0:00:05 - (Matt Widdoes): Welcome to Growth@Scale. I'm your host, Matt Widdoes. This is a podcast for leaders who want to bring sustainable, predictable, scalable growth to their businesses. Every episode, I sit down with world class growth experts across product marketing, finance, operations, you name it. The hope is that these conversations will give you real, actionable advice for building and sustaining company growth.

0:00:33 - (Matt Widdoes): Welcome to another episode of Growth@Scale. Today we're joined by Jason Parmar, a monetization expert. He's got over a decade of experience largely focused on ad monetization and revenue, and he's held roles at Microsoft, King, Activision Blizzard. Welcome, Jason.

0:00:49 - (Jason Parmar): Thanks for having me, Matt.

0:00:50 - (Matt Widdoes): I'm so excited to have. You know, so, I think this is a topic that many people are curious about and have limited experience in, and it's a huge piece of growth. But before we dive into ad monetization know, for people who don't know, tell us, who are you? Where have you been? What do you do?

0:01:07 - (Jason Parmar): Sure. Well, I've worked across London and Berlin over the years and almost exclusively focused on revenue. Started out at Microsoft learning how a big corporation works before moving to digital advertising, where I also went to AOL, Yahoo and Oath and Verizon Media, which was part of mashing together all the old media brands together, but really got my teeth cut in launching a new business from scratch on ad monetization at King.

0:01:35 - (Jason Parmar): They'd been acquired by Activision Blizzard for about $6 billion and wanted to start a new revenue stream in addition to just in-app purchases. So, learned everything from scratch there to scale that business, actually, and in a lot of different roles within ad monetization. And probably, I should say, I'm a little bit more nerdy than your average ad-mon on sales lead in that I do have a degree in chemistry, but I think it's pretty transferable because it's quite an analytical degree and I hope that will make sense.

0:02:14 - (Matt Widdoes): Yes, I mean, it's really similar to user acquisition. I think you need to have a high EQ as well as IQ, and it's data and analytics-driven, but it's not in and of itself data science per se. But yeah, there's a lot of other things like softer skills that get mixed into that. And I think even on sales, you could say, when done correctly, should be super data-driven and requires a decent amount of, if nothing else, self reflection, but certainly some analytics.

0:02:41 - (Matt Widdoes): And so, for people who don't know what ad monetization is, can you just walk us through, high level, what is it? And then we can kind of dive into some of the details of what makes it great.

0:02:52 - (Jason Parmar): Sure. In the simplest way, ad monetization is essentially making money in your application or website or product by running advertising. I'd say the big part that's different to what it sounds like is that in today's world, that is largely, almost completely automated, and we call that 'programmatic'. So there's lots of advertising technology involved and not perhaps the old ways where you would sign paper contracts. So when we talk about ad monetization, we're really talking about the most efficient way to make revenue for your product using advertising.

0:03:30 - (Matt Widdoes): Yeah. And so for people who, to piece that together, you think of yourself in a mobile game, you have lost a life. You can click a button to watch a 30 second ad and get another life. That's ad monetization at its purest form. And I think a lot of companies, I think the other way that particularly in mobile companies, make money is through what is called IAP, or In-app purchases. That's buying lives with money.

0:04:00 - (Matt Widdoes): And the vast majority of mobile users don't end up converting on in-app purchases. So they either just wait until the timer's run out in a day and they've got more lives, or they can accelerate that with a dollar or $100 or watching a video or 100 videos. And so it's a key piece of growth that is often overlooked. It can make up a considerable amount of the revenue. For apps like Words With Friends, it's 90 plus percentage of the revenue, versus other apps where maybe it's more marginal. What have you seen in that, kind of broadly speaking, in what percentages ad monetization can represent? Because I think for anybody who's currently got an app that's just selling in app purchases and hasn't introduced ads for any number of reasons, they're leaving a lot of money on the table, and they're also leaving some engagement on the table because there's some people who, if they don't have that option to watch a video, they're just going to disappear. And maybe that's the last time you see them. So can you walk us through, through kind of the range that you've seen or that other companies you're aware of have seen and what that can mean to a company's growth at the end of the day?

0:05:09 - (Jason Parmar): Absolutely. And it's a great point and great question. Traditionally, a lot of applications, mostly games, but also, other applications focus on one thing. They focus on in-app purchase subscriptions, paying users basically, or purely just advertising, especially with hyper-casual games, traditionally, and what we've definitely noticed is a big increase in what we call 'hybridization'. A great example of this, actually, as you kind of mentioned, would probably be King, actually. Traditionally, since the launch of the iPhone, predominantly, almost entirely in-app purchase-based, meaning I run out of lives and I just really want to complete the level, I'm going to pay. But the reality is much over 90% of users will never make a payment.

0:05:56 - (Jason Parmar): Now, those users have value because they help grow the network, but actually that's a lot of potential revenue you might be leaving on the table. And as you said, there's actually some really interesting examples we might be able to go into later, which show that giving those users the choice of watching an ad might actually increase engagement and not only benefit your revenue, but also grow your user base. So the benchmarks have been increasing in terms of, in advertising as a percentage of revenue for companies.

0:06:24 - (Jason Parmar): Traditionally, it has been where companies do both, around 10%, but that is actually increasing. And we've definitely been noticing more and more companies who only do in-app purchase will do more in-app advertising. And it complements the offering because it provides the optionality to a lot of consumers who might switch off otherwise. And you can keep them in the game and actually even drive in-app purchases through in-app advertising, which seems very counterintuitive, but I've seen happen.

0:06:56 - (Matt Widdoes): Walk us through that. I'm curious to hear more about that. Using leveraging ads in your own game to drive IAP.

