Mobile has spent the last decade perfecting paid acquisition, and often, paid UA comprises 99% of a mobile game’s marketing budget. PC and console gaming generate tens of billions of dollars in revenue, yet most studios have never tried performance marketing once. PC and console have always played in the field of brand, festivals, press, influencers, and content creators due to a variety of reasons that we unpack in this episode.
Alexandra Takei, VP of Platform Revenue at Medal, sits down with Jules Baculard, founder and CEO of AdZap, to unpack why PC performance marketing has lagged behind mobile and what it would take to build a real acquisition stack for Steam, PlayStation, and Xbox. The conversation covers the structural challenges: thin desktop inventory, cross-device attribution, Steam’s limited data visibility, long purchase windows, and the uncomfortable reality that many Steam purchases are never played. Jules explains how AdZap built their performance platform using mobile ad infrastructure, Steam’s companion app, landing pages, signal engineering, and retargeting to drive qualified traffic and wishlist growth. The episode also digs into real case studies across indie and Ubisoft titles, incrementality testing, CPI benchmarks, and why wishlists still matter, even if they are an imperfect KPI.For any studio trying to make every marketing dollar work harder, this is a tactical look at PC performance’s next frontier
We’d also like to thank Medal.tv for making this episode possible. If you're a PC gamer and want to clip your moments or a studio, publisher, or marketer looking to reach a high-quality gaming audience and get your game in front of the right players, check out all Medal has to offer at https://grow.medal.tv.
This transcript is machine-generated, and we apologize for any errors.
Alexandra: What's up, everyone? And welcome to the Naavik Gaming Podcast. I'm your host, Alex, and this is the Interview and Insight segment. Every category in media and entertainment has sort of figured out paid acquisition —E-commerce, social apps, mobile games, streaming services. TikTok scaled to a billion users partly on paid. Mobile gaming studios at the top of the market run UA budgets that are probably like 99 if not 100% of their total marketing spend and can do so confidently because they have good estimates on customer ECPI, DCPI, and LTV.
And there's a ton of creative optimization to an insane level. And then there's PC gaming, a market that generates tens of billions of dollars in y- in revenue per year, where a s- typical studio maybe allocates around 90% of their budget to production and treats marketing kind of as that last minute oh shit budget.
Until lately, which is out of necessity since it's getting a lot more focus. But it's generally kind of been a bit more fuzzy and hocus pocus cauldron. The dominant go-to-market strategy is Discords, Steam wish lists, and hoping the right streamers or group of micro-influencers picks up your game.
And if you're rich, you do brand, big trade festivals, pay for big spotlights or press. And if you're trying to build... bring over the mobile performance playbook, you'd hit a wall immediately because 30% or more of Steam purchases are actually never played, identifiers don't work the same way, there's limited desktop ad inventory, and there's generally a lack of structure and data footprint, you know, years of historicals by geography or by genre, to inform just how much money you should sink into PC performance.
But I think here's the thing, and the problem isn't that PC performance marketing doesn't work or that it isn't needed. It's actually, I think, more... needed more than ever. And just like mobile, there's a lot of content, and we need to try and match the right games with the right, with the right players.
Just the problem is that there isn't actually the right stack for it yet. So today we're discussing PC performance space as a new entrant alongside brand, influencer, and other types of marketing for games. I'm particularly passionate about this problem from lived experience, having run a studio and thinking about how to deploy a very small marketing budget and where I'd want every dollar to work for me.
But I don't wanna buy transitive KPIs like Discord members or wish lists. I just want people to buy my game. To do so, I'm welcoming Jules Baculard, founder and CEO of AdZap. AdZap is a full service PC and console performance marketing agency that spent two years building the ad tech infrastructure that PC probably didn't have before.
Jules and his team have worked with clients ranging from indie studios all the way up to Ubisoft, and they've been trying to prove that you can acquire PC gamers at performance prices. Jules, welcome to the pod.
Jules: Thanks, Alex. Great to be on the show.
Alexandra: Awesome. I'm so glad to have you. This is gonna be a really fun one.
And I know we've been working on this for a long time, getting this together. So before we break down the PC performance problem, like the delta between mobile and PC, the tech that you guys have built, and maybe share some real case studies, give me the short version of who you are, where you're coming from, and what made you wanna work on this problem.
Jules: Sure. So first of all, I wanna say you really did sum it up quite nicely. Like, the- there are... It's a very big problem with multifactorial kind of issues, and I think you're, you're spot on. So first off, yeah, I'm, I'm Jules. I've been in the gaming industry now for around 10 years. Started in the PC console business at Ubisoft, but then worked at various startups, most recently a company called Homa Games.
You might be familiar. It's a mobile games publisher, hypercasual. So really at Homa it was all about scaling very simple game experiences from zero to hundreds of millions of installs through performance marketing. So really aggressive ads, trying to use data as much as possible to, you know, target the right player base for the right game.
And basically AdZap, this is a company that's about a year and a half old, and we started as a publisher. So, the initial pitch was actually not an ad tech company. It was, "All right, we're coming from hypercasual." So Homa and my co-founder was from Voodoo, the leader in the hypercasual space.
Let's,, we, we published together games to over 500 million installs. Let's apply the same framework to the Steam platform, 'cause Steam at the time was growing and is, is still growing very much with the indie segment getting more and more popular. And so, the pitch was simple. Same thing, get the same indie teams we had been working with for years on mobile, get them onto Steam, and scale their games with performance.
So we got an a16z speedrun investment off of that thesis, and for the better part of 2024, we scaled a few games on Steam. We published a few games. And then we realized, the, we hit the wall that you just described, which is performance doesn't really work that well on Steam. So, it's impossible to get a proper CPI to get your ROAS.
Steam doesn't give you any data, you know, on the actual store page, like what's happening on the store page. It's basically very hard. It's a great platform to distribute your games, but it's overall very hard to work with and to control, to have agency as a publisher. And so, we decided to switch, to pivot, and to build these tools the, the tools that we wanted to have as a publisher but build them for the, for the overall, for the whole market, 'cause we realized that nobody else had these tools and everybody else had that same problem.
And that sure, PC console gaming isn't like mobile where 99% of your spend will go to, to ad dollars, and maybe we can talk more about the differences between the two markets, but there is a big need to, to unlock that programmatic channel, so yeah, I would say to, to reframe it a little bit, the problem statement is that Steam is a black box.
Currently publishers try to optimize on the Steam algorithm to try and get Steam to feature their games, which is, which is a good strategy. But to us it feels like the App Store 10 years ago where you try to game the ASO, but it's getting more and more crowded and so you need another channel, which for mobile has been U- has been UA and Apple Login more broadly.
So yeah, that's why we're here.
Alexandra: Makes sense. Yeah, yeah. It's a really good analogy to think about the App Store back in like the early 2000, uh, 2010s, , where ... And I think that is exactly what it is. I mean, Medal, we're working with studios all the time and they're just like, "I need to do this thing to game the algorithm."
And we're like, do we even know if it actually works like that? Like, it's, it's just all like a mystery, you know? Like, I think like you know that like frontloading Steam Next Fest is pop- is like, because the, because like the, the ether speaks it so. So, anyway, I think you're addressing a really important problem.
So, before we go on to like the foundation, right? Like we've talked a little bit about the problem at a broad, a broad level and, but we're gonna unpack like the reason why and, and what can be done. Is there anybody else that is working on this problem too?
Jules: It depends. So, there are a few problems work- a few people working on the game discovery problem from different angle.
