Big Huge Games: Tim Train

Today we are joined by Tim Train, CEO of Big Huge Games 2.0. Tune in for some Big Huge Lessons on the difference a great LiveOps strategy can make.

Transcript

Copy Transcript Copied

Speaker 1 0:05

Hello, I'm James Gwertzman. I'm Cristin Cox . Welcome to the art of live ops podcast.

Speaker 2 1:09

[inaudible]

Speaker 1 1:10

Hey Kristen . Hey James. So today we're talking to my good friend Tim train. Yes . Big huge games. I am Tim train . I'm CEO of Big Huge games. The second incarnation of that studio we make dominations which should just past its four year anniversary. We just announced 50 million installs on the title and he, he's had an interesting journey down the path of live ops. Yeah. I mean he was, you know, a founder for big huge games, which had a lot of great success as more of a traditional product focused game company for many years. Then sadly shut down. And then he went to Zynga and got his hands dirty with live ops and got his hands, you know, really figured out what , what that sort of social mobile live ops story looks like. And then came back to big huge games and restarted it. And their game dominations dominated. I mean they had a really big successful launch. Yeah, I think it would be really interesting talking to him about what he's seen as far as working sort of on both sides of the fence as a developer that was not focused on live games, transitioning to a developer, totally focused on live ops one especially because you know there , thanks to the success of dominations, they were acquired by Exxon and next one we all know it was one of the leading companies globally in, in doing live ops effectively on PC, on mobile now. And so it'll be interesting also to hear his perspective on what, if anything has changed now the depart the next on. Yeah, let's dive in. All right.

Speaker 2 2:44

[inaudible]

Speaker 1 2:44

well, I, I wanted to talk to you today because you had spoken at a mobile cabal meeting, I think it was three or four years ago, and you give a talk on sort of the live ops, you know, you kind of report on live ops the first year, I think it was a nominations and I remember there was a chart in there where you had been showing how your usage as we said, it's going down and you added live events and everything. I turned around and I still remember that chart. I still talk about that to my colleagues as you know, the impact of live ops can happen in a game . And it was really my first kind of really wake up call to just how important this stuff could be. And so I'd love to hear you if you don't mind, kind of retell a little bit that story and what happened with that first year. Sure. Yeah. So yeah, the way to answer that question is to talk a little bit about the history of the second incarnation of acute games. Uh, we founded the first incarnation of the studio in 2000 , uh, with the founding group , uh , some of which are in the second incarnation , uh, ran that studio until the , uh , a little bit before the Curt Schilling implosion shut the whole thing down. Uh, I had spun out to Zynga East , uh , before then , uh, before the actual implosion and we had developed frontier Ville and after the Kirk Chilean pollution , we then , uh, there were the remnants of that studio went on to found an Epic game studio. And both the Epic studio and the Xingu studio closed within a couple of weeks of each other. This meant that we had about 110 developers who are all on several months of severance. And so we were able to grab up a lot of the people that we really wanted to work with a , and then we founded the studio that would evolve into big, huge games. Second version. Uh , and so, you know, our angel round of funding basically came courtesy of the HR departments of Epic and Zynga, which was , was really nice. And so when we started the studio, however, we had a very clear vision for the product that we wanted to make, which for very early on was be the age of empires for this generation of strategy games. We've been making strategy games for a very long time. We'd seen several weaves of them come through the market and we knew that there would be a million fantasy versions of a clash of clans, the kind of dominant a game in the combat city builders genre at the time. But we knew that we could make the history version the through the ages Virgin of a combat city builder, and that we knew that it would find an audience. What we didn't know was anything about unity mobile set asynchronous PVP or having a game that was actually something that people would care to hack a, this was not a thing that we hit that we had a really had to deal with very much before in our, in our previous career. Uh, and so we started the studio with I think 10 people , uh, many of whom had worked for the original big huge games. And the rest of the work forbs and get yeast . Uh, and it took us about 18 months to get the game into soft launch and another six months to actually get it, you know, all together our numbers were really good and we were really excited about it. And we came out and , uh , this would've been almost exactly four years ago. Uh, so , uh, what does that 2015 now if I'm doing my math right? Yeah . And uh, we had double featuring from Apple and Google back when that meant something , uh, and shot up to a giant number of DAU. I think we might have maxed out at something like two, two and a half million DAU. But our tech did not stand up under scale. And so we had all kinds of crashes and, you know, server problems and downtime. We had hackers who were hacking into the leaderboard and changing everyone's name to their clan name as a cute, funny joke for everyone to wake up to and start panicking about. So we spent that first six months after the game launched, trying to fix all the stuff that we didn't know. We didn't know when we, when we launched the game. Uh, and so, you know, people were, that was the one time in the studios , uh, uh, history , uh , to that point that we actually had any kind of crunch because when you wake up in surprise, everyone's a name is changed to the name of this particular clan . Uh, it tends to lend a certain urgency to your efforts that you know , uh, doesn't allow for, for work life balance. Um, and at the same time we really wanted to get the game out of Asia. So we were also working hard towards building a version of the game that could, that could work in Asia. Uh, and we had also not quite hit feature parody with some of our key competitors. And so we were working on a key feature called world war that was clan versus clan battles Alliance versus Alliance battles. Uh, and so , uh, that was a very busy year for us. Right . But that's what we did for the first year after we launched the game. Then once we felt like, okay, we've got to get me out Asia and like we got all the fires under control and we still had like all these people who were still religious to playing the game, our retention numbers special, longer term retention numbers were really good. Uh, and so then it was kind of, okay, how do we keep these players around? How , how can we start thinking about this game less in terms of two week chunks and more in terms of how is this a game that's around for 10 years? How is this around for 15 years? And so that was when we started to really look at what the kind of state of the art was in live ops in the industry. Uh, and we started building tools and features that would really help support live ops. And so one of the things that was that, that, that we failed out . I talked about this in the talk that you're referencing. We hit assumed that if we copied the business model of the best competitor that we would get their numbers and that, you know, we would, we would see to say , you know, just by kind of having cool systems that people wanted to engage with on a long time, that we would see the kind of revenue that they saw this turned out not to be true. Uh, and some of that may be scale that they got their first test with the most dist , uh, and just, you know, grabbed up a bunch of audience. So it made you that they were just, you know , better at kind of weaving together tight systems, you know. But for whatever reason, we did not see the same kinds of numbers even on a per user basis as our competitors. And we probably were too generous with the game economy such that dominations has really great longterm retention, like really good longterm retention. But it doesn't monetize that well. And so we kinda , you know , keep people around for a long time and that's great. And there's a lot of them. That's great. But boy, it sure would be nice if people you know , actually would like to pay us a little more money for that.

