What's the deal with iOS 14? Patrick and Isaac discuss what led Apple to make changes, how their new policies impact Facebook, Google and the advertising industry as a whole, the shift towards a cookie-less world and what all of these changes mean for how consumers receive, experience and interact with ads.

Full Episode Transcript:

Isaac Rudansky: Hello, and welcome to the How to Hide a Dead Body podcast. I’m Isaac Rudansky, founder and CEO of AdVenture Media, one of New York's fastest-growing digital advertising agencies and I'm joined as always with the great Patrick Gilbert, AdVenture Media’s executive director and author of Join or Die: Digital Advertising in the Age of Automation, available in paperback and hardcover and digital e-book now on amazon.com and anywhere books are sold. Patrick, how are you?‍

Patrick Gilbert: I’m doing great. How are you, Isaac?‍

Isaac Rudansky: I'm doing good. It's been a busy few weeks. I know we're here to talk about a topic, which is a hot topic right now. It impacts a lot of our clients. It impacts probably most people listening to this podcast in some way or another. And that is iOS 14, privacy, Facebook's response. How that's going to impact Facebook, Facebook advertisers, Facebook publishers, or publishers that rely on Facebook, and Google's position and role in all of this. So, it's a triangle of complications of mystery, and politics. 

I think we bring an interesting angle to the conversation because many of our clients are Facebook advertisers, and we rely on Facebook's reporting. There's really not that much clear, obvious information that's been released directly from the major players, i.e., Google, Facebook, and Apple. So it's been up to people like us who are sort of in the trenches, so to speak, to try to parse this information out and to figure out what it means for our people and ourselves. 

So, why don't you kick us off here and set the stage for what we're going to be talking about?‍

Patrick Gilbert: Sure. So we almost recorded something on this topic, probably two months ago. This has been something that's been in the news, and it's been in the industry conversations for a while now. 

It's been about a year since Google originally announced that they're going to be moving to a cookie-less environment. So, this idea of a cookie-less world has sort of been a buzzword for the last year. It's been really hard for us to get our hands around everything, and only recently have we made some headway. That's why I'm finally confident to talk about it. 

We rarely talk about these types of topics. News stories related to the digital advertising space, at least as far as the podcast is concerned. Because I think a lot of these things, they don't have a lot of shelf life, and they don't really impact a lot of what we do. Yes, a new update here or a new story will impact some of the day-to-day of what we work on. 

But everything that's happening right now with iOS 14, with Facebook, with Google, with privacy, with cookies, is not just a quick news story. This is changing the entire landscape. It is something that will be around for a long time. And it's forcing us to think differently about how we're developing campaign strategy. So, therefore, it makes it much more important to discuss. 

Even from a broader perspective, the idea of data and privacy rights is all over. Last night, I was driving home from work, and I tuned into a Clubhouse room where a bunch of former athletes and people that work in sports journalism, were debating the privacy topic for athletes, and what sort of data do athletes have access to? Honestly, everyone's just asking questions that no one else has a really good answer to. 

There are so many different people that are involved in this conversation. It impacts advertisers. It impacts agencies. It impacts the athletes on the ground floor. It impacts C-level executives. It impacts society as a whole. So I think what's really important is we all have to establish more common ground. And I think that's what we're trying to do with this episode is trying to take a lot of this information and make it digestible for the average marketer to understand. 

So, we have a few recommendations as far as what you should look at after this podcast, and we'll give those recommendations toward the end. But I think what our real goal here is like, “Okay, there's a lot going on. It's a lot of technical jargon. It's a lot of broader societal debate. But what can we as AdVenture Media do, to help decode a lot of this jargon for the average person so that they can go on and contribute to this conversation in a more sophisticated way?” 

My main takeaway and I'm curious what you think, Isaac is that there's a lot of warning signs going out and people are hitting the panic button. There are certain folks that this certainly shouldn't be doing that for, that level of panic. But I would say, by and large, the majority of advertisers, particularly the ones that we work with, really don't have much to worry about. 

This whole transition will change a lot of what we do, but it's not going to make Facebook irrelevant. It's not going to diminish the value of Google as an advertising platform. We just need to understand how this is going to change and adapt. But do you have any other broader takeaway thoughts, before we dive in? ‍

Isaac Rudansky: Yeah. I want to also clarify a couple of things. You were talking to me about this a few months ago, how it's a good idea to show an audience where you're going at the beginning of a presentation, and then you get them there. So let's just try to set expectations for the audience now where we're going, because there's a lot of different places in this debate that we could go. 

I don't think we're going to get hyper-technical. You and I are not developers. But we're going to try to be as open and as practical as possible with what is actually happening, and what does it really mean? But let's just define a couple of brief things. 

You used the term ‘a cookie-less world’. Now, a lot of people who like cookies don't want to live in a cookie-less world. So, we have to describe what you mean by a cookie. Just very briefly and simply, a cookie is an alphanumeric string of numbers and letters, that is placed on either your device or in the cloud when you go to visit a website, let's say for example, or you open up in-app. 

Let's say, I was using this example the other day, I'm on a diet. It's not apparent yet, but it's early. I only just started. So I use this app called MyFitnessPal. I went and splurged for the premium. This is actually relevant. I went and bought the premium version for 20 bucks a year. So, I'm not exposed to ads. This is important because the majority of the impact that these changes are going to have are on in-app ads. I want to just define what an in-app ad means as well, while I'm defining what a cookie is. 

An in-app ad is just as it sounds. It's an ad you see when you're inside an app. Typically, you see a lot of ads in games, and you see ads in apps. But imagine you download an app. It might be the Spotify app. It might be MyFitnessPal. So, MyFitnessPal has lots of ads until I paid 20 bucks a year to remove ads. So, there's a cookie placed on my phone or in the cloud that identifies me not by my name, or my address, or any of what we call PII. Personally identifiable information. Then that alphanumeric string is shared with lots of other companies and websites and ad platforms. 

As I continue to use Internet-connected devices, and I visit other websites, and I watch certain YouTube videos, and I listen to certain Spotify playlists, my cookie sends rich information about me to all sorts of different people and companies, but for example, an ad network or a company like Facebook, or Google that will allow an advertiser to identify me as a certain type of person interested in certain types of things and certain types of products. That's really what the cookie functions to do, at a very, very simple level. 

There's been a lot of debate around cookies. Are these cookies revealing too much information about me? Are they being shared with too many people? We hear the word creepy, right? The non-technical users say creepy a lot. It's like the ads could be so targeted, that it's almost creepy. What was it, Patrick? You were showing me the other day. We were talking about going somewhere. You were going upstate New York somewhere. You don’t remember this?‍

Patrick Gilbert: We know we had a call with a client that's in Montana. And then I started getting Instagram ads that say, “Visit Montana,” and it was the weirdest thing.‍

Isaac Rudansky: Right. And by now, probably everyone listening has had their experiences, and that's related to cookies. That happening is part of the cookie technology because the advertiser who then places the “Come visit Montana” ad is paying to leverage, to access the information stored, or paying to access your cookie, basically. 

Okay, so that's a hopefully simplified understanding of what a cookie is. If you happen to ever hear the word pixel, pixels are tightly related to cookies because pixels are the piece of code that places the cookie on your browser, or on your device, or in the cloud and in large match tables in the cloud.