Obviously, there are people handling it from the short form side, like a huge short form platforms. There are people working on the attribution side, which we- Mm ... we can talk about, and we started as an attribution. So, the first pivot after being a publisher was being an attribution company, and we pivoted again into a performance company, which we are now.
And there are obviously... So yeah, there are people unlocking new channels for games, people building ad tech tools for games, and then agencies, which we're, somehow competing with, sometimes competing directly with, and sometimes we partner with, - Mm ... 'cause they're our distribution partners into clients, and- Hmm
and we serve as their performance arm. But yeah, I would say that's kind of the landscape right now.
Alexandra: I see. Okay. And just to make it clear to the audience, what to you is the difference between attribution platform and a performance platform?
Jules: Sure. So, when you sell attribution, you sell visibility and analytics.
So, it's like, okay, can you tell... The marketing director might tell you, "Okay, where are my wishlists coming from? Where are my installs coming from?" It's so it's more about precision of the actual attribution framework. When you sell performance, you resell, uh, cost-effectiveness. So, you sell a marketing channel.
Hmm. So, it's like I'm selling to CMOs, like, okay, I can get you wishlists for a dollar in the US, cost per wishlist, or I can get you- I see ... installs for, for that much. So, it's really about, like, okay, my, my... Our focus is how to get the costs down for our clients while keeping a scale interesting enough for them.
Alexandra: Hmm. And one could also extrapolate the differences mostly, a measure- an attribution platform is retroactive, right? It is looking- Sure ... backwards at your performance, right? A performance company is a bit more forward-looking, where you're saying, "Hey, like, this is where you should deploy your dollars given this efficient CPI." Makes sense.
Jules: Exactly. Okay. And I, I would say attribution is one tool that a performance company can use. And so, we built our own attribution internally because, we need it to increase our performance. Makes sense. But it's not the only thing, we use.
Alexandra: Okay. Cool. All right. So, I've often said, um, we'll start at the foundation of the problem, that PC performance marketing doesn't exist, and we've kind of just agreed in that extent that it doesn't.
But there's a lot to unpack here in terms of why. So first, I think there's four ch- ch- categories I wanna talk about for, like, five minutes or so, which is the first is budgets, the second is streamers and influencer marketing, the third is lack of desktop inventory, and the fourth is the technology and data footprint.
We've mentioned all of these things, but we'll go into each of them, each of them individually. But as mentioned, a typical PC studio puts, like, 90% of their budget into production and, like, not a lot into marketing. I've seen a lot of studio P&Ls and, like, this, this tracks, but why has this not changed in your opinion?
Is it cultural, structural, or something else?
Jules: Yeah, it's a, it's a good question. I still don't have the exact answer myself, to be honest, but my, my take on this would be the way I see PC console gaming is more akin to e-commerce than it is to mobile gaming. So, in mobile gaming, you would spend actually very little of your mark- of your overall budget to production.
All of it will go to marketing, and all of that marketing budget itself will go to ads. It's very simple. On gaming, I feel like PC console gaming, I feel like it's closer to e-commerce, where a lot of it has to go to product development, for sure, and then some of it will go to marketing. Where I am confused, and this is also why we think the market we're on will have to grow at some point, is that every single product on the internet, a digital product, so mobile games, or even e-commerce, fashion products, anything, all use performance marketing as their number one, marketing budget.
And for PC console, not only is marketing a small share of their overall budget, but also ads are a very small share of the overall marketing budget. People will spend a lot more on brand or on influencers, as opposed to performance.
Alexandra: Okay. And sorry, and just to say here, performance for you is ads, or are there other ways to do performance that is not ads?
Jules: It's an interesting question. I would say performance, yeah, when I say performance, usually it's programmatic ads. So, ads- What do you mean programmatic? ... can be not- Okay ... , performance-based, like if you buy on the CPM basis, for example. But performance, I would say, is the, is the action to buy ads optimizing with a programmatic- Programmatic engine that optimizes based on results. Yeah.
Alexandra: Cool. Okay. So that's problem one. I'm also confused, like you are. - Yeah ... so there's that. But maybe it's because they don't have anywhere to sink it into because the technology isn't good, so maybe we'll get to that in a second. But all right, the second thing, streamer and influencer marketing.
The streamer economy, I think, is actually part of the answer here. So, you know, I think you actually mentioned, like AAA companies won't spend like anything on paid media... Sorry, paid performance programmatic, as you just said, but they'll pay like 40, 40K for an influencer. What do you think that, like when you look at those eCPIs, how do you think about that this is influencing the problem?
Jules: Yeah, I think, I think this is a huge problem. I think the influencer channel has been a great channel. I mean, it was the Apex Legend kinda distribution advantage when they launched. It's been, it's been a huge channel. It still is, but I think it's largely tapped out. Like, streamers are so expensive now, and it's becoming, and there's so much competition for attention that it's becoming more and more unreliable.
And I think the core problem for me for, with regards to the influencer marketing if I were a CMO, is first of all, yeah, the price. So, to give you an idea, like CPM pricing for influencers versus ads or performance is 100X or 200X higher. Really? So it's, like, a lot. Yeah, like it's... If you... It's, it's so much that influencer agencies have started reporting cost per viewer per hour, as opposed to cost per, like, CPM, which is the traditional advertising metric.
Because if you, if you count, if you use that metric, then a viewer who watches several hours of stream can be counted several times and so yeah, it's kind of crazy. And the other, the other angle of influencers that I'm not a fan of is the repeatability of it. So even if you have an influencer campaign that's successful, which obviously happens, then it's hard to reinvest those dollars in the same place.
You, you don't know that it will work again. It's, it's harder.
Alexandra: Yeah, you don't know
Jules: that. As opposed to performance where it's, it's really repeatable and scalable. Like if it worked once, you can invest a few, a few dollars here and there in the performance channel. And if it works, you know you can scale it, and you know it will work.
Alexandra: Hmm. Okay. All right. Moving more towards that, to the performance section. Desktop inventory, right? So, like what's the issue with desktop inventory? Mobile works pretty well because the advertisement is already in the same surface that the player already is. Yeah. So, walk me through this challenge.
Jules: Yeah.
And so, attribution actually works quite well for desktop to desktop, uh, flows. So, if you advertise on desktop inventory like YouTube or Twitch, and then you use a traditional off-the-shelf, attribution solution like IP, IP address, uh, that kind of, that kind of thing, we used to have a product like that.
It actually works quite well. So, you will be able to attribute fairly well the traffic that goes to your game and know where your traffic is coming from and so on. The problem is that it's super expensive, so, you won't be able to get performance. And this is again, the difference between at- attribution and performance is that sure, you can, you, you will have, you know, clear as day kind of view on where your traffic's coming from and whether a campaign is profitable or n- or not, but the campaign will never be profitable because the channel itself is very expensive.
'Cause I think it's like 95% of ad inventory is on mobile. Pretty much everybody that's, you know, hanging out on the internet is on their phone, on social media. And so you're bidding with a lot of other players on the same small sliver of inventory. Right. And so that's why it's super expensive. And I would also say that the ad engines are not the best.
So if you bid on... YouTube probably is the best 'cause it's Google Ads. But other than that, if you bid on others, like desktop networks, they're not super good at optimizing, like figuring out like who your audience is and optimizing and showing the ads to them, as opposed to like a company like Meta, which is probably the leading, actor in the space.
And if you feed data, the right data and the, the right signals to Meta, they will find, at scale an audience for your game for sure.
