speaker-0: I'm smart. And I'm right. And together we are smart and right. Smart and right. Yay.
speaker-1: The ener the energy levels are slightly different here, aren't they?
speaker-0: Yeah, I mean I I I've been I've been a bit poorly the last couple of days, mate, so you might not see the best of John Wright today. You might see blackluster. Instead of the usual, you know, happy go lucky ADHD hyper person, you might see me leaning into my chair a little bit.
speaker-1: Well, I think I'm on my fourth coffee, so I I'll I'll I should be able to bring the energy for the both of us.
speaker-0: I'm I'm good. I'm
speaker-1: Why why are you ill, John? Why are you ill?
speaker-0: mate, I I think it's a combination of of jet setting last week. So I was in Barcelona for three days for the set of then which was which was great. I'll talk a bit about
speaker-1: How was that?
speaker-0: do you know what mate? It so very funny enough there were three events in Barcelona last week all at the same time. So there was Google's App Growth Summit, Global App Growth Growth Summit. there was App Growth Summit, which is a completely different event. It's like A c a completely different thing, even though like if Google and them have the same name
speaker-1: Were people there by mistake? Like did they get by conferences?
speaker-0: But but then there was also set Uetopia. And Very different.
speaker-1: Name. I like that.
speaker-0: Yeah, it it really, really good name, really good branding. We was in the Met Barcelona, which is up in the hills. Never been there. The most the best view of Barcelona I've ever seen. Like and all in all, mate, I've c I've come to a bit of a conclusion over the last year or so, which is like the the best events that I'm going to, the ones where I get the most from it, where I met the most interesting people, I get the most interesting insights or or information, are the invite-only events. Yeah. Anything that's like kind of mass like mass applied where you have to get a ticket. Like obviously you there's it's it's a different kind of thing you're going for. You know, you're going for for for quantity, not quality. But but the set event was amazing, man. Like I said, it was You know, it was only kind of a one day conference, but we we got there on the on the the night before we had a welcome dinner with all of the participants and there was you know senior directors, VP, CEOs from some of the biggest companies in in gaming. And and for and for once it was really tailored to marketers and and UA people, right? So everyone there was kind of a matter expert, which w is amazing because You know, sometimes mate, like I mean, me and you have probably done talks at P G C where we're doing a talk on rewarded and seventy five percent of the people in there are not not interested or applicable, right? Like you're
speaker-1: Yeah, there are other conference leaders. There are other conference managers.
speaker-0: But but like it so so for for once to have like an isolated group of serious professionals who know the the the subject matter and Yeah, exactly. So part some
speaker-1: Not recorded, I presume. That makes a difference. I mean, when you've got when you've got people from some of the larger like certainly public companies and that kind of thing, they're not inclined to talk about anything if there's if there's cameras there or if there's like p people, reporters in the audience and that kind of thing, are they? I mean I I certainly wouldn't be.
speaker-0: I i exactly, dude. And I think within the more intimate event, because again there was like you know, it was a effectively six talks, right? So there was breakfast, two talks, lunch, sorry, sorry, breakfast, two talks, break, two talks, lunch, two talks, finished. And then like dinner and stuff. And and each of the talks went on for about forty five minutes, so quite long. some of them were, you know, presentations and and lots of data highlighting a number of interesting things. predominantly was like focusing on the kind of AI creative brute force strategy which is which is ever apparent, right? What we're seeing is you know the percentage of cre or the number of creatives is going up, the percentage of hero creatives and and success to to hero creative is going down. And you know it was it was very interesting to see that we still got we still got
speaker-1: We said that before and we still haven't done it. So we need we need to
speaker-0: List to add it to the list. But do you know what? One of the most interesting talks, so there was you know, I'm not gonna name people by by name just because I don't know if they want that in if like to know they were part of this discussion, but there was me, someone very senior from Zingar, and someone very senior from Plateika. And there's a team on stage and they're presenting and they're basically saying that effectively more is better because the hit rates are so low now, if you don't test at a higher velocity then the likelihood of you finding a new winning creative every month and then having an incremental uplift is going to be very, very, very low, right? It might take you one in three months, one in six months. But, you know, interestingly, the the the three of us, so the person for singer, the person for Playton and myself, like we were sitting in the crowd and we just kind of jumped in. Because it was an intimate event, like we you know you can do that, right? When there's only like three rows of people, four rows of people, and they're sitting you know, four meters away, you can actually interject as and call bullshit. Yeah. And not that I was being that aggressive, but
speaker-1: That sounds kinda brutal, to be honest.
speaker-0: Do you know what afterwards someone come up to me and went, Wow, you went in and I went, No, I just want to create a debate, right? Like I I like debate, I like confrontation. I live for this.
speaker-1: Shit.