0:07:03 - (Jason Parmar): Yeah, I'll give a practical example, and this is kind of an interesting one. One of the most profitable placements in Candy Crush is what's called the end-game placement, which is where you run out of moves at the end of the level and you get the option. You might be really close to completing it. You might have been stuck in it for weeks, let's say. It's the highest-converting payment spot placement in the app. And one thing we tested with advertising was, can we put an ad there? A rewarded video ad, which says, we'll give you some moves, we'll randomize it, if you watch an advert in a fair-value exchange, which, when users get told that they opt in, they very much like these types of ads. 'Rewarded video' it's called.

0:07:46 - (Jason Parmar): And the thought would be, why would we give up the most profitable part of the business? We want people to pay for this. And it seems quite counterintuitive. But when we ran the numbers, it was pretty fascinating. What had happened is most people had never tried because they'd never paid in the app before. So by offering the placement with an ad, people had tried it for the first time. And the key was we limited it. We only let them do about once a day or once a week, depending on the user pool.

0:08:12 - (Jason Parmar): And suddenly people were like, they really like the ability to have that option. And so later in that same day, they were more willing to pay for it. And actually, what you've got there is an increase in engagement, an increase in revenue from ads, and an increased revenue from payments.

0:08:30 - (Matt Widdoes): I was just going to say… And an increase in the customers delight and joy.

0:08:34 - (Jason Parmar): Absolutely.

0:08:35 - (Matt Widdoes): Yeah.

0:08:36 - (Jason Parmar): So that's a really good example of it being used in action in a way that seems counterintuitive that most people would never even consider.

0:08:44 - (Matt Widdoes): Yeah, that makes total sense. And I'm glad you brought up King. So that's where you and I met. I was leading user acquisition, you were leading the ad monetization work and really building that from the ground up. And nobody likes ads, right? By and large, particularly with the culture at King and very player-first focus. I know there were a lot of- 'enemies' isn't the right word, but people who were staunchly against it, who were saying, what are we doing? We're selling out. And I think one piece that really resonated with me back then when those decisions were being made, is that people are looking for this.

0:09:23 - (Matt Widdoes): One, our competitors have it. But two, there is a huge subset, maybe the majority of users, that want that outlet where they're like, well, how am I supposed to get free hearts in your game? You don't even let me watch a video, right? So they kind of expect it. And so a lot of the conversation back then was around, well, how do we do it in a way that's not intrusive so that we can appease the internal people as well as our player base, who maybe do not want to see the ads. Right. They're just like, what are you doing?

0:09:52 - (Matt Widdoes): But how do you introduce it in a way where people can still find it? And let's walk through some of those implications and that thought process. And how can people, I think oftentimes people tend to think it's very black or white. You've got to have it one way or the other, and you can bury it and have a link somewhere that says, 'Watch an ad', and that's where that lives. You can segment it so that we're not showing it to payers at all and they never even see it because they're known payers and they've been paying for many months or years.

0:10:21 - (Matt Widdoes): But what are some of those considerations on how to introduce ad monetization in a way that it is additive to the user experience, additive to the revenue, and not something that all of a sudden feels like we've changed the entire landscape and experience of the game?

0:10:38 - (Jason Parmar): Yeah, I think firstly, it's most important to set expectations in testing and say we shouldn't be against testing anything, especially on a small user base and knowing where your boundaries are. So if I was to use the example of King, it's probably the hardest one because the context there being the CEO had got on stage and said a few years earlier, we'll be the only major games company to never offer ads in game.

0:11:02 - (Jason Parmar): Fortunately, we found some examples of users actually asking for it in reviews, which helped us. But you're not going to find it harder than that in my opinion, or where I found. So one of the main things to do is actually to find out those boundaries with the team. So what are their expectations and what are their major concerns and why? Critically, it's almost always why? Because they'll say almost always it's detrimental to the user experience and almost always show it not with data but with anecdotes. And even with data they'll often find some bad example or something.

0:11:34 - (Jason Parmar): So I've always approached it as well, let us test it then, and have pretty restrictive conditions on your testing. So small user base, non-payers and also making the point on not all users are specific to one market. So often, in that case, the examples we took were, what are some countries that you don't care about that we can test this on that you don't care about?

0:12:01 - (Matt Widdoes): From an In-app purchase perspective.

0:12:03 - (Jason Parmar): Yes, exactly.

0:12:04 - (Matt Widdoes): Right. Where we're not going to be cannibalizing something you thought would happen otherwise, but instead where it's just like, okay, we've already written them off, from a revenue standpoint as far as purchases go, spending $5, spending $10.

0:12:17 - (Jason Parmar): Exactly. A good example of this probably is if you look at the statistics on India as a country versus the US, I think it might be close to 10-to-1 users might pay in the US, whereas it's probably 150-to-1 in India.

0:12:31 - (Matt Widdoes): Or higher.

0:12:32 - (Jason Parmar): Or even higher. Yeah. And so, hey, can we test that market because we don't really acquire profitably in that country because we don't run any advertising there. Okay. You get that permission, you go through, you find your boundary conditions, you set out, okay, have you got any strict rules on the types of advertising? you kind of go through those hurdles to do some basic appeasement and basically then it's just iteratively testing. So lots of different formats. Rewarded video, by far, if they care about the user experience one.

0:13:01 - (Jason Parmar): Is one I would focus on because statistically, rewarded video is the most liked form of advertising. If we were going to be more negative about it. It's basically the least disliked, essentially the most enjoyed, or the least disliked.

0:13:17 - (Matt Widdoes): It's the most tolerated.