Alexandra: Cool. All right. And the last one, which is you tried to bring over some mobile UA stuff to Steam, right? As you mentioned, and you were like, "Wow, this isn't working." You know, you want to maybe install an SDK, maybe do fingerprinting or reconcile the ad exposure, like to the in-game behavior.
Yeah. And what challenges did you encounter beyond, with, in a little bit more detail than Steam black box?
Jules: You mean like if we tried to do... Yeah, 'cause when we started, we had the, we had an attribution SDK. So, we, we were like, "Okay, let's do the same thing that we had on mobile," which is you put an SDK on your mobile game, and then you try to link between what happens on your, like, on the ad side, on the store side, and then inside the game to kinda get a full picture of the funnel.
, First off, on Steam it's harder 'cause the store itself doesn't send you any data, so you, you don't even know what's going on inside the store. You can know what's, ... So usually what publishers do is they, they run an ad, people get redirected to a landing page where the IP address, for example, gets fingerprinted, and then when they launch the game, the IP address again is fingerprinted, and so they're able to reconcile, okay, this person that's playing the game and that's spending money in my game, especially for free-to-play games, is that person that saw this ad on this campaign.
So, this is kinda how it works. The problem with that is, as you said early on the podcast, 30% of people who buy or activate a game or install a game on Steam will never play it. So, you will never see a signal coming from inside the game for that. So already even if your attribution is perfect, you're missing 30% of the traffic.
Then you're, you're running into other problem which is if desktop inventory doesn't work, you have to bid on mobile, and this is where the IP address signal doesn't work. So, if somebody's in the metro, you know, browsing through TikTok, they see your ads, and then they go play on their PC, you will never be able to attribute them.
And so yeah, this is kind of where it breaks. And so, this is why we decided even though the attribution... I guess one learning for us when we had our SDK was that, - You know, it, it worked. So, if you, if you do desktop to desktop attribution, actually the attribution is quite clear. But it doesn't work- Mm in terms of performance. Uh-huh. And so, and so this is why you're not able to scale spend, and this is why we decided to fully focus on mobile, and to not rely on game signals. So, all the signals we rely on are pre-Steam. So, because from the ad network side, everything is pretty clear. You have a good amount of data that you can work with.
Same thing on the landing page if you send gamers to your own web properties. But then as soon as it gets to Steam, tracking gets broken, and it's not... it cannot be recovered from the game itself.
Alexandra: Hm.
Jules: So, our approach is maximize the amount of qualified traffic we can send to Steam, and then let Steam do the conversion, which it does actually quite well.
Alexandra: Okay. Cool. And we're gonna talk a little bit more about your technology in, in a moment, but I think that's a really useful kind of insight about the differences between where the marketing occurs and, and how, and, and where you can actually do clean attribution. As you said, if you're desktop, desktop, it works.
If it's mobile, it's mobile, it works. It's crossing the two that's really challenging. Yeah. But let maybe let's talk a little bit about mobile and PC and, like, why you can't just, like, copy/paste the same strategy that you did on mobile on PC. I'm gonna reiterate that on mobile, like, for our audience, is that, like, basically the experience to me is you see an ad, you hit install, and you're in the game.
It's, like, on one piece of hardware. Usually install is free, so it happens easier. There's not a lot of friction there. So can you just... But whatever, that's maybe, like, the, maybe the, the caveman way of explaining it. Can you explain to me exactly how mobile attribution works in a few sentences? And then we'll do the same thing for PC.
Jules: Sure. So on, on mobile, attribution relies mostly on fingerprinting. So, I'm not, I'm not a mobile attribution expert, so I won't dwell on that too long. But it's, it's basically fingerprinting. So, you, as you said earlier, you play... You see the ad on your phone, you play on your phone, so you get game signals, signals from the ad network, and you're able to...
There are frameworks like Skad Network or, like, others that you can use to attribute quite cleanly even with IDFA deprecation. On a PC console, the core problem is the cross-device nature of the flow. So, if you see an ad on your phone and you play on your PC, it's another... or your console, it's another device, and so it's, it's very difficult to, to have a clear tracking.
I would say that's one, one issue. And then the other issues are that it's a different... Again, it's more akin to e-commerce, so it's a different sales cycle. Mm-hmm. Especially for premium game, a gamer might wishlist a game and then purchase it, I don't know, two months to even a year later. So, it's very hard to have to attribute, right?
'Cause your attribution window, in mobile gaming, typically attribution window is seven days. You can push it to 30 days on PC, but still, I don't think it would capture, uh, enough- Yeah ... of the demand you're driving. So, there's that, and then there's the fact that, of course, like, a lot of people buy the game they never play, and so again, it's harder to attribute on, on PC from in-game signals.
Alexandra: Okay. Got it. So, the differences are kind of that there's attribution differences, right? There's consideration period differences. You know, I might s- like you said- Yeah ... like, I might not buy a game for a super long time, and the longer the window is, also the more chance there is for there to be something inaccurate. You know-if it's like I saw an ad, right, like 30 days ago, I might have had ... If, if it's 30 days, I might have had a lot of time to talk to my friend Patrick, who also told me to buy the game. Exactly. Yep. And then you're like, "How do I attribute?" Right? Whereas the reason why it's seven days for mobile is because, like, it's trying to keep it tight so that you can really do that last-click attribution.
And the last thing is that you said that, you know, there's a dearth of desktop inventory. You have to cross device from where mobile, where the marketing actually occurs, which is on your phone, and then, you know, on TikTok, on Meta, or wherever you are on your phone, and then you go to, to your PC, and that, and that breaks it.
- Yeah ... so my question, I guess, in that, in that d- in, in that direction before we move on to your, your tech stack is, so you're running most of your PC game campaigns mostly on m- mobile inventory. Do you think that you'd want more desktop inventory? You said desktop inventory has good attribution if you're going desktop to desktop, but it's expensive because there's not a lot of it.
So, if there were more desktop inventory, would that help solve this problem? Or is mobile just-
Jules: Definitely. For sure. For sure. And this is why we're opening our own DSP. So, we're gonna be able very soon to be bidding on any, not only any mobile app or game, but also any website that, you know, opens their inventory up to any DSPs, to the open web, basically.
And I'm very excited for that because, as you said, like, typically desktop inventory is the best inventory for PC games, but it's usually very expensive and doesn't work. And so, with being a DSP, that could enable us to buy per impression kinda thing, which enables us to be, bidding at competitive costs still on desktop.
So, this is something I'm quite excited about.
Alexandra: Okay. Cool. All right. So, let's go a little bit more into the technology that you guys built. Um, you actually already alluded to this, but in my opinion, some of the desktop ad tech is in the Cretaceous period, but like mobile's in like, you know, 2040. Um, and so mobile- Yeah
advertising is obviously a lot younger. It grew up with the advent of mobile in 2013. It's already a, it's already a more advanced system. So, I wanna spend a minute on actually what's more specifically missing on the PC side, what tools exist on mobile that just simply don't have the PC equivalent.
So, can you walk me through like your signal engineering approach? The traditional attribution stack doesn't work. The desktop to inventory is too thin. So, what did you guys build instead at Ads Up?
Jules: Yeah. So, I think one of our big learnings, so I guess for less, in the last 18 months, we ran ads for around 100 games from like pre-launch campaigns, demo campaigns, free to play, premium, all sort- all sorts of genres, and so we learned a lot.