In-app advertising is where the biggest impact is happening here because I want to just define, or at least constrict the hysteria a little bit. These changes are happening in iOS 14 … Okay, so maybe we'll wait because we'll describe what is iOS 14 doing and come back to that. So, I wanted to just define that at a simple level, and then just one last thing. 

There are four key categories of in-app ads that customers or app users are typically exposed to. Rewarded ads. Those are these really annoying ads that you beat a level of Temple Run, and now I need to wait 60 seconds to watch a video before I could continue. Those are often also called interstitial ads. Those ads often show up in a different format where you could choose instead of having to buy more coins or buy more life or buy upgrades, you could watch video ads for 60 seconds, and you'll get those upgrades. So, those are ads. 

This is part of what I think that all these changes are going to be ultimately good for the industry. Because Google writes on their AdMob website. AdMob is the most popular product that exists, that app developers use to place ads inside their apps. They wrote innovative ad formats like rewarded and native ads help balance user experience with monetization to keep users happy and engaged. No one is happy when they have to watch 60-second ads to like get to the next level of a game, or open up my meal counter so I could track the calories that I ate. Okay. 

Then you have native ads. They exist online also. But when they're inside an app, you could be scrolling through. We could go back to my example of MyFitnessPal. MyFitnessPal has a blog feature inside the app with articles and tips for recipes and healthy living and whatever. Native ads would be sponsored placements that look like a typical post. They’re sort of disguised and camouflaged.‍

Patrick Gilbert: Basically, most Facebook ads are made to look like native ads. That's essentially the easiest way to…‍

Isaac Rudansky: Yes. Right. Exactly. It looks like it's something in your newsfeed. ‍

Patrick Gilbert: Yeah. ‍

Isaac Rudansky: Which reminds me of a really funny story. When I was sitting with my brother-in-law, we were looking at Reddit shower thoughts. Reddit shower thoughts are really funny. They're really deep. It's like a shower thought, like something really interesting, but a funny observation about the world. One of them was like when sharks smell blood in the water, it's like you walking past your neighborhood, your neighbor’s barbecue. Like, smelling a barbecue. It's like the same feeling, like shower thoughts.‍

Patrick Gilbert: Right.‍

Isaac Rudansky: Sometimes they take a second to really get. He had never seen them. So he was scrolling through and scrolling through and he's cracking up and laughing. And we're like, whatever. Then he's stuck on one for like six minutes. He's like, “I don't get it. I don't get it. Like, what is so funny?” Then he's like, “Oh my God, it's an ad.” It was an ad for a mortgage. It's a mortgage product. We were cracking up. ‍

Patrick Gilbert: That’s funny.‍

Isaac Rudansky: So that's native, right? Then you have banner ads, which everyone knows. Rectangular ads that, by the way, are very insidious, because I don't know how large the average human thumb is. My thumb cannot hit the X button on a banner ad inside an app. 

So I always cost the advertiser money. And this goes back to the part of why I think this is good for the industry in general because these are not good ad products. So I end up pressing X. Not only do I cost the advertiser money by clicking on an ad that's not relevant to me, but I also have a poor experience, because I end up then being redirected to another site and I have to come back to the app. Those are banner ads.

Then interstitial ads, where you're scrolling through images of an actor on IMDb, and every third image is an ad, right? That's what would be considered an interstitial ad. So those are just to give a visualized, practical sense of what we're talking about here, and what's going to be impacted. And those ads have become very, very targeted because Facebook collects and Google collects a lot of data. 

I really like that example from Foxwood Digital. He said it's like a dam. Apple being the manufacturer of the device that these apps are installed on, they sort of operate the floodgates. They could put a stop to or open up the flow of information that takes place from within these ads to the advertisers, to the companies like Facebook that want to use them and to monetize and to target their ad products based on the information. 

So Apple is in a very, very strong position. While Google and Android and the Google Play Store are certainly large, I don't know the exact percentage. Maybe you know Patrick, like what percentage of app users or apps installed. Apple still has the lion's share, I think as far as app activity in the world or in the United States goes for sure. Google's definitely not nothing, but this does not impact. 

So everything we're talking about here doesn't really impact all of the advertising that's taking place inside apps on Google phones, right? So, that's an interesting point. Everything we're talking about is only impacting app developers who are making money from their apps that are being downloaded on Apple's phones. Because this is an iOS 14 update, Apple is sort of taking the charge here. 

As of right now, if you're making money, and you're advertising inside apps that people are using on a Samsung Galaxy, which is an Android phone, there's no real change right now, although there has been chatter that I've seen about Google taking certain steps also, but we'll see how that plays out. Right now, we're talking about iOS 14. So that's just my … I'll kick it back to you now.‍

Patrick Gilbert: Okay. And I think one very important caveat of everything you just described, when we talk about apps, it's all apps. It's not just a Temple Run game. It's your Facebook app and your Instagram app. 

So if you're going to be serving ads inside of the Instagram app, which it's like 90-plus percent, 95% plus of Instagram engagement is done within the Instagram app, as opposed to the desktop experience. So, that's why this is funneled through there. Most people just think of Instagram as a product, and don't really think of it in terms of, “Well, this is actually a product that I access through my Apple device.” And that actually comes back to the crux of why this happened in the first place. So‍

Isaac Rudansky: Can I just give a caveat on your caveat? ‍

Patrick Gilbert: Sure. ‍

Isaac Rudansky: Because I think that’s an interesting point. Facebook's app itself, the data Facebook collects on you when you're inside Facebook, or Instagram because they own Instagram are actually not being impacted.‍

Patrick Gilbert: Right. So, we’ll get to that.‍

Isaac Rudansky: You're going to recommend that podcast, Episode 292. But that's something that I learned from that, that iOS 14 updates don't impact the first-party data Facebook collects from you as long as you're interacting within Facebook's app, ‍

Patrick Gilbert: Right. However … Sorry, we're totally going off the rails here, and our promise to the listeners here that might be kind of confused, we'll circle back and put a bow on all of this. But to that last point, Isaac, the majority of the data that Facebook collects and uses to both report on advertising campaigns, and also optimize for advertising performance, is collected outside of the Facebook ecosystem. It's collected from third-party sites, which is why all of this is so crazy, detrimental, whatever word you want to use. 

Okay. We just covered a lot of technical definitions and details. Let's try and walk through the story of what's actually happening here. ‍

Isaac Rudansky: Let's do it. ‍

Patrick Gilbert: Apple. Everyone's familiar with Apple. They're a pretty big company. Okay, so the whole thing here with Apple tracking is that … So they have the App Store, which is essentially the way that people that use Instagram on their phone are accessing the Instagram interface. When you go into an app on your Apple device and make a purchase because it's through the app store protocol, Apple is charging a 30% tax to the company that administered it, that is kind of monetizing that event. So let me give you an example. 

When you have … When you sign up for an … When you download the Netflix app, and you sign up for a Netflix account on your Apple device, Apple will forever take 30% of your subscription that goes back through your Netflix account, which is why a company like Netflix is very much incentivized to try and get you to sign up for your service on Netflix.com directly. Because the Netflix app is technically being run through Apple, so they want to put their hand in the … What's the expression that I'm totally blanking on? Hand in the bucket?‍

Isaac Rudansky: Hand in the jar?‍

Patrick Gilbert: Hand in the jar. That's it. Hand in the jar. Apple's putting their hands in the jar to collect a 30% tax when you sign up through the app. But if you go to Netflix directly and sign up for your subscription, they will not take a fee on that. 

