Equation of Excellence

Rabah (00:02.008)
All right, folks, three, two, one, and we are live with one of the smartest men on the planet, humans on the planet, in my opinion, the proprietor and brains behind mobile dev memo amongst a bunch of other awesomeness. Eric, I've been just chomping at the bit to get you, like, some of the stuff you post, some of the stuff you write is just so inspiring and impactful, just brain -jarring that I've been.

wanting to get you on the show forever. And then finally, Rishib connected me, shout out to my boss, and then we made it happen. You were gracious enough to give us some of your time. So Eric, welcome to the show.

Eric (00:43.726)
Yeah, well, thank you for the various sorry, the very generous introduction, I would, you know, probably elevate Rishabh the MIT PhD above me. But but you know, that's, you know, he's your boss. So you'll have to deal with the consequences of that.

Rabah (00:53.816)
His CV's video game.

Rabah (01:01.848)
See, I told you guys, big brain, big brain. So for people that don't know you, how'd you get started? What do you do now? Let's give some people some color kind of to your background and then I have just a ton of questions to jump into.

Eric (01:17.326)
Yeah, so I mean, just kind of start from, you know, graduated from college, you know, studied finance and computer science in college, first job out of undergrad was in oil and gas investment banking. Didn't last very long in that and didn't really like it. And I, I had always, you know, wanted to work in tech. And so I just did that, got a job at a startup, thought that was pretty fun, but

felt that, just kind of felt a sort of compulsion to go to graduate school. So I went to England for graduate school, I went to the University College London. I was to study economics, I received like a full, you know, kind of full scholarship to do that from the European Commission, when when the UK was still in the EU. This is a while back, did that. And then and then as part of that program, it was kind of like a a cross

Rabah (01:56.472)
Yeah.

Eric (02:15.342)
disciplinary program. So I ended up doing a year at it was a master's. So I did a year in England at UCL. And then I did a year in Estonia at the University of Tartu, which is Yeah, it's actually one of Europe's oldest universities actually founded by the Swedes. So the Swedes are interesting because when they when they would colonize places, they would like build universities and educate the population and, you know, build infrastructure. Anyway, so did so got

Rabah (02:25.56)
Really?

Eric (02:44.11)
finished the degree there, wrote my thesis and that's actually where Skype was born in Estonia. So was very lucky to get a job at Skype after graduating. This was still coming out of the GFC. So like there was no possibility of getting a job in finance at the time. And, you know, Skype was like this ascendant tech story and you know, it was a very exciting company to be able to work at. I joined them as like an analyst just to

you know, kind of generic analyst, worked there until the acquisition by Microsoft. And what I, what I really appreciated about Skype was that it employed the freemium model, right? So Skype was, you know, VoIP product and people may not be familiar with it depending on how old they are, but, it was kind of like one of the original VoIP products, right? So you download Skype and this was, you know, kind of this, you know, this was 2010, 11. So smartphones existed and, and the iPhone existed. and it was a couple of years old, three years old at that point.

Rabah (03:18.776)
Yep.

Eric (03:41.774)
but it wasn't ubiquitous. And so people still primarily consumed media on their desktop computers. And so that was the most popular app that we had was the windows version of Skype. And so it just allowed you to make VoIP calls, right? For free. And if you, if you could pay by essentially like renting a phone number, so you'd get a dedicated phone number. So when someone called that phone number, it would actually call your, your Skype app, right? I call it an app. It was a program on your desktop for the most part, but,

I was there when we started expanding into smartphones and actually we had an offsite in Amsterdam one time when they did the first ever VoIP call from an iPhone to an Android. Like pretty exciting. And some people might remember that we actually had a partnership with Facebook where you could install the Facebook plugin to your Facebook chat and you could do Skype chat. So that was the original version of like Facebook messaging or calling was actually using Skype's technology. And what was cool was that that offsite, you know, there's like, I don't know how many people worked with Skype at the time, probably like,

Rabah (04:19.32)
huh. Yeah!

Rabah (04:25.002)
Do remember this.

Rabah (04:32.312)
Yes!

Eric (04:40.686)
thousand or something and maybe more, but we were all in this big auditorium and we Skyped Mark Zuckerberg. So he came up on like the big screen and the CEO was having a conversation with Mark Zuckerberg. And anyway, I think Sky, I think Facebook ended that partnership after like a year and they were just like, we got to build this ourselves because you did have to download a plugin. It wasn't, it wasn't, you know, sort of native anyway, but was fascinated with the freemium model. And once Skype got acquired by Microsoft, I didn't think it'd be that exciting of a place to work anymore. And I was living in Tallinn, Estonia.

Rabah (04:57.912)
Yeah. Yeah.

Eric (05:09.934)
And I saw that there were these gaming companies across the Baltic sea and Finland doing all this interesting stuff with the freemium model, mostly for Facebook, canvas games. And so I just really thought that this freemium model, when smartphones became ubiquitous, because I mean, it was heading in that direction. I thought that that would be the business model that any sort of software used, right? Because if you've got everyone with a computing device available to them at all times, it would just make sense to employ the freemium model where you would sort of exploit that large TAM.

exploit the ease of distribution and then, and then sort of capture value from users commensurate with their, the utility that they find in the product. Right. And that would be a better way to monetize, better way to make money than just gating content or gating, gating functionality behind a fixed price point. So I moved to Finland, I got a job with a gaming startup. It went under. I went to another gaming startup. It also went under in Finland. And then I decided to start writing a book.

called freemium economics. I'd kind of learned a lot, spent a lot of time thinking about the freemium model. So I started writing this book. Meanwhile, I also got a job at a company called Wuga, which was a sort of scaled casual game developer in Berlin. They were pivoting away from Facebook canvas at that time, which all gaming companies were to mobile, right? Like, so this was around 2014. This was like this, the sort of inception of the mobile gaming category as we know it today, right? So before 2014, you had a lot of, or 2012 really was the actual inception point, but 2014 was when the industry was sort of like,

was, was, was maturing, evolving in that direction. So 2012, you had clash of clans launch, you had candy crush saga launch. you had a game called, mobile strike launch. And these were the, you know, these are still billion dollar friends. Well, clash of clans and candy crush are still billion dollar franchises, but like this, this really proved the model, right? It proved that these games could scale. They could, you know, attract, you know, millions, tens of millions, hundreds of millions of users.

Rabah (06:43.368)
Yeah, canned across, yeah.

Eric (07:06.574)
They could grow into like billion dollar products. And then you saw like an entire industry sort of blossom in that, in that 2012 to 2014 period, I think was when, when, when the sort of the industry really like, it, it, it experienced like that, that explosive growth. And so I moved to Wuga in 2014 was finishing my book around the time that I started and, and Wuga had, you know, it was a casual games developer, like, shifted to mobile. We had a big hit game called jelly splash.

And, and then published a book. So, was there for a couple of years. Then I moved back to Finland, went to Rovio, which is the company that makes Angry Birds. I launched Angry Birds 2, did the marketing for the movie, left Rovio, started a company that, that built an analytics product for marketing teams, sold that to another mobile game developer in the U S and San Francisco. So I moved back to the U S was there for like an earn out. And then, just when my son was born, I just kind of left and

Rabah (07:49.176)
Wow.