And one of the learnings was, okay, the Steam app is actually super popular. So, I don't know exactly the number of downloads that the Steam app has, but it's between 100 million to 200 million, so it's a huge audience that has the Steam app installed, more or less the same for PlayStation. Mm-hmm. And it is, even though it's not really heard of, like typically Steam is thought of as a PC first or desktop first product, actually the app is not that bad.
And so, we use this, the app as the core funnel to drive people to Steam. We don't drive people to Steam on, on PC, we drive people to Steam on mobile directly. Okay. And so we've built signals to, detect basically, so what, what we've built really is a conversion API. So, it's a- Mm ... a user sees an ad on their phone, on TikTok, for example.
They will get to an Ads Up hosted landing page where we would do the typical fingerprinting that mobile game, company do mobile games company do, but with tailored to Steam. So basically we're able to predict that this person will be a Steam gamer, that they will have the Steam app on their phone, and what many other typical digital marketing signals And so we know that when we send somebody to Steam, it's very likely that they have the Steam app, and thus they will already be logged in, they will already be able to do, like wish listing or purchasing actions, uh, straight from their phone.
And so that was the core hypothesis behind the product, and it's been, it's been fully realized. So, we're able to get really good performance on cost per wish list. So, $1 per wish list in the US is typically the number I communicate to, to clients. It's our median performance across games. But also on campaigns that are...
Like, games that are already live. We have really good results on, on these as well. Okay, interesting. And so basically the fingerprinting, yeah, it's, it's very... It's a bit technical, but it's like anything your browser gives you is something that you can use to figure out, like, for example, the IP, but there are many other things, like th- that you can use to, to figure out, okay, what, what this person does or, or has potentially on their phone.
And so, we use the data we've accumulated to build a conversion API, to figure out whether or not people are Steam gamers, and we also have that for, for PlayStation and Xbox.
Alexandra: Okay. So, to some extent actually, you're sort of cheating is the wrong word, but you sort of cheated in the sense that you're not actually- Yeah
building any of the tech for PC, you're just like, "Oh, Steam is a mobile app, so I'm gonna do it like I do mobile." Yeah. Is just kind of what you're doing. So, you might be missing some people who have Steam on desktop, but at least you can do what you can for the, you know, the 100, 200 million people that have Steam Mobile, and the same thing is true for either Xbox or the PlayStation mobile app.
Jules: Yeah, exactly. And I believe that Steam, for new Steam accounts, the Steam app is mandatory for 2FA. So, every new Steam gamer- Mm. Yeah, that's a good point ... , that's logging onto the platform now has the Steam app. And yeah. Yeah, yeah, prob- so yeah, we, we... There's probably, a lot of people tell me, like, "Yeah, I don't have the Steam app on my phone," 'cause usually those are developers that have been on Steam since, like, 2002 or 2003.
Yeah. I have it, so, - But- ... there's
Alexandra: that ... you, you have it? Oh, so there you go. I have it. I use it for the two-factor and, well, I also just like-
Jules: Do you browse?
Alexandra: I don't... Sometimes I do, but it's not really, like... It's more-- It's not browsing, I would say. It's more specifically looking. Okay. Which is, I think, how a lot of people use Steam on p- I use Steam Mobile in the same way that I use Steam PC, - Right
which is specifically looking, which is another, to your point around discovery, and, you know, we've talked about this even with m- you know, metal-related things. Like, that's a whole separate thing. But yes, I do use it and I have it. So that's, that's- Yeah ... that's really interesting because essentially what you've done is you've said like, "Hey, I'm just gonna use the same mobile architecture and just rely on the fact that people will have the mobile parallels of these companion apps on their ph- on their phone."
Okay. So how do you actually turn that something, turn that into something that, like, TikTok or Meta can optimize on? Like, and, and maybe I'm not even saying that right. But like- No. Okay. Yeah.
Jules: Yeah, so this is exactly the signal engineering part that you were mentioning. So, the way, the way I see it, there are two types of advertising environment.
One is the open web, which is a billing system that you bid for, to, for impressions basically in a real-time fashion. So, this is something we're, the, an environment we're going towards. But most of the ad inventory, I would say, are between, are caught between walled gardens. What are called walled gardens.
So, it's basically environments where you don't really have much to work with. So, Meta lets you optimize on clicks, so you can do cost per click campaigns, optimize on reach, so views. Or you can optimize on other more custom things, like you can send them a value, and they will optimize on sending you users that have the highest value, and that value you determine however you want.
And so, what we've built is a conversion API that, based on the users we receive on our landing pages, we send Meta a value, and we get them to optimize on sending us relevant traffic. Mm. And so that value obviously i- i- this is the, I guess, our core piece of technology is that conversion API and that, that targeting model that we built using all the game campaigns that we ran in the past
Alexandra: Okay.
And so, then when you set that infrastructure up, I presume there's a lot of, like, everybody in that loop needs to, like, hold hands, right? The, the studio needs the right landing page. The influencer needs to use the right link. Like, all of that sort of stuff. How do you make sure that that is actually airtight versus some of the chaos that I think can happen when you're kind of pulling a campaign together across multiple different channels?
Jules: Yeah, so this is why we made the deliberate choice to do everything internally. Right. That makes sense. So, when a studio works with us, we do everything. We do creative- You're like,
Alexandra: Don't touch it, guys.
Jules: Exactly. Yeah, yeah, yeah. 'Cause we've, we've heard- Step back. Yeah. And we, we started, as I mentioned, as an attribution company, so we had this SDK that we were trying to sell and then we were running into, like, privacy, you know, conver- endless conversations with, like, the DPOs and, and even, like, people having to integrate stuff into their games.
It's, it's tricky 'cause they need to have bandwidth. So, there's a lot of tricky questions that are all solved if you're just on the web using our own infrastructure. So, for studios it's very easy. It's like we need your media budgets to start running the campaigns and we need assets, and that's it. Then we set up the landing pages with all the tracking ourselves.
It's all automated. We run the campaigns. We have a dashboard with the performance you can check in real time, and you can also see it in your Steam backend. And we can also do creatives to optimize the campaigns ourself.
Alexandra: Mm.
Jules: So yeah, it has two main advantages. One is,, we can get set up really quickly, so within days.
There's no need of like, as you mentioned, like, this coordination. And then two is we, given we own all of the funnel, we can optimize it really, really well. So, you can up, gain, you know, cost per wishlist points on creatives, on obviously the targeting engine, but also how you set up the campaigns on meta.
Campaigns are not the s- set the same way on meta, Reddit, TikTok, Android, iOS. You know, based on the user's device also sometimes the targeting model changes. So essentially it's a, we have more to play with because we, we manage everything internally.
Alexandra: Okay. All right, so let's, let me just summarize. First, you basically have built a different ecosystem that centers around the mobile apps, companion apps of PC, of, sorry, of Steam, Xbox, and PlayStation.
There's probably a bunch of gamers who don't have Steam on their phone. They're just lost to you, essentially. You've just forgotten them to some extent. You are trying to do this thing where you're talking about, like, a pre-Steam storefront funnel, right? Where you're trying to just funnel the correct people to the landing page, and you're intercepting the users before they, like, actually make a purchase.
So, you actually don't ever know whether or not someone has installed the game, right?
Jules: Correct. Yeah. Okay. Because it's not our goal. Our goal is to... It depends. I mean, we can run some in- install, cost per install campaigns for free-to-play games, for example. But usually our goal is either, yeah, wish listing or getting people to purchase a game.