So, over time, 30% is a lot of money. If you remember from this past summer, there was a big news story about Epic which is the company that controls the game Fortnite. There was a big clash where Fortnite, or Epic, was trying to take Apple to the Supreme Court and all these other things, because 30% is just egregious, so on and so forth. 

So this has been a big topic in just the digital commerce space for a very long time. And a lot of companies, Facebook, in particular, have been trying to figure out ways to get back some of this revenue that they've just been giving back to Apple all of these years. Okay, so that's kind of like the very beginning of all of this. 

Now, the second part of it now is when we talk about Facebook directly. They've been drifting out of their lane, in terms of pushing the boundaries of this rule. So, Apple says any monetization through your app needs to … We'll take a 30% tax on it. Also, no companies are allowed to have their own app store within their app. That is, accessed on an iOS device. Meaning there cannot be an app store within Facebook that looks like the Apple App Store. Apple will kick Facebook off their platform if that happens because that's basically them trying to do the same thing. They're trying to monetize this whole thing. 

So, Facebook has drifted out of their lane by essentially creating their own sort of App Store experience and not directly monetizing through the iOS app. They've been pushing the boundaries of this over the years. They have their own Facebook shops. They have all these different game interfaces. That is basically an app store. But they're pushing the boundaries to what that really means, and it's basically been pissing off Apple. So, Facebook has been able to generate more revenue from their users, and they've squeezed Apple out of the picture. So obviously, this pisses off Apple. 

At the same time, the Apple App Store over the last 10 years, I didn't even realize this until I started researching this topic, the Apple App Store is virtually worthless at this point. If you remember 15 years ago, or maybe that's too long of a stretch of time, but 10 years ago, when I first got the first iPhone that I had, the Apple App Store was everything. The featured list on the App Store, they were kingmakers. It was everything to be featured on the App Store. Now, I can't tell you the last time I went to the App Store. If the App Store itself on Apple is nothing but just an annoying step that you need to take when you go to download an app that you probably heard about from a Facebook ad. 

So it's like, oh, generally you learn about new apps because you saw an ad for them on Google or Facebook. You click through that. It brings you to the Apple App Store, and it's like, “Oh, this is so annoying. I have to download it,” and now you're able to access the app. So, the Apple App Store itself has essentially become worthless, which diminishes the power that Apple has, which further pisses them off. With me so far, Isaac?‍

Isaac Rudansky: Yes.‍

Patrick Gilbert: Excellent. So in June of last year, June of 2020, Apple announced the plans for the iOS 14 update, which will include new restrictions in terms of data privacy. So now, everything that I just spoke about in terms of the Apple versus Facebook feud, and why Facebook's drifting out of their lane, that's not written anywhere. Apple didn't say that they're doing all this, because of that. That's just speculation, and most people that have been writing about this topic have sort of agreed that that's what's led to this. 

These ads started months ago, where they started talking about how they care about your privacy. Apple figured out that it would be very advantageous for them to take a position that they care about your privacy more than any of the other big tech companies. And they can blame Facebook for basically everything that's happened in society, and they can blame Amazon for other things, and they can blame Google for other things. But they, Apple, care about your data, which is partially true. 

If you remember, a few years back there was this big debate whether or not Apple should let the FBI decode terrorist phones. So they have sort of cared about this, but never to the extent that they've gone about this over the last six months. They found that to be advantageous from a PR perspective, from a marketing perspective, to say, “We’re the privacy people.” So, that's what they’re going off of. 

iOS 14 has a new set of rules and regulations as far as what data can be tracked, including forced use of what's called the IDFA, which is an identifier for advertisers. The IDFA is a prompt that users will see that will say, “Facebook.com wants to be able to track your data. Do you grant them permission, yes or no?” That prompt is essentially the IDFA, allowing users to choose whether or not they can be identified and tracked by the app that they're downloading. 

This has been around for a long time, but it's never really been forced upon everybody. Now, as part of this new wave of what Apple is enforcing, they are making that mandatory. So in June, when this came out, Facebook was like, “Okay, not a big deal. I'll just not show the prompt, like whatever. It's not that big of a deal.” 

Apple eventually came out with what's called the ATT, app tracking transparency. I keep forgetting what it actually stands for. App tracking transparency, it's basically their stance saying that anybody that's going to track any data on an iOS app, needs to show the IDFA prompt and give users the option to not have their personally identifiable information tracked. ‍

Isaac Rudansky: Even more, it's by default now, the app developer has to not track any information. And the IDFA prompt is giving the users to opt in to tracking. That's when Facebook got so up in arms, because why would you … If I'm already tracked into tracking, maybe I'll just leave it or X out of the prompt to be like, “Okay.” But now that I have to click a button to allow tracking, that's when Facebook realized that most people probably won't allow tracking because they don't really understand the value of the tracking. It just sounds creepy to the average person. Tracking doesn't sound good to anybody. No one wants to be tracked. Yeah, go ahead.‍

Patrick Gilbert: I forget what this terminology is, offhand. But there actually is separate terminology that Apple uses when they prompt their own users to opt in to tracking. I forget what it is, and offhand, and I'll look it up and I'll put it on Twitter or something.‍

Isaac Rudansky: What do you mean ”their own users”? For Apple-owned apps?‍

Patrick Gilbert: Apple-owned apps. So when you go to download Instagram now, or when you go to refresh Instagram, you're going to get a prompt that says, “Do you give Instagram permission to track behavioral data?” or whatever it is. It makes you say no. The terminology that Apple uses to show for their own apps is much lighter, per se. That, by itself, is just completely … It's infuriating to me.

Separately, I think that it came out the other day, there's an antitrust lawsuit in France they're bringing against Apple because they're basically doing this and they're exploiting their own customers for all this stuff. 

But Eric Seufert, who runs a site called Mobile Dev Memo has been sitting in the middle of this entire privacy debate and has probably been the most influential and outspoken educational person that's really helped. His content has helped me learn about this. He had a really good point where he was like, why can't Apple's terminology say something like, “Do you allow Instagram to track your data? Data tracking allows ads, or allows brands to keep their content free”? Why can't you just explain? Why? Do we like advertising? Do we hate advertising as much as we like … Or do we hate advertising more than we want to pay for things? Generally not, because that's why the advertising industry exists, to begin with because we're not-‍

Isaac Rudansky: This goes back to your whole point in the beginning, which is why Eric Seufert’s point is, to me, fine, but not really relevant. Because Apple is looking to find a competitive advantage to what their business is. And fair enough, right? They're positioning themselves as a hardware company that has really good privacy controls. When the FBI wants to crack the password of a terrorist’s phone, we won't do it, because there's nothing more important to us than privacy. And that's their marketing thing. That's their shtick. 

Look, they're one of the world's largest companies with smart people. So when they get up in court, they're going to say, “Look, we don't have to use the same type of language in our apps, because our apps, like we collect … Inside one of our apps, when you're using Keynote on your phone, you're already inside our privacy vault. We're not doing much with your tracking, other than crash reports.” I'm just giving examples. ‍

Patrick Gilbert: Yeah.‍

Isaac Rudansky: But a third-party app, we have to be much, much more careful for the safety of our users. It's like I own a house. Imagine that's the device. Now, I have a nanny who has been part of my life since I was four years old. My family trusts her and she's awesome. Okay, so we give her the alarm code. Now we have a new babysitter. She looks the same. She acts the same. She has the same title, but she doesn't get the alarm code. It's like, “Well, why? They're the same.” And Apple will say like, “No, they're not the same. One of them is part of our family. The other one is not.” It's like Apple's going to make that claim. 