speaker-0: So I want to create a debate. and and and yeah, so basically I was going, the the issue is that because AI is is increasing the quantity so much, you know, you're not really reviewing the creatives that are being tested in the same way that you did previously, right? So yeah, you're creating more. More slop essentially. And it does and and slop doesn't mean that it's bad looking or ill performing or whatever. It's just it could just be random, right? Like it doesn't have like the the human-based input on every if you're building it was something like they were testing 700 playables a month, and I was like, there's no way that a person like is is involved in every one of those iterations, it's impossible, right? It If unless you're a huge organization which this company
speaker-1: That's a team of that's a team of people. I mean that
speaker-0: You you would need you would need fifty people, you know, to to to do this. and so, you know, what I was basically saying and what what the person Zinger and the person Plateak were saying is like if it was just if that was getting built by a dedicated team rather than being produced in AI, we think that if we did it that way, which is more human led, the percentage of to to success to HeroCreate would be much higher. So you wouldn't need to test as many because the qualitative approach would be better. Yeah, because it'd be more sort of human first interaction. And so we had a big debate about that. And obviously the team on stage were like, we understand where you're coming from, but you know, from our perspective, it is kind of like a brute force strategy now. You've just got to test as much as possible. And and, you know, we basically agreed to disagree. But I think there's really Adam, I think there's just basically middle ground, right? I think, you know, like testing You know, two thousand creatives a month is not effective because unless you have dream games level scale, there's no way the algos of the networks are gonna give enough impressions. You know, there's not enough time and and and impressions for these to be tested all effectively. And there's probably creatives that go live that never actually get the amount of impressions they need because there's so much, you know, going through the system. Because if you're testing two thousand craves and all your competitors are testing two thousand craves, And yeah, we spoke about this a couple of weeks ago about discoverability. So it's not just about discoverability in the atmosphere, it's about discoverability on the network, Adam, you know? Yeah. So so it comes back to all of that, mate. And for me, I would still go with like, you know, using AI to increase production, but I think that you still need a very strong human led strategic pick of what creatives, what concepts, what iterations and what. Don't let the AI do the creative thinking, essentially. But but yeah, mate, that was that was the big discussion.
speaker-1: I I like you said right, pick a subject. So I for today's conversation and bearing in mind we're already eight minutes into today's conversation. I
speaker-0: It was a great conference.
speaker-1: I wanna talk about AEO.
speaker-0: Okay. So app event optimization, we're going up we're going meta, yeah?
speaker-1: No, we're not. We're going we're going far away from creatives. So what what we've been looking at a lot or you are looking
speaker-0: Joking.
speaker-1: at a lot over the last little while is it's answer engine optimization. so actually looking through through AI apps and and chat GPT specifically for this one.
speaker-0: What you're effectively talking about, Adam, is more people are using LLMs to do search than Google now, right? People are are moving to asking.
speaker-1: Yeah, yeah. I mean, like it's funny, we've spent probably the last twenty years trying to optimize like search engine results, you know, as everyone knows, SEO and then AEO for the for the App Store optimization. ASO.
speaker-0: Yeah, so
speaker-1: It's been great. But I think we you know, now now with the way people are adopting Chat GPT or or there there are other AI agents out there. I think it's it's kinda changing and some of the results that we've been seeing are really interesting. And I I wanted to have a conversation specifically about I wanted to get your take on what you What you think of like I it the funny thing for me is it's like a you've gotta if you really want to succeed in this, you gotta you kinda gotta drop everything and and move into optimizing optimizing the the you know answer engine optimization. I don't like the term, but I think we need to find another term for that one. But I think if we do you as a gaming company and and gaming I've not seen being particularly strong on sort of chat GPT ads, bearing in mind it's very early days and and you know it's it's kind of experimentation time. But i it's kind of do you drop everything to try and optimize the the where you sit in those results of, you know I wanna find another game that's like I like XCOM and I don't know, Midnight Suns. I wanna find another game that I enjoy playing. Do you try and optimize your your where you land in that now? Or do you it'll be way too early? What do you reckon? Is is it something that we should look at?
speaker-0: Do you know what, mate? It's actually we we did a big exercise on this about three months ago where we we went to both Claude and Chat GPT and we just asked it a bunch of questions, right? And so like it was like what is the kind of best motocross game on mobile? And and fortunately fortunately for us, like Mad Skills Free, our biggest game came up first, right? But we also asked it a bunch of other questions, like What are the top sports games on mobile? Yeah. What are the top racing games?
speaker-1: So when you when you were very specific, it came up with yours, but when you kind of went broader it sort of threw in other stuff.
speaker-0: Exactly, dude. Yeah, we we we we we obviously when it came down to motocross games as the the biggest name in motocross games, we were very fortunate to get the the larger success rate, right? In terms of people actually being fine like finding links to our games and stuff like that. So we completed this assessment and we were really happy about when we were focusing on it from a very niche perspective. But the problem is we already get a large amount of organics you know, through through Apple and Google. We're already well known.
speaker-1: Seems be right there. I wanna I wanna dive into that or something.
speaker-0: And then but the and the games are super the games are super well known within the motocross community. So what we're what we're trying to do is find users outside of that golden cohort, Adam, right? Like so yeah, it's great. Anyone who knows about motocross or motorbikes or turborilla is gonna come to us. That's great. But those are the people they're gonna
speaker-1: That don't you want to try and find.
speaker-0: Exactly, mate. It's about widening the net, right? And increasing the audience size. And so from in from that perspective, obviously optimising towards having, you know, higher rankings in in sports games globally, etcetera, would be the thing that we would want to focus on. arguably, you know, capturing those people in the kind of you know the ideal cohort or ICP effectively means that their the performance would be higher. But again, they wouldn't come to us anyway. So it's it's better to to widen the net. The the issue is there's not a really a lot of understanding of of how to do this right now. Like how do you actually No like there's
speaker-1: Yeah.
speaker-0: there's there's there's one particular company called App Tweak where they've got a product which allows you to understand where you actually come up in the search criteria. So if you again if you ask a question
speaker-1: So what what people have been searching basically?
speaker-0: Exactly, exactly. So it's the same way that you would use Google to understand where you are in in in search ads, right? Or or AdWords, sorry. And and so they give you some some
speaker-1: 'Cause I'm presuming like Google it changes based on Geo and all sorts of different
speaker-0: Yeah, of course.