And in that case, as you mentioned, like 30% of players who purchase a game never play it or install it. And so, our goal is usually to get people to Steam, and then Steam does the conversion part pretty well.
Alexandra: Okay.
Jules: So, once we send people- So do you track when
Alexandra: someone purchases a game?
Jules: No. It's, it's hard for us to do that.
Right. Okay. Because of the attribution window issue we mentioned. So-
Alexandra: Okay. Exactly.
Jules: Okay ... user might wishlist a game and then purchase it a year later. Even Steam doesn't track that very well. Mm. Mm.
Alexandra: Okay. Yes, okay. That's what I was trying to clarify. So, you don't actually track whether or not someone has purchased a game or installed the game.
So, in a different, in a different way, it's still not exactly the same thing of ECPI, a cost per install, as what it is on mobile. Like, it's like CPW, cost per wishlist, or-
Jules: Yeah ... cost per qualified traffic or anything. So, so yeah. So, I can elaborate on that, what we, exactly what we see on the Steam back end.
So, Steam offers an attribution tool called Steam UTMs, which we rely on quite heavily. So, for wishlist, typically they will give you a good picture of the wishlist you're bringing to, from each campaign that you, you run. And also activations. It has an activation step. So, you know for free to play games, this is especially useful.
You know how many people activated the game, which means how many added it to their library. 'Cause from mobile, you can add a game to your library, and then it will be in your library. And so usually you then go to your PC and then install it.
Alexandra: Mm.
Jules: So that's, that's more or less, like, the, the value that we share to our clients.
And then the other way of measuring our impact is lift studies. So typically, kind of more traditional retail, the way they do a, you know, retail, measurement, which is like you figure out a baseline per country of like wishlist you were getting or sales you were getting, and then after a campaign, you measure the lift that actually happens in the relevant geos.
And potentially you keep one or two geos that are in control. This is how we measure lift in live games that we advertise for.
Alexandra: I see. Makes sense. Okay. And now obviously the game has to be out for some sort of substantive period of time- Right ... to have historicals. All right. And that was kind of actually going to one of my last questions on this section, which is kind of like how you guys think about proving incrementality.
So like how do you know that your campaign drove a Steam install, not just organic one that would've happened anyway? This is obviously a very hard question, I think across the whole marketing space. I don't think anybody has the perfect answer, but mobile seems to have cracked it somewhat a little.
But tell me about how you think about that.
Jules: Yeah, yeah. This is something we usually, we usually get, and I think that's what's, you know, - keeping marketers from investing a significant amount in performance, 'cause it's always hard to have a clear attribution, no matter what you do, even if you have the best state-of-the-art techniques.
On our side, we like to start with clients on the wish listing because this is where there's a clear picture. It's as clear as day. Usually there's something, there's something that we call the ads up effect where the day that we start campaigns, we get clients sending us, like, they really see the impact.
Like, we get clients sending us screenshots of like, "Is this bots?" Like we, we've had that in the past because they really see a spike in their wish list. Given it's a free action, people really do it, freely on their phone. Um- Mm ... and so yeah, there's really a day one impact of the campaigns we run on the wish list side.
And usually we become the number one wish list source on their Steam dashboard. This is actually very common. So, on wish listing, the, we don't really have those, those discussions usually. But they are on the installs and sales front, it's always, it's always tricky 'cause it's the sales cycle is tricky.
So, for example, if you advertise, we, we've tried running campaigns in every setting. So, if the game is full price, usually it doesn't work well because Steam gamers only buy games when they're on sale. And so, when the game's on sale, all of a sudden the sales of the game increase, like, a lot. But then, you know, is it the marketing?
Is it because the game's on sale? It's tricky. So, the way we figured out how to do it is when the game goes on sale, we usually keep a week, like the first week is control, and we know that the first week is usually bigger than the second week 'cause people get the notification on the first week. And then we advertise on the second week in some select countries and we fig- we see how, you know, what's the decay.
Is, 'cause you're gonna have some, let's say everybody, so we ran a campaign, for example, for Voyagers of Nera, an early access title. And so, we, we like, we measure like all the countries, how much they were selling in week one, how much they sold in week two, what was the decay, and then for some geos actually we sold more in week two than in week one.
Mm-hmm. So that was kind of the, the spread that we, we brought.
Alexandra: Okay. Yeah, and we're gonna, we're gonna actually talk about that campaign in a second. I just wanted to kind of address that. Like, it's a, I think it's a hard question across the whole space about incrementality and what you spent your money on and what would just have been organic, but you, and you sort of never truly know.
But I think it's, it, I think it's, you know, a question that you probably have been facing and I wanted to kind of get your thoughts on it. I mean maybe this is actually a good opportunity to jump into that, right? Like, I mean we're gonna talk about one of your smaller one, Might & Magic first, but well let's talk about the Voyagers of Nera one.
This was a game. Can you tell me about the budget? Can you tell me about whether or not the game was free or not? Can you tell me about like what your objective and what your goals were, what the client came to you with and sort of how did you approach the problem?
Jules: Yeah, so Voyagers of Nera was a game, it's still a game in early access, that's a premium game, survival crafting.
And so, they came to us with the, a pilot campaign that they wanted to run 10K, $10,000 and they wanted to see exactly that. So, 'cause basically it's a game that's very well, well-rated on Steam, very positive rating, has a great community, and they wanted to see if they could lift the long tail revenue of the game, basically bring the game to more people without...
'Cause they rely on Steam obviously, but it's not enough. Like, the, the game is, is super cool and it could be brought to more, more audiences. And so yeah, the test was basically what I, I just said. So 10, $10,000 for the second week of the- Of the sale ... it was the, the sale, the Christmas sale, and see what kind of, what kind of uplift we could drive.
And so, on the first one, I think we only did it for the US, and we measured, like basically we, they sold more or less the same thing in the second week than in the first week in the US while other geos decayed like 60%, something like that. I mean, we have a, yeah, case study on the website with all the numbers.
And then they did another test with us on, on further sales, so now they're regular customers. They, they... Basically every time the game goes on sale, they, they try to work with us. And every time we do the same thing, which is, like, 10K the second week of the sale. And now we do more, more geos. So, the, the second one test that we did was with, I believe Germany, US, France, and then some geos, like in France, I think we almost two X'd, on week two versus week one.
Alexandra: Hmm. Hmm. Okay.
Jules: So, there's clear incrementality. It's, it's more, it's more complex typically than the wish listing campaigns, which are, like, obvious and, and more easier to sell typically on my end. But we can clearly see a good impact on driving additional incremental revenue for, for clients.
Alexandra: Okay. Got it. I mean, I think, and we'll talk about, we'll close this podcast with a little, a little debate about wish lists, but- Sure
I mean, I think that this is, like, obviously the valuable thing, right? Like, there's a... I've looked at so many games. Like, there's one that I was looking at today. I was like, "How? Holy shit, these guys have so many wish lists, and holy shit, they've sold so little units." Like, it's just like- Yeah ... it's so tragic, you know?
It is, yeah. They work so hard to build up this thing, and, like, all of their campaigns are running around it, and then, like, the thing that they actually want, which is, like, revenue in the door, is just like-
Jules: Not coming
Alexandra: in. Yeah ... in a dumpster. But anyway, but okay. So, this is specifically- Yeah, yeah,
Jules: I have some horror stories with that, too. Yeah, yeah.