Speaker 3 9:12

Uh huh .

Speaker 1 9:14

And so when we started building out live ops, we had a few goals. The first one was to be sure that people had a lot of reasons to keep coming back into the game on, you know , a couple of times a day, several times a week, and you know, extend kind of those, those engagement loops throughout, you know, a full month of time. Uh, the second goal was to really move people around the box of the feature set of the game. And so that means that like, well, if people aren't engaging in this, we want to be sure that we're kind of pointing them to the direction of this feature that they may have forgotten about or that they weren't, that , that they weren't that interested in. Uh , and then the third thing was to make a little more money as we, as we , uh, you know, tried to be sure that we were economically viable on a long period of time while UAA costs were continuing to rise. And so the UAE LTV arbitrage problem had not quite manifested itself, you know, three or four years ago. But we , you just keep looking at the CPIs going up and up and up and you kind of have new choice if you're going to be viable, but to, you know , continue to try to, to , to figure out ways to give value to players that they are willing to pay you for. So , um, so yeah, so then we started building all these lives, live ops features. We sorta knew what we were doing, but we didn't really know what we were doing at the time. We kinda, you know , started with weekly events that would give out cool units and things like that. Uh, we, the world war system itself was a kind of lives or repeatable event that , uh , you know, players could engage with on a 48 hour cycle if they were just going to blast, you know, from one world war to the other. Um , [inaudible] and so that's when you saw the numbers really start to rise. They still didn't rise enough to outpace the UAA LTV arbitrage problem. Uh, but they at least kept pace with it such that, you know, we still have 35 people , uh, working happily on the game to continue to tell the story of human history. So nice as you're going through that journey , um, a lot of the dads we talked to who are getting started in live ops , um, get kind of tripped up by analytics. You know , they either start just looking, they collect all the data, but they don't know what to do with it or they're not sure which things to look at. What did you guys find most valuable? What did you guys learn about looking at data while you're doing all that? So we were very fortunate that one of the 10 people that we got early on , uh, was a woman named [inaudible] who is our lead product manager, lead data analyst . And she's pretty much would be in the conversation for one of the industry's best analysts , uh, and was a really fortunate thing for us and was really meaningful as we tried to kind of, you know, figure out what to do with all of these ho Jillian's of, of DAU that we're, we're trying to again, figure out how to keep interested in the game for, for a 10 year long timeframe. Um, so it's, it's a little bit hard to tell new devs that like, yeah, just go find one of the industry's best analysts and , uh , then you'll be fine . It's good advice if you can take it. Right. Exactly. Uh , I guess one of the things is that there is a little bit of diminishing returns at how deep you look at analytics, which doesn't mean that you, you know, diminish returns doesn't mean zero returns. It just means that you can kind of spend 80% of your time chasing information that isn't really gonna yield super meaningful results. And that for most people , uh, you know, just looking at your basics, like what features are users engaging in? How do your users segment themselves , uh , how would you target, you know, particularly sales and things like that to people. Like you would get a good chunk of the value of data analytics just from looking at those things. And yes, you could go super deep into player segmentation and likelihood to churn models and all these kinds of things that are very complicated and that do require any industry best data analysts to, to really , uh, figure out. Uh , but it's not a thing that, that , uh, everybody you know , needs to do, especially not right out of the gate. Or if you do, you probably have bigger problems then you know, how to optimize for an extra, you know, five or 10% , uh, that you might get out of a , a super deep data dive for . Right. [inaudible] just mentioned , you know, one of the industry's best analysts, and I'm sort of, my curiosity is piqued by that. So like what , what does the world, what does one of the world's most analysts look like? You know, so what one of the things that's come up for example is that analytics is often knowing what questions to ask. Yep . And I'm guessing them that this, this person, your team is probably asking really good questions. Like what are some examples or can you think of any like insights that came out of out of her or her team around things that were going to help make the game better or different or maybe things that were not necessarily intuitive. Right? Uh, a good chunk of what an industry best data analysts does is be able to look at the data dispassionately. And you know, when gen hires on new data analysts, you know, she actually would consider game , uh, experience and background and a passion for games to actually be something that is maybe even a net negative. Uh, and she really looks for it . I think that this is very smart on her part looks for people that are , uh , uh , they just love data. They're just really just passionate about. Because I think it's really easy if you're a game player to any great degree that you're going to get in there. And you're going to be like, well, I got , what I want to see is that players like this really fun feature that, you know, may or may not generate a lot of revenue because I liked that feature and I really like playing that feature . And so Jen , it really emphasizes that you need to look at data as , as kind of this holistic a piece of the puzzle. So that's one of the pieces that, that uh, inform step one of the other things. This isn't exactly answering what, you know, genius insights did. We did we get out of the data, but Jen is also got a really strong technical background, so she understands how the data is gathered , uh , in ways that have turned out to be super important in various ways. Like when sometimes , uh, our publisher next on might look at the data and they'll come up with different answers to certain questions than we would. And a lot of times that's actually at a , how is the data collected, you know, on the ground or whatever metaphor you want to use. And because Jen really understands the pipeline soup to nuts, she's able to kind of tease out, well , where is it that they're seeing different kinds of data and maybe they're right or maybe you were right and they need to, they need to know that. So there's that. Another facet is that Jen is really good at understanding how to visualize data and transmit and communicate data to people. So, you know, one of the things that she talks about is , is her burning he treated of pie charts , uh , as a way , uh , that, that often data gets misrepresented. And that is not a particularly useful, the statue look at that. So she's been really good at educating the whole studio even when sometimes the studio hasn't necessarily wanted to be educated about a , about all of the, the aspects of that, but about how data can get visualized , uh, and, and how you then, you know, draw inferences from the data. Right. So I'll