speaker-1: information they have about you and everything else. I'm I'm guessing it's not the same answer for everyone.
speaker-0: Well the the th the thing is with like Google and and and you know stuff like that's looking at the metadata on the website and the relevancy score and stuff like that, right? So how relevant is this website to the question that's been asked and there's there's loads of things that go into like why you are higher up on the search criteria after the paid section, right? So you've always got those first three links which are paid and then you've got organic first. Yeah. And and and normally like look, if you're talking about SEO in particular. If you're not on the first page, your discoverability isn't nil, right?
speaker-1: I was discovered a long, long time ago if you're the first page of Google results, yeah.
speaker-0: But the but the the fact is that'll be applicable to AEO as well. Right? Like if you're if you're not like people's attention spans have never been so small, but they want instant gratification, they want instant up So like
speaker-1: And and chat
speaker-0: they're gonna click on the first thing.
speaker-1: It normally gives three to compare, like three it normally finds three like if you ask it about a product or if you ask it about a game or you know, I suggest a a game that I could play next with I enjoy X, Y, and Z. It normally gives you three to choose from or three options. And I guess to if you're not one of those options, that kind of rules you out of it completely, right?
speaker-0: E exactly, dude. So whereas before you might get people scrolling the first ten options on Google, again now the focus is I'm actually just running a a query now to to prove this. But it's essentially, yeah, I would say if you're not in that
speaker-1: Who's the best MP?
speaker-0: I'll do it after. We'll do it I'll I'll I'll do it in real time. We'll see what happens.
speaker-1: I'm a little bit scared now, but
speaker-0: It's not apps flyer, we're fucked. Okay, like shout out to our sponsor.
speaker-1: We'll have to cut this bit.
speaker-0: Thank you for sponsoring. It's it's still running, but when it when it pops up, I'll let you know what it says. But but yeah, dude, like you know, so so App Tweak again is the only product that I've heard of in the market which is actually allowing you to understand where you are in in like the hierarchy, you know, for the 'cause 'cause remember, like each product
speaker-1: Do you think that's gonna come in more and more? Like an analysis or
speaker-0: I think I think an analytical tool that lets you understand your position in AEO is gonna be very important in the future because you're gonna see a shift of people like I don't have the numbers, right? But I would say in the last two years, people that are using Google and and demographically as well will be very interesting, but the amount of people using Google for search versus L and for search, there has to have been a migration. I also know when I speak to people from like Gen Z, etc. Is it Gen Z? Yeah, Gen Z. They're the the younger ones and what's what's your below millennials? Gen Z.
speaker-1: Anyway, let's let's go with Gen Z.
speaker-0: Gen Z? think they're called Gen Z. Gen X? I Gen X is the older one, think. Anyway, whatever the younger denig the young the younger generation than me, right? If you look at the percentage of people using LLMs for for for like using LLMs in general, obviously there's a higher adoption of AI in younger people anyway. So there has been a bigger migration from from traditional SEO to AEO, i.e. the people using LLMs to get answers, right? Yeah. And
speaker-1: So imagine that.
speaker-0: what's the best motocross game on on mobile? And it came out Mad Skills Motocross free would be my overall pick. Particularly for a competitive physic based racing on mobile, it's handling, track mastery, and multiplayer events make a strong case. Though best is ultimately subjective. And then it went two alternatives Dirt Bike Unchained by Red Bull and SMX, which is a Supermoto versus Motocross. So like again, so that's it's giving you three options, right? And
speaker-1: Yeah, and if you're not one of if Yeah, but if you're not one of those three, then you're you're nowhere now. So it's not strategic
speaker-0: And also.
speaker-1: results, it's it's like the top three.
speaker-0: And but again also you've got to look at the language, right? Th the there was a strong, strong insinuation for mo MSM free, right? Which I'm very happy for because it's my game. But but then it's giving you the alternatives just in case maybe you already have MSM free. But the language is super strong, Adam, in terms of like it's it's it's funneling you to that top choice.
speaker-1: Also I think like like you were talking about, what was it, Gen Z? Gen Z?
speaker-0: We'll go with Check it out.
speaker-1: What you were saying there, I mean, I've I've met a lot of people who are using it almost as a as a friend now. Like through through anything that they have to deal with, they they'll start talking to like messaging it as a friend. And then if obviously if your friend's suggesting you play this game or this game, you're more inclined to play it because it's from a it's from a seemingly trusted source, which is a bit strange, but
speaker-0: It's what is acting like a companion? Yeah. Like Yeah. Okay. Good good news, Adam. What's the best MP in mobile games? Apps Flyer would be my default pick.
speaker-1: Well that was lucky.
speaker-0: Yeah, and then adjust singular. Which no no surprise. No surprise. Phew
speaker-1: Yeah.
speaker-0: So someone's been doing some AEO at Apps Flyer.
speaker-1: Or or we're just incredibly lucky. It could be either.
speaker-0: But I I I do you know what dude the thing is I
speaker-1: What do you think about this whole like you know, you've got you've got the AEO side of it, which is the equivalent of SEO, but then you've also got the ads side coming into stuff more and more now. Like, do you think it's it feels like it's it's very, very early days and it feels like it's it's you know, there's there's that experimentation that needs to happen with it and everything. But would you focus on it right now?