Alexandra: Yeah. So, I mean, and again, like, it's what you can do, but we'll, we'll talk about that at the end. But this is pretty cool, and I think, like, that's sort of a, a good case study of saying like, "Hey," like, through Meta, through TikTok, through the demand sources that you're deploying, you're able to do localized performance opportunities to these certain geos that move the me- you know, meaningfully for PC units, right?
Which is, like, the actual north star of, like, what people want.
Jules: Exactly. And this is really why we're doing what we're doing, is, like, we think there's just a lot of good games out there that aren't, like, that have good ratings and are not being discovered enough by, and, like, shown enough by Steam 'cause there's limited real estate on, on Steam.
And so, what we enable studios is first to regain control over their distribution 'cause they can activate that lever. It's in their control.
And two, it's a scalable, repeatable channel. So again, as opposed to influencers, once we run a campaign, you know that it works, and you can just pour more dollars in it, and it just scales.
Alexandra: Yeah. Okay. So now let's flip over to the other case study before we talk a little bit about the business of performance for you guys in general. But-
Jules: Yep ...
Alexandra: this one was for Might & Magic: Fates for, for Ubisoft. So set the scene for me here. Um, what was the mandate? What infrastructure did you have to work with?
Was, like, the game free, et cetera?
Jules: Yeah, so this was a TCG game from Ubisoft under the Might & Magic, uh, license, and they initially came to us for wishlists. So basically, uh, yeah, we wanted to acquire a wishlist for them, which, which we did., And, it's, by the way, like, it, it was a card game, so it was probably one of the, the trickiest games we worked on, because card games are notoriously hard to, to advertise, 'cause they're all UI, right?
Like, for, for us, we see the same cost per wishlist dispersion that we saw on mobile. Like, basically the marketability, some genres like simulation or first-person games are super, super marketable 'cause the ad is basically you get straight into the gameplay and it's very immersive. And some UI-heavy games, like card games for example, are typically harder to advertise.
But for this game it worked quite well, so we were around, yeah, like one, $1 per wishlist or something like that. And then for their launch they decided to, - To continue with us, and so when the game launched we used the retargeting,, function- Hmm ... of all the, the campaigns. So basically this is something I didn't talk about that this is something actually I think is very valuable.
Alexandra: Yeah.
Jules: The wish lists we acquire are not just a number sitting in the Steam back end.
Alexandra: You know
Jules: who they are. They are an audience that we can then retarget.
Alexandra: Yeah.
Jules: And so, and by the way, it doesn't even have to be a wish list. So once somebody gets into our landing page, we know that, you know, they might be a Steam gamer and whatnot and that they're interested in the game, and we can retarget them at will for every milestone of the game, be it a demo, be it, you know, an expansion DLC, the game launch, whatever.
And so, this is what we did for, for Might & Magic, and it worked super well. So, we were at $6 per install, cost per install in tier one countries, using, so using the incremental method that I, I outlined earlier.
Alexandra: Mm-hmm.
Jules: And, and they were happy, so basically they continued working with us,, a few, a few weeks after the launch, and for, for a few months we worked on, on the game.
And we basically, yeah, we spent what Ubisoft had budgeted for the game, and we were profitable on that, on that acquisition.
Alexandra: Cool.
Jules: How did- And I think this is because of the retargeting aspect of it.
Alexandra: Right. Right. I mean, that's definitely very valuable. How did the studio... I mean, this is just a question, right?
And I mean, it's, it's hard to answer, but obviously this is a free-to-play game, $6 Steam install, right? Like, that's a pretty, like I mean obviously mobile ECPIs are depending on where the country are from could be somewhere between like, you know, $3 to $5, right? Like, that's, that's on the very high side. How do they feel given that the game was free but it was a $6 CAC, and do they feel they were gonna be able to offset that with some sort of in-game monetization?
Um, what was the-
Jules: Yeah, yeah.
Alexandra: Okay.
Jules: Yeah, so the game is free to play, in-game monetization, uh, and they were definitely profitable on the $6 cost per install. Uh-huh. Which is why they, they continued, uh, piling on- Okay ... UA budgets.
Alexandra: Cool.
Jules: Typically, I wouldn't consider $6 cost per install to be expensive. So, in mobile, like obviously- You wouldn't.
Okay ... on, on... No, I wouldn't. Okay. On hyper-casual mobile, yeah, sure, we were looking at, like, 20 cents cost per install, something like that. Yeah. 50 cents. But that's hyper-casual. Not
Alexandra: for a PC gamer. Yeah.
Jules: Not for a PC ga- even, like, so this is something I learned working at this company, is that, so first off, like some casual match three companies on mobile buy traffic for $80 an install, like 80 bucks, because they know that they will recover that amount over, like, the five years that this person is gonna play that match three.
It's not uncommon for, for, yeah, for companies like match three companies to buy people like $60, $80 cost per install. Obviously, those are like ROAS campaigns, so targeting whales or people who will monetize in those games, but still it's, it's, it is very not uncommon. And the thing I learned, at this job is that actually free-to-play games monetize better on mobile than on PC Which is interesting, 'cause I would have thought that- No, it makes sense
you know, PC gamers are more hardcore, they tend to spend more. But that's not true, actually. So, t- PC gamers definitely tend to play more, but they don't spend as much. And LTVs, from what I've seen, LTVs on mobile are higher on the free-to-play side than on mobile.
Alexandra: Yeah. No, that makes complete sen- I don't know.
I could maybe coming from the Blizzard background, like, nobody, everyone. Like, the mobile people, they're like, "Ooh, a couple dollars here for these gems, a couple dollars there." Yeah. Whatever. Like, you know, also it's, it's also maybe like the geo, right? Like, the Asias, right, they typically spend and game on, like, play games and love spending money on mobile.
Sure. Right? It's their console platform. The PC players, they're like, they play a lot and they're like, "By God, I won't spend any money." Yeah, yeah. Like, "I will make it through League of Legends without spending a dollar." Sure. Or, "I will get through all of Hearthstone and I won't spend a single dollar on any card packs."
Yeah. So maybe that's actually part of the hardcore-ness. Um, but the, it's, it's funny that that's showing up in, in your guys' numbers. Okay. So, I wanna talk a little bit about the business of, of you, right? You know, you're, you said, I think the other day you said you're, like, eight people. You know, serving, like, 10 to 15 clients per month.
You know, some of the people that we talked about, like the Voyagers of Nera was clearly, like, an indie game. Yeah. And then Ubisoft, Might & Magic. How do you guys, like, structure this engagement economically? And, like, how does it actually, like, work and, and, and what do they get out of it?
Jules: Yeah, so we tested different formats, and the one we have now, as I mentioned, is full service.
So, we take care of everything, we get set up super quick, and there's no commitment in terms of number of month we have to work together and so on. It's, ad hoc campaigns. So, it's really like an agency business model where you come in with a budget, we deploy it, and we take a cut of budget, uh, that is 20%.
And then that covers everything, including creatives if... That's usually an optional. So, some, some developers are very touchy of other creatives. They don't want us to touch their, their art or their brand, which is totally fine. But it does improve performance if we can iterate on creatives and, you know, this creative has, like, a super good CTR, like, let's make variants.
Even if it's a, if it is, like, changing the button on the creative. Like, that definitely has an impact. So, we also handle creatives internally.
Alexandra: Okay. How do you, like... But, but, but imagine that you're having, like, 10 to 15 games a month. How are you doing that, like, at scale with eight people?
Jules: Yeah, so that's the part where the platform comes into play.