Speaker 4 16:09

capturing kind of the early days of free to play, coming to casual gaming. And it's interesting because there is that we were in a, I'm remember jewel blitz was the first free to play game that PopCap worked on and it was interesting because we had a real tension early on between us, call it the live ops team and the game design team. Cause the games on team was used to a very driven, very creative approach to design and the animal and the, and we're making decisions that would take several weeks to add features. It might be really cool and I heard cool lot. And then the live ops team meanwhile was looking for optimizations like Hey guys, we can boost number X by by Y percent if we could just make this change, you know , quick , please make this change. You're leaving money on the table. And then there was a sensory interesting kind of caused a little bit. We eventually resolved it but I think early on different, they're being driven by different motivations and that was causing conflict. Did you ever see them ? How do you , how do you integrate data into, or how do you integrate some of these insights into your pipeline so that you feel like you're making the right decisions ? The studio ,

Speaker 1 17:10

uh, that is a thorny question. Uh, I, you know, cause I, we kind of cut our teeth at Zynga when they were also figuring out , uh, things about live ops and free to play and how to, you know, make a, a product, product, mood ties, that kind of game design , uh, in ways that were sometimes ultimately harmful to their, to their overall ecosystem. But , but very often, you know, found new ways of, of looking at how to engage people. Um, and so there's kind of the ideal answer to this. And then there's the practical answer to this. And the ideal answer from my perspective is that people who are great product designers, like if you're just looking at it like this is a product, should be equal measures, awesome game designers and awesome product people. And that's, you know, that would kind of be the ideal that you have the quiz Zack hotter rack of product and design , uh , that can, you know, that can, that can lead you forward in that way. Um, the reality is that people coming from a product background and people coming from a game design background are very often different people that the game design people, you know, a a failed their econ final because they were up playing D and D all night or whatever the latest video game release was. Whereas the , uh , product people may have got their MBA because they're super hard charging. You want to be successful at whatever they do. And just that mindset makes it very difficult to have the one true person that can, that can do both. But if you set as a standard for your studio culture that like, Hey, like we shouldn't have this tension. Like, we can acknowledge that there is attention and it's finally when you talk about it openly and whatever, but the ideal is that you product people get better at understanding what actually makes a fun game and why is it the people want to be in this game for four years, five years, 10 years down the line, and that you design people get to the point where when you design a feature, you're just , it's second nature to be like, well, how do we, you know, drive whatever a, a metric that we're trying to, to drive with this, whether it's retention or a engagement or revenue. Uh, and so that's kind of how we've resolved. We acknowledge there's tension. Sometimes the game designers feel like the product people are pulling too hard on , uh, the revenue and sometimes the, you know, the, the reverse is true that, you know, the game design people are not thinking about money at all. And that's just crazy, right? Uh, but, you know, we've kind of, we've muddled our way into progressively better places over the years that we've been doing , uh , live up .

Speaker 4 19:39

I totally agree with that. I think I've found on my team's pairing together a great designer who's really gonna take care of empathy for the player and thinking about fun with a great product manager. That's where you get the amazing synergy.