speaker-0: I don't know man, it's it's a case of like I don't I don't know if I've seen any ads up to yet. Is it like it's probably not live in the UK, it's probably just live in the US or something like that now. So but it will be rolled down, right? Yeah. Because they need they need money. Like all of these big labs are hemorrhaging money, right? Like Anthrophy owners like six hundred million in debt or something, right? So it's a case of all of these companies need to start
speaker-1: got the user base. So it's y you know, it's like it's like Facebook back in the day. You know, Facebook was was huge monetization.
speaker-0: But they've only got a user base 'cause it's they've only got a user base 'cause it's cheap, right? Like it's it's value it it's it's always about value versus cost. And at the moment the value is very high and the cost is very low. If my clawed ninety nine dollar a month you know membership or whatever co like s you know agreement was to go to a thousand dollars, I I wouldn't pay it, Adam, right? It's only because for a hundred dollars I'm getting so much value. So yeah, if they start increasing the price and the same as with with ads like adding
speaker-1: You'd have to go back to syncing again.
speaker-0: Yeah, I mean yeah, I I still believe
speaker-1: Yeah, but are they under the assumption that you can't that you would struggle to go back?
speaker-0: I mean I I wouldn't 'cause I I believe in we have to I had a big debate about this the other day about like, you know, not losing the ability to to have like to to to think for yourself and and be cognitive, right? Like
speaker-1: We did we did with satnav. We let we let satnav come in and now people can't read maps, like the Gen Z as you were saying. So trendy.
speaker-0: Yeah, look look, man. Yeah, look, I just I just I don't know, like I wouldn't I would like someone was saying the other day about it was part of the set off site, right? They were saying, Are you doing an AI assessment for hiring someone now? So if you're hiring a creative person in UA, are you asking them to do a test which involves a p an AI component to to make something? And most people were going, yeah, we've factored it in. And my thing was like, okay, but You know, like a lot of these kind of like assessments and and offline questionnaires or whatever you have to fill out are being answered by AI now, right? Like the people are like to get the AI to do it. So I was like, I think you need to to strip it back to bare bones and have like a a session where they have no ability to use a phone, an AI or a calculator. Like I think you still need to check someone's
speaker-1: It's almost taking all tech away and seeing what the capable
speaker-0: I'll it's I c I've called it a caveman test. I'm like, this is the caveman test. Like we're in a room together, show me you can deal with a problem without assistance, you know?
speaker-1: But it it is, it's almost show me your thought pattern. It's l it it's
speaker-0: Yeah too.
speaker-1: understanding of how they approach problems in the first place or or you know, how they approach design in the first place it makes a lot
speaker-0: Yeah.
speaker-1: of sense.
speaker-0: I think problem solving is a very, very important skill in life in general, and I think AI is a tool. And so you should be able to like sometimes you can use the AI to help you come up with a solution, but I think more often than not it's better to think of a a possible solution and get the AI to refine it, right? That's the way I would like to to position it. But but who knows, dude. But look, the the the the big question for me is I still don't know How people are positively influencing the search results on AEO yet. And I think that's the thing that we're still figuring out as an industry and as not just our sector or vertical. I mean, I think everyone, like every business in the world, is trying to figure this out, right? So so that's what we've got to do next. We need to come up with like a playbook of like what are you doing to to to get picked up. But then look, Adam, we we launched a game last week.
speaker-1: We should get we should get someone on here at some point who has a very, very good understanding of all of this. Maybe maybe somebody from
speaker-0: But mate, I'll give you
speaker-1: Maybe like Sam Sam from
speaker-0: Yeah.
speaker-1: OpenAI.
speaker-0: but no, but mate, I'll give you one last example, right? Because we're gonna have to wrap up in a sec. But we we soft launched a game in Canada last week. We ran UA for seven days. At the end of the test was Monday this week, so two days ago. I then went on to go like Gemini, Claude, and Chatter GPT. Right? And I said, What can you tell me about the name of the game, right? And it's a very unique name, okay? Even it's not something that's generic. Yeah. and Obviously, we launched the game on on Google Play. So Google already had, like in Google Gemini, already had like something about the game. It was like, this is the games by Turbo, these are the things that it focuses on, these are like the the USPs, blah blah blah. But but Claude and ChatGPT didn't know about the game, right? So so obviously, you know, they they hook do are accessing huge amounts of information, but They are still not getting a hundred percent of everything, right? Like 'cause obviously we did a very small test of like five thousand users in Canada and and the LLM couldn't tell us anything about the name of the game, even though it's a very obscure name. Yeah. Whereas Gemini could because of Google Play. So Google Play is probably connected to it because all of the Google infrastructure is, and that's why it picked it up and give and it could give me an answer. Amazing. But but that also makes me think about scraping now, like for games and and you know, if people are building like prompts to be like, okay, what you know, what new games are being submitted to the store? Is there a trend? You know, at the moment we've got this like smash trend in hybrid casual, right? There's a smash fest and all this other stuff. I'm thinking are studios and publishers just writing these scripts. So now they instead of like scraping the app store, they're just using L L to do it for them and then they're saying this is the
speaker-1: Sure, they they trend right.
speaker-0: build a game. You know, in this trend
speaker-1: Yeah yeah yeah. Yeah. And and even understanding the industry, everything else, I I mean I've seen so much recently of of agents gathering information from all sorts of different locations and everything else. I think it's it's a really interesting time to be honest. I looking s forward to seeing what else happens.
speaker-0: For sure.
speaker-1: Cool. cool. All right. Well, thank you very much. We should we should do this the same time next week.
speaker-0: Yeah, exactly. Cool. Right. Thank you everyone. Peace. Bye.