To Eric Seufert’s point, which is why can't you just encourage people that tracking is anonymous? Educate them. Because Apple doesn't want to educate you about tracking. Apple wants the competitive advantage of more people relying and needing Apple's ecosystem and Apple's devices. And Facebook's a real threat, Google's a real threat to Apple in other ways that maybe we understand some of it or not all of it. These are complicated businesses that overlap in complicated ways. 

So yeah, Apple is not going to be encouraging you to opt in to help Facebook's advertising machine, do any better.‍

Patrick Gilbert: You're right. And it's infuriating to me, because honestly, people really take advertising for granted. This is totally me on my soapbox, and I'm very biased on this, but…‍

Isaac Rudansky: It's your podcast, man. This is your soapbox. ‍

Patrick Gilbert: We take for granted that things are available to be free, because advertising exists. And advertising can exist if you're willing to give up the fact that you advanced to level two in Temple Run. Your credit card information is not being siphoned into Temple Run and into Facebook, whatever. Okay. I'll close the soapbox out, at least for now. I'm sure we'll circle back to it. 

So as far as what this means for Facebook. So now, really, everyone that has an app is being impacted by this. But Facebook is an app that collects a lot of data, and across multiple, multiple apps, not just theirs, and uses that data to help create a monetization engine in advertising. 

Going back to a previous point, and this is an important piece. The majority of the data that Facebook uses, is not collected inside Instagram and is not collected inside Facebook. It's collected from third-party websites. It's generated from walmart.com, and it's generated from the random Shopify site and your WordPress account, and that person's Magento site and every other website on the Internet. Because what happens is when people opt into Facebook advertising, and there, they put the Facebook pixel on their website, they have to agree to terms and conditions. And as part of those terms and conditions, it allows Facebook to grab your data. 

Okay, so now say you're a Shopify store. You're a small business and you're running a T-shirt company, and you say, “I want to run Facebook ads and I want to track conversions and create remarketing pools. Retargeting pools, rather, from my Facebook things, so I need to put the pixel on my site.” Okay, great. You're telling Facebook now that you're giving Facebook access to your data, so they can use that to better inform their own optimization algorithms and reporting and all that good stuff. 

The same happens with Google Analytics. Every single website on the internet that has Google Analytics on the website, including AdVenture Media's website, is helping give data back to Google that isn't just for reporting. Google doesn't let people get Google Analytics tags for free because they want to do good for the world. They let you create a Google Analytics account for free because they want to be able to use the data that you're getting on your site. 

Sure, Google Analytics is a very useful tool. People use it to make incredibly important strategic decisions. But it's a mutual … Instead of paying Google for access to Google Analytics, you're basically paying them in terms of your users’ data. That is a very overlooked piece of this whole puzzle. Because as these things get complicated and people opt out of apps, you have to realize that Facebook in particular is extremely reliant on the data that they're acquiring from everywhere else on the web, not just Instagram and Facebook.‍

Isaac Rudansky: Should we talk about the types of clients we work with, like direct response e-commerce businesses, service-based businesses, and talk about the impact to how Facebook reporting might change and what that really means?‍

Patrick Gilbert: I would say let's get back to that toward the end. ‍

Isaac Rudansky: Okay.‍

Patrick Gilbert: So the next thing here, we covered where this data is being taken away. So to one of your earlier points, Isaac, any information within Facebook that you can glean from Facebook is extremely important, as well as your first-party data. So if you have a customer list of your customers, and you can leverage that data in Facebook, that's going to be extremely valuable. These are some things that are going to change. 

But as far as how all of this works, and what's going to happen when these cookies are deprecated, I totally didn't really understand any of this until Google came out with their announcement last week about their privacy sandbox and their FLoCs and all this other stuff that's taking place. Isaac, do you want to explain that?‍

Isaac Rudansky: Okay, so you walk us through that for a minute. Walk us through Google, like, how they make money through advertising. What is a FLoC, and what does it mean for them to not have cookies? Go ahead.‍

Patrick Gilbert: Cool. So Google has their own advertising platform, as probably anyone listening to this knows. It's Google search, but it's also they have their own Display Network advertising. So, ads that can be served on the New York Times website, and within apps. The Temple Run app, and all these other things. ‍

Isaac Rudansky: What I mentioned earlier in the episode, AdMob, which is the most popular exchange for in-app advertising is owned by Google. It's a Google product.‍

Patrick Gilbert: Right. So, they certainly monetize as well. What makes Google unique, compared to Facebook or any other platform that's being impacted by this, is that Google has much more first-party data than anyone else. So when we said earlier, anything within the Facebook app is technically Facebook's first-party data. So if you're interacting with Instagram … By the way, we use Facebook and Instagram interchangeably, because they're essentially the same thing, just a different interface. 

When you follow certain accounts on Instagram, Facebook is using that data to give a profile of who you are, because that's their data. That's first-party data. As an Instagram user, you're agreeing to that. What they can't do is pull from a third party. Now, that's limited, though. There's only so many things that you can glean from inside Facebook and inside Instagram. Google, though, has access to a lot more data that's their own. 

So if you use the Chrome browser, which I think last I checked, is the most popular browser. Maybe I'm wrong, double check me on that. Chrome is an extremely popular browser, whatever. It's one of the top three. A lot of people in the world use Google Chrome as their browser of choice. Anything in there can be tracked by Google. 

Anything inside of an Android operating system, which is the number two market share, is being tracked by Google. Anything within a Google-owned app, like Google Maps, or Gmail, to an extent. Google doesn't read your emails. They track certain behavioral patterns. Anything in YouTube that you do, it's owned by Google. So, they get to use that to their advantage, whereas Facebook doesn't have this huge amount of data sets to work with. So they're at a significant advantage. 

Google came out last week, in early, early March, depending on when you're listening to this. It was early March, saying, “Here's our plans for the cookieless world. We're actually going to get rid of cookies altogether. And we're going to specifically rely on server-side applications that are not going to upload any data to the cloud.” 

In Isaac's previous example, where he had that cookie assigned to him, that basically was taking information and feeding information back as he navigated to different websites, Google is now saying, “We are only going to look at data, when you're in the moment interacting with a given website. And when you leave that website, when you leave Google Chrome, we're not saving any of your data. And we have our own systems to make sure that we're doing the right thing.” So that's basically them saying, “We're not going to track you. We're not going to follow you around without your consent. We're going to play nice, and we're going to meet all of the outcries of all the people for privacy.” 

Now, this sounds great in theory. But what they're also doing here is the way that they're still going to make advertising-informed decisions is by assigning individuals to what's called a FLoC. And it stands for a Federated Learning of Cohorts, which is honestly... just think of a FloC as a buyer persona. It's probably the easiest way to look at it.‍

Isaac Rudansky: I'm thinking of geese.‍

Patrick Gilbert: It’s sort of like a FLoC of geese. So you gave your example earlier that you downloaded a fitness app or a dieting app, whatever. So that's like a data point that Google can know about you to better inform where you should fit in the grand scheme of things. But let me give a different example here. 