Speaker 1 19:56

Yup . I had a , a meeting here at GDC earlier this week with, with Elka from Supercell and I had a bunch of questions for him. There were mostly about the nuts and bolts of how they build their teams. You know , I've read the interviews where they kind of talk in broad strokes about, about their philosophies, but I, you know, some of that stuff I'm like, really? And so one of my, you know, one of my questions for him was just like, so really you don't have product managers? Like there's just no product managers at super. So, and he said, he basically said that no, but there always is one person on the team that really has a knack for monetization. And so I don't know whether that's just, you know, they all have just, every single person is amazing because of the water in Finland or something like that. Free day . Yeah. Yeah. Must must be that they, they breed them from birth or natural economists I guess, but, you know, anyway, but he acknowledged that was still the person who had to be the voice of how is it that we make money on the team, even if they didn't call them a product manager, maybe if their background was more of a design side of things. So that was, you know, an interesting kind of subtlety to add to the, you know, we have no product managers at Supercell. Right. So this , this is a follow on that one more question about that. I mean, you, you guys are now part of the Exxon and next on obviously has a huge background in building these kinds of games from, from Asia. How has that, or has that at all affected your, your team, your culture? Are you learning? Are the lessons being transferred over from next on that you have sort of adopted in , in, in , uh, in pride into your team? Uh, that's been an interesting dynamic was next on, partly because Asian games seem to run in kind of slightly different ways than Western games. Uh, and there's a bit, there's a bit of a tension there between , uh, cause domination is one of the very few , uh , Western games that's crossed over and done well in Asia. Uh, and part of that we think is because a history game looks like a history game anywhere in the world and that people know who Cleopatra is regardless of where they live. And so there's a universality to history. That's one of the reasons that even though it's never gonna drive the numbers that that fantasy does, but that it's always going to have a cool audience of people there. So I'm sorry. Domination is one of the few, few games that's actually made the jump to Asia in a sustainable and successful way. Um, but Asian games very often are built around like the monthly bumps of the mobile phone cycle where, you know, you kind of get some new premium currency on the first day of the month and then the design of those games is then built to kind of suck all the excess currency out and they run a bunch of live hops and sales and things like that. Right around the first of the month. We don't really have the kinds of sinks and we were a little more now, but, but you know, definitely when we, when we started, we didn't have the kinds of sinks that would allow for a bunch of different ways for people to spend all this new premium currency. And so , uh, that is, was kind of a challenge for us. And , and it was, it was a , you know, very interesting thing to wake up , uh, you know, on the first of the month and all of a sudden we see a spike in our revenue where we like, what's going on. We didn't run any, we didn't run anything. And then like, after a couple of those say, Oh, it's the first of the month in Asia, this is, this is, you know , thing that naturally happens. Uh, so that's a thing that's been really different. Uh , the other thing is I think that because there are all, they build in all of these currencies that they can, that they can, you know, use to, to kind of dream, dream , uh , different, different economies with , uh , it means that when there is a customer service incident, they're much more free with like, here's $10 of premium currency because $10 of premium currency is not actually, you know, $10 of, you know, actual meaningful premium currency. For us. $10 per human currency is 10 actual dollars. And like we can't just give whatever a hundreds of thousands of $10 in premium currency every time we have a 20 minute outage. Right . Which would be more the kind of customer service way and you know, then that can cause some conflicts around you're not treating the customer right, which from the perspective of how live games are run in Asia, you know, might be correct. We just can't, you know , cause we also, the other thing is that we can't give, you know , $10 only to the Asia players. You know, we have to do it worldwide or people feel like there's, there's an unfair advantage. And so we're just like, we can't wreck our economy for a full month, you know, a a bit to try to , to try to conform to, you know, one segments of the player base is a customer service expectations. So that's been a , that's been a tension and again, it feels like that was a thing that we just didn't know. We didn't know and built the game. Uh, and I've kind of learned and for our second game that , that we hope we'll be able to talk about , uh , more soon. Uh, you know, we've tried to really be much more mindful about being sure that we're able to have all the levers to, to , uh , to, to, to be a great global game in that way. I'd love to go back a little bit. You've mentioned it like the things you didn't know, you didn't know. If you could just highlight a couple of those things that now when you look back, if you could go back for it four years in time and talk to yourself, what were the, what are the things you wish you could tell your ear or for yourself ? It's a good question. Um, well, the, some of the foundational mistakes that we made were that we probably needed a little more server expertise , uh, when we built the team. And you know, if you guys have been a in the job market or you know, a pink tissue, the John Merck , it's true for engineers are hard to come by everywhere these days. Um, but one of the things that was really a mistake for us was that we built a client authoritative model. Actually, we didn't actually build it. We used a piece of tech that kinda came to us as part of the next on deal that had been used for a previous game, but a game that had never really scaled that far , uh , as far as we did. And so that tech was super rickety. And , uh, it was, it was not server authoritative, which it should have been for, for any kind of, you know, security purposes. Uh, and that, that, you know, we're still paying for that , uh , for that decision today. Um, but that was a, that was an unknown unknown. Yeah. Um, one of the other things that we were probably a little bit idealistic in thinking that when we designed our economy that if we made the game feel more free that we would drive sufficiently meaningful retention changes that , uh, we would, you know, we, we would see a good, good return in that way. Um, so one of our year where we had our, our , uh, our , our product pillars, you know, we actually had really early on in this game, sometimes the product colors have to emerge from your prototype process. But we had a really clear vision of what we were making. So one of our pillars was be 20% more free than the competition. This was when games like clash of clans and other things