speaker-1: His bike.
speaker-1: The ener the energy levels are slightly different here, aren't they?
speaker-0: Yeah, I mean I I I've been I've been a bit poorly the last couple of days, mate, so you might not see the best of John Wright today. You might see blackluster. Instead of the usual, you know, happy go lucky ADHD hyper person, you might see me leaning into my chair a little bit.
speaker-1: Well, I think I'm on my fourth coffee, so I I'll I'll I should be able to bring the energy for the both of us.
speaker-0: I'm I'm good. I'm
speaker-1: Why why are you ill, John? Why are you ill?
speaker-0: mate, I I think it's a combination of of jet setting last week. So I was in Barcelona for three days for the set of then which was which was great. I'll talk a bit about
speaker-1: How was that?
speaker-0: do you know what mate? It so very funny enough there were three events in Barcelona last week all at the same time. So there was Google's App Growth Summit, Global App Growth Growth Summit. there was App Growth Summit, which is a completely different event. It's like A c a completely different thing, even though like if Google and them have the same name
speaker-1: Were people there by mistake? Like did they get by conferences?
speaker-0: But but then there was also set Uetopia. And Very different.
speaker-1: Name. I like that.
speaker-0: Yeah, it it really, really good name, really good branding. We was in the Met Barcelona, which is up in the hills. Never been there. The most the best view of Barcelona I've ever seen. Like and all in all, mate, I've c I've come to a bit of a conclusion over the last year or so, which is like the the best events that I'm going to, the ones where I get the most from it, where I met the most interesting people, I get the most interesting insights or or information, are the invite-only events. Yeah. Anything that's like kind of mass like mass applied where you have to get a ticket. Like obviously you there's it's it's a different kind of thing you're going for. You know, you're going for for for quantity, not quality. But but the set event was amazing, man. Like I said, it was You know, it was only kind of a one day conference, but we we got there on the on the the night before we had a welcome dinner with all of the participants and there was you know senior directors, VP, CEOs from some of the biggest companies in in gaming. And and for and for once it was really tailored to marketers and and UA people, right? So everyone there was kind of a matter expert, which w is amazing because You know, sometimes mate, like I mean, me and you have probably done talks at P G C where we're doing a talk on rewarded and seventy five percent of the people in there are not not interested or applicable, right? Like you're
speaker-1: Yeah, there are other conference leaders. There are other conference managers.
speaker-0: But but like it so so for for once to have like an isolated group of serious professionals who know the the the subject matter and Yeah, exactly. So part some
speaker-1: Not recorded, I presume. That makes a difference. I mean, when you've got when you've got people from some of the larger like certainly public companies and that kind of thing, they're not inclined to talk about anything if there's if there's cameras there or if there's like p people, reporters in the audience and that kind of thing, are they? I mean I I certainly wouldn't be.
speaker-0: I i exactly, dude. And I think within the more intimate event, because again there was like you know, it was a effectively six talks, right? So there was breakfast, two talks, lunch, sorry, sorry, breakfast, two talks, break, two talks, lunch, two talks, finished. And then like dinner and stuff. And and each of the talks went on for about forty five minutes, so quite long. some of them were, you know, presentations and and lots of data highlighting a number of interesting things. predominantly was like focusing on the kind of AI creative brute force strategy which is which is ever apparent, right? What we're seeing is you know the percentage of cre or the number of creatives is going up, the percentage of hero creatives and and success to to hero creative is going down. And you know it was it was very interesting to see that we still got we still got
speaker-1: We said that before and we still haven't done it. So we need we need to
speaker-0: List to add it to the list. But do you know what? One of the most interesting talks, so there was you know, I'm not gonna name people by by name just because I don't know if they want that in if like to know they were part of this discussion, but there was me, someone very senior from Zingar, and someone very senior from Plateika. And there's a team on stage and they're presenting and they're basically saying that effectively more is better because the hit rates are so low now, if you don't test at a higher velocity then the likelihood of you finding a new winning creative every month and then having an incremental uplift is going to be very, very, very low, right? It might take you one in three months, one in six months. But, you know, interestingly, the the the three of us, so the person for singer, the person for Playton and myself, like we were sitting in the crowd and we just kind of jumped in. Because it was an intimate event, like we you know you can do that, right? When there's only like three rows of people, four rows of people, and they're sitting you know, four meters away, you can actually interject as and call bullshit. Yeah. And not that I was being that aggressive, but
speaker-1: That sounds kinda brutal, to be honest.
speaker-0: Do you know what afterwards someone come up to me and went, Wow, you went in and I went, No, I just want to create a debate, right? Like I I like debate, I like confrontation. I live for this.
speaker-1: Shit.