0:13:18 - (Jason Parmar): Exactly. And in that case, because you're being pretty transparent, you're giving people something that they wouldn't otherwise maybe pay for. And what you might find is some interesting things. So one example we had, I'm bringing King up again, but it's just a really interesting one, is it resulted in the first time ever that King were able to acquire users profitably because they suddenly could turn a profit on the advertising revenue they could make from users who previously they would never acquire because their only revenue stream was in that purchase. So the whole system was geared towards making revenue for those users. But there's a really nice halo-like double effect here. One, revenue has gone up, but two, they grow the user base significantly and more profitably than just traditional UA, because they're now optimizing on two metrics, in-app purchase and in-app advertising. So those kind of stories win over the studios a lot because they're like, well, you've just pushed my user growth metric in addition to my revenue and average revenue-per-user metrics.

0:14:21 - (Jason Parmar): So knowing what they're incentivized by, I'm talking about game studios here, but we can also apply this to almost everything. And working very closely with your cross-functionals and just hearing them out. And I address this as, I know I'm the evil revenue guy, and I know you hate advertising because you've seen it done badly, but it doesn't mean we have to do it badly. Like give me your biggest concerns and I'll be transparent with you up front.

0:14:42 - (Jason Parmar): And if you let me test and have free will over this small cohort, I can show you something and I think that's the best way to go into it. And then you build trust with them and they give you a little bit more and a little bit more to play with.

0:14:55 - (Matt Widdoes): Well, I think in addition to the revenue growth and the user growth, you're also getting international expansion. And with a brand like Candy Crush, that's proven to be kind of this in the kind of cultural zeitgeist, you have this ability to potentially recreate that fervor that existed in 2013 or whenever we launched, into a whole new country that's maybe been less exposed generally. And when you think about ignoring or not even taking a shot on goal at a company the size of India, right?

0:15:31 - (Matt Widdoes): You're leaving so much on the table. And I think because of the nature, this may not be as true as it once was, but they're also oftentimes not using things like 5G. They're actually almost always connected to Wifi. They don't have the same kind of cellular plans that are available in other parts of, certainly in the West. And you can have this situation where basically everybody in the market, well, let's just zoom in on mobile gaming for this.

0:15:58 - (Matt Widdoes): They're all ignoring India because they don't monetize in traditional in-app purchase. And so that means from an auction dynamic on the user acquisition side is that they're very inexpensive to acquire because they're not very valuable on the other end of that. And so, the magic trick there is, if we can make $1.50 out of them through ads and we can acquire them for $0.40 at massive scale, why is that not good?

0:16:27 - (Matt Widdoes): That's the same as acquiring them for $4 and making $15. Right? So what's the difference? It also from a - I would imagine certainly from an ad seller's perspective, where we can now start to say to advertisers, particularly large brands like Coca-Cola or somebody that's international, oh yeah, we have share in India. We have placements that if you're trying to promote something in India, we have those users, which is super valuable. And I would imagine, and correct me if I'm wrong, but when somebody's doing some global Christmas campaign, or maybe Christmas isn't the best example, but when they're doing some massive campaign for the Olympics, let's say, that is kind of internationally recognized and it's Coca-Cola, you might be able to bundle in some of that Indian inventory into a bigger package that says, hey, look, we're going to give you this many total impression share.

0:17:14 - (Matt Widdoes): The blend is going to include 8% India, which you told us is important to you. And you might actually even be able to kind of better monetize that through packaging it with other stuff. Is that accurate? Is that kind of how it ends up playing out sometimes on the other end?

0:17:26 - (Jason Parmar): Yeah, precisely. I mean, you've got, with CPG's particular, any global brand specifically as well, who wants to push into the next big market. They love it. They're like, we can buy this campaign globally. And in terms of efficiencies for you as a publisher, it's significantly better, right? Because you're able to reduce the total number of your sales force because you'll be able to sell these global deals.

0:17:51 - (Jason Parmar): There is actually another small benefit that comes through is you're growing your network of users in total, which, if you're a public company, helps you with your monthly active users. In Activision Blizzard's examples, it actually makes quite a significant know, probably even a factor into Microsoft acquiring them as a company as well.

0:18:12 - (Matt Widdoes): And so, I'm curious, as you think through, like let's say I am the CEO of a massive, mobile, or maybe I'm a blend, maybe I do mobile, I do console, I do everything, like let's use Activision Blizzard as an example, because we lived Activision Blizzard King and we've never done ad monetization. We're huge public facing, care about growth, we care about expansion, we care about profitability, we care about all these things.

0:18:42 - (Matt Widdoes): And we've never done ad monetization in a meaningful way. Maybe we've toyed with it here or there, but it's not a significant focus. Maybe the team is only like 4 people. We haven't really, you know and the global team is 5,000 people or 15,000 people for that matter. What are considerations that need to be made? If you're building it from scratch, outside of testing into it and some other things, what advice would you give somebody who's at that kind of massive scale, who's trying to get it right and is either starting from a cold start where they haven't done anything whatsoever, or they've kind of never really given it that much attention and now they've seen the light and they're ready to do it, right, but they don't know where to start?

0:19:22 - (Jason Parmar): Yes, great question. I'd actually say, I mean, Disney and Netflix are kind of both in that area right now, actually bizarrely non-gaming companies. But in the context you say, let's say we're [Activision/Blizzard], your first thing is probably to…

0:19:38 - (Matt Widdoes): You could just as easily use Netflix or Disney for that matter. But yeah, maybe the game element is unique, I'm not sure, but let's talk about it more generally. And then maybe if there's a way to bifurcate that where we think, hey, and then if you're in mobile gaming, then there are these other considerations that you can make because of the nature of the game or the nature of kind of the user experience, which it's more bite-sized than something like Netflix or Disney.