So, we have an engineering team which t- like, half the team is basically engineering, and everything is automated, from campaign setup, landing page setup, analytics. Basically, as soon as we have a new client, like, the, and we have the assets, the landing page and the campaigns get rolling with very, very little human intervention, and the data, you know, goes to our dashboard and our clients can, - Hmm
interact with it live. The human part comes into more, like, the strategy of, like, okay, let's launch this or that campaign, and interacting with the client. Like, the back and forth with the client is where that takes a long time of, like, okay, we gotta figure out what the needs are or how they wanna orient the, their next campaigns and so on.
But the campaign management itself is all automated, and this is something that took us a long time. Mm. Like, last year we didn't have all those automations. And, and it also has problems of, like, human errors. Typically, you don't wanna have a human error when you run, like, I don't know how many campaigns we run in parallel, but probably in the hundreds.
It, you know, it can, it can mess up very quickly.
Alexandra: Yeah, makes sense. And then that also kind of goes to my other question, which was gonna be around, like, why does a studio not hire someone in-house versus using you guys? Like, what's, like, your response to that?
Jules: Somebody in-house
Alexandra: will- I feel like people feel that way about agencies sometime.
They're like, "Why don't I just do it myself?" Sure.
Jules: Yeah, the thing is, like, the, the, the technology we built is really our edge and our moat. Like, if, if you're... Sure, anybody can run ads on Meta, but we, we've, we've done it in a way that we're sure, for sure we're getting the best performance. And 20%, if you consider, it's not that much.
Like, if you, if you diminish your cost per install or cost per wish list by, like, a 3X factor or 5X factor, then paying 20%... and then you don't even have to pay payroll, right? Like, the payroll, the HR, the having a person in-house, then it might be better to have somebody external like us.
Alexandra: Cool. And then, also earlier you mentioned that you might be moving to doing some real-time bidding.
What does that unlock for you- Yeah ... and your team?
Jules: Yeah, so this is really exciting for us because it unlocks, I, I think... So right now, we're sitting at around, like, $1 median cost per wish lists for our clients. And I would say the biggest bottleneck to get that down is, the walled garden. So, Facebook obviously has a huge, keep a huge margin on the advertising dollars that we spend with them, and it's very hard for us to, to go beyond.
Like, it's, it's a black box as well, basically. Like, we can feed it signals, and they will, the algorithm will optimize. But if it thinks that the performance maybe is too good, maybe it will keep some margin for itself, and we will never know, right? Uh, on real-time bidding, you have a lot better control over the economics.
Mm-hmm. And so that's, that's number one. Like, we think we can get to 10 cents per wish list with real-time bidding. And we also get access to new inventory. So, as I mentioned, like, desktop inventory, niche websites that we can bid on in a programmatic fashion with our own algorithms and, uh, and bidding engines.
And then the, all the mobile apps, like any mobile, like AppLovin, I used to be at Houma, we were spending like 10, 20 million a month on AppLovin. It was, I mean, obviously it's, it's hard to see all that money go away, but there was a, a reason why we were spending that much on AppLovin. It's, it's, it's a really, a really good network.
And so, if we start bidding, we can, start bidding on AppLovin, we can start bidding on any mobile game or app with a much, better view on, you know, on the users themselves. So, we can know, for example, have they spent any money? We can start bidding on PC niche, PC apps or games or pixel art games or whatever, targeting only spenders.
Like basically we have more, control over how we're spending the advertising dollars. And again, it increases our moat 'cause then if you're a studio and you wanna do this, then all of a sudden it's not just one or, or three like marketers that you need, it's like 20 people with a big engineering team.
One interesting thing that I learned on the job, I'm, I'm coming more from a gaming background than an ad tech background. And meeting with my CTO, Daniel, who comes really from the ad tech side of the spectrum, he worked at Criteo, if you're familiar, a huge, uh, ad tech firm. Yeah. He worked in, in Israel in, at various DSPs.
And it's basically, it's a, it's a fascinating world and it, it's, but it, because it powers all of the internet we know, right? But it is also very, very different from what we know in gaming. So those two worlds are, are very niche in their own ways and very like, people are very specialists of their crafts, and but they, they don't talk to each other very much.
And so, it's rare that I talk to PC console executives and they're familiar, like intimately familiar with how the ad tech space works. -
Alexandra: Yeah,
Jules: definitely So yeah, I think that's, that's why it's worth it for us to go that route.
Alexandra: Yeah. Is it, is it, is it Criteo? I thought it was Criteo.
Jules: Criteo or Criteo.
Alexandra: Criteo? I'm not sure how they pronounce it. I'm, I'm probably saying it wrong the whole time then. Okay. Well, I learned something today. Okay. Makes sense. I wanna move on to some of our conclusionary topics. You know, I think that there's a couple questions that we could, we can talk about. But the first is actually around PC performance marketing you said won't actually look like where mobile UA is, right?
Where it's, like, 99% of the budget. So where in your opinion do you think it lands? Like, what's the actual realistic ceiling on how much a PC... How much of a PC studio's marketing budget should go to paid performance if you were to make, like, a recommendation, and where do you think this actually settles in the next, like, five years?
Jules: Yeah, it's a good question. So, my metaphor for PC console gaming is e-commerce basically. So, I think in the end an, a gaming studio or publisher P&L will look very much like an e-commerce brand. So, they will have a lot of product development, but a lot of their marketing, so marketing will take a larger share of their overall revenue, and eventually within marketing, performance will take a larger share.
And I think it will go... So right now, yeah, I think less than 10% of publishers' budgets are, spent on performance, something like that., And I would say it will increase from 30 to 50 depending on the game genre. I would say that, that will... But obviously it's finger in the wind. We'll, we'll see where that lands, but that would be my, like, within the next five years, that would be my guess, yeah.
Alexandra: Hmm, okay. What is, ... All right, so we talked about with budget. I wanna talk a little bit about Steam. First, what is something that Steam could do to make your life easier to do PC performance marketing?
Jules: Well, Steam, so it's funny. I, I was at the GDC, it wasn't this year, it was last year, at Steam's, at Valve's happy hour.
And I, 'cause they used to have Google Analytics on the Steam pages. So, you used to have a much better view of what was actually happening on your Steam page and, you know, understanding your audience and whether or not this Steam page conveys what you're trying to convey. And it turns out that they removed Google Analytics and included their own tools, which are worse.
And I asked them, like, "Are there plans to reintegrate, Google Analytics?" A- and they told me that, for privacy, that was the number one reason why they took it out was privacy concerns, and that they wouldn't, turn it back on. Which is, which is fair. But yeah, I don't expect Steam to open up privacy-wise.
I think it's a terrific platform. I'm really excited about all the new things they announced, the Steam Machine, SteamOS. But as a, yeah, as a, a environment for advertising, I still, I think it will still remain very hostile. This is why we exist.
Alexandra: Okay. So, no f- no notes besides go back to Google Analytics.
Jules: Yeah. Yeah. Well, I, I don't think they're gonna build tools for, for advertising unfortunately. Yeah. I mean, obviously the best thing would be to have an IDFA or some kinda, something like that.
Alexandra: Yeah.
Jules: But I don't think that will ever happen. Even on mobile it doesn't exist anymore.
Alexandra: Right. Of course, yeah.
Okay. Two, who do you perceive are your competitors in this space?
Jules: It's a good question. Well, I would say right now we're mostly competing for budgets with other agencies. Mm. But we feel that we have a very differentiated value proposition because it's very much tech-enabled. And we're also very comfortable competing.