speaker-0: So I want to create a debate. and and and yeah, so basically I was going, the the issue is that because AI is is increasing the quantity so much, you know, you're not really reviewing the creatives that are being tested in the same way that you did previously, right? So yeah, you're creating more. More slop essentially. And it does and and slop doesn't mean that it's bad looking or ill performing or whatever. It's just it could just be random, right? Like it doesn't have like the the human-based input on every if you're building it was something like they were testing 700 playables a month, and I was like, there's no way that a person like is is involved in every one of those iterations, it's impossible, right? It If unless you're a huge organization which this company
speaker-1: That's a team of that's a team of people. I mean that
speaker-0: You you would need you would need fifty people, you know, to to to do this. and so, you know, what I was basically saying and what what the person Zinger and the person Plateak were saying is like if it was just if that was getting built by a dedicated team rather than being produced in AI, we think that if we did it that way, which is more human led, the percentage of to to success to HeroCreate would be much higher. So you wouldn't need to test as many because the qualitative approach would be better. Yeah, because it'd be more sort of human first interaction. And so we had a big debate about that. And obviously the team on stage were like, we understand where you're coming from, but you know, from our perspective, it is kind of like a brute force strategy now. You've just got to test as much as possible. And and, you know, we basically agreed to disagree. But I think there's really Adam, I think there's just basically middle ground, right? I think, you know, like testing You know, two thousand creatives a month is not effective because unless you have dream games level scale, there's no way the algos of the networks are gonna give enough impressions. You know, there's not enough time and and and impressions for these to be tested all effectively. And there's probably creatives that go live that never actually get the amount of impressions they need because there's so much, you know, going through the system. Because if you're testing two thousand craves and all your competitors are testing two thousand craves, And yeah, we spoke about this a couple of weeks ago about discoverability. So it's not just about discoverability in the atmosphere, it's about discoverability on the network, Adam, you know? Yeah. So so it comes back to all of that, mate. And for me, I would still go with like, you know, using AI to increase production, but I think that you still need a very strong human led strategic pick of what creatives, what concepts, what iterations and what. Don't let the AI do the creative thinking, essentially. But but yeah, mate, that was that was the big discussion.
speaker-1: I I like you said right, pick a subject. So I for today's conversation and bearing in mind we're already eight minutes into today's conversation. I
speaker-0: It was a great conference.
speaker-1: I wanna talk about AEO.
speaker-0: Okay. So app event optimization, we're going up we're going meta, yeah?
speaker-1: No, we're not. We're going we're going far away from creatives. So what what we've been looking at a lot or you are looking
speaker-0: Joking.
speaker-1: at a lot over the last little while is it's answer engine optimization. so actually looking through through AI apps and and chat GPT specifically for this one.
speaker-0: What you're effectively talking about, Adam, is more people are using LLMs to do search than Google now, right? People are are moving to asking.
speaker-1: Yeah, yeah. I mean, like it's funny, we've spent probably the last twenty years trying to optimize like search engine results, you know, as everyone knows, SEO and then AEO for the for the App Store optimization. ASO.
speaker-0: Yeah, so
speaker-1: It's been great. But I think we you know, now now with the way people are adopting Chat GPT or or there there are other AI agents out there. I think it's it's kinda changing and some of the results that we've been seeing are really interesting. And I I wanted to have a conversation specifically about I wanted to get your take on what you What you think of like I it the funny thing for me is it's like a you've gotta if you really want to succeed in this, you gotta you kinda gotta drop everything and and move into optimizing optimizing the the you know answer engine optimization. I don't like the term, but I think we need to find another term for that one. But I think if we do you as a gaming company and and gaming I've not seen being particularly strong on sort of chat GPT ads, bearing in mind it's very early days and and you know it's it's kind of experimentation time. But i it's kind of do you drop everything to try and optimize the the where you sit in those results of, you know I wanna find another game that's like I like XCOM and I don't know, Midnight Suns. I wanna find another game that I enjoy playing. Do you try and optimize your your where you land in that now? Or do you it'll be way too early? What do you reckon? Is is it something that we should look at?
speaker-0: Do you know what, mate? It's actually we we did a big exercise on this about three months ago where we we went to both Claude and Chat GPT and we just asked it a bunch of questions, right? And so like it was like what is the kind of best motocross game on mobile? And and fortunately fortunately for us, like Mad Skills Free, our biggest game came up first, right? But we also asked it a bunch of other questions, like What are the top sports games on mobile? Yeah. What are the top racing games?
speaker-1: So when you when you were very specific, it came up with yours, but when you kind of went broader it sort of threw in other stuff.
speaker-0: Exactly, dude. Yeah, we we we we we obviously when it came down to motocross games as the the biggest name in motocross games, we were very fortunate to get the the larger success rate, right? In terms of people actually being fine like finding links to our games and stuff like that. So we completed this assessment and we were really happy about when we were focusing on it from a very niche perspective. But the problem is we already get a large amount of organics you know, through through Apple and Google. We're already well known.
speaker-1: Seems be right there. I wanna I wanna dive into that or something.
speaker-0: And then but the and the games are super the games are super well known within the motocross community. So what we're what we're trying to do is find users outside of that golden cohort, Adam, right? Like so yeah, it's great. Anyone who knows about motocross or motorbikes or turborilla is gonna come to us. That's great. But those are the people they're gonna
speaker-1: That don't you want to try and find.
speaker-0: Exactly, mate. It's about widening the net, right? And increasing the audience size. And so from in from that perspective, obviously optimising towards having, you know, higher rankings in in sports games globally, etcetera, would be the thing that we would want to focus on. arguably, you know, capturing those people in the kind of you know the ideal cohort or ICP effectively means that their the performance would be higher. But again, they wouldn't come to us anyway. So it's it's better to to widen the net. The the issue is there's not a really a lot of understanding of of how to do this right now. Like how do you actually No like there's
speaker-1: Yeah.
speaker-0: there's there's there's one particular company called App Tweak where they've got a product which allows you to understand where you actually come up in the search criteria. So if you again if you ask a question
speaker-1: So what what people have been searching basically?
speaker-0: Exactly, exactly. So it's the same way that you would use Google to understand where you are in in in search ads, right? Or or AdWords, sorry. And and so they give you some some
speaker-1: 'Cause I'm presuming like Google it changes based on Geo and all sorts of different
speaker-0: Yeah, of course.
speaker-1: information they have about you and everything else. I'm I'm guessing it's not the same answer for everyone.
speaker-0: Well the the th the thing is with like Google and and and you know stuff like that's looking at the metadata on the website and the relevancy score and stuff like that, right? So how relevant is this website to the question that's been asked and there's there's loads of things that go into like why you are higher up on the search criteria after the paid section, right? So you've always got those first three links which are paid and then you've got organic first. Yeah. And and and normally like look, if you're talking about SEO in particular. If you're not on the first page, your discoverability isn't nil, right?
speaker-1: I was discovered a long, long time ago if you're the first page of Google results, yeah.
speaker-0: But the but the the fact is that'll be applicable to AEO as well. Right? Like if you're if you're not like people's attention spans have never been so small, but they want instant gratification, they want instant up So like
speaker-1: And and chat
speaker-0: they're gonna click on the first thing.
speaker-1: It normally gives three to compare, like three it normally finds three like if you ask it about a product or if you ask it about a game or you know, I suggest a a game that I could play next with I enjoy X, Y, and Z. It normally gives you three to choose from or three options. And I guess to if you're not one of those options, that kind of rules you out of it completely, right?
speaker-0: E exactly, dude. So whereas before you might get people scrolling the first ten options on Google, again now the focus is I'm actually just running a a query now to to prove this. But it's essentially, yeah, I would say if you're not in that
speaker-1: Who's the best MP?
speaker-0: I'll do it after. We'll do it I'll I'll I'll do it in real time. We'll see what happens.
speaker-1: I'm a little bit scared now, but
speaker-0: It's not apps flyer, we're fucked. Okay, like shout out to our sponsor.
speaker-1: We'll have to cut this bit.
speaker-0: Thank you for sponsoring. It's it's still running, but when it when it pops up, I'll let you know what it says. But but yeah, dude, like you know, so so App Tweak again is the only product that I've heard of in the market which is actually allowing you to understand where you are in in like the hierarchy, you know, for the 'cause 'cause remember, like each product
speaker-1: Do you think that's gonna come in more and more? Like an analysis or
speaker-0: I think I think an analytical tool that lets you understand your position in AEO is gonna be very important in the future because you're gonna see a shift of people like I don't have the numbers, right? But I would say in the last two years, people that are using Google and and demographically as well will be very interesting, but the amount of people using Google for search versus L and for search, there has to have been a migration. I also know when I speak to people from like Gen Z, etc. Is it Gen Z? Yeah, Gen Z. They're the the younger ones and what's what's your below millennials? Gen Z.
speaker-1: Anyway, let's let's go with Gen Z.
speaker-0: Gen Z? think they're called Gen Z. Gen X? I Gen X is the older one, think. Anyway, whatever the younger denig the young the younger generation than me, right? If you look at the percentage of people using LLMs for for for like using LLMs in general, obviously there's a higher adoption of AI in younger people anyway. So there has been a bigger migration from from traditional SEO to AEO, i.e. the people using LLMs to get answers, right? Yeah. And
speaker-1: So imagine that.
speaker-0: what's the best motocross game on on mobile? And it came out Mad Skills Motocross free would be my overall pick. Particularly for a competitive physic based racing on mobile, it's handling, track mastery, and multiplayer events make a strong case. Though best is ultimately subjective. And then it went two alternatives Dirt Bike Unchained by Red Bull and SMX, which is a Supermoto versus Motocross. So like again, so that's it's giving you three options, right? And
speaker-1: Yeah, and if you're not one of if Yeah, but if you're not one of those three, then you're you're nowhere now. So it's not strategic
speaker-0: And also.
speaker-1: results, it's it's like the top three.
speaker-0: And but again also you've got to look at the language, right? Th the there was a strong, strong insinuation for mo MSM free, right? Which I'm very happy for because it's my game. But but then it's giving you the alternatives just in case maybe you already have MSM free. But the language is super strong, Adam, in terms of like it's it's it's funneling you to that top choice.
speaker-1: Also I think like like you were talking about, what was it, Gen Z? Gen Z?
speaker-0: We'll go with Check it out.
speaker-1: What you were saying there, I mean, I've I've met a lot of people who are using it almost as a as a friend now. Like through through anything that they have to deal with, they they'll start talking to like messaging it as a friend. And then if obviously if your friend's suggesting you play this game or this game, you're more inclined to play it because it's from a it's from a seemingly trusted source, which is a bit strange, but
speaker-0: It's what is acting like a companion? Yeah. Like Yeah. Okay. Good good news, Adam. What's the best MP in mobile games? Apps Flyer would be my default pick.
speaker-1: Well that was lucky.
speaker-0: Yeah, and then adjust singular. Which no no surprise. No surprise. Phew
speaker-1: Yeah.
speaker-0: So someone's been doing some AEO at Apps Flyer.
speaker-1: Or or we're just incredibly lucky. It could be either.
speaker-0: But I I I do you know what dude the thing is I
speaker-1: What do you think about this whole like you know, you've got you've got the AEO side of it, which is the equivalent of SEO, but then you've also got the ads side coming into stuff more and more now. Like, do you think it's it feels like it's it's very, very early days and it feels like it's it's you know, there's there's that experimentation that needs to happen with it and everything. But would you focus on it right now?
speaker-0: I don't know man, it's it's a case of like I don't I don't know if I've seen any ads up to yet. Is it like it's probably not live in the UK, it's probably just live in the US or something like that now. So but it will be rolled down, right? Yeah. Because they need they need money. Like all of these big labs are hemorrhaging money, right? Like Anthrophy owners like six hundred million in debt or something, right? So it's a case of all of these companies need to start
speaker-1: got the user base. So it's y you know, it's like it's like Facebook back in the day. You know, Facebook was was huge monetization.
speaker-0: But they've only got a user base 'cause it's they've only got a user base 'cause it's cheap, right? Like it's it's value it it's it's always about value versus cost. And at the moment the value is very high and the cost is very low. If my clawed ninety nine dollar a month you know membership or whatever co like s you know agreement was to go to a thousand dollars, I I wouldn't pay it, Adam, right? It's only because for a hundred dollars I'm getting so much value. So yeah, if they start increasing the price and the same as with with ads like adding
speaker-1: You'd have to go back to syncing again.
speaker-0: Yeah, I mean yeah, I I still believe
speaker-1: Yeah, but are they under the assumption that you can't that you would struggle to go back?
speaker-0: I mean I I wouldn't 'cause I I believe in we have to I had a big debate about this the other day about like, you know, not losing the ability to to have like to to to think for yourself and and be cognitive, right? Like
speaker-1: We did we did with satnav. We let we let satnav come in and now people can't read maps, like the Gen Z as you were saying. So trendy.
speaker-0: Yeah, look look, man. Yeah, look, I just I just I don't know, like I wouldn't I would like someone was saying the other day about it was part of the set off site, right? They were saying, Are you doing an AI assessment for hiring someone now? So if you're hiring a creative person in UA, are you asking them to do a test which involves a p an AI component to to make something? And most people were going, yeah, we've factored it in. And my thing was like, okay, but You know, like a lot of these kind of like assessments and and offline questionnaires or whatever you have to fill out are being answered by AI now, right? Like the people are like to get the AI to do it. So I was like, I think you need to to strip it back to bare bones and have like a a session where they have no ability to use a phone, an AI or a calculator. Like I think you still need to check someone's
speaker-1: It's almost taking all tech away and seeing what the capable
speaker-0: I'll it's I c I've called it a caveman test. I'm like, this is the caveman test. Like we're in a room together, show me you can deal with a problem without assistance, you know?
speaker-1: But it it is, it's almost show me your thought pattern. It's l it it's
speaker-0: Yeah too.
speaker-1: understanding of how they approach problems in the first place or or you know, how they approach design in the first place it makes a lot
speaker-0: Yeah.
speaker-1: of sense.
speaker-0: I think problem solving is a very, very important skill in life in general, and I think AI is a tool. And so you should be able to like sometimes you can use the AI to help you come up with a solution, but I think more often than not it's better to think of a a possible solution and get the AI to refine it, right? That's the way I would like to to position it. But but who knows, dude. But look, the the the the big question for me is I still don't know How people are positively influencing the search results on AEO yet. And I think that's the thing that we're still figuring out as an industry and as not just our sector or vertical. I mean, I think everyone, like every business in the world, is trying to figure this out, right? So so that's what we've got to do next. We need to come up with like a playbook of like what are you doing to to to get picked up. But then look, Adam, we we launched a game last week.
speaker-1: We should get we should get someone on here at some point who has a very, very good understanding of all of this. Maybe maybe somebody from
speaker-0: But mate, I'll give you
speaker-1: Maybe like Sam Sam from
speaker-0: Yeah.
speaker-1: OpenAI.
speaker-0: but no, but mate, I'll give you one last example, right? Because we're gonna have to wrap up in a sec. But we we soft launched a game in Canada last week. We ran UA for seven days. At the end of the test was Monday this week, so two days ago. I then went on to go like Gemini, Claude, and Chatter GPT. Right? And I said, What can you tell me about the name of the game, right? And it's a very unique name, okay? Even it's not something that's generic. Yeah. and Obviously, we launched the game on on Google Play. So Google already had, like in Google Gemini, already had like something about the game. It was like, this is the games by Turbo, these are the things that it focuses on, these are like the the USPs, blah blah blah. But but Claude and ChatGPT didn't know about the game, right? So so obviously, you know, they they hook do are accessing huge amounts of information, but They are still not getting a hundred percent of everything, right? Like 'cause obviously we did a very small test of like five thousand users in Canada and and the LLM couldn't tell us anything about the name of the game, even though it's a very obscure name. Yeah. Whereas Gemini could because of Google Play. So Google Play is probably connected to it because all of the Google infrastructure is, and that's why it picked it up and give and it could give me an answer. Amazing. But but that also makes me think about scraping now, like for games and and you know, if people are building like prompts to be like, okay, what you know, what new games are being submitted to the store? Is there a trend? You know, at the moment we've got this like smash trend in hybrid casual, right? There's a smash fest and all this other stuff. I'm thinking are studios and publishers just writing these scripts. So now they instead of like scraping the app store, they're just using L L to do it for them and then they're saying this is the
speaker-1: Sure, they they trend right.
speaker-0: build a game. You know, in this trend
speaker-1: Yeah yeah yeah. Yeah. And and even understanding the industry, everything else, I I mean I've seen so much recently of of agents gathering information from all sorts of different locations and everything else. I think it's it's a really interesting time to be honest. I looking s forward to seeing what else happens.
speaker-0: For sure.
speaker-1: Cool. cool. All right. Well, thank you very much. We should we should do this the same time next week.
speaker-0: Yeah, exactly. Cool. Right. Thank you everyone. Peace. Bye.
speaker-1: His bike.