0:20:04 - (Jason Parmar): Absolutely. I think the first thing is, know your context and where your USP is going to be, because that's what's going to appeal to advertisers and they're going to be split into two buckets. Realistically, you're going to have your brand buyers who are predominantly going to be sold to if you have an on-the-ground sales team, and your performance buyers who are going to be outcomes based and largely linked to install purchase campaigns. So knowing, firstly, where you're going to be very successful, and the second thing is like your quality bar and maybe your boundary conditions, for example, you might go, under no circumstance do I want game ads running for whatever reason, whatever reason that is, you need to know.

0:20:47 - (Matt Widdoes): You're saying, because I don't want to sell ads to my competitors.

0:20:51 - (Jason Parmar): Yeah, that's normally the first one from a gaming standpoint that comes up.

0:20:55 - (Matt Widdoes): Yeah, I don't want my number one competitor selling… We're basically selling away our users for pennies. And the juice isn't worth the squeeze when we consider the fact that we might be losing that person. Now, the counter to that, and this is coming from a UA person, so maybe this is why, because I want to buy those users, is that we're coming for them either way. And they're still using Facebook, they're still using platforms where I can advertise them.

0:21:22 - (Matt Widdoes): And if the distinction or the connectivity between that user and your game is so weak that they could be pulled away forever with an ad, it's just a matter of time anyways. I'm sure nobody on the games teams want to hear that, but what is your reaction to that?

0:21:38 - (Jason Parmar): I mean, I think you're right. I think those people who say that are using anecdotal evidence and not real data, because in the example, like you just said, those users are probably going to be pretty sticky. But also why you're going to make maybe 10, 50 times more revenue per user. With the users who see those ads.

0:21:57 - (Matt Widdoes): 10 or 50 times? That seems way higher.

0:22:02 - (Jason Parmar): Potentially, it could jump upwards of, I've seen CPM's jump upwards of like $200 for the most valuable users in that area.

0:22:09 - (Matt Widdoes): Oh, wow.

0:22:10 - (Jason Parmar): So, they can be really profitable for you. And that's why I always say, the testing element works, but when you're starting at the start, you basically have to pick your battles. Maybe a good example of how we ended up approaching it at King in the end was we kind of saw it as taps and we almost, within the first year, kind of had a 5 year plan with kind of taps that we could use. For example, one of them that was turned off at the start was game ads ever coming on. So that was a tap that we just left off and we knew it was there, and we kind of one day will touch that number of…

0:22:48 - (Matt Widdoes): And when you say tap, you mean like water taps? Is that the analogy, or is tap something else?

0:22:53 - (Jason Parmar): And I'll say those taps.

0:22:54 - (Matt Widdoes): Okay. I just want to make sure.

0:22:55 - (Jason Parmar): Sure.

0:22:55 - (Matt Widdoes): Okay, got it. I just want to make sure it wasn't something, just a phrase I wasn't familiar with. Okay, so you've got these different water faucets that you can turn on and off. So that's one. We'll never run ads, period. That's a big one.

0:23:09 - (Jason Parmar): Never running ads, period, is probably the giant one at the top. And then after that, there's like, 5 or 6 smaller ones. And they can be stuff like plate areas where you'll never show ads. So, for example, a lot of people will be fiercely against app-open ads, which is an ad that opens right as soon as you open your app. That's kind of a big one most people won't touch, but that's a potential one. Frequency of ads that are going to be shown, and lots and lots of more examples like that.

0:23:37 - (Jason Parmar): So lots and lots of these taps. So you want to find what you're allowed to do with your boundary condition.

0:23:43 - (Matt Widdoes): And a ton of segments. Right, like, we're never do payers, never do high value payers, never do people who are high retention. You could really split that up. And then I think you had kind of touched on it briefly, but we'll never sell into casual gaming, we'll never sell into match three gaming, or we'll never sell into X company, because we're neck and neck and we don't want to play nice. So, yes. Okay, so you're saying you really need to understand where those boundaries are.

0:24:12 - (Jason Parmar): Yeah. Knowing where your boundaries are means you can focus. So if you're a massive company, like, let's say, ab, in this example, you'll go, okay, what are your boundary conditions that we can test in? And that helps because you know what you can do to be successful. And once you know that, you can kind of move forward with those elements. So let's say it is direct sales only, which is somewhat restrictive, but you can start with that.

0:24:33 - (Jason Parmar): The benefit is, in 2024, as of now, you actually have a lot more control with kind of your technology layers you can pick to have control over your advertisers a lot more. So you would basically start that with what technology I'm going to kick off with. And the big trend that I think is almost kind of complete now is everyone will use something called a unified auction. When they pick what's called that ad server, the technology that enables you to show ads in your game, you'll pick one that is, in a simple way, a unified auction is essentially a very efficient way of trying to get the highest paying ad without latency issues in your game. So that's your kind of first big decision.

0:25:18 - (Matt Widdoes): So it's real-time inventory. Real-time auction dynamics.

0:25:21 - (Jason Parmar): Yes, real-time auction dynamics. And the key thing that's probably different to before, without going into too much detail, is before you had this concept called 'waterfall', which would mean I call Facebook, then I call Google, and I do it individually each time. Each part is adding a cup, 10 or 100 milliseconds.

0:25:39 - (Matt Widdoes): Milliseconds.

0:25:39 - (Jason Parmar): And they add up if you're doing 30 of them. And the new system, a unified auction, means we get all of their top bids and put it in one. And let's say it's ten times faster. That's quite important because it means that your users in, say, countries without extremely high-speed internet don't have any speed issues. And that is quite a critical one, because the user experience does degradate if your user experience and ads is slow or sluggish and it feels like it's slowing down.