had been in development for awhile . Right . And so, or sorry, live for a while . And so their economies, as you got deeper in the game, just, you know , became progressively more grind . And so we were like, okay , well we're, you know, if you're going to compete on price features or quality, right marketing one Oh one pick one of these to, to compete. And we're like, well, you know, we think we're gonna make a great game but we're not really going to be able to compete on features because they've been out for two years and have just been able to, you know, to continue building on that. So can can be done that, what are the things we can compete on is price so be 20% more free in the competition. And that's obviously a completely hazy, field-based thing. There's not like, we just are gonna apply a 20% algorithm to everything we have . I guess we could do that, but that wasn't how we're thinking about it. We're just thinking about it like, let's be a little more free than the competition. It turns out that what that meant was the actual bullet point turned out to be, make 20% less money on a per user basis than the competition. But that would be the mathematical conclusion. Exactly . But the, the, you know, there , there's a little bit of bet I talk to people all the time about this. At this year's mobile cabal , I spent like a half hour talking to some very passionate , uh, guys who were running a new startup who had the exact same , uh , you know, theory that we did and rise actually likes , they said , Oh, Tim train , you must talk. Uh , because, because you , uh, cause we, we, we, we faced the same basic thing, but there's this idealism that you want, right? That players, you know, will kind of just, you know, hand you over more money just because your game feels like it's more generous and more free. And it just turns out that, you know, when you just look at the data and the people that are actually paying you money, there is no segment of people. There does not a larger meaningful segment of people that is willing to pay you not that much money. Right. Right. So you know , the people are either willing to pay you a lot of money, right ? Or you might get like your starter pack people who you know, buy a really good valuable thing at the beginning of the game but never give you money, you know, anywhere else. And you do have some people that are in that in the , you know, in kind of the middle of the middle bracket that way. But you can't build a business model out of them. Right . Unfortunately they're kind of more tertiary, you know, revenue than they are than they are meaningful to your bottom line. So that's the thing. I wish I could be like, you should be less idealistic. And yet that came out of what you were saying just now. I want to kind of emphasize point is we've also heard game. Oftentimes you hear people talk about monetize your whales and really, you know, you know, and I think there's often, especially early on, there's almost this maybe the attitude of like your players are like walking bags of money. Your job isn't to extract the money. And I didn't hear you saying that. I think that that's a thing that we've really kind of a theme has been developing over these conversations is the successful games really are thinking about how do we provide great value that our players get really, really enjoy, you know, wanna spend money on, but it's because they're really getting that value out of it. Yeah. Uh, so rate up as a random thing. You know, we basically tried to ban the use of the word whale in our studio. Yeah. And so, you know, we don't, we don't like it for a number of reasons and the main one is because it's disrespectful to our, to the, the people who are, you know, are kind of most important customers. So we call them high value players , uh, which we shortened to HVP, which has an unfortunate dyslexic, a way of turning into HPV. Sometimes in a , in our product review meetings, oops. Ah , yeah, exactly. But yeah, HBPS is what is what we , uh, what we try to, we try to refer to the people internally that way. Sometimes people coming from other, other studios , uh, you know, have, have , they have a little bit of a cultural adjustment there. But anyway, we think labor's important in that way and that and that we wanna you know, we wanna we want to STEM that. Um, so some of it is just being sure like, yes, we're designing features that are like, Oh, this is going to really appeal to somebody who likes paying a lot of money and , and getting their value for that. It is really important to think about that value. You know, obviously they're not just going to hand you money over for something that they don't perceive as valuable. The other thing is just being sure that you're actually surfacing things to them because especially now that we're four years old, we have a lot of things that we can surface. And so one of the most important switch ups for us was actually targeted sales. Like if you only ever pay us in hundred dollar increments, right? Which is the most valuable, you know , if you're just looking at it at a discount basis, like that's the most valuable way to pay us. If you're, if you're just buying a crown sell packages, then we pretty much only want to show you like when we're showing you in , in our , our , our daily Papa , we pretty much only want to show you a hundred dollar packages. You know, why, why waste their time on a $5 package? But then there's a lot of people that are like, Oh, $100 that's way too much money. I would never spend $100 on that and they only want the $5 package . Right ? And so that was one of the important, kind of a middle tier or things that we did on our roadmap was implement a sales segmentation in that way so that we could surface things to players that they would actually be interested in. Right. Yeah, that makes sense. And some of it I imagine for you guys was a big market shift. You guys had come from making games that were either that PC focused or you know, I think we would call [inaudible] AAA, right? Big, big premium experiences to doing mobile. And you talked a little bit about your technology's struggles there, but what about that, the struggles you guys experienced in shifting over to a completely different demographic? Uh, yeah, that, that was our AAA background has definitely been something that's informed the DNA of the company in good ways and in , in ways that we've struggled with sometimes. We did have kind of that transitional experience at Zynga, right where frontier Ville was, you know, a massively successful product for them. Back in the day, I think I'd , I'd heard this, I don't, I don't know this, but I'd heard that it ended up generating about $750 million over the lifetime of the studio. And we , when we left, it was a third biggest money maker at Zynga behind a Farmville in poker. Uh, and so the other thing about frontier Ville is it crushed per user spending records at Zynga that they , that we made, I can't remember, but it was something like three or four X what the next closest game did on a per user basis. And so , uh, and you know, the real innovation or frontier Ville now a lot of people, you know, remember it these days, but it took the model that was Farmville and it added quote real game design to it. Uh , so you know, Brian Reynolds was the guy that , that, that, you know, mostly was the , the creative visionary behind that came . And so from his perspective, he was just kinda taken game design one Oh one bag of tricks stuff and kind of putting it onto the Farmville model and whether that's class or lightened narrative or exploration or a , you know, clear and compulsion, the kind of doers that I think he actually has a patent on the like sort of, you know, little things that you can tap on and click. Uh , he also, I'm pretty sure that he invented the visit your friends , uh, frontier farm feature that you can kind of help them out and harvest their hearts from. And you see that in , in now in games like heyday and things like that. That's the only way porn has that, you know, we, we did learn a lot about , uh , a free to play and the market, you know, with, with the frontier Ville in that way. And so we weren't exactly starting from zero on that. Um, so you would have thought that we would have been better about , uh , you know, not just listening to our most inner generous cells that just wants to give the entire game away for free and then never have enough money to pay our rent. Uh, and , uh, but, but, but we still, you know, when we started our company, we were like, okay , this is mobile. And it's like, dude, like we all are in love with these new strategy games that are out there for this platform. Right? So let's do it different. And you know, that, that was a consequence of our AAA background. It was a consequence of, of you know, liking to kind of tinker with game design models but more with an eye towards how do we make this more fun, not having to make this more fun. And how is it that people, you know , uh, see enough value to, to want to give us money for it. Do you find that the players that are spending the most time with your game now are the same kind of players that you had when you were making AAA games? Or do you find it is like totally different audience? That's a good question. Uh, we haven't done any kind of deep dive into that. Meaning we haven't actually done like an audience study that would say, you know, did you play age of empires? Did you play resignations back in the day? Um, anecdotally like, just when I read reviews, you know, like, especially in the first several months of, of the game, you know, I pretty much read every single review that was on an app store. Uh , and a lot of people call it out to, you know, this reminds me of civilization. This reminds me of age of empires, which was exactly what we were looking for when we designed the game. And so I think there is a lot of overlap that way. You know, as our lives become increasingly fragmented and you know, harried and whatever, people maybe only have five minute chunks to play, but if they can get 70% of the fun of a game, like in age of empires , uh , in the way that they would have defined it back in the day, but instead of an hour and a half a game, they can do it in five minutes, then know that's much better because otherwise they just wouldn't be playing these games at all anymore. Right. Uh , so, you know, there's that , uh, I think that people are kind of adapting more to these, these uh, these long tail kind of games that you just persistent world I guess kind of games is what I mean by that where you're just, you know, you're just kind of continuing to build over many, many years and you develop relationships with your Alliance members and all this kind of thing that we see are really organically happening in the game. Uh, so the short answer to the question is definitely many of our players are people who used to play those games and maybe don't have time for them anymore and they delight in having the same, you know , basic kind of experience on , on the mobile phone even if it's not the exact same thing. Yeah, I mean, I think , um, I've had a couple of people in the industry, USA to me over the last several years, like what happened to the RTS genre? And I think my answer is often it's huge. It just changed a little bit or changed a little bit. It's on, it's on phones now. That's where that is. And so yeah, it's kind of interesting to hear that. Yeah. You are seeing people say I was a big fan of, you know, HOA Pires and things like that. Right. That's all . You know, it's interesting cause , uh, I w I went back and played rise of nations when it came out on steam , uh , not that long ago. And resignations was developed by the, the original big huge games . Yeah. And I was like, this is pretty fun. Right. But you can't go home. Yeah. Uh , you know, people have this, this memory of kind of the best moments of RTS I think in the pulse pounding this and the narrow reversal that and all that kind of thing. And there's things are cool and we're cool. Uh , but you know, that kind of really long econ building phase in the beginning of every game. Like this is why these games took whatever an hour, hour and a half to play. And they, I don't have that kind of time very much anymore. It's a to kind of do that. Uh, so anyway, I think the people are very fondly remembering games that might not actually hold up. I think the real classic RTS model kind of said all it had to say there just wasn't that many interesting variations left on how is it that a gather resources build an RV and go beat up on an opponent. Yeah. Um, I'd be curious to see more experimental variations. This may be geeking out too much on RTS stuff, but uh , I think of the, the, the central thing about RTS is that it puts tension on my bandwidth as a player, right ? That's the central thing that defines an RTS is that way I talked about this is there's, at any given time there's 10 things that you could be doing in the game, right? You only have the bandwidth for three. If you get really good at the game, maybe you can do four or five. If you're a pro player, you can do six or seven, but you can never do all 10. And that's the one thing that RTS has that no other game hasn't. So when people talk about like clash Royale as an RTS, I'm like, that's not RTS cause there's no bandwidth. There's a bandwidth tension, but it's cool. It's great game. I think Crossrail , maybe like the best game that's on mobile right now, other than other than dominations of course. Uh, but , uh, you know, certainly it's really innovative, really polished came , but it doesn't have bandwidth tension. So it can't possibly be in RTS . Right. But I also think that you could have some like kind of , you know, thought exercise this, but, but I've never really tried to see if a team wants to, it wants to try it. You know, he's like, could you get that been with tension in completely, you know, other ways that aren't in any way related to a military model or base building model or whatever. And you see a game like overcooked is actually kind of like, like an RTL wide bandwidth tension , uh , a kinda game. Right. And I sit right on the other side of the wall from where our Nintendo switches. Uh, and so, you know, I , I hear our team, you know, after, after hours just being like [inaudible] know , chop, chop, chop job , like , you know, that kind of thing. They're , and so, so I'm like, okay , that's kinda like a co op, you know, being with tension game, I wonder what else you could , you could do to play with, to play with that kind of tension attention . So anyway , that's my an interesting, I think space that could be explored more. I mean, obviously the MOBA genre explored one direction

Speaker 4 39:37

very clearly, but there's a lot of other things too, which would be cool. Yeah . I want to talk about the fact that you've been live now for four years. Yep . And , and still going strong, you know, I mean, I , I dunno if you can share numbers. My guess is your numbers are probably, you know, as good as it's been along the way or are pretty, pretty close to it . So , uh , when you're building a community, this going to be engaged for that long. Uh, it's a very different design exercise than , than , than , you know, deal. It's a complete opposite extreme for the package. Cause this case we had eight hours of gameplay and we're done. So what, as you're now in that kind of like the really the deep ply of the games that were , you know, they're really deep part. What , what lessons are you learning now? Three, four years in that, that are changing how you design, build, operate, manage the game.

Speaker 1 40:20

Right. Uh , I can talk a little bit about numbers, although I haven't actually vetted this with the next one . So , uh , so, but I, I can't imagine that they would, they would have much of an issue with this. But , uh, we just looked at our , our month one retention data, meaning like from the first month of the game, and we still have 1.4% of players playing the game. Uh, the first leaders . Exactly. Now that's the golden [inaudible] don't seem golden cohort this coming quarter , which is a frustratingly a thing. Like why is this still a thing? Yeah . Uh , but still is a thing. Uh, and so , uh, yeah, so we're, we're really excited that there's that many people that are, that are, that are interested in playing the game. The other way that I can , that I talk about metrics without actually, you know, kind of telling people is that, you know, we keep a team of 35 people profitably employed. And so that's, you know, that's one of the other, the other ways of life that's actually pretty good. You know, team developing features and feature sets for a game that's four years old and that again, that we , we expect to be a 10 15 year lifecycle for the game. Uh, and who knows, you know, many of these games have been out for a long time and showing no signs of slowing down. So, you know, we hope it's a long time. Anyway, those are some rough metrics that I can share that, that are, that are, that , that , that can help bracket, you know, what we're , how we're doing from a, from a success perspective. Uh, as far as how you think about the game play, I mean, you do, you have to be aware that your core gameplay will get old. Uh, and even, you know, even if you designed a really fun game and a really great game, you know, after four years it does get a little bit , uh , a stale. So one of the things that's great about human history as your subject matter is there's an inherent progression built into the story of human history. You never need to tell anybody why getting to guns is cool. You never need to tell anybody why getting an airplane is cool. And so we just launched the space age, which comes with the offense of missile silo, which is basically a big blue shit up button that you get to use, you know, for, for a , your battles. And so that's an example of a thing that like, we have this kind of inherent thing that we can, that we can build onto . That's, that's the story of human history, right? And so that, that very much changes the core loop . The game gets much harder from a , an offer. If you're on offense, it gets much harder as you progress through the game, which we think is fun. You're kind of increasing the challenge curve. You know, you give more defensive buildings, give more defensive kind of actions that the, the , the uh , defendant basis can do. And that, that's fun, that it kind of continues to , to, to challenge your skill curve. Uh, and sometimes our players, you know , think that maybe we go too far with that, that it's , it's actually, you know, too hard now to, to five-star base or whatever. Right. Uh, when you're into the cold war age or the space age , uh, you know, my personal experience of it is that when we've made the game harder in that way, I've had more fun with it. You know, they used to be whatever, six or eight months ago, I could just five-star pretty much everything in my sleep. And it wasn't really playing combat. I was more sort of farming, if that makes sense. Cause it , cause there wasn't any challenge to it. And then we kind of looked at our numbers and we're like, let's do a rebalance. And I have a lot more fun with that , with the combat game now, even though it's harder and I have to spend more time on it because if I'm collecting my stars every day , I can't, I can't just do it in three battles. I needed to, you know , have four or five battles that, but I can do it in. So anyway, that's basically how we think about it. Uh , we are planning some things that I can't get to into for later this year that are going to significantly alter. I wouldn't, well , I says , not altered, but significantly augment , uh , what you do in the game on a daily basis. The, the quarter loop of it , uh , that we're really excited about and then feel really, really innovative and different for us . And so hopefully at GDC, you know, 2020, I'll be able to be like, well, here's this feature and here's how it worked and this was, this is awesome. And so we, and you just casually tossed off a significant rebalancing. Yeah. I mean, how was that tricky to pull off? I mean we , we were, another conversation we had earlier is how upset players can get when things change . Yes. And how would you rebalance the games inefficiently and not upset your , your, your golden cohort for example? Uh , so that's a great question. And uh , it was a significant effort. We had a designer, Brad Fisher , who's a really great kind of nuts and bolts and especially as it's grown into a really great balance designer , uh , working on that for a number of months. Uh, and there is a tradeoff between , uh, wanting to, to do what you know is right for the game on the longterm , even though you know that when anytime you disrupt people's play patterns, like I am used to playing the game this way, now I can no longer play the game that way. Uh , there's going to be a attention about that. And then there's actually like internally when you're developing a prototype, one of the, the, the kind of Sid Meier rules, and this might actually be a Brian Reynolds rule, I'm not sure, but we call it feature inertia, which means that, you know , if you're playing a prototype and you're like herpes time zone with the prototype and then the designer changes anything about the core thing that people are going to be mad about it and you have to let feature inertia bake in for like three or four days at least before you can accurately gauge whether that change was a good change or not a good change. Because if you , if you change like you know , like whatever all tanks is the winning strategy and I'm like the master player at the all tank strategy and the designers like, look, you can't, can't just have it be the tanks are the only way to roll its own phase and then they make it that, you know, he turned down, turned down the tee . Like that player is going to be mad. Like whether that's in the wild or whether it's internal. And so you kind of just have to acknowledge that that's going to be the thing. I think that we could have done a better job of engaging our community about it in advance. Right . Uh, and, you know, we, we, we, I think that was definitely a learning for us is that, is that we should have been more , uh, proactive in explaining why we were doing this. Uh, and , uh, maybe a little more back and forth or incremental changes over time instead of kind of w we did try to incremental change things, but we still probably could've been a little bit better about that. And then we did actually roll back some of the changes or, or tweak some of the numbers in response to community feedback and also the data. Like, you know, if community feedback had been at odds with the data that we were seeing, then we would've had a tougher decision to make. But, you know , there were some of the things that community highlighted that we validated with data. Okay, fine. You know, we'll, we'll make some changes that way. So short answer is we faced this problem on a prototype where 10 people are playing it, right? Your inertia is important as an important, you know , design consideration when you're, when you're developing games and, but, but we probably can and should have engaged the community in a, in a, in a better way than we did on the balance. We've heard from some other people how difficult that can really be to actually manage a community on mobile because there is so few options to actually speak to them directly. How do you guys handle that? Um, we've handled it in a very straightforward way. Meaning I'm not sure that we have a very sophisticated approach or understanding. We're just like, well, CADA Lake back in the old days we got you know , forums and we liked to talk to our users on our forums and we try to engage in a back and forth on that without, you know, with keeping things by while keeping things respectful and you're not being able to talk too much about what our future roadmap is or anything like that because we don't want to promise things that we don't end up being able to deliver that kind of thing. So we're kind of old school about it. I think that, you know, in the, in the crawl, walk, run, you know, formulation of things were like somewhere between a crawl and a walking on how we and how we, how we kind of deal with the community. I don't think that we're really adopting the best practices of where things are these days. Right. But we are really excited about our second game and feeling like we , we hope to be able to, to kind of move more into a walk and hopefully break into a trot . You know, it's at some point in the next , uh, the next year, 18 months on community. So that's how we think about it. Cool. One quick thing is you mentioned early on Gilt sounds skills are an important part of the game, properly honor the alliances. Yep . And just curious to hear any, cause it gives her something that we see , uh, not a lot of games taking advantage of and more now. So , uh, any thoughts on, on designing game when you have a strong Guild structure? Yeah , uh, I mean we definitely think that that that alliances for dominations drives a huge amount of player behavior and engagement and, and in ways that feel like it's almost the best of what games can do, especially in this mobile and fragmented world, which is just connecting people. And we're always surprised at the organic , uh , way that people create communities that are links to the game, but external to the game and the use of discord and things like that, that kind of meta alliances. And I kind of always wished that we could, that we could wrap that into the actual game. Uh , and you know, I , we just, it just never ends up being a high enough priority on the road map to all the other things that we could be doing to try to, you know, create kind of that metal metal Guild or feeding your finger alliances that like, well you have to, you know, you can come into, you know, a , you know, clan one a before you, you know, earn the right to go to class to me and you know, all the way up to the like, I wish that we could , we should, we could put more resources towards towards people in that game is to my knowledge, nobody's yet done that. But I'm sure that somebody will. Uh, and so yeah, I mean we think this drives a huge amount of, of continued player engagement in the game. Uh, we do see differences in the numbers. I couldn't speak to the, the, the real specifics of it. This might be a good GDC talk of behavior player behavior of people who are in guilt versus player behavior that are not in guilds. Uh, in fact, I think I'm going to write that down and be like, this is this GDC is 20, 20 times. Yeah. And so , uh , but we do see a meaningful difference in those things. And this is again, just on an intuitive basis. Like we have more fun in the game when we're, you know , when we're ready to go on , even when we're prototyping a game or our second game that's been in development for, for about two years now. Uh, when we do natural play throughs where we're trying to, you know, well, let's just see how this feels over a weekend or over a couple of weeks or a month or whatever the Guild is, is a, is a critical component of that. Like actually be getting in a Guild and chatting and trash talking and whatever. Even in our internal prototypes, you know, it is something that meaningfully changes the, the fun of the game. So the last question is , uh, one of favorite sections of these interviews that we've been doing is the train wreck section. [inaudible] we love asking people for , uh , for their , their tales of , of when things didn't go as planned. Uh, any , any stories you can share with us over the life of the, the , especially around live ops, live ops , you know, disaster disasters . Hm . I mean, one of the things that I think a lot of people run into a bunch is when they aren't careful about how they design their tool sets for live ops, it's very easy to accidentally give away. Yeah . Whatever. $1 million of currency and say, we did have this , uh, incident and it might not have been, that might have been like only five or six months ago where someone changed a number in some tools somewhere realized that they had done the wrong thing, reverted it, but not before it got propagated out to the wild and their revert change didn't. And then again, it's like we wake up, we look at our numbers and we're like, wait, why are we spiking in purchases at three in the morning in Russia? And, Oh no, we looked at, you know, people are just, you know , buying this one super cheap pack. That was incredibly valuable. Uh, and so that was the most recent thing that I can, that I can think of that was a real uh, kinda kind of , okay , let's build a war room and get some, how'd you him stuff happened . Did you roll that back by by sort of climb back the currency ? Did you find new sinks you could put in the games are going to soak up that extra currency that got put out there. We didn't call back. We haven't made a practice of climbing back anytime. Things like that have happened. Partly because it's not really fair to people who have spans or having span probably cause really complicated to figure out like, well was that missile silo upgrade because of the, you know, the cheater currency. And mostly because it's our bad, like we were the ones who made a mistake that's very different from cheating. Like gaming the system is a, is a thing that like, okay, well we should have built a system that could be gamed. And to us that's very different than, than actual cheating. And so yeah, to try to club back just said are bad. We need to, we need to fix our tools. Was, was mainly our thing to, to be sure that there was confirmation dialogues and I, and all that kind of stuff. But yeah. So. Cool. Well thank you so much for coming. This is great. Yeah. Great . I really enjoyed trying to get checked in . Thank you so much.

Speaker 2 52:33

Thanks [inaudible] .

Speaker 1 52:39

Thanks for listening to the art of live ops podcast. If you'd like to, you heard, remember to rate, review and subscribe so others can find us and visit playfab.com for more information on solutions for all your live ops needs. Thanks for tuning in.