would like permission to track you across apps and websites owned by other companies.” Users will get to pick yes or no. Very clearly, most people will just say no, and that’s really the problem. Even right now, 30% of people have said no to having LAT enabled. When it becomes completely opt-in, I suspect that number is going to be no more than 10%. To give you guys some sense of the context around what sort of opt-in rates you could expect, push notifications, which are opt-in right now on iOS, have an opt-in rate of 40 to 60%. GDPR, which I think is very poorly enforced and very poorly thought out – just has a ton of loopholes – has 70 to 90%. I don’t think IDFA is going to be anywhere close to that. It’s going to be far, far less than 40%, I would say. I don’t think GDPR is any sort of comparison for what sort of percentages we can expect in the opt-in. 

How did we get here? How did we get to this point where IDFA is no longer a valid option for marketers? Until 2012, Apple had something called the UDID, which was a persistent identifier, which was problematic because this was persistent for a user. Apple said, “Right, we’re going to have an IDFA, and LAT was added in 2012 as a part of this. Except it was buried within the user’s device, it was really hard to find it. If the user turned LAT off, marketers would know that it’s off, but it was up to the marketers whether to do anything about it. Just kind of crazy. In 2016, Apple said, “We’re going to make LAT such that if a user turns LAT off, a marketer cannot target them because their IDFA is shown as zero.” 

The next step was 2019 When Apple said, “Oh, on kids’ apps, we will not allow the presence of any SDK or analytics or cracking systems at all.” I think that was a trial balloon by which Apple said, “Look, this is the way things could be going forward.” They took the next big step this last Monday. Now, as a part of the recent changes, there is something that Apple calls the SKAdNetwork. 

I think the SKAdNetwork is very rudimentary. It can take the place of what the MMPs have today, but really it’s an API that allows an advertiser to track ad networks, source apps and conversions from them – really that’s it for each ad that is run by an advertiser. If you think about it, it’s far, far less sophisticated than the MMPs and a lot of the in-house infrastructure that a lot of the more sophisticated advertisers have. If you think about it, Apple does not have a great track record when it comes to ad tech – iAdd and Search Ads are just terrible products. What could happen, is that the SKAdNetwork API can be built on top of. I do think a lot of companies can build some interesting products on top of the SKAdNetwork. 

I would like to look at what happens to the different elements of the ecosystem: MMPs, Facebook and Google, DSPs, SDK ad networks, Android, and you and me – marketing and UA teams. What happens to all of us? First, I also want to demarcate what we do know and what is TBD. MMPs in the current incarnation are based on IDFA — they just can’t work because there’s no other way they can get personally identifiable information. What is possible is they can build different kinds of tooling based on the SKAdNetwork because they do have a lot of historical data on which channels are working and which channels are leading to incremental performance. Definitely, I think they can build different tooling based on SKAdNetwork. Now what’s also TBD is whether campaign level, aggregate data, because SKAdNetwork is giving us aggregate data. Is this going to result in the same level of performance as right now? I don’t think so. Performance will drop. 

Next is Facebook and Google. We do know that Facebook and Google control very powerful user graphs. They sit on massive amounts of purchaser data, but Facebook SDK in its current form relies very, very heavily on the IDFA. It can work without the IDFA, but it’s going to be much less powerful. It’ll be helpful to imagine one world that is similar to web-based businesses today where they operate and scale on Facebook and Google without an MMP equivalent. I think this is going to be very similar to the world we might find ourselves in. Facebook and Google could show less targeted ads because Apple search does allow you to show ads targeted to LAT on users — just that these ads are very untargeted, and that’s certainly possible. What happens if the ads are untargeted – their ROASes are going to be worse, their CPMs are going to be worse, performance is not going to be great, spends are going to drop. That’s a very real possibility. 