0:26:08 - (Matt Widdoes): Well, and to your previous point, that's oftentimes where these are being consumed. Right? So if our premise is that we can make a lot of money out of ad monetization in India and a waterfall is going to make that experience really poor, then that's going to be a massive impact to the test and massive impact to the overall. So, yeah, those milliseconds matter. Ask Amazon. Milliseconds matter on any sort of purchase decision, for sure.

0:26:35 - (Jason Parmar): Yeah, I have an example, actually, of this on latency. I worked for a very large football app company called OneFootball out of Germany, and they had a huge amount of users in LatAm, football-crazy countries like Brazil, Argentina, Mexico. And we started to say, oh, we could really monetize in LatAm. That's our growth area. Okay, cool, let's go after it.

0:27:03 - (Matt Widdoes): As many others before have had thought as well.

0:27:06 - (Jason Parmar): Yeah, they've got their own issues over there with kind of how tax works for corporations and there's a lot of factors to consider there. However, one of the things that actually really hurt us or restricted it was, we had optimized the business so much towards Germany, UK, France, that the latency issue that we'd let build up in the app just meant, I think it was something like only 1-in-5 ads was even showing because we just built up all of these technical inefficiencies over years.

0:27:38 - (Jason Parmar): So these unified auctions, platforms like picking your ad server is actually kind of more important than you would think. So that's kind of that first step you would do to have a highly efficient system there. And after that it's signing a lot of partner deals, basically, once you've got those boundary conditions and picking where you're going to show the ads, pick a medium traffic spot. Context matters, and from there it can really vary and we can go as deep or high as you want, but that's how you kind of are starting out.

0:28:08 - (Jason Parmar): Hired like 1 or 2 adults, people to kick that off, and someone who works with a product manager and can measure maybe LTV, and then you're at the races after that.

0:28:18 - (Matt Widdoes): Well, and I'm curious, two follow-up questions to that. So one, what is the inventory level that somebody might need to sell direct? Right. Because at some point you're like, I don't even have enough inventory to sell. Where do you kind of place that bar? And what's kind of the minimum? And then what's something where you're like, if you have this much, you're ready to go, that's plenty.

0:28:39 - (Jason Parmar): That's a really great question. I think it really does depend, but I think you need to be in probably over, at least over 10 million, I think probably. Direct sales… there is some context to this. You can get away with a lot less, but your experience you're selling has to be pretty premium. So there's probably one big question you need to ask yourself, which is, are we going to do bespoke what's called 'native' advertising? Which means we're going to build something very unique for an advertiser.

0:29:12 - (Jason Parmar): And we kind of would group that under what's called partnerships usually. Or, are we going to do standardized formats? So the IAB (Internet Advertising Bureau) have all these standardized ad formats. So the whole ads industry is pretty standardized. And if you're in that bucket, you almost always are going to go down this programmatic, automated route in which you mostly only really need operational people and not necessarily direct sales.

0:29:39 - (Jason Parmar): However, there is a benefit to having direct sales, which is you're probably going to make about 2-to-3X the CPm's you should roughly be able to sell at if you have a decent brand product. And that's what you can do with a ground direct sales team. If you have the partnership piece and you can do bespoke stuff, for example, you say, we will integrate your brand directly into our game. Fortnite are probably the best at this.

0:30:06 - (Jason Parmar): Or, there was a funny example with, I think Jumanji 2 in Candy Crush or The Emoji Movie back in the day, another funny example. But those can really benefit. I mean, you need direct sales for those and the margins can be extremely healthy. So I think the threshold is around that 10-million mark because you want… Under that, if you're doing standardized ad formats, you need something really special, I think, to stand out.

0:30:35 - (Jason Parmar): And really, if you're getting to the 100-million user mark, that's when you're really talking… You should be really looking at it.

0:30:42 - (Matt Widdoes): And on that number, are we talking MAU? Are we talking DAU?

0:30:47 - (Jason Parmar): About MAU? Yeah. Okay.

0:30:49 - (Matt Widdoes): For anyone who's not… Yeah, I was just going to say for anyone not in mobile, that's 'monthly active users' or 'daily active users' for that matter. Okay, so if you have 10 million monthly actives, then that's reason enough to go direct. And you probably want to be hiring people who've sold that type of inventory before, because so much of that is like Rolodex based, right, you're not calling into Coca-Cola to be like, hey, who wants to buy some ads?

0:31:13 - (Matt Widdoes): Although I would imagine your phone also rings. But if it's a new program, as is in this scenario, it's going to require some getting past that kind of cold start.

0:31:23 - (Jason Parmar): Yeah. To be clear, you could do it at a million or lower even. But I would say, like, if you're at that 10 million mark, you should definitely start to consider it.

0:31:33 - (Matt Widdoes): Okay. And then, just because the ROI is so much higher because you're taking it out of auction and into direct, if we all know that, why would Coca-Cola or somebody else pre-buy instead of go auction one? I guess it's probably to lock in their price. So that, because they're trying to make these deals 6 months in advance of something, know they're making the deal far out. And so they want to have some ceiling, essentially.

0:31:53 - (Matt Widdoes): And they don't want to be bid out. They have their own goals. And so they don't want to be in a situation where suddenly they thought they were going to run whatever campaign with targeting whatever metrics, even if that was just impressions. And all of a sudden they got iced out auction-wise because they didn't lock in a rate. Is that why you're saying, hey, it's kind of like a flight where it's like, if you buy it 4 months out… you're going to pay a ton if you buy it last second. Sometimes there's deals, but sometimes there's no seats left.

0:32:18 - (Matt Widdoes): Is that the premise?