Like, this is a space where, you know, you see this- It's performance, I mean. Yeah, exactly. Yeah. So, so you see it. You, you know, you see it. Like, you, you run two campaigns side by side, and then you see, you know, white and black, wi- which one is the best. And so, we're very comfortable competing. I think it's, it's just we don't have the deep publisher networks, that other agencies have 'cause they've been around for, for a very long time.
Well, I, I, I also say agencies, it depends which agencies. So, I would say, I would specify and say performance agencies. There are a lot of agencies that we work with and are partners with us, and we are their performance arm, and I think that's a very smart choice 'cause we don't have the ambition to go into influencer, to go into brand, and into events, and I don't personally believe that performance will be the only marketing effort for publishers going forward.
I really believe that publishers have to have a marketing mix that is balanced, and that performance is only a part of it. It is another string to our customer's bow, essentially. And so, I think, yeah, for agencies, it's, it would be... I- it's a good strategy to, to partner with us as their performance partner and then to handle the rest where they're, uh, better than us.
Alexandra: Cool. Okay. And then finally what we'll do is we'll close with the, the wishlist debate.
Jules: Right.
Alexandra: So, you're clearly, you're basically optimizing around like a, what I would call like a peripheral KPI, right? Like, I think the, the, the peripheral KPIs that I think are big in games right now are the following: your Discord size, your wishlists, your number of social media followers.
Right? And a lot of these things are very transitory, right, in they're sp- they're meant to be like leading KPIs, right? Like, you hope that because you have more wishlists or that because your Discord is bigger, that you will sell more units. So, I guess my question for you is like do you think studios should chase wishlists, and, and why?
You're obviously helping them chase wishlists because they're willing to pay for it, but, you know, like, in a, as, in an existential sort of macro market level way, what are you thinking about this, this, this problem?
Jules: Yeah, so short answer is, yeah, I think they should definitely chase wishlist. I think there are obviously other KPIs they should follow, but it's, it's very important because I would say it's the, the most important real estate on Steam is the wishlist tab.
So, people, when they wanna purchase a game, they usually go, like, they, when they wanna discover new games or whatever, they, they wanna play a new game, they usually start with their, wishlist tab, and we see it on, customers' data. So you can see where your, Steam page provenance is, you know, the- where the traffic is coming from within Steam.
Steam does tell you that, and usually a large part of the traffic is coming from the wishlist tab if they have a big wishlist provenance. So yeah, I would say I would say it's important. It's not the only thing, obviously. But one thing that I wanted to touch on, this is something I get often. So, when we got started, people were telling us, "Okay, uh, paid ads don't work on this team.
Don't bother." And then when paid ads finally worked, and we were chasing wishlists, people told us, "Well, you know, sure, you're getting a dollar per wishlist, but paid wishlists don't convert as well as organic wishlists." This is typically something- Hmm ... that I get at every sales call with clients.
Like, well, you know, and it does make sense intuitively. Like, somebody that watched an hour of your game on stream might- has likely, you know, more probability to convert into an actual sale than somebody who just clicked on an ad, right? Well, it turns out that after looking at, you know, s- we've worked with over 100 games now and some games that we scaled to 100 plus, 100K plus wishlists, we were the, the only source of wishlist they had.
Like, nothing else, all paid. And, and the conversion was in line with what we see on the organics. Like, we didn't see a... Basically- Hmm ... I think there's a big myth in gaming right now is that, like, paid wishlists don't, don't convert. I think that that's a big myth. I think there are other factors. And to touch on your, on your earlier example, like, there are definitely horror stories of games that have, you know, a million wishlists and they, you know, turn out not having too many sales.
And in that case, usually the leading factor is those Steam reviews at launch. So, if your game is reviewed mixed or, or very negative or whatever at launch- Yeah ... it's done. You can have, like, a million
Alexandra: wishlists- Which is a product problem, which is objectively- It is a product problem ... you deserve that, right? If, if, you know- Yeah ... like, you, that's a, that's a, that's a thing that marketers can't control, right? I think that- Exactly ... that's also a really hard thing about generally this space, is, and, and I think we actually encounter as well, like, on what we're doing at Metal. But, like, it's, you can't control what happens when someone arrives at your store, right?
And there, and there's a bunch of people that are like, "Game is not, game is, uh, okay," right?
Jules: You know. The game's not good. Yeah. So- Exactly. And you've done- And it's already been read on the, on the Steam page ... everything
Alexandra: you could have done for your job. You could have gotten them all the wishlists in the world, the highest quality people- Yeah, right? But if that's what happens on launch, like, it's kind of like the-
Jules: Exactly ...
Alexandra: and, and, and objectively that's fair, right? Like, if you, you know, everybody has a right to buy the product they wanna buy, so,
Jules: yeah. Yeah, and there are other factors that, that we found were, were very interesting in wishlist conversion.
Obv- the, the other two, so you have the Steam reviews, which are by far the number one impactful, uh, variable. Other variables are how old the wishlist is. Hmm. So typically, if the w- game was wishlisted seven years ago, person might not even be on Steam anymore, so they might not- ... you know, buy your game. They don't...
Like, some games have been around, like, in pre-launch for, like, years, like close to 10 years. And they have a million wishlists, but sure, like, th- those wishlists will, will never convert. And the other one is country So we've had games with, like, huge influx of wish list from tier three countries, and they have same, like, wish lists in the hundreds of thousands or in the millions.
And those wish lists typically don't, don't convert super well. Hmm.
Alexandra: Okay. But yeah, like w- Are you finding the same dollar cost per wish list across the US or is that a blended number across geo- No, this is y- Across geo-
Jules: All the numbers I shared are US. So, if we go to even tier one, it's lower. And then if we go to, like, tier three l- or like a, like a Brazil, for example, or, you know, other type of countries that are, like, big Steam- Got it
gaming countries but are less expensive, it's like, yeah, it's much lower.
Alexandra: I see. I see. Okay. Makes sense. Yeah, I mean, I think it's just a general macro question. I mean, I, I honestly, I don't fault anybody for chasing wish lists. I just do wonder when I think about the macro PC space as a whole, right? Like, is it really, like, leading to the best outcomes, um, for people?
And, um- Yeah ... and sort of, you know, you're optimizing for a pre-funnel KP- KPI, but it's, you know, one of the few that you can. But, Jules, this was such a pleasure. I think this is... You're working on a very important problem, something that I, you know, when I was... We met when I was looking for a PC performance solution.
Sure. 'Cause I was committed to, like, you know- Yeah, I remember ... I wanted to, I wanted to work. So thank you so much for coming on the show. If anybody wants to get in touch with you, how can they do that?
Jules: Well, we have our website, website zap.ad, or you can add me on, on LinkedIn.
Alexandra: Very cool. All right, guys. As always, that's my, that's our show. If you have feedback or ideas, hit me up. I'm always open. And with that, I will see you, guys, next time. Thank you so much.
If you enjoyed today's episode, whether on YouTube or your favorite podcast app, make sure to like, subscribe, comment, or give a five-star review. And if you wanna reach out or provide feedback, shoot us a note at [email protected] or find us on Twitter and LinkedIn. Plus, if you wanna learn more about what Naavik has to offer, make sure to check out our website www.naavik.co there. You can sign up for the number one games industry newsletter, Naavik Digest, or contact us to learn about our wide-ranging consulting and advisory services.
Again, that is www.naavik.co. Thanks for listening and we'll catch you in the next episode.