The other things to note around Facebook and Google is that they do have an alternative in-house device graph that they could make work just without an IDFA. They could use the IDFV basically because if Facebook has an MAU of 2 billion, it has unique identifiers on each of them. That’s an IDFV that is still present, and that is very, very powerful for Facebook to use to inform its Facebook ID, which is a different unique ID. They could use different forms of fingerprinting. They could use contextual aggregate data that doesn’t involve IDFAs because they are still sitting on a pile of interest-based data – data from Facebook’s earlier days, so they do have a lot of data. What could also happen is Facebook and Google can strike side deals with Apple to run ads via SKAdNetwork. That is a very real possibility. The important question is what happens to ROAs and CPM without user level data – it is going to go down.

The next important sector is DSPs. What we know is that the DSPs cannot rely on device graphs. They can do retargeting via IDFV, but really, it’s going to be challenging to see how it might be operationalized because the IDFV cannot be used across different companies’ apps, so there’s no competitive advantage to be had with IDFV. That is something that’s still TBD. I suspect that without user level data, if DSPs purely relied on contextual signals, performance is not going to be great. CPMs are going to drop, and ad monetization revenues are going to drop. There is going to be a challenging time, there is going to be a tough time with ad monetization. SDK ad networks just can’t operate in the way that they work right now. Header bidding and unified auctions are meaningless because you can’t have user level bids. What could happen would be contextual, non-personalized ads. Again, the CPMs are not going to be great, ad monetization revenues are not going to be great, so this is going to be a challenging time. This is going to be a tough time for a lot of players. 

What about Android? Android has no plans to make it opt-in yet. They do have an opt out of ads personalization feature, and that opt-out rates are far, far lower than iOS. It’s about 2 to 3% as compared to 20 to 30% on iOS. Google faces the same anti-privacy pressures, though, and they have been known to follow what Apple has done. I do realistically expect that Google will have something very similar and comparable to a no IDFA regime.

What about marketing teams? How are marketing teams going to change? A lot of our work is going to go from making deterministic decisions to probabilistic decisions. We’re not going to be able to make clear, quantitative, precise judgments, but more of judgments and decisions at a campaign level, at a segment level. This is going to be more incrementality focused rather than last click focused. That’s the reality of how things will be. The era of large media buying teams is going to give way to marketing strategists because we’re not going to be working out a ton of Excel sheets much like 2015-2016. We’re going to be more like traditional marketing organizations. 

Yes, there have been headlines such as marketing is dead, CMOs are going to die, user acquisition is dead. I don’t think that is true. I don’t think that will come to pass because mobile apps are an 80 billion industry. Mobile marketing spend is going to touch north of 200 billion. If I recollect correctly, mobile apps will be there, marketing will be here. We will just need tools for a post IDFA world. You will just need very different tools compared to what brought us here, and not everyone will pivot successfully. 

There are a lot of opinions floating around in this post IDFA of a world. I would say ask questions. Ask yourself, “what are people’s incentives?” Over the next couple of weeks and months, we will be waiting and watching and learning. We will be aiming to share first-hand experiences of our own and of people we know. Please stay tuned, we will keep you all posted about what happens and what changes in this very, very monumental time. It’s a potentially very difficult time for a lot of us. 

I will conclude with this quote by Arthur C. Clarke: When a distinguished but elderly scientist states that something is possible, he’s almost certainly right. When he states that something is impossible, he’s very probably wrong.” I don’t want to say something is impossible or something is going to die. I think what lies ahead is an era of a lot of change, a lot of opportunity, and I look forward to keeping you all posted on what happens next. Thank you 

A REQUEST BEFORE YOU GO

I have a very important favor to ask, which as those of you who know me know I don’t do often. If you get any pleasure or inspiration from this episode, could you PLEASE leave a review on your favorite podcasting platform – be it iTunes, Overcast, Spotify or wherever you get your podcast fix. This podcast is very much a labor of love – and each episode takes many many hours to put together. When you write a review, it will not only be a great deal of encouragement to us, but it will also support getting the word out about the Mobile User Acquisition Show.

Constructive criticism and suggestions for improvement are welcome, whether on podcasting platforms – or by email to shamanth at rocketshiphq.com. We read all reviews & I want to make this podcast better.

Thank you – and I look forward to seeing you with the next episode!