The funny thing about FLoCs is it's very hard to give an example that doesn't make you seem like you're judging people. But let's just assume for a second that there's a woman named Mary that buys a lot of organic food. She has young kids at home and she is into clean, non-pesticide cleaning products. There’s all these different attributes that make Mary unique, very similar to the buyer persona thing. 

Now, Google can track Mary's behavior, as she's interacting with Google's apps. And Google will slap a label on her and say, “Okay, she's Organic Mary.” And instead of having a name, it's like a long string of numbers. That's the FLoC. Great. Now, the next time Mary goes to a different website, Google can know that Mary is Organic Mary. She's in that FLoC. But they won't know how she got added into that FLoC. They won't know that she's added into that FLoC because she was recently buying organic products for her kids‍

Isaac Rudansky: That’s what I don't get. Can I ask a clarification question? ‍

Patrick Gilbert: Sure.‍

Isaac Rudansky: There's Mary. Let's make the labels a little simpler because we don't want to … Mary is the person, and let's just say the FLoC name is Organic Charlotte. ‍

Patrick Gilbert: Okay.‍

Isaac Rudansky: Okay? Because you're going to have a lot of people that … That's the idea here, right? We're moving away from an individual into …‍

Patrick Gilbert: Right. Groups.‍

Isaac Rudansky: A FLoC of a group of people. So now Mary, who's the actual person, has to have taken X number … She has to have done X number of things, before Google is comfortable or confident enough in statistical terms, to apply her to the Organic Charlotte FLoC. With me so far?‍

Patrick Gilbert: Yes. And the Charlotte thing is probably confusing. Let's just say Organic Mothers, as far as what this FLoC is. A group of people. Google says it's going to be at least more than 1,000 people. So, Organic Mothers is what the FLoC is called. ‍

Isaac Rudansky: Okay. Great. Sounds like a great name for milk. Patrick, if we ever sell AdVenture Media and start a milk--‍

Patrick Gilbert: Organic Mothers?‍

Isaac Rudansky: Yeah, we'll start a milk company. We're going to call it Organic Mothers.‍

Patrick Gilbert: Perfect. Yes. So you're basically saying like, where does that first FLoC come from? I don't really know about how the [inaudible 00:42:10] works.‍

Isaac Rudansky: Right. No, no, no. So, I don't need to know how … Mary has to do enough things, but Google's tracking those things that Mary's doing. Are they no longer tracking Mary as she goes through websites, and with a cookie? Is it because Google just has enough connected devices, that there's enough data points that Google already owns, that they no longer need to use an actual cookie on your hard drive that doesn't know it from their own first-party data? Is that it?‍

Patrick Gilbert: Yes. The difference is when you have a cookie, that information is being uploaded to the cloud. A cookie is basically a digital ID or a name tag. Every time Mary shows up to a new website, the website goes, “Oh, it's Mary. Okay, we know we can …” Okay, let me be clear here. It's not the website doing that. Every time Mary shows up to a website, Google can use their Google Analytics tag to know, “Oh, this is Mary. We've seen her before. We know that she was buying organic cookies last week,” which is very different. So, that's the individualized tracking that we currently have with cookies. 

But in the FLoC system, Mary will show up to your website, and Google will say, “Oh, this is an Organic Mother,” and make a bunch of assumptions about who this is, who Mary is, based on that. Now, if she buys something, if she whatever, that information in that session is not going to be uploaded to the cloud and stored forever, as part of her digital name tag.‍

Isaac Rudansky: Alright. So it's still a little confusing, but it sounds like to simplify it, Google has gotten big enough or pervasive enough that they have enough touch points where they no longer need to use this cookie technology. Because also the cookie was something which was shared amongst lots of different companies. It was bought. It was traded like a stock, in a way.‍

Patrick Gilbert: Well, hang on. Cookie information is not being sold. Personal data has never been sold by Google or by Facebook or by anybody. That's a big misconception with all this data stuff. None of these companies are selling data. They sell ad inventory that uses the data to place the ads.‍

Isaac Rudansky: Right. So what I mean is, let's say Mary, she bought cookies last week, and that information stored within her cookie. Then she comes to a website. Let's say she's using Safari. Maybe it's AdRoll or Criteo, or there's some other tracking tag that could read the cookie. But Google, which owns DoubleClick, which is another huge ad exchange, is licensing the right to read that cookie to place a bid in the auction, right? 

They're not reading Mary’s personal information. They're paying, I guess you can call licensing fees to all these other partner technologies to read this cookie about Mary and decide how much to bid and what ads to bid, and so on and so forth. It's a whole complicated group of technologies that are involved in that process. 

So now Google's saying, “Look, we don't really need that cookie anymore, because we'll know enough about Mary to put her into a specific enough group of people. And we could just do away with cookies and our advertising will be just as effective.”‍

Patrick Gilbert: It’s group-based targeting, not individual. ‍

Isaac Rudansky: Relevant, yeah. But I guess my point of confusion is still like, there… Well, I have two points-‍

Patrick Gilbert: I’ll give you a different example. Maybe this will clear up some of the confusion, right? So let's say Mary, she's in this cohort. She's in this FLoC that's Organic Mothers. Google doesn't know how she got there in real time, right? Google doesn't know how you got onto a FLoC, because it's not tracking the history. It's just tracking real-time information. So now you get a label slapped on your head. So now when you show up to this website, it's like, “Oh, cool. Organic Mother. Here's how we value organic mothers.” 

But now let's say you buy something on that website. Google's now learning. It's observing in real time, what's called a privacy sandbox. That's the terminology. They're using the server-side application to look at what's happening in real time, and say, “Oh, she bought something. She's not just an organic mother. She's an organic young mother, and that’s a different FLoC.” So now Mary's cohort FLoC has changed. So now she gets a new label slapped on her head, that now the next time she sells through different websites, she's no longer just Organic Mother. She's in this different FLoC altogether. 

So it's using the machine learning capabilities in the data that it's ingesting. It's still tracking information. Google is still tracking information about what you're doing and who you are. it's just trying to refine that over time to get better information about who you are. Right.‍

Isaac Rudansky: Right.‍

Patrick Gilbert: Or what group best fits your buying habits. 

Isaac Rudansky: So I think the way to describe it is Google still knows as much about you, as they always knew, because they have to. If they know enough about you to put you in FLoCs and to change you from FLoCs to FLoCs and to put you on multiple FLoCs, and so on and so forth, they have to know all these individual things about you. But their claim is that we're no longer using an individual profile, when we decide to show you what ad. We're going to use your group identity. ‍

Patrick Gilbert: Exactly. ‍

Isaac Rudansky: So remember, I posted this to our company Slack the other day. And I'll just reiterate this point here because I don't know if it made any sense, and I still don't know if it makes any sense. But what I think my take on this is that it's really no fundamental change to what Google has always been doing.‍

Patrick Gilbert: Nope, and that's why it's brilliant.‍

Isaac Rudansky: Right. Again, look, this is not … I don't find it evil or wrong, that companies are looking for angles with which to market themselves. Great. Sometimes you have to please the public. You're a large company, and you want to make your customers happy, and you want to play nice. It's a big balancing act, and they have smart people figuring all this stuff out. 

However, here's why I don't think it's any different. Because it's like, “Oh, wow, you were able to use individual data points from a person's entire Internet history,” or however long the cookies last. I'm not a cookie expert. I told you, I'm on a diet.‍

Patrick Gilbert: This is the second time you've made a cookie joke.‍

Isaac Rudansky: It’s still not funny?‍

Patrick Gilbert: It’s thrown me off entirely. It's still not funny. Maybe the third time. There's see, ‍

Isaac Rudansky: We'll see. We’ll see if we can have a vote. ‍

Patrick Gilbert: Okay. ‍

Isaac Rudansky: Where was I? Okay. So Google's like, “Okay, we're not going to use…” So people are like, “Well, I was able to … The whole idea of Internet, of online targeting was individualized targeting. Individual, individual, individual.” And now Google's going to groups it’s just like group identities are not going to be as relevant and as targeted as intent-based. It’s like, no, Google has always been doing group targeting. 

Let's say I want to figure out, because what is the really interesting value of internet targeting? Predict the conversion rate. That's the genius behind all of these technologies. That's really the ShangriLa that these companies are trying to get to, is how accurately can I predict your likelihood to convert after you click an ad? Because that will get you to the exact economically true bid on behalf of your advertiser.

Now, predicted conversion rate is not a perfect science because you have to predict human behavior. So it's all about predicting human behavior, based on what you know about that person. But it's also based on what you know about people. So if you're hungry, Patrick, right? Let's say you're hungry. ‍

Patrick Gilbert: I'm always hungry. ‍

Isaac Rudansky: I will … This is not a cookie joke. I'm being fully serious here. 

Let's say you're hungry, and I was asked by my psych professor, like, “Can you predict how Patrick will respond to his hunger?” I will have a much better time … I will more accurately predict how you respond to your hunger by looking at how other people, not you, how other people have responded to their similar situations of hunger, more accurately than I'll look at all of your past behaviors that you manifest in other situations, like when you were not hungry. 

I'll learn more about your personal likelihood to react to being hungry, by looking at other similar people's reactions to being hungry, than I will about learning about your previous behaviors, like your movie preferences and tastes and all of your other previous behaviors. Which is interesting, because I can learn more about you by looking at other people than I can by looking at you. 

So Google has always, and that's just fundamentally true. Similar people will behave in similar situations similarly, whereas an individual will manifest a huge range of behaviors based on circumstance. Meaning we're more alike than we are different than each other, especially when you have a FLoC of an age group, and an income bracket and a geographical location and a time of day, so on and so forth, which is what Google is probably using to identify groups. 

So all I'm saying is that when Google looks at predicted conversion rate, how will you react to this ad at this time, they're looking at how other people like you have reacted to similar circumstances. And they always have been, which is why I think it's probably very little or zero fundamental change to how Google has been trying to figure out predicted conversion rate all along.‍

‍Patrick Gilbert: I agree. 

I think from a machine learning perspective, it actually simplifies the entire equation. Because instead of looking at the largest data set that is totally incomprehensible, and trying to figure out which are the signals that matter most, if you're trying to look at group behavior, it's easier to look at a trend of all these different signals together in mass, and then make a prediction based on what the entire group is. It's not necessarily better or worse, it's just different and likely simpler, and can do it at scale. 

People really underestimate the power of data and machine learning, and that's really what it is. So the fact that Google came out over this past week, or whatever, and said, “Hey, listen. We're just going to do this entirely. Look at us,” they wouldn't have done so if they weren't so confident and excited in the solution that they have here. That's really good for them, and it's good for a lot of our clients that things weren't worse for them. 

So, let's bring this back to Facebook. ‍

Isaac Rudansky: Okay.‍

Patrick Gilbert: Because now, what's interesting to me is that I've understood this entire thing a lot better once I can compare the differences between what's happening with Google and what's happening in Facebook. 

So everything that we just discussed regarding Google, essentially gets to the point of like, Google is going to be fine. They have these FLoCs. They have this different privacy sandbox thing they're going to get. They're not going to be deprecated by privacy legislation, this, that or whatever, even though they probably should be. Whatever, that's a separate topic. But they're probably going to be fine, and their advertising is going to continue to be powerful. 

Why is it so different for Facebook? It comes down to what I said earlier. It's the lack of first-party data. It's the lack of being able to track data across those third-party sites and their own limited amount of first party data that they have access to. So, does this mean is this a nail in the coffin for Facebook as an advertising platform? I would say no. I'm still extremely … Maybe bullish isn't the word, but we're certainly not going to start pulling our clients’ ad spend off of Facebook. 

The difference here is that most businesses have not been utilizing all the different tools inside Facebook that are really going to be taking away. Most businesses are using pretty simple interest targeting and email lists and a 1% lookalike audience and all these different things. They're not leveraging 300 conversion actions, that value that have different incremental values at every stage, depending on how far you're advancing into a game. That's very sophisticated stuff that is really going to get complicated in the future. 

But if you're just a standard, if you're an e-commerce business, if you're a SaaS company, if you're a law firm, if you are an educational institution, you're going to be able to be fine. It's going to be challenging, but you can be fine. Here's basically the technical aspects of what we mean by this and what needs to be done in the future. 

Facebook, as they move away from the pixel and the cookies and everything that they're doing there, we need to now lean on what's called the conversions API, which is abbreviated to CAPI. So whenever you hear people talking about CAPI, they're referring to this thing called the conversions API, which is essentially their own version of their own browser within a browser. 

So when you go to Google Chrome, the traditional sense is like, okay, you go to a Shopify website, and they have the Facebook pixel on there. But the information is only being fed into the pixel through a cookie. Then once it gets into the pixel, it goes into the cloud, and then Facebook can then pull it down. 

When you're using the conversions API, it's what's called the server-side application, that it's pulling that information directly from that browser, and directly depositing it into Facebook in real time. So this is similar to what we were talking about before where Google's not storing this information over time through the cloud. It's directly going back to Facebook. So, that's good. That's a good sign. So you have to install the conversions API to really track any data moving forward. That's number one. 

The second thing that they're implementing is that you need to verify your domain within your Facebook Business Manager. If you've ever worked at the Google Merchant Center, it's a similar type of idea. Facebook's trying to cut down on the amount of nefarious websites that are advertising with different domains and all this other stuff. So that's step two. 

Basically, you can only have one ad account per website domain, which is going to be complicated for advertisers that use landing pages, and lead pages and all these different things. I don't really know what the next thing with that is. But again, not a lot of information out on this yet. But for the standard business owner, you need to verify your domain in the Facebook Business Manager. 

The third implication is that advertisers will now be limited in Facebook, to eight conversion events. Historically, you are allowed to set up as many conversion events and track as much individual event sequences as you would like. Now, Facebook is limiting, is capping that number at eight. I have no idea why it's eight, but it's eight. 

Most businesses that we've worked with have never needed more than three. Your standard e-commerce company is tracking purchases, maybe add to cart, maybe something else. Your standard lead gen business, if it's a software-as-a-service company, you're tracking lead, then marketing qualified lead, sales qualified lead, close deal, something like that. So you're tracking a few different events down the sales pipeline. You have up to eight now to be able to track. 

Now, with that, advertisers will have to choose the order of importance of how they value these different events. Specifically, you have to choose your most important event, and I'll explain why this is relevant in a second. For an e-commerce store, you will choose purchase because that's probably the most important thing to you. You probably value your purchase events more than your Add To Cart events. For the SaaS company that has a multi-step conversion flow, that's a little bit more of a complicated situation. Depending on your conversion volume, you might end up only being able to track leads, unfortunately. 

So, you have to do that. You have to set the order events. Now, what this means. When people, users, opt out of tracking, it impacts the amount of data that Facebook can track on these people. Data is used in two ways: It is used to report back to the advertiser on performance. It is also used to fuel optimization via machine learning. 

So when people opt out of tracking, they're basically opting out and saying like, “I don't want to be tracked anymore.” Okay, fine. However, with conversion API, with your verify domain, and when you set up your one of eight conversion actions and choose which one you convert on, which one you care about most, you will still be able to, at the very least, report on the conversion actions that make that make the most impact to your business, that mean the most to you. 

So you will still be able to see for your e-commerce store, how many Facebook ads led to a purchase. You won't get that information in real time, and you won't be able to track it down to the ad set level, and you won't be able to do the 28-day view anymore, or 28-day click. It'll have to be within a seven-day time horizon. But for the most part, you will be able to track all of your customer purchases that came back, whether or not somebody opted into or out of tracking. And that's good news for advertisers. That’s good news for brands. You will be able to at least report on the amount of revenue that your Facebook ads are generating. 

The second consideration is that it impacts optimization, because really, Facebook can't use that data to fuel and improve your optimization over time. So, that's going to get more complicated. That just makes your job as an advertiser harder. Remarketing pools are going to be smaller because people are opting out of tracking. Interest targeting is going to be less accurate. The predicted conversion rate algorithms are going to be a little bit hungrier, I guess, so to speak. They're not going to have as much data to make accurate conversions. It's going to make your job harder, but it doesn't necessarily mean it's impossible. 

So, there are two different dynamics here. You have reporting, which can be solved regardless of whether or not people opt in or out. And then you have optimization, which will be impacted, but not necessarily thrown away altogether. Isaac, do you have anything to add on either of those points?‍

Isaac Rudansky: No. I think that there's still a lot of opacity around how this is different between iOS apps and, let's say, like using Chrome or a web browser on an iOS device. Some of these measures that Facebook has taken, how exactly do they relate to this iOS 14 update, specifically? We just don't know the inner workings of it. 

For something like conversion windows, for example, where Facebook's going to be reporting conversions on a seven-day click window and a one-day view window, and you used to be able to do it on a 28-day window, why is that change happening? My best guess is because Facebook needs to have a reporting system that just works for regular people. While the conversions being reported are not totally impact, they're only partially impact. Meaning the ones that would be tracked through iOS devices as opposed to, let's say, desktop conversions would be eligible still, for accurate 28-day window converting. 

Facebook's making decisions that are just more global, like, “Look, we're not going to have separate desktop, not iOS 14, in this.” Facebook's making a lot of decisions that take this … Meaning, iOS 14 is impactful enough that Facebook's making some global decisions, to simplify and make reporting still make sense for the regular person without too many complications or fracturing the reporting too much into 20 different buckets. ‍

Patrick Gilbert: And it's possible that these are conversations that took place behind closed doors between Facebook and Apple. That may have been a concession, right? I was shocked when I found out that you can still report on conversions for people that opt out, given the ranking. I was surprised on that. Maybe that was a result of a concession that Facebook made by saying, “Look, for anybody, we're not tracking 28 days.” ‍

Isaac Rudansky: By Apple into Facebook. Right.‍

Patrick Gilbert: Yeah, yeah. Like, “We're not going to do this. Can you give us this data? Can you allow us to tap into this aspect of the IDFA?” So, that's probably what happened. ‍

Isaac Rudansky: Yeah.‍

Patrick Gilbert: So, the other side of this is, the other part of the concession is that this data is going to be delayed. So when people do opt out, you will only get data, conversion data, three days after the purchase takes place. 

You made the analogy earlier, Isaac, of opening the dam and letting the data flow back through. Apple will only open the dam three days after an event takes place. It makes it less individualized. It makes it more like in a group. They're going to basically batch a group of conversion data back to your Facebook system to obfuscate it a little bit, as much as they possibly can. 

So if a conversion takes place on Monday, and a user buys a product on your site, even though that user has opted out of Facebook IDFA tracking, on Thursday you will be able to see that conversion action reported on your Facebook platform. Which is different, though, if a user converted on Monday on an Android device or on a web app. That will come back immediately. 

So, this is going to really confuse a lot of advertisers and their clients, when they're looking at, “Hey, look, we spent $10,000 yesterday. What's going on?” There's a huge variance in terms of what's being reported in the platform. Again, because we're not getting a lot of our data until three days after. Which I'm personally excited about because after this learning curve takes place, it's going to force clients to not just be so crazy, as far as individual one-day performance. It's going to force people to look at a longer time horizon, which is what we've learned all along. 

So, I think there's not a lot. We still don't know a lot of information. We still don't know a lot of what's taking place here. But what I can say is that, for the most part, most advertisers will be okay. Facebook and Google are still extremely valuable platforms. Google, obviously, probably more than Facebook in the future. But Facebook, they still command a lot of attention. They get a lot of eyeballs, and they still have a very powerful advertising platform. This is going to force them to have to get better. So I think that there's good things that come out of this. I'm also excited because I think this is good news for agencies. This whole thing is becoming very challenging, which makes the value of an agency tenfold. 

Then just the whole reason that this is a little bit different is because this is really impacting the mobile gaming industry, which is like a whole separate conversation. Those people that were able to track your advancing to level two in Temple Run, you bought for 40 cents a new skin in Fortnite, or however much. All those different incremental things that you were able to track on a 28-day horizon, that now goes away. That whole industry is being completely turned upside down. 

But if you don't run a mobile gaming site, and if you're just the standard advertiser that runs ads on Facebook, you will be able to figure this out, or your agency will be able to figure this out. It'll just take some time and a little bit of new habits and some weird reporting modules and correlation studies to be able to prove the efficacy of your campaigns.‍

Isaac Rudansky: Love it. I guess either I'll close this out, or you'll get the final words, but I'll go on my soapbox for a minute. I think I feel similar to you. Overall, I'm happy about these changes. I think it's a great thing for the industry. And here's why. 

There's a balance between … Let me start with this. Good advertising used to be a result of solving complex problems, figuring out business problems, and figuring out how to economically communicate the value of your products and services to customers. It was about value. What mobile advertising has become, is this really out-of-control proliferation of arbitrage where 95% of global app revenue is from the mobile gaming industry, and that's a huge number. 95%. 

If you think about what's happening, there's a whole conversation that could be had about children who are, by the way, not typically the customers that you want to target anyway, because they're not buying anything. They certainly are not interested in being exposed to a product, but that's not even what's happening in this world. In this world of arbitrage, you're playing a game, and you're being advertised another game. And when you get that game, you're being advertised another game. 

So now, a whole army of developers worldwide are trying to find the next permutation of something mindless. They're spending a ton of money on user acquisition, but just enough that the amount of ad revenue that can get by selling you the next more mindless game is like a one penny margin. And that's now a $60 billion industry. That's not advertising, right? That's not advertising. That's, to me, exploiting the allure of antisocial behavior in children and teenagers and adults actually, if you look at probably the Candy Crush Saga user pool, and exploiting it for profit, with nothing more than just another flashy appeal, and another app download one click away. 

I'm not saying these are bad people. I'm not. I'm not saying I wish or root for the demise of people making an income by creating mobile games. I have nothing wrong with it. But as an advertiser and in the advertising industry, I think it went so far in the other direction, that it no longer even is advertising. Like I said, it became a whole scheme of low-cost arbitrage. 

If you're a product or your service, or you're a b2b, or you're selling goods, and you've been using Facebook ads, you're still going to be able to do that. You're going to be forced to not use all of these technologies as a crutch as much as you've used them in the past. We've seen that where brands get really … They get excited about these individualized targeting capabilities that these ad networks tout, and it blinds them from fundamental flaws in their business and their product. It diverts attention and energy away from improving value, and figuring out how to communicate that value economically. That's what advertising is about. 

I'm all for leveraging FLoCs, and personas, and interest groups. None of that's going away. None of that's going away at all. If you're somebody who's willing to think critically about a business and figure out where this market is, what do they want, how you're going to provide them value, you're going to do very, very well, like you said. I think it's a very good thing for agencies, but also for advertising. 

Listen, I think it's great. There's so many apps. There's too many apps, because they're all free, and they're all being monetized by this arbitrage model. So, more apps should be thinking, and I'm not saying that … Look, broadcast TV with really good ads. To me, the Super Bowl is a great model, because people really think for a long time about how to make these ads enjoyable and useful and interesting. 

Back in the day where print advertising reigned supreme, I had a magazine, and I was able to read interesting ads about new products that I didn't know existed. And they were informative. We actually have one from Leo Burnett, with a beetle hanging up framed in our wall. That's advertising. 

There's a place for free supported by advertising, like public utilities and things that are really useful, like broadcast television. Facebook, for example, by the way, it's free. So that goes into a whole other example of everyone complaining about targeting. But there's also a place that more app developers should think about creating better apps that people are actually willing to pay a few dollars a month for. That would be nice. I'm happy to pay a couple dollars a month for a really good app. I don't need 1,000 mindless apps on my phone that are monetized by advertising. 

Then I'll end with a story that I thought of, as I was thinking about this over the last couple of weeks. I remember reading this story in Claude Hopkins book, My Life in Advertising. I hope I'm not infringing on any copyright laws here by reading a paragraph from it, but I bought enough copies that I should be good.‍

Patrick Gilbert: You literally gave out 30 copies at a speech that you gave. ‍

Isaac Rudansky: Yeah. ‍

Patrick Gilbert: A few years ago, as if it was your own book.‍

Isaac Rudansky: Yeah. So, Claude Hopkins eventually left Grand Rapids where he did very, very well with Bissell. That's a separate story. And he went to work for Swift & Company. They sold in those days a product called Cotosuet. It was a lard. It was lard that was used as a replacement for butter. It was a rendered animal compound lard. Cottolene was its biggest competitor. They had the biggest part of the market. Here, I’ll read what he says:

“We opened a sales office in Boston and started advertising in New England. We had hardly started when Mr. L.F. Swift came to my desk one day. He said, ‘Father is very nervous about this money spent in advertising. He considers it an utter waste. The results so far are not very encouraging. You have been here nearly six weeks, but our sales on Cotosuet have hardly increased at all.’” 

“I had no need to explain to him. He knew that advertising had hardly started, but I saw that I had to help him out by making some quick showing. That night after dinner, I paced the streets. I tried to analyze myself. I had been a great success in Grand Rapids and I was making a fizzle here. What were the reasons? What did I do there in a different field and a different company that I could apply to my new company's problems? At midnight …” Look, he's pacing the streets. It's midnight. He's thinking he's analyzing. “At midnight, I thought of an idea.” 

So, he eventually went and approached Rothschild & Co., I'm going to skip paragraphs, which is a huge department store. And he negotiated with the marketing manager of this big department store that on the 3rd or 5th story, he's going to have a bakery, and he didn't even have the bakery yet. Build a huge cake and he was going to use Cotosuet for the cake base instead of butter, and he was going to give out samples. People would have to climb the stairs, whatever, to the 5th floor of the department store. He would give out samples, and the claim would be like, “Look, if this is better than butter, then it's got to be better than Cottolene and the other competitors.” That was his idea and he was going to make a whole show about it. 

“My idea was to make a cake with Cotosuet in place of butter, then to argue that a product better than butter was certainly better than lard. Mr. Jones, who was from Rothschild, accepted my proposition. Then I went next door to the H. H. Kohlsaat & Co., bakers and ask them to bake the cake. I told them to make the special tins which were necessary to decorate the cake in a magnificent way, and to build it as high as the room. They did so.” 

He eventually went. There was a huge sea of people outside the department store barricaded by police. “During the next week, 105,000 people climbed four flights of stairs to that cake. The elevators could not carry them. There, I had demonstrators to offer samples of the cake. Then we had prizes to offer to those who guessed nearest the weight. But every guest there had to buy a pail of Cotosuet.” 

Then he himself went with people from Swift & Company to all the other states on the eastern seaboard, including New York and a bunch of other ones, and he came up with a different variation on this plan. And by the end of the year, Cotosuet had the lard market and Cottolene was out of business, basically. 

So to me, it's like this was a man pacing the streets, thinking about a problem. Analyzing himself, his own flaws. Thinking about the product value, how to communicate this, and he came up with a solution. Now, that's a far cry from where we are in advertising today. That's the other side of the gamut of what in-app advertising today has become. 

So I think any movement back in that direction, back towards the center, where you could use technology and you could use complex problem solving is, to me, a very good thing. So, that's my overarching feeling about the iOS 14 updates and how it's going to impact the industry.‍

Patrick Gilbert: Very good. I would say for further reading, if you are a marketer or advertiser that runs ads on Facebook, I definitely recommend a podcast called Ecommerce Influence. Look for episode number 292. They talk about iOS 14 and this whole update. They talk about a more tactical level with some optimization techniques. 

At least one of the guys is from a company called Foxwell Digital. They have a blog that has a lot of really good, useful content. So if you're running ads on Facebook, and you need some optimization techniques, and more descriptions about what conversions API is, and all that other good stuff that we talked about, check out that website. We'll link to it. 

The other thing I'll say is, if you're interested in his overall topic and the privacy debate as a whole, Eric Seufert, he runs Mobile Dev Memo. He also has a podcast you can check out. He goes through this stuff pretty in-depth. Then from an overall tech perspective, Ben Thompson has been covering this stuff on Stratechery in depth. So, there's a lot more to come on this. 

Again, we're still in the early stages of what this entire transition and new world is going to look like. But to Isaac's point, I'm excited about this. It's going to be hard. It's going to be challenging. And there's probably going to be some frustration and crying along the way. But I'm excited for the opportunity to be able to go into this new era. I think it'll be better for users. I think it'll be better for advertisers. And I'm excited to be a part of it. Thank you for the opportunity. I think that's it.‍

Isaac Rudansky: One more important question for you, Patrick. ‍

Patrick Gilbert: Yeah. ‍

Isaac Rudansky: Where's the best place to hide a dead body?‍

Patrick Gilbert: Second page of Google. Or inside of a cookie. I don't know.‍

Isaac Rudansky: Thanks for listening, everybody, and we'll see you on the next episode.