Eric (08:04.846)
tried to figure out what I wanted to do next and started actively angel investing and thought that that was pretty fun. And so I started a fund, called Hercules capital invests in precede stage companies in the mobile ecosystem. So nowadays, now my sort of professional life is Hercules capital, mobile dev memo, which is my sort of trade blog on the digital economy. I do some consulting, usually for like larger companies. and that's base and then some advisory work. So I kind of, I feel like I have a.

a field of vision that encompasses the entire spectrum of scale in the digital ecosystem from the very earliest stage companies that precede when they're raising the first money on an idea through to very large companies that are operating at tremendous scale that bring me in to help bring operational efficiency to their marketing team.

Rabah (08:55.128)
Man, that's super impressive. Did you enjoy overseas or, cause where are you at now in the States? How did I not know you're in Austin? my gosh, we got to run this back and we'll do it in person. I'm in New York right now, but my gosh. this is, my dreams are coming true.

Eric (08:59.982)
I'm in Austin.

Eric (09:03.95)
I knew you were in Austin, so I don't, yeah, I don't...

Yeah, well, yeah, we'll do that. We should next time Rishabh is in town, we should go a beer or something.

Rabah (09:15.288)
yeah, we'll do a triple. Yeah. Okay, a few questions. What's the thesis for the fund?

Eric (09:22.734)
So it is actually very thesis driven, right? So I raised the fund principally to invest in the companies that I feel like are best positioned to either navigate the privacy changes that have, you know, that have sort of dramatically altered the digital landscape or to instigate, you know, new methodologies for navigating them, right? So like on the content side, I invest mostly in companies that I feel like are best positioned.

to do well without the IDFA, right? To do well, to be able to distribute their products without needing deterministic identity. And so that is, you know, usually content that meets a very specific need for a group of users that would actively seek out that need, right? So no longer can you just market a generic mobile game, right? Like a sort of me too, generic, undifferentiated mobile game using performance user acquisition, because that just doesn't...

you can't work anymore because you can't target known spenders. So that's out of thesis. But on the infrastructure side, it's like a lot of privacy enhancing technologies, measurement, like probabilistic measurement. Fermat, I'm an investor in Fermat. Risha was a reader and he said, hey, we're raising this round, but we think we're in your thesis and would you want to invest? And I very happily did and I'm very glad that I did because that's been marked up significantly.

Rabah (10:46.296)
Yeah, let's go. Yes, it has.

Eric (10:49.646)
But so, you know, infrastructure that I feel like helps companies navigate these privacy challenges, the frictions and then content that I feel like kind of natively navigates that.

Rabah (10:59.416)
Yeah, no, I love that thesis. Let's jump in a little bit. So you've been in the space a little over a decade, mobile gaming, advertising, et cetera. What do you think the biggest changes you've observed in the mobile user acquisition space? Because you kind of touched on it, but I'd love to just dive deeper in there because you talked about the iOS changes, basically the deterioration of identity, targeting, et cetera. But yeah, just open up that big brain and give the listeners kind of a peek in there.

Eric (11:27.854)
Yeah, it's interesting. So I wrote and let me just give me a second. I'll pull it up. I wrote a piece about this a while back. And let me see.

Rabah (11:31.32)
Yeah.

Rabah (11:37.528)
Your essays are sensational, by the way. And if you guys are not subscribed to the pod or to the paid, you need to get over there because it is, it's, it's top class, man. You're a, it's really challenging, I think sometimes to be a very intellectual writer that's entertaining. And I think you balanced that incredibly well. I think, I mean, cause I also follow Ben a ton as well. You've been on his podcast and stuff. And I think you guys are too, I've just so well written, well spoken, like it's.

Eric (11:59.854)
Yeah.

Rabah (12:06.68)
I'll quit gassing you up, but go subscribe. It's fantastic.

Eric (12:10.158)
Yeah, I mean, on that, you know, Ben is like a unique person because, you know, I was guilty of this. Like you could look at Ben and you could say, there's no way he could possibly understand all of these subjects that he writes about. And so it's not worth reading, right? You could just look at his output and the breadth of his output. It's impossible. It's impossible for any human being to actually understand deeply all of these topics to a degree that would be interesting to read.

And if all I'm going to do is get a super superficial like overview of these topics, well, I'd rather go just read a news. Like I'd rather go read a news site because at least that'll be shorter. Right. But that's not, that's not the case with Ben. Ben is a mile wide and a mile deep and he's a very unique person. He deserves all of the success he has because he is unique in being able to do that. And early on, you know, when I first started mobile dev memo, it was really a companion website to the book. Right. So the idea was in the book, I.

Rabah (12:39.672)
Yes.

Rabah (12:50.136)
Yes.

Rabah (12:54.488)
He really is.

Rabah (13:00.952)
Yes.

Eric (13:08.59)
I kind of walk through these different like analytical approaches to, to, to managing a free to play business, a freemium freemium business. And what I wanted to do with the website was like, take every single one of the chapters and provide essentially like a model, like a spreadsheet model or like a Python model to help people, you know, to help people through that exercise. And that's early on the website. I mean, that's kind of what it was. It was like me sharing code and you know,

walking through like math equations and stuff. And it was like very, very niche and it gained some traction. And it was like, it was reasonably popular within a certain segment. Right. I mean, I think if you walked into like a user acquisition team at a mobile gaming studio, chances are good that at least one person would have heard of the website and read it. And what I feel like I kind of hit a crossroads at one point because I, it peaked. Like there was just, or it felt like I had reached, I had reached this sort of like ceiling of what I was capable of doing with that scope.

Rabah (13:50.072)
Yes.

Eric (14:05.23)
Right. It was like, okay, well, if this is all it's ever going to be, which is like, you know, mathematical models for assessing, you know, the performance of a free, free to freemium title, mostly free to play. I mean, I keep interchanging those two because, you know, these tools, these tools and methodologies were only really relevant for like mobile game developers. It's like, that's a pretty small audience. So I really want to spend time doing that. Right. Cause like, I mean, my wife read me the riot act one time, like it was,

Rabah (14:22.52)
Same, same, but different.

Eric (14:34.35)
I used to basically take every single Sunday and that was it. That was my Sunday, I was writing. Because I felt like it was really helpful for me in my daily work too. Like, okay, I'm really thinking deeply about these topics. I'm going to write about it because that helps me sharpen my understanding and helps me clarify the concept in my mind. But then I'll just publish it and I'll make it available, right? And then there's probably knock -on benefits to that. And so one time she was like, I don't want this to be our life that every single Sunday is a work day for you. It's basically off limits.

Rabah (14:38.168)
Yeah. Yeah.

Rabah (14:47.928)
Yes.

Eric (15:02.926)
And I was like, okay, you're right. and so like, I need to find a way to distribute this across the week, right? so that I'm not just concentrating on Sunday, but I was just like, okay, well, is this worth it? Right? Because if my wife's pissed off, I'm taking a lot of my free time to do this is, and it, and, you know, the maximum possible audience of this is like 2000 people's worth doing. And so I decided to broaden the scope, right? It wouldn't just be freemium and it wouldn't be more like the analytics. I did be, I think more of the, the entire funnel, the entire spectrum of the funnel. Cause I was working in user acquisition at the time.

Rabah (15:13.816)
Right.

Eric (15:31.95)
but I was mostly talking about like product analytics. And so why don't I open this up to the, to the user acquisition side of this because a that's just much bigger, right? It encompasses more people, more product categories. but B I felt like that was becoming a lot more, of, of an area that was ripe for bringing that kind of analytical rigor, because this was around the time that Facebook was really taking off. Like this is probably like 16, 17, right? And so if you think about the history of Facebook, it's advertising business.

Rabah (15:55.896)
Yes.

Eric (16:01.358)
Explo - first of all, it's it, you might remember this or maybe not. I don't know how old you are, but like, remember they had, they didn't even have a native app. I mean, they weren't interested in, in, in, in app ads. Like mobile was not a big component of the revenue for a long time. It was all the right -hand side ads and they weren't doing that well. Right. And then, actually the, it's the CEO of Instacart now who left from Facebook. I'm blank on her name, but she was the one who said, okay, we need to actually build an ads product for the app.

Rabah (16:05.4)
I'm old, I'm old so I remember.

Rabah (16:17.912)
Yes.

Eric (16:29.902)
Right? Because all the engagement shifting to the app, it's shifting out of the web, it's shifting into the app. We need to be there with ads, relevant ads for, for, for people's eyeballs when they're using a product on their mobile phone, which increasingly is, you know, the, the, the, the bulk of their engagement. And then what they did was 2016, they launched a EO and that was the game changer. That was let's do conversion targeting. Let's do conversion optimization against the things that advertisers actually care about. And then 2017 was VO. Right. And VO was essentially the first.

Rabah (16:44.184)
Yes.

Rabah (16:53.784)
Yes.

Eric (16:58.542)
version, the first iteration of advantage plus, they even said that they even said when they launched advantage plus for app campaigns, that was just rebranded VO, but they took that model and they because they had to post ATT, right now I'm skipping ahead a little bit. So VO is 2017. But that was the tidal wave moment for digital advertising for for SMB advertising for bringing all these SMBs onto the platform being able to buy revenue. When you could do that, when you could buy revenue, and you could do that efficiently, and you turn that into a

Rabah (17:00.92)
Yes.

Rabah (17:15.288)
Yes.

Yes.

Eric (17:28.014)
virtuous cycle of more revenue equals more data equals better targeted ads equals more revenue equals more data. When you could do that, you essentially exponentially grow the business and you create new portions of the economy that couldn't exist before. D2C couldn't exist without AEO. It just couldn't. D2C was born from AEO, right? So that's what Facebook did 2016 -2017. I was like, this is really fascinating. Like these algorithmic systems that some people are really scared about.

Rabah (17:32.888)
Exactly right.

Rabah (17:37.496)
Yes.

Rabah (17:46.36)
Absolutely right. It's exactly right.

Eric (17:55.886)
They're like, no, I don't want to entrust you. And you think about now with Advantage Plus, people are like, I don't know, it's black box. 2016, I mean, this was like, you know, the, this was like aliens coming to visit earth. Like, I mean, this was unheard of that you could do this. And so I started writing about that and the blog kind of exploded because of, wow, this is happening. This, this is, this is a sea change. Like, this is a totally structural transformation of the way that digital advertising takes place. Right. And so then the blog expanded from there and then ATT.

Rabah (18:00.664)
Yes.

Rabah (18:14.36)
Yes.

Eric (18:24.526)
And then that was the other sort of like big explosion of, of, you know, sort of, awareness for the blog was when I started covering that because, you know, no one was really covering it or when they were covering it, like the press, like those were tech press, they were very differential to Apple. They're like, isn't it so great that Apple's doing this stuff? And I was like, well, okay, you can reserve your moral judgment. but, you know, or, and I was like, you know, I'm, I'm happy to, to have that conversation, but we need to have the other conversation too, which is what is this going to do the economy of the internet? Right. And.

Rabah (18:24.856)
Yes.

Rabah (18:52.856)
Exactly right.

Eric (18:54.126)
And if we want to have, if we want to get into the ethical considerations here, well, let's actually rigorously interrogate what Apple is really doing here. How much more protection are they giving to users? Any? Cause my argument is none. They're using all the data. They still use the same exact scope of data is still being used to target ads, but only Apple can do it. Right? Only Apple can do it. And they don't give you a genuine choice for that scare screen. And when they do it themselves, they call it ads personalization.

Rabah (19:07.735)
Exactly right.

Rabah (19:14.584)
Exactly right.

Eric (19:21.326)
And they get a lot more real estate to explain the benefits of it. Right. That's so anyway, that was the other sort of big, you know, kind of, audience expansion moment was ATT, but, but really the, the first, the first moment, which where I said, okay, look, I can either keep writing about freemium analytics, product analytics, and how to optimize freemium products, or I can expand this to, to, to the digital marketing subjects that are, you know, particularly relevant now because of these changes that are happening. So I did that blog grew, you know, I would say pretty meaningfully, probably like two or three acts. And then with ATT, it probably like five.

Rabah (19:21.592)
Yes.

Yes.

Rabah (19:49.368)
Yeah.

Eric (19:51.758)
So that's kind of the history, but if we go back, that's history of the blog, right? Which I think probably maps to the history of the mobile space. But if you think just about gaming, I mean, first of all, you had 2008 App Store launched, right? 2009 IAPs were introduced, obviously a big sort of watershed moment. Then kind of 2013, 2012, 2013, free to play becomes the dominant business model for mobile games. Another big sort of, you know, a sea change.

Rabah (19:51.768)
Yeah.

Eric (20:19.086)
2014, 15, you get this games as a service framework. And keep in mind, gaming is probably a year to two years ahead of the rest of the consumer tech space in terms of like business model, distribution, all that kind of stuff. Then you had like 2016 kind of subscriptions. The subscription fees were dropped on the mobile platforms. And that's when you saw a lot more companies moving into mobile with consumer, just sort of generic consumer products. I call them utility apps that were.

Rabah (20:28.248)
Yes. Yes.

Eric (20:48.526)
monetized by subscriptions. And then you had 2020 with 21 is I, you know, IDFA deprecation through ATT, right? So I think those are kind of like the big watershed moments. And then 2016, 2017, the conversion based optimization that Facebook introduced, but Google did it too with mobile. They had a thing called UAC, which you, again, these are pre, these are precursors, right? To the exact same products that have been rolled out now broadly. P max, you can draw a straight line from, from UAC to P max. You can draw a straight line from,

Rabah (21:09.624)
PMAX, the ASCs, yeah.

Eric (21:17.198)
a VO to, and what they call AAA, right? So automated app ads. You can draw a straight line from automated app ads to, to Advantage Plus. So 2016, 2017, that was another big sort of like transformative moment in mobile.

Rabah (21:31.16)
No, man, that's such an interesting, I guess segue segueing off of that with Metta. The stock is absolutely ripping. It took a little bit of a haircut, but I mean, the numbers are just atrociously good. What is that kind of signal about the state of mobile advertising, you guys are acquisition? Like, is that a bellwether for anything or is that the company is just rigorously doing awesome because they they did and we'll get into the metaverse stuff as well. But once Zuckerberg kind of got off the metaverse tee,

I mean, the stock has just been screaming ever since.

Eric (22:02.99)
Yeah. Well, I mean, you know, I pushed back a little bit on the, a little bit of a haircut comment because it went down to 93, from, you know, the, the previous all. So it was three at three 79 and August 1st, 21, right? You know, ATT gets rolled out in just before that in July 21, and then it went down to 93. So it's, pretty significant haircut. They went bald for a while.

Rabah (22:12.664)
Okay, that's fair. That's fair. So yeah.

Rabah (22:20.664)
That's fair.

Rabah (22:27.288)
That's fair. I mean, but it's still up 40%. Yeah, it's still up 40 % this year, right? So I mean, it's still...

Eric (22:35.214)
yeah, this year, you're to date. Let's see where's it at. You're to date, it's up. Yeah, yeah.

Rabah (22:37.976)
Yes.

Give or take 40, right?

Eric (22:44.43)
80 % but yeah I mean it's on fire right so what they do so you might remember that comment that Zuck made I mean I kind of this is I don't know I feel like this stands out for me let me I don't remember exactly when it was it was an earnings call I wrote about this recently called I mean a piece called Metis Turnaround but you know Zuck made a comment when the stock was near the all -time low it was October 27th

Rabah (22:44.6)
Yeah.

Eric (23:12.462)
2022, right? So there, there is an, is an earnings call. and he said, those who are patient and invest with, with us will end up being rewarded. And this was like days before the stock hit its all time low. And it actually gapped down after that call. and what he was referring to was like the wholesale transformation of the company. Right. So I think a lot, you know, the metaverse stuff, that was a distraction. I mean, I, my sense is they've completely abandoned that. I mean, I think.

Rabah (23:33.496)
Yes.

Yes.

Eric (23:40.654)
You know, they could be the market leader in VR, but who cares? I mean, that's not going to be a big market. I mean, I've been in games for too long for anybody to convince me that VR is going to end up being an important, meaningful hardware form factor for consumers. It just, it just can't be like, I've heard that argument so many times and it's just never come to fruition because that's not how people want to engage with content. They want to be aware of their surroundings. No one, I mean, the Apple vision pro, I mean, that's, I think proof point. Anyway, so well, yeah.

Rabah (23:43.576)
Yes.

Rabah (23:49.048)
Yes.

Rabah (24:07.)
have one and it's useless. Or not useless, but it's like the utility you get from it versus the economic cost is just so divergent. Like it's just not even near worth buying. I mean, it was the most expensive piece of computer hardware I've ever bought. It was almost $5 ,000, like all in, and I've used it 20 times maybe, 30 times. You know what I mean? It's just not there.

Eric (24:21.678)
Right.

Eric (24:28.59)
I think the singularly compelling use case for that is error travel. That's it.

Rabah (24:33.144)
Yeah, the two big kind of 12 out of 10 features were the extension of desktop where you can, you basically put this on, now you have this huge desktop that's connected to your laptop and then consuming media. But outside of that, it's just not where it needs to be in terms of that. And I think kind of to your point in terms of mobile stuff, if you don't have the developers on board,

then it just becomes really challenging. Where there's just no real big incentive to develop for Apple Vision, they haven't made it easy, and so it's just, you start to get this like cool horse, but not a big cart to pull.

Eric (25:11.63)
Right. Well, the killer app has to be a game. It just has to be.

Rabah (25:15.384)
That's a really good point. That's a really good point. That's a really good point. That's a really good point.

Eric (25:17.23)
And I mean, that was true for smartphones. I mean, Xbox didn't take off until Halo. Like you just need, you need a killer app and it's got to be a game. And why is that a better surface area for engaging with a game than a smartphone or whatever, a console, anything like, but really a smartphone. I mean, the thing with smartphone is the dominant gaming platform. It's the dominant gaming hardware. It's just people don't think about it that way because it's not men playing those games. It's women.

Rabah (25:33.4)
console or PC Master Race.

Rabah (25:40.824)
Yes.

Yeah, that's fair.

Eric (25:46.222)
Right? Like I, I was going to write this big article one time about why it's it's, I won't write it now because it's the, you know, the kind of it, there's the sentiment doesn't really exist anymore, but there was a lot of, hostility to free to play gaming when it was, when it was sort of gaining traction in the West. Right. And the reason was, I think, because it was women playing them. It was women playing those games and you're not a gamer. I'm a gamer. You know, I'm a gamer because I buy a new console every two years.

Rabah (26:10.04)
Yeah.

Rabah (26:15.256)
Yeah. Yeah.

Eric (26:15.374)
And those games are made for me. And now you've got developers making games that aren't for me. And I resent that. And I hate those games. I hate those games because they're, you know, they're exploitative and they're psychologically manipulative, which has all been proven to not be the case. They hate those games because they're not made for them. The people that consider themselves gamers and have to open up the tent to a whole new group of people, they don't like that. And that's why they were hostile to the free to play gaming category. but you know, the smartphone is, is by far.

Rabah (26:19.96)
Yeah. Yeah.

Rabah (26:25.912)
Yeah.

Yeah. Yeah.

Rabah (26:37.784)
It's so fascinating.

Eric (26:45.23)
far and away the largest platform for gaming. It's just people don't think about it that way because they don't think of these apps as bonafide games. A game is Call of Duty. A game is GTA. It's not Candy Crush, right? Anyway, yeah, so.

Rabah (26:45.918)
absolutely.

Rabah (26:55.032)
Yeah, exactly.

Yeah. Yeah. That's such an interesting take. I've never heard that. It's awesome.

Eric (27:04.878)
But I think the metaverse stuff, I mean, for Zuck, I mean, it was just something that they needed to do to show that they were doing something. And I think that the rebrand was more to do with the bad press around the Facebook files. But I think the metaverse stuff was just, hey, we have to show that we're doing something. We have to show that we're taking decisive action. And so this is what we're doing. But because they couldn't really highlight to investors all of the change that they were making in the background because they wouldn't understand it. Like even to this day, I mean, I'll talk to...

Rabah (27:30.84)
Yes.

Eric (27:32.686)
I mean, some of the smartest people on the planet and like, they're just, this is, it's not that they can't understand it. It's just that they've never been exposed to it. It's like, no, this is how ads work. This is how ads are served and targeted. Right. And so, you know, what they did behind the scenes was they, they sort of, transitioned the friends and family graph to an open graph, right? They opened up a lot more.

Rabah (27:57.656)
Mm -hmm.

Eric (27:59.79)
content, they opened up every user then was sort of could be exposed to a much, much larger pool of content. They transitioned to short form video, right, which is just more engaging and they boosted engagement that way, right? So there's more stuff to show you. And the stuff that we're showing you is more engaging. So we'll keep you on the, in the app longer. And if in doing so, in doing so we'll, we'll show you more ads, right? So we don't have to ramp ad load in the feed. If you just watch more stuff, if you spend more time watching stuff, but in order to do that well,

Rabah (28:18.712)
Exactly right.

Rabah (28:22.648)
Exactly right.

Eric (28:28.814)
they had to build the ML classification systems to curate video content, which they never had to do. Right? The platform was text and image -based. That's much easier to classify for relevance and to rank than video. Video is much harder to do that. Well, how did they do that? Well, they say it's AI, but I mean, you know, this is maybe just a semantic argument. I don't think it's AI per se, but it's just, you know, heavy duty machine learning. And what they did was they ran those models. They made the models much bigger and much more all -encompassing because the models were very sort of

Rabah (28:36.856)
Yes.

Rabah (28:52.12)
Yes.

Eric (28:58.03)
purpose specific before, right? And then they ended up just taking all of the data they had available to them as inputs in a much larger model and running that on the GPU clusters that they bought in order to do the classification efficiently, or at least, you know, to make it more efficient from a relevancy standpoint. So that's how they boosted engagement. Then on the advertiser facing side, they built out the advantage plus suite, which they call AI. Maybe that I feel like

AI, I feel like when you talk about AI, what you really mean is you're emulating human activity. It's not just machine learning, right? So like, I think if like if a person did that before and now they no longer do it, that to me is AI. But I mean, I think people could disagree about that. But with the advantage plus, they just made campaign optimization that much more efficient, right? So what you had before advantage plus was, well, I've got a bunch of manual levers, and I'll put in a bunch of inputs and even with VO, right with VO. So the structure of a VO campaign is

Rabah (29:28.792)
Yes.

Rabah (29:34.84)
Yes.

Rabah (29:40.472)
Yes.

Eric (29:51.246)
I'm not giving you a bid and it was like, it was a fundamental deviation from the way digital advertising was purchased prior to that. I'm not giving you a bid. You tell me how much, you tell me what your ROAS target is. You don't give me a bid. So I might end up spending on a CPA basis, CPE basis, I might end up spending $100, $200, right? Whereas your bid would have been 10 and you get scared. But the reality is I'm bringing you ROAS, I'm bringing you your ROAS target, right? And so that's, that's,

Rabah (29:59.896)
Yes.

Rabah (30:15.032)
Yes.

Eric (30:20.174)
That's ultimately what you should care about. And so let me figure everything else out. You don't have to tell me what your bid is. You don't have to tell me anything, any targeting parameters. I will find the right audience for you and I will deliver the row as that you need. And the only, the only question is what the scale will be on in terms of your budget. And then all you really have to do then all as the advertisers, all you really obligated to do in order to drive performance is to experiment with creative. Right.

Rabah (30:34.232)
Yes.

Rabah (30:44.376)
Yes.

Eric (30:45.23)
And so advantage plus just took all of those manual levers away, push them into this machine learning is very sophisticated. Performant machine learning models that we're going to optimize your campaigns for you. Don't worry about that. We'll bring you to optimal performance. Right. So they, so they, the, the data that they lost from the IDFA deprecation is lost forever. They can never regain it. And that, that the targeting will forever be less relevant. The targeting will forever be, will forever be sort of diminished, right? They can't regain that.

Rabah (30:56.056)
Yes.

Eric (31:14.734)
targeting efficiency, what they can do is make sure they're not leaving any money on the table in terms of campaign optimization. So you bring that to sort of like the theoretical optimal level, which they did. And then the next step is to even take away creative production from the advertisers that will do that. We'll do that for you. Now that's, that's further down the road, but so advantage plus made campaign optimization more efficient. you know, the, the open graph plus short form video boosted engagement kept people in the app for longer, watching longer, so they consume more ads.

Rabah (31:18.136)
Yes.

Rabah (31:28.92)
Yes. Yes. Yes.

Rabah (31:42.072)
Yes, some more inventory.

Eric (31:44.174)
And then on the measurement side, rolling out, so Cappy is old Cappy predates ATT. Cappy was built for in response to ITP on Safari, right? But Cappy predates ATT, but they rolled that out. They sort of pushed the engagement of that post ATT. They pushed the adoption of Cappy post ATT so they could get the conversion events. They're not deterministically matchable, but they can be matched with IP. And so they're able to do better conversion matching. And then they rolled out AEM for app campaigns.

Rabah (32:02.488)
data back in. Yep.

Eric (32:13.262)
So they did like those three things. I mean, but that's all in the background. Like how are you going to sell average? Are you going to sell investors on that? That seems tough. Like that's, that's very in the weeds. They really have to understand these systems in order to believe that you're going to be able to do that. And they have to understand that, you know, how long of a process that is and how expensive it is. Right. So they did all of that while kind of pitching while this metaverse narrative, which I don't think they ever really were fully bought into.

Rabah (32:19.544)
Yes, it's a hard sell.

Rabah (32:28.952)
Yes.

Yes.

Eric (32:41.518)
But it was like, hey, look, we are definitely doing something. See, look, metaverse.

Rabah (32:46.936)
Fair. That's a fair, that's the best explanation I've heard of it. I thought it was just, Zuck had a bunch of money and he was into this and he was driving the boat. He has X million voting rights, so it's like the board is pretty much impotent to him and so he's gonna do what he wants, but that's a really good explanation. Kind of shifting a little bit to the metaverse as well, you had a great combo with Matthew Ball and you guys talked about the potential economic models of the metaverse.

How does that work when you have, like what's the economic function for digital sort, like resources when you have, you know, essentially no scarcity?

Eric (33:25.646)
yeah. Well, I think, I mean, it maps pretty closely to the idea of, of, you know, zero marginal production costs with freemium, right? So, and then just digital products in general, right? And like, this is also like intersects with, with Ben's aggregation theory, but, you know, so like that's the, so the, the, the beauty of freemium, and I talk a lot about this a lot in the book, it's actually like the sort of,

It's the crux of the model is like we with the digital product, you have no marginal production costs. Like it doesn't cost you anything to produce one more copy, right? There's no physical distribution. It's all digital. And so you can use that to your advantage in, you know, distribution because you could reach everybody with no cost to you. And so if you can reach everybody with no cost to you, why not not reach everybody with no cost to them? Then you've absolutely maximized the number of people that can access your thing and your product. And so then what you want to do,

Rabah (34:01.14)
Yes. Yes.

Rabah (34:14.2)
Yes.

Eric (34:24.878)
is make a product that a lot of people want to access, right? So you want to make a very sort of broadly appealing product because if it's niche, you're probably better served. You might be better served. I mean, you have to run the analysis, but you might be better served for just a fixed price point, right? But if you're very broad, like let's say, you know, you go to this theoretical extreme, everybody in the world could potentially appreciate this product and it's free and it costs you nothing to make every single copy of it. Well, then there you go. You've got utterly free access to a market of the entire world, right?

Rabah (34:27.576)
Yes.

Eric (34:53.71)
and then, you know, then you're really just competing for eyeballs. Then you're competing for attention, and trying to get your, your thing in front of people. And so then that's just an advertising exercise. and then on the backend, you're trying to make sure that, you know, for the people, you know, to whom this product is, it is, solves a need or has, you know, you know, a significant amount of utility, you're able to monetize them.

Rabah (34:56.984)
Yes.

Eric (35:23.15)
commensurate with that utility, right? So what, you know, I talk about in the book, like this is 95 % rule. It's just a rough rule of thumb, but like, you know, for freemium product scales, 95 % of the users will never pay. Right. And so what you, what that's fine. and maybe that's a good thing because maybe that's a feature, not a bug because those 95 % a could serve as like, you know, social proof, right? You just got a lot of users. Hey, this must be a great product. They could be like viral,

Rabah (35:25.304)
Yes.

Rabah (35:52.536)
Word of mouth, distro, yep.

Eric (35:53.134)
touch points, you know, provide word of mouth. They could be cannon fodder like in a game. so, or, you know, if you think about like an app, like a Tinder or something, they could just be people to match with. Right. So like the there's value in having a lot of people, even if many of those don't pay. And then what you have to do with the 95 % is you have to, you have to develop an economy within this product such that, a very long tailed LTV distribution is produced. There are people that can spend a lot of money, they get a lot of value out of the product and they match the value that they get with how much they spend. Right.

Rabah (36:03.288)
Yes.

Rabah (36:18.584)
Yes.

Rabah (36:22.424)
Yes.

Eric (36:22.862)
cause if you don't do that, what you want to do is you want to sort of minimize consumer surplus with freemium. You want to make sure that there's very little gap between what a person would be willing to pay and what they end up paying. You want to make sure that those almost map one to one and everybody wins, right? Because they're getting a lot of value and they're, they're paying what they think the thing is worth. you're not, you're just not leaving money in the table. That's the goal of freemium. And so like, if you apply that to like the metaverse where, you know, it's not just a tool that helps me with,

Rabah (36:32.152)
Yes.

Rabah (36:35.576)
Yes.

Eric (36:51.566)
some problem or serves as entertainment that is otherwise a deviation from my day to day, but it ultimately is my world that I inhabit. then you just, you just, you just find, pain points or, you know, when you find problems to solve with the digital products that are much more valuable, right? Like if, like, if you're thinking about,

Rabah (36:59.992)
Yes.

Rabah (37:14.392)
Yes.

Eric (37:17.966)
You know, I can't, I can't make houses on a freemium, but I can't be a freemium based, based construction company. Cause I, there's a lot of costs incurred with me building a house for you, but in the metaverse I can. And if that just becomes a much more important part of, you know, our existence as humans is like this space that we inhabit in digital worlds. Well, then those things can be monetized with the freemium model, right? You could be a freemium, monetized construction company. You could be a freemium monetized law firm.

Rabah (37:29.432)
Yeah.

Rabah (37:39.672)
Yes.

Rabah (37:44.696)
Yeah.

Eric (37:47.342)
right? Or whatever. And so it's shipbuilder, right? But you can't do that in the real world because there are extreme costs associated with doing those things. Maybe less so on the lawyer side, but I still don't think it would be a good idea. But you know, if the digital world becomes much more of a simulacrum of the real world in the in especially in terms of like how we spend our time and where we spend our time then then yeah, you could you could all of those endeavors could be monetized with the freemium model.

Rabah (37:47.512)
Yes.

Rabah (37:58.36)
Zing!

Rabah (38:18.04)
No, I love that, man. So well said. You were pretty early on talking about how the shift towards probabilistic attribution, attribution is such a quagmire in terms of mobile marketing. Can you kind of explain to people out there what probabilistic attribution is and why you think it's becoming more and more important?

Eric (38:36.686)
Yeah, well, probabilistic attribution is really just a method for assessing the performance of advertising campaigns using data sets that can't be directly linked. Right. So like with deterministic, so the, you know, the kind of, I don't know if I would call it like the, the diametric opposite, but the, an alternative approach to attribution, to probabilistic attribution would be deterministic measurement. Right. So I've got,

Rabah (39:04.568)
Yes.

Eric (39:05.198)
A user ID that is persistent and it is consistent across context. And I can just match what a person does in context A with what they do in context B. That often involves me sharing data with some other company, right? So context A is my app and context B is somebody else's app in order for us to, to connect those two data sets that, you know, that were emitted or artifacts of this user's activities. Then I need to share the data, pool the data somehow with, with, with a separate company.

Rabah (39:15.704)
Yes.

Rabah (39:30.648)
Yes.

Eric (39:32.43)
Well, probably, probably listen. And so that company might be another app, but it could be an advertising platform, right? Well, with ATT, you know, you deprive the ecosystem of the IDFA for the most part. and so what you end up with is I've got a person, I've got a user in my product, that has some paper trail, some, some, bank of, of activity that I have total oversight over. I know who they are because they have like a user ID in my data warehouse.

Rabah (39:36.92)
Yes.

Eric (40:01.198)
And I know what they've done because I've just been able to instrument everything they've done in my, my product, but I have no ability to match that with some other service, for that same user, because I no longer have a unique identifier that, that, that to map it with that is available to both me and that other person. Right. And so it's probabilistic you'd say, okay, well, and like from the advertiser's perspective and say, okay, well, I can no longer just send events back to Facebook that are tagged with the FB ID or tagged with the idea of Faye because.

Rabah (40:01.304)
Yes.

Rabah (40:19.576)
Yes.

Eric (40:30.798)
I don't have the IDFA and the FBI ID is, is excised in a lot of ways from my, my visibility with, you know, first party cookies. And so what I'll do is I'll just say, okay, I don't know what that person did. There's no way for me to definitively know what this person clicked on on that other property to get to my, to get to my property, to get to my app or my website and then make a purchase. Right. I can't know that it's impossible to know that now.

But what I'll do is I'll just look at my levels of ad spend across all channels and my levels of revenue. And I'll try to sort of map those sets of data, those disparate sets of data at a higher level of aggregation together to decide, okay, well, this is when I spent a dollar on this platform, it made a dollar 50. Now I don't know what that, how that dollar 50 was constructed from the, from, from the bottom up, right? Like that this user made this purchase and this user made this purchase, but I have a sort of problem. And so then I just use probabilistic methods to do that sort of.

Rabah (41:05.848)
Yes.

Eric (41:29.39)
attribution, right? That's one approach. Another approach would be, when I spent a dollar, it had a $1 .10 of commensurate revenue that I sort of like probabilistically can attribute to it, right? So that would be more like an incrementality approach versus like a media mixed model, which is what I was just talking about before. So really, just you're looking at causal impact.

Rabah (41:43.896)
revenue.

Rabah (41:53.976)
Yes.

Eric (41:58.03)
like what's the impact when I spend this dollar, what's the impact, right? And so that's just uses a different set of data. It uses like very high level aggregated data. It doesn't use any sort of user level data. and it looks for like statistical,

Rabah (42:05.976)
Yes.

Eric (42:13.902)
statistically robust links between ad spend and revenue or something or some outcome that you're pursuing, right? So that's, it's just using statistical methods to assess that versus what you believe to be deterministic identity.

Rabah (42:37.432)
Yeah, so basically the too long didn't read. It's ultimately looking for really high correlations and then trying to lean into those correlations from the past data that you've been tracking.

Eric (42:49.454)
Yeah, you can say that. Yeah, sure.

Rabah (42:52.184)
Man, that's so awesome. Okay, a couple more questions. Are you bullish on Snapchat? Do you think Snapchat has any chance of taking any advertising dollars away from meta or TikTok?

Eric (43:05.358)
So I don't think they have to take dollars away. I think if they improve the platform, they'll, you know, the cool thing about digital advertising is like, it's not a zero sum market, right? I mean, if I'm providing value, incremental value to an advertiser, they don't, they're going to want to invest as much as they can in my platform until that's not true anymore. Right. And that doesn't come at the expense of another platform. Like this could be net new opportunity. And so what you end up doing usually with, as a, as a, as a digital advertiser, it's like, if I'm an Ecom, I'm looking at like AOV and I'm thinking about,

you know, what like the singular purchase experience is worth to me. And I'm just in and I'm buying that and I'm like recouping it very quickly. But I could I might also be looking at like a lifetime value. So I'm buying a user, they're going to make an initial purchase, but then they're going to come back and make another purchase in two months or in three months. And so I'm looking at that sort of like predicted lifetime value. And that's how I'm bidding right now with like a lot of app developers. That's the only way they really think about this is like, what's the predicted lifetime value over a year or two years, because it's gonna be a much more sticky experience. Ideally, you habituate the user to your product.

And so they're in there every day. And so you're front -loading the revenue, right? So you're front -loading the expense and you're recouping the revenue over time. And so, and a lot of, you know, and the amount of the timeline over which you're comfortable doing that just depends on your financial situation. Like if you have a lot of cash in the bank, then you're okay to pony up for like a year's worth of revenues and then recoup those over...

Rabah (44:12.12)
Yes.

Rabah (44:15.48)
Yes.

Rabah (44:24.344)
Right.

Eric (44:31.278)
you know, a year and then make your profit in the in like the next year after that or whatever. Right. And so like, if I see that a new channel is unlocking better return on ad spend, I'm just gonna push more money into that channel that that money doesn't necessarily have to come from some other channel, right. And usually it doesn't, I want to sort of invest as much as I possibly can in every channel, because that's just top line revenue growth. And the thing is, you know, there are a couple of companies are obviously cash constrained, there's realities that you have to confront with this.

Rabah (44:34.392)
second year. Yep.

Rabah (44:54.104)
Yes.

Eric (45:00.27)
But there's also a lot of facilities that you can use to borrow against your expected LTV. Right. And so, you know, you're going to be cash constrained to some degree. Right. But there are ways that you can avoid having to go off balance sheet for the expense of buying like a year's worth of revenue or usually buying like

Rabah (45:05.816)
poster.

Eric (45:25.966)
10 months of revenue that or buying 12 months of revenue that where the advertising spend gets recouped in nine months or something. And so, yeah, so for that reason, I mean, there's, there's, you know, it's not zero sum, it's Snap, if it gets better with its DR product, they certainly have users that Facebook doesn't, right? It's specifically Snap. And so that just unlocks more growth for all the advertisers. And I mean, it's a great outcome. And I actually wrote a piece published today called

Rabah (45:32.192)
Right, where your payback period.

Rabah (45:44.568)
That's fair. That's a really good point. That's a very good point.

Eric (45:55.054)
the rising tide in digital advertising. And I was talking about, I think a big contributor to the quarters that snap and Pinterest had, and actually their revenue growth over the last couple of quarters is Facebook getting better because when ATT happened, everyone, there was a flight to quality. Everyone was in trenchment mode. They were just, they just needed to get Facebook to work. they were focused on, you know, Facebook performance deteriorated, right? For a lot of companies significantly, you know, dramatically.

Rabah (46:06.456)
Yes? That's a really good way to put it.

Rabah (46:20.696)
Yes.

Eric (46:22.382)
And, but that was the critical channel that they had to fix. They had to sort of make that work. And so as Facebook has improved its platform and delivered more performance for advertisers, they're getting more comfortable experimenting and shifting budget to other places, right? Because they, A, they're just making more money. They can grow the team. They have the headcount to do that. They have the ability to build the infrastructure to do that. But B, there's just more of a level of comfort, that, you know, the advertising ecosystem hasn't just been laid waste to, right? And so, you know, my sense is that the Facebook tide.

lifted all boats and they allowed other advertisers to have the luxury of experimenting and spending more money and diversifying away from just Facebook, especially in Ecom. Ecom was devastated by ATT. I mean, mobile gaming was too, but Ecom to a much greater degree. I mean, there's just a graveyard of Ecom company. I mean, you look at like Blue Apron, what was the shoe company? Is it Adams? Allbirds.

Rabah (47:02.936)
devastated.

Rabah (47:07.576)
Devastated.

Rabah (47:12.632)
Yes. Adams, Albers is still not doing well. You're exactly right.

Eric (47:19.054)
these comp and then you could you could trace that to ATT. I mean, it's just there was you know, and obviously these companies were not like very healthy. But and look, I mean, if you build a company that is entirely dependent on Facebook, well, okay, you kind of did this to yourself. But you know, the Ecom space was just devastated by ATT.

Rabah (47:22.488)
Yes.

Rabah (47:30.776)
Yes. Yes.

Rabah (47:38.712)
It was it was absolutely a cambering explosion of apocalyptic death. I mean it was it was super challenging and the That is such an interesting thesis though because my pushback was gonna be Yes, but bandwidth right like if I only have enough bandwidth to do one or two channels I'm gonna do meta Google, but that's so right that like okay. My Facebook is crushing now Google's kind of set it and forget it

So now where do I open up? I gotta read this essay. That's really interesting. I love that thesis.

Eric (48:12.494)
Was it's a totally agree and I'm actually a huge advocate for channel specialization and focus, right? I think it's actually really bad when people are like, so there's two approaches, right? And then we're coming up on time, but there's two approaches and I wrote an article about this, like called, I've written two kind of on this topic. One is called opportunity cost and user acquisition, I think. And the other one's called building a channel.

composition strategy or traffic composition strategy. And there's two approaches basically. One is waterfall. I maximize, I max out my best channels and only when I do that, you know, cause usually ROAS and budget are inversely correlated. So the more budget I spend, then my ROAS decreases, right? But I've got some sort of target. Like let's say it's 110 on 90 days, right? So I hit 110 on a 90 days. I can't spend a single dollar more because that would, that would, then my, my ROAS, it would break my economics. My ROAS would descend below my target.

Rabah (48:51.064)
Yes. Yes.

Rabah (49:01.464)
break my economics.

Eric (49:05.486)
So then I moved to the next channel and then I maximize that. Right. And so what you're doing there is you're minimizing operational complexity. but you're also, you're also minimizing ROAS, right? So, cause the other opportunity would be I just diversify and I spread my money around every single channel. I'm getting the max possible ROAS on every single channel. Right. But that, that increases operational complexity. I need, I need to, I need someone who knows how to run snap and I need creative team that knows how to build creative for snap, tick tock, whatever.

Rabah (49:12.504)
That's.

Rabah (49:17.464)
Yes.

Rabah (49:30.488)
Yes. Creative eggs.

Bingo.

Eric (49:35.278)
And so my sense is for most advertisers, the trade -off is not really worth it. You'd rather just concentrate and specialize and become an inculcate a genuine expertise in Facebook. Cause that honestly will give you more ROAS. Right. And if so, and the thing is like, well, then if you're taking the waterfall approach, which everyone did post ATT, then they concentrate everything on Facebook. They just try to get Facebook to work. They became channel experts. The channel got better and now they've hit their limit. They're maxed out. You know,

Rabah (49:56.12)
Yes.

Eric (50:05.294)
they'll be able to grow that steadily over time, but it's not, you know, it's not meaningful growth like quarter of a quarter. So now we'll onboard Snap again. Now we're onboard Pinterest again, right? And so I think that's what's happening.

Rabah (50:12.632)
Yes. Yes. This is such a good thesis. That's such a brilliant thesis. I love it. Okay. Two more questions, because I know we're up on time. So the first one is real quick, you're so inculcated in marketing, the space, everything. What do you think most CMOs do wrong?

Eric (50:30.334)
don't get me started. I mean, most CMOs, I think the CMO role is mostly like antiquated, right? Like I think it's an anachronism. It's an anachronism from like many decades ago. I just don't think it's fit for purpose, especially with digital companies, companies that make digital products. I mean, look, saying that there are great CMOs, but I think the way that the CMO role is conceptualized for most companies is ineffective.

Rabah (50:31.448)
Okay.

Rabah (50:39.576)
I agree with you.

Rabah (50:44.728)
Yes.

Eric (50:59.31)
Right? A CMO essentially should own the outward facing growth initiatives for the company. Right? And a product manager should own the inward facing growth initiatives for the company. Because growth is the ultimate goal. Right? I mean, and so, you know, a lot of CMOs, they don't actually think that way. They think that they should own like brand vision or

Rabah (51:10.232)
Yes.

Yes.

Rabah (51:22.904)
Yes.

Eric (51:25.198)
like the way that I, you know, the thoughts that are conjured up when someone recognizes your brand, but I don't think so. I think they should want to get people to buy your product. Right. And like I keep my marketing shelf is really interesting because it's a book called modeling, extremal events, which is, you know, it's a, it's a, it's a book about like insh like pricing insurance premiums. But what's, what's,

Rabah (51:50.232)
That sounds like a great one to take to a party.

Eric (51:52.942)
But what's an extreme event? What's an extreme event? A purchase. A purchase. When you're running, when you're running direct response campaigns, a purchase is an extreme event. That's been very helpful in helping me like kind of clarify some of the ways I think about digital advertising. I've got putting auction theory to work. I've got auction theory. I've got numerical recipes in C. I've got data science for marketing analytics. I've got the universal economics and I've got more at home, but these are my like sort of everyday, you know, I'll pick that up and reference it a lot.

Rabah (51:56.728)
Yes? Yes?

Rabah (52:22.232)
Yeah. of course. Come on. Use the G.

Eric (52:22.254)
And then I've got Ogilvy on advertising. And that book was written, I think in the, in the eighties. And he says something really like what, what, you know, what, what, brings me back to that book and what makes me think it's like actually, you know, relevant in today's time is he says something is like, my goal as a marketer was never to win like an award for the best creative. It was never to, you know, to, to, to get, accolades from people, you know, for the, you know, the, the billboard or the magazine.

I mean, I'm, I'm bastardizing this horribly, but my goal was always to make a sale. Right. And like, and I think the CMO has missed that. They think their job is, to fly around the world and, and go to Cannes and to just like be a quote unquote, be uncrit be creative. And, you know, my sense is that your job should be to drive sales and drive revenue growth and that. And so how do you do that? Well, for digital products and for actually most products, it's performance advertising. Now.

Rabah (52:57.176)
Yes.

Eric (53:20.59)
Performance advertising encompassing brand ads, right? So my conceptualization of performance advertising is not that it sits opposite to brand advertising. Those are not two separate things. Performance advertising is a model driven framework for spending money on ads and bringing more than that back. And brand advertising sits within that. Direct response sits within that. Delayed responses within that. I'm not saying all your budgets should go to Facebook for like a 30 day payback period. What I'm saying is you should have a model.

Rabah (53:23.096)
Yes, of course.

Rabah (53:40.376)
Yes.

Eric (53:48.91)
that with like a high level of reliability tells you how much money you're going to get back when you put a dollar in. Right. And maybe some of the effect of that dollar going in is like better brand awareness that leads to sales later on or leads to higher AOV when the people do ultimately purchase. But I want the model to tell me that. And it's not just gut level intuition that, Hey, I think this advertising will make people much more aware of our brand and more likely to buy it than our competitor. No, I want a model to tell me.

Rabah (53:53.72)
Yes.

Rabah (54:01.176)
Yes.

Rabah (54:07.288)
Yes.

Rabah (54:17.848)
And this is coming from the CMO myself two times. I think that's exactly right. That is spot on. I honestly think the better pairing, depending on the org structure, but having like a chief brand officer and a chief revenue officer, I think is a better split than having a CMO. Cause you sometimes get people on both sides of the spectrum versus finding that really unique person that has the analytics, but also the creative and like, you know what I mean? It's like a really hard.

It's almost like the president, right? Where you think of like the US president, such a hard job, because you have to be super charismatic, but you also want this legal wonk that can understand policies and all this stuff. And it's very rare to find that person in one, but that's a brilliant articulation. I love that. Okay, last question, because I know we're up against it. What's your, I end every podcast with this question. What's your function for excellence? What do you think excellence is a function of?

Eric (55:13.326)
Good question. Yeah, I started thinking a lot about that lately, right? Like, how do I, you know, how do I approach what I do in like a systematic way that is auditable, right? And to help me understand if I'm doing, if I'm getting better over time. I think it's, for me, I guess it probably is very context dependent and probably very personal, but...

For me, it's like achieving a good balance of, you know, first of all, doing things like I'm very bad at doing things I don't want to do. And so, and I just, I just end up performing poorly when, when that happens. So it is being able to, and you know, this isn't, I've got the privilege of being able to do that now, but I didn't always in my career, just picking projects that I want to do. So that first of all, it's that it's being really good at betting and saying no to stuff, even if it pays well that like, I know I'm not going to enjoy it. I'm not going to do a good job.

Rabah (55:48.824)
Yeah.

Rabah (56:07.768)
Yes.

Eric (56:08.942)
and the other is, is so it's just, it's, well, actually I would say it's not achieving a balance in that. My achieving excellence is actually disproportionately doing stuff that I want to do. and not trying to achieve any balance in that, but then, you know, achieving balance across like work, personal life and enjoying my family and, you know, taking a lot of time to build a relationship with them and enjoy the time that I have together. Cause that motivates me to work harder, right. To provide for them and to, to create those opportunities.

Rabah (56:31.223)
I love that.

Rabah (56:35.224)
this.

Eric (56:39.086)
Not that you can't do it without money, but it makes it easier, I think.

Rabah (56:42.68)
Yeah, it does. It needs more energy. No, that's a beautiful answer, man. I'm going to re -listen to this pod 10 times. There's so many things on here. For a couple of things, VO is value optimization for people following along at home that didn't know that. And GFC is the great financial crisis. But Eric, this has been incredible. Where can people find you? Where can they follow you? Are you taking consulting clients? How can they get on your, how can they get you on their cap table? This time is yours, my friend.

Eric (57:09.87)
Yeah, so, Heracles capital it's HR CLS dot VC. just hit me up on Twitter, LinkedIn. you know, thank, appreciate that. Yes. I'm just Eric Sufer at Twitter. you know, send me an, send me, send me a Twitter DM, email me at Erica, Eric at HR CLS dot VC and then mobiledevmemo .com. You know, there's a subscription tier, eight 99 a month. I've got enterprise packages if you want to get multiple seats. so you get access to the content feed.

Rabah (57:17.624)
Great feed, great feed on Twitter. I love your Twitter.

Rabah (57:32.056)
the best.

Eric (57:39.055)
And you get access to dark pool, which is my app app store app economy market intelligence tool tracks, you know, app store rankings, but I've also got a search tool that allows you to look at and very easily compare rankings across platforms, across categories, multiple apps, across countries. Yeah. It's a mobile dev memo .com backslash dark pool. I've got a dashboard product coming up really soon. That'll show you like the biggest jumps and declines, on a daily seven day, three day basis. So.

Rabah (57:55.576)
Whoa, I didn't know this. that's awesome.

Rabah (58:06.328)
wow.

Eric (58:08.654)
yeah, dark, it gets you. So the subscription gets you access to dark pool. It gets you access to the content.

Rabah (58:14.52)
Amazing. Eric, this was everything and more than I expected. They say never meet your heroes, but you absolutely came through. Man, I really appreciate you taking the time out of your day and all the eloquent, thoughtful responses. I really appreciate you, man. And yes, when I'm back in Austin, we'll get some dinners together. We'll get Rashib on the horn with me, you, and him, and we'll do an in -person one as well.

Eric (58:34.766)
Awesome, thanks for by. Appreciate you having me. Take care.

Rabah (58:36.216)
Thanks, man. Appreciate it. Thanks so much. All right, folks, that's all we got. Make sure to go to firm .commerce .com, book a demo so my boss loves me and I can have some more probabilistic attribution so I can do all the cool brand stuff Eric is talking about. And then make sure to subscribe to our newsletter, Geometry of Growth. You can go subscribe right at firm .commerce .com and then go get you your mobile dev memo subscription. It is absolutely worth it. All right, folks, that's another one in the books. We'll see you on the flip. Bye -bye.