How and Why Players Play Games w/ Troy Skinner: The Art of LiveOps S3E1

Troy Skinner is a researcher at Immersyve who studies both games and game designers to learn why people play video games and what keeps them engaged.  He works with behavioral scientists and psychologists to operationalize these insights to make game design decisions. 

In this episode, Troy discusses the importance of providing players with consistent growth and long-term goals and he uses an analogy comparing game genres to making pizza.  He discusses the intrinsic and extrinsic goals of players and how collecting and analyzing these data points can help developers determine appropriate rewards and monetization strategies for their games. 

Transcript

Copy Transcript Copied

00;00;07;13 - 00;00;23;23
Crystin Cox
I'm Crystin Cox. I'm the Director of Live Operations for Xbox Game Studios Publishing. I've got 18 years of experience in the industry. Pretty much spent my whole time as a game developer working on large scale LiveOps games, things like MMOs like MapleStory and Guild Wars 2.

00;00;24;05 - 00;00;46;16
James Gwertzman
Hi, I'm James Gwertzman. I'm currently a general partner at Andreessan Horowitz, focusing on the game industry. Previously, I was a co-founder and CEO of PlayFab, which focused on building game services for game studios. I also ran the gaming vertical for Microsoft, which is how I met Crystin. And I'm really excited for this third season of the podcast, as we continue expanding the group of people we talk to from the industry. Who we're talking to today?

00;00;46;25 - 00;00;49;03
Crystin Cox
Today we're going to be talking to Troy Skinner.

00;00;49;04 - 00;01;07;29
Troy Skinner
I started as a game developer working for various companies like Visual Concept, Vivendi, Warner Brothers and Monolith. But halfway through my development career I decided to become a researcher. For the last ten or 15 years, I spent more of my time trying to figure out why people play video games. In that time, I've done over 500 studies.

00;01;07;29 - 00;01;28;06
Troy Skinner
Many of those are on live games, and my preoccupation is trying to get to the bottom of what engages players. And a lot of the people I work with are actual behavioral scientists who started in psychology or studying motivation. They bring world class science to understanding why people do anything. And then I kind of focus on how we operationalize those insights to actually make design decisions.

00;01;28;09 - 00;01;36;17
James Gwertzman
And I've read Troy's bio, he is a researcher. He studies games and helps game designers be more effective.

00;01;37;02 - 00;02;00;21
Crystin Cox
Yeah, I know Troy. I've worked with him a little bit at Xbox. And he is a researcher he works for a company called Immersive. And they do player motivation profiling. They dig really deep into trying to understand why people play games and specifically why people play games for long periods of time. So lots of interesting implications for Liveops.

00;02;00;21 - 00;02;08;21
James Gwertzman
Got it, so if I'm a game designer trying to build a game that will hopefully pull people away from other games and get them to play my game. This is going to have some of the some of the answers.

00;02;09;01 - 00;02;13;11
Crystin Cox
I think Troy is the closest thing I know of as a true expert in engagement.

00;02;13;14 - 00;02;14;25
James Gwertzman
All right. Well, let's get to it.

00;02;19;23 - 00;02;45;14
Crystin Cox
So Troy and I have worked together at Xbox Game Studios. It's always awesome, Troy, to get you to come in, take a look at the stuff that we're doing here and give us that outside perspective that really comes from a little bit more of a grounded research place around what our designs are going to operate like and why people may or may not actually want to play our games long term, which is sort of the essence of LiveOps games.

00;02;45;14 - 00;03;09;06
Crystin Cox
You've had a really fascinating sort of point of view on the growth of LiveOps in the last ten years. And I'm sort of curious to hear over that time, especially since you've been a researcher. But certainly going back to your time as a developer, how have you sort of perceived live games as having changed? Like what have you seen the evolution of LiveOps games looking like from your point of view?