0:32:19 - (Jason Parmar): That's an excellent analogy, yes, it's the guarantee, and it means their measurement, their attribution systems, where they're saying, we want to measure brand uplift. We can do that because we've got guaranteed inventory. We are running an offer. Yeah, we're launching a new brand. The time matters to us that this happens, therefore we don't want to miss out. And you can trade off brands against each other with that approach as well.

0:32:45 - (Matt Widdoes): Makes sense. And people are oftentimes, again, we'll use the holidays as an example. Everybody wants to be there. So that's a perfect kind of like moment where you're going to find some competitiveness. And then in that example from OneFootball where you were running ads to expand into Latin America, and you realized because of some of the technical debt, many of those ads weren't showing. I'm curious, how long did it take to realize those ads weren't showing? Were you kind of beating your head against the wall being like, why isn't this working for a long time? And then eventually did some digging and found out you had a technical issue? Or was it, like, apparent from the start?

0:33:18 - (Jason Parmar): It was relatively apparent quickly when I started. They hadn't focused much on their advertising technology. They were much smaller, and they had done just direct sales. I think the German market is slightly less technologically-advanced than the rest of Europe and America. So they hadn't focused on it. And when they brought in some new meat to look at, we noticed it relatively quickly because we kind of noticed based on the user numbers and looking at the data, we dug around and found it, and we're like, okay, there is a lot of technical debt. I think we need to move this up as number one ahead of, say, new formats or lots of other ideas that would help the direct sales team.

0:33:59 - (Jason Parmar): But, fixing the technical debt, and also changing the ad server alone would, I think at that time, I predicted over 50% increase the revenue we were making, which is actually quite significant. So it was kind of a quick realization and almost a, I'm kind of surprised no one noticed this earlier and realized how significant this is. But it is probably a lot more common than people realize, because if you don't have those experts in-house and you haven't hired them, you probably wouldn't notice.

0:34:30 - (Matt Widdoes): Yeah, if you don't know what you're looking for, you're not going to find it. And so that all tracks. And I think it sounds like fixing some of those technical elements as well would have knock on effects for the rest of what you're running anyway. So it's kind of like sure, future-proofing you anyways, even in the local markets where things are serving just fine, but as it stood, blocking you from an entire continent, essentially.

0:34:54 - (Matt Widdoes): And so I think one thing that has surprised me over the years is even for companies that have ad monetization stood up, and they largely run against in-app purchases as well, how often the user acquisition teams are disconnected, or are not even really considering the revenue coming from ad monetization. I think also, even more so, because it's not a revenue number, disconnected from things like retention or what we call K-factor, which is kind of some of these referral quotient. So you play all the time, you're in a group of people that play all the time, you never pay, and you're in a group of people who never pay, but you guys, on average, have brought ten people into the network to play with you who do pay, and treating that user as just another non-payer, right, and not really factoring that in. And it gets into things, when we think about it on the user acquisition side, as far as especially really understanding our LTV and our ability to scale, which all of that has an impact on what we're willing to pay on acquisition or what's often called CAC, is huge, because this kind of comes into this premise of essentially a quality score. So we can say, hey, look, you get… revenue is the bottom line. So we can count ad monetization… at minimum, you should be counting ad monetization alongside your IAP so that we actually know what this user is with full stop.

0:36:19 - (Matt Widdoes): So that gives us at least a picture on hard revenue. And then we have these other things that we value, like additional, bringing additional people into the network or playing other games of ours. So if you play four of our games and you have a super high brand affinity, we like that too. So that's maybe worth ten cents in made-up revenue or value of some underlying thing. And really using all of those conditions, you know, if you come back every day and you play consistently, that's valuable. If you refer people, if you watch ads, if you spend money, let's not just call IAP the end all kind of gold standard, because there's a lot of other things that you value.

0:36:57 - (Matt Widdoes): And so oftentimes, I find that's overlooked or under-appreciated, and in almost every case not factored in to the numbers that the hands on keyboard media buyer is seeing when they're reviewing last week's cohorts or information from a few days ago. And it really speaks also to the importance of data science and the predictive abilities of data to tell us what a user or a cohort of users are going to be worth long term and really looking at that holistically. With that in mind, I'm curious one, if that's also been your experience, just anecdotally, if you've seen that across lots of companies, where even on the most basics, if we go back to your faucet analogy, the first one is, you should definitely be combining IAP and ad monetization stuff.

0:37:41 - (Matt Widdoes): But, do you see similar things and is that your take broadly?

0:37:44 - (Jason Parmar): Absolutely. I think you've actually hit something that's like a big trend. It happened a lot in '23, and probably will happen a lot this year, is 'hybridization monetization'. So what we mean by that is more and more companies move into having both In-app purchase and in-app ads. And I would say that your ad monetization team should sit with your user acquisition team. The reason for that is actually, there's these kind of concepts that are now coming out or talked about more. For example, in in-app purchase, you talk about whales a lot.

0:38:15 - (Jason Parmar): That's one phrase people use, superVIP or whatever. But you can also talk about ad whales. There's a concept of people who watch a lot of ads that are useful, to the point where you have consumers who sometimes even complain in forums saying, oh, I don't have as many ad opportunities as before, can I please have them back, for example. And the team sitting together means that they can work on the UA monetization loop pretty much better. And in-app advertisers should very much be part of it, because you can have users who do both.

0:38:47 - (Jason Parmar): They purchase and they watch ads. So I've definitely noticed it. I think more companies are doing it because they're catching onto it. As we said, there's huge network effects, benefits, and actually a lot of models on lifetime value that need to be adjusted and adapted to actually account for the advertising-watching users. Now, there are some who won't watch any, which is fine, but giving them an option is massively going to help your user growth. And that's why it's such a cool, great cycle to have. And I think that's why I think voodoo is probably a good example from the hyper-casual market moving towards hybrid a lot more as well.

0:39:24 - (Matt Widdoes): It's so funny to hear that that's an emerging trend because I feel like we've been doing that for 8 years or more, but I feel like sometimes gaming is ahead of the curve a bit. Yeah. And it's funny, you mentioned the ad whales premise because it got me thinking, and I don't know if anybody's ever tested this, but I imagine you could, and I would imagine in certain cohorts it would work, but imagine a world where somebody, one of the things that you could purchase for a dollar is more ad opportunities.

0:39:53 - (Matt Widdoes): You know what I mean? Somebody who's like, okay, if you want to see an ad, you want to see an ad after every play, that's a dollar and you get that for a month. It's a subscription to ads.

0:40:01 - (Jason Parmar): Yeah. And if it's like with rewarded video, let's say it's gold bars or whatever thing you need in a game that you're doing well in, if it's like, you get a guaranteed opportunity to have this reward at the start. Earlier in the podcast, we talked about the end game placement for more moves. That's the most valuable part. If it was like, you can pay to always have an ad opportunity to get that, I think genuinely that might actually work. It'd be a really fun thing to test because that also seems so counterintuitive and it'd just be fascinating if that's true.

0:40:34 - (Matt Widdoes): Yeah. I don't know if it would apply universally, but you could see things that people have scratched their head about in the past where they're like, how is that ever going to work? And then it becomes a standard. I could see that as being one of those things. You mentioned from a trend standpoint, that's something that you're seeing, that kind of hybridization on the monetization side, I'm curious, any other stuff that you're seeing or thoughts, particularly in ad tech, big tech kind of stuff? Generally?

0:41:01 - (Jason Parmar): Yeah, there's quite a lot. I think, to summarize it, maybe simply you have, regulation is probably the biggest one you've probably seen. Chrome is going to deprecate third party cookies, which has a big factor to play. Meta is adjusting to the DMA. Amazon is getting rid of third party ad serving. Google might at some point divest some of its double clicks. So there's a big push for independent ad tech, but also in terms of ad monetization for apps, I have seen more companies push into trying to do some innovative formats to help.

0:41:36 - (Jason Parmar): After Apple's ATT, A lot of the industry struggled a while, and some of the opportunities that have come out of that have been an increased adoption of what's called app-open ads, as well as a kind of new format revolving around audio, rewarded audio ads. So let's say you're doing something and you can listen to an audio ad while you're in app. So I don't know if you're reading something or you're playing a game, you can just have the audio part played and that could be rewarded or non rewarded. But there's 3 or 4 new companies that are offering that. And these things are kind of new-ish in '23, but I think in '24 will be even more important.

0:42:16 - (Jason Parmar): So we'll keep an eye on the regulation part. I think there probably will be more and more happening, and attribution feels like the one part of the industry that needs the most help. So keep your ears out for a lot of maybe medium-mixed modeling. You might hear that term a little bit more.

0:42:32 - (Matt Widdoes): Yeah, it's coming back.

0:42:34 - (Jason Parmar): Yeah. So those ones are probably front and center of my mind of what I think is big there. And also one other thing that I've seen has changed is in the last year, we've seen a lot of kind of subscription offerings. In fact, I actually helped launch an ad-free version of an app. It's an extremely small bucket of users, but if you're more profitable with them to get subscription ad free, it is potentially an option, I mean, Twitter/X probably the main example, but shortly after they did it, we saw it with Meta and Snapchat+ also go that way, so we might see more of that. I don't know if it's going to work because it's not particularly scalable, especially on the massive platforms with lots and lots of users, rather than the more premium users who go for it. YouTube Premium probably being maybe actually the best example.

0:43:28 - (Jason Parmar): I think their numbers are skyrocketing as they have significantly increased their ad load and more people pay to get rid of ads. That is actually kind of an interesting trend where it's kind of like going full circle, right, to actually reducing the amount of ads for a set number of users, but as long as it's significantly more profitable, it's kind of a smart thing to do. So I do think we might see more of it in 2024, potentially.

0:43:55 - (Matt Widdoes): Yeah. It's funny how many things we've seen, I'd say even just over the last 2 or 3 years that are kind of coming full circle. Things like one of the main things that I think of is in that space is in lifecycle marketing and email, push, SMS, and the value of that and how, I think back to 2015 or so, and if we had just given that proper attention alongside what we were doing, because Facebook was so good that it was like, nobody had to, it was like easy to get lazy. And now people are trying to find the pennies in the couch cushions again and they're discovering like, whoa.

0:44:31 - (Matt Widdoes): To no surprise that, wow. By engaging lapsed users and doing these other things within lifecycle to make sure that we are keeping people engaged while the user acquisition team is dumping money in at the top and bringing all these people in, it really helps to walk around the party and say hi to people and welcome them. And when they start to walk out the door, be like, hey, one more thing. I don't know if you saw the photo booth over here, but we have some other things to do over here.

0:44:55 - (Matt Widdoes): How much that's coming back, and so, I guess with the direct sales piece, because timing is so important, do you think that'll go? I mean, that could very easily go digital, right? Where somebody could just say, hey, I want to place a bid for December and I'm bidding on inventory. Not sure that the publishers would want to do that because you're essentially selling futures at that point. And whether you want to put that up to an auction dynamic, I guess you could with some floor that says, hey, for anybody that wants to bid on December revenue now, it's this CPM and come and get it and minimum bid is this. And we're going to close that purchase window out 6 months in advance, so that'll close in June. But do you anticipate any of that stuff happening? I think some of that still will require a human presence to go sell it and negotiate and think through bigger deals, especially for brands that have more than one property or have other opportunities, like co-marketing on billboards or any number of other things that might play into a broader campaign. But any sense if like, and I don't know if that's already exists, if people are able to buy futures on these platforms already, but do you think that might be something we see eventually?

0:46:03 - (Jason Parmar): Yeah, I mean, I think the product mostly exists in Google's ad tech stack. Programmatic guaranteed would be what it's called. Now, you still need someone on the other side of the publisher to say yes or no to reject, which requires an ad ops person. I don't think it will go away. I've always thought the entire last 10 years of my work, I thought salespeople would be gone. When I started at Microsoft, I kind of thought that from day 1, and then I moved into sales and quickly realized maybe it's a bit overblown. I do think it will get a lot more efficient, for example, with that programmatic guaranteed offer, like you said, buying futures instead of going and doing a whole big process that will take significant amount of hours.

0:46:49 - (Jason Parmar): It can now be as simple as, I know I want this semi-reservation product. I can submit and do all of the billing programmatically, essentially meaning automated, and it goes through, and all someone needs to do on the op side is just go, yes, just approve it. You still kind of need someone to evangelize overall, but you could definitely be more lean. So programmatic guaranteed is the Google example. There's lots of other examples in the marketplace and actually a lot of what's called reservation products. So, you know, at the start, earlier, I talked about partnerships.

0:47:21 - (Jason Parmar): There's a lot of products that exist in market, kind of these takeover ones that are traditionally actually signed semi-paper IO we would call it, even if it's sent over email. Those kind of are all moving into self-serve platforms or in the programmatic ad tech. So I do think more and more will go that way. But, it's more of an efficiency piece, and we probably have seen that very recently with actually Google simplifying their direct sales teams very recently actually, cutting back in their investments there. And largely they called it because of the rise of AI, now, I don't think that has anything to do with LLM's. That is just an efficiency piece to say, hey, we're able to sell and do a lot more with less people now, and it's been coming for a while, and the technology is in a place where it's exactly like that. So you can get away with actually having a relatively small direct sales team who are pretty efficient and kind of can go to the right place at the right time rather than traditionally, where you would have huge teams covering every single agency at all times, going through every single deal.

0:48:29 - (Jason Parmar): More and more of those deals will go automated, which means you've got 2, 3, 4X efficiencies in your direct sales business, which actually makes it more appealing to start them well.

0:48:41 - (Matt Widdoes): And the pricing of that inventory is not just a salesperson pulling a number from the air, it's based on historical, predetermined… So it's like I don't really need to be sold. And there's not always a ton of room necessarily for negotiation in a meaningful way. And I think for media planners going out looking for this inventory, leveraging these tools to be able to see like, okay, where is the heat?

0:49:03 - (Matt Widdoes): And I love that price, I love that app. I'm going to just buy that cold. Don't need to see anything else. Let me click a button. I see, so and so has this inventory over here, it's close, but I need to negotiate that we're a big buyer, we'd like to buy all of it. We might be able to get a better rate on that. So let me go in and let me have six conversations versus, to your point earlier, 30 conversations just to figure out what your inventory is and what that might be priced at.

0:49:28 - (Matt Widdoes): Then go do all that work on what I want to just buy straight up without negotiating or whatever. So I can see that human element as well, not being fully removed, but the role being slightly different and really focused. I would think, if I was running that team to say, okay, that's great… But now your role is not just to take 5% off of the price on this. It's more to kind of build a lasting partnership so that when Coca-Cola or whoever is coming back, you're just always on with them. You're finding opportunities to sell in inventory that they weren't looking for, or you're finding ways to partner with them on a bigger branding experience that we're saying, hey, we're going to go billboard and print and we're going to do this other, we're going to ride along on your Super Bowl commercial, co-branded, whatever that might be, where that becomes more of a partnerships role and less of a straight, know, sales function and more of a relationship function.

0:50:17 - (Jason Parmar): I mean, you can kind of do that with Google's ad tech. If you can build a publisher profile and say, here's my contact info for the sales rep, here's what we have. Kind of do a high level summary link away so the buyer can just read through without having to talk to them when they're ready to say, hey, you know, we're going to buy quite a lot. I'm going to contact this person and say, can I actually negotiate a discount because we're going to buy three of these huge activations.

0:50:43 - (Jason Parmar): I want something bespoke and suddenly you've done some pretty amazing lead gen there in the system and they can also just buy automated. So it almost has free lead gen in it. And I think it might be actually relatively underutilized in the industry as a whole. So that is an opportunity there. And as you said, it can also lead gen into partnerships more broadly. Hey, contact for more bespoke custom things that we can't sell yet in an automated way.

0:51:11 - (Jason Parmar): These come into stuff where a lot of manual work's done. Say we're going to really integrate you mash up our logos together, do something bespoke

0:51:22 - (Matt Widdoes): We're going to put a ranch bottle at the top of Candy Crush.

0:51:25 - (Jason Parmar): Yeah, exactly.

0:51:26 - (Matt Widdoes): Hidden Valley Ranch.

0:51:27 - (Jason Parmar): Great example. Yeah, exactly.

0:51:29 - (Matt Widdoes): Yeah. Okay, cool. Well, Jason, thank you so much for the time today. I'd love to do another one where we go super deep on the technical side, but I am excited to kick this call off. And I think many of our listeners are going to be really excited for some of the things that we talked through. And I'm sure there are tons of follow up questions I didn't get to that people would have loved to have heard answered. But maybe we'll do that another time in a few months.

0:51:51 - (Jason Parmar): Sure. Well, thank you so much for having me, Matt. Loved it.

0:51:54 - (Matt Widdoes): Yeah. Looking forward to the next time. Thank you for tuning into another episode of growth at scale. We hope that you found Jason's insights on monetization as useful as we did. We'll see you next time.