00;03;09;13 - 00;03;51;23
Troy Skinner
Yeah, it's a big question. The first thing that kind of occurs to me that I found super interesting is I started making what I would call like linear first person shooters, like games like Fear. And so, you know, we would have a multiplayer mode tacked on, but we didn't really have a strong conception of what Liveops was. And so I was in development, in transition to research in the time period where lots of people were trying to take genres that were not MMOs that were not kind of obvious liveops genres and trying to figure out how to reconfigure them to make them engage, you know, in a live environment.

00;03;51;23 - 00;04;15;26
Troy Skinner
And so that involved lots of you know, tinkering under the hood with game design and game genres to kind of turn them into something, you know, to hybridize them or turn them into something that was still true to kind of their...where they started from, they were still shooter or they were still an open world game or what have you, but have other characteristics that would make them compelling to players long term.

00;04;15;26 - 00;04;38;12
Troy Skinner
And a lot of what you find out if you study engagement is the key to it. There's a lot. But the two biggest keys are can you get people to form long term goals and can you give them consistent growth? And the when I started making games, a lot of the goals of the game was to finish the story and so those things weren't set up to last forever.

00;04;38;13 - 00;04;59;12
Troy Skinner
And so, you know, kind of reconceptualizing what a game that starts from that foundation reconceptualize it, how you would make that into something last much longer and still be valuable and entertaining. It's kind of the thing that feels like its changed the most to me. And obviously that's you know, that's kind of the the old, old view of what it looked like, you know, 15 years ago.

00;04;59;12 - 00;05;12;21
Troy Skinner
But I think that story is still being told today because everyone is now trying to make games that lasts forever and from wherever their creative starting point is. And it's just it's a real challenge to figure out how to make that work in lots of different cases

00;05;13;16 - 00;05;28;19
James Gwertzman
Well, and I've heard one description I've heard of of sort of the shift to LiveOps and the shift to games as services is that this is really about now monetizing engagement as opposed to sort of the One-Off monetization we used to do when you bought the game. And so, you know, you talked about engagement a couple times.

00;05;28;20 - 00;05;40;13
James Gwertzman
I think that's a really key thing. I want to dive a little bit more deeper into engagement itself. So when you talk about engagement, what are you measuring? Like what is it just does a player play the game for a couple hours every day? Like what? What is engagement to you?

00;05;40;24 - 00;06;04;21
Troy Skinner
Yeah. So that's great question. So we start to be vaguely nerdy for a moment. We start from a base of what's called self-determination theory, which is the leading theory of engagement motivation worldwide. And the people that founded self-determination theory, they're the people that moved away from Skinnerean behavioral psychology towards essentially what you would think of as intrinsic motivation.

00;06;05;02 - 00;06;21;05
Troy Skinner
So there's three intrinsic needs that all humans have. They have the need to be competent, to be effective, to grow they have the need to be an autonomous so that they want to feel like they're the ones in charge of what's happening in the game or what's happening in their lives that won't be controlled by someone else, by design or by their parents.

00;06;21;05 - 00;06;53;04
Troy Skinner
Whatever. And then there's a need for relatedness, which is people want a meaningful connection with one another. So we started from that base. And then what we did is over the last ten years, we've broken motivation down to over 135 different components. And so when we test the game, we'll have an outcome metric like longevity, like how long do you intend to play the game or an outcome metric, like, you know, some value designation from the player, about how much they value the experience or how much they think it's worth spending the game or how long we want to play it.

00;06;53;04 - 00;07;13;27
Troy Skinner
etc. etc. But then we have the 135 components of motivation and we'll run linear regressions and essentially it's like reverse engineering, the formula of what engagement is in each gaming genre. So you if you test a match three game or you test an MMO or you test a battle royale, you test a mobile or test a trading card game, those all have different formulas.

00;07;14;04 - 00;07;35;01
Troy Skinner
So at a high level, we measure engagement by looking at how long the player intends to play the game because we feel like a prediction of the future is more reliable than a backwards looking behavioral read on what they've done. Though we'll look at both. And then we're looking at, well, what are the recipes that build engagement in each game?

00;07;35;05 - 00;07;42;28
Troy Skinner
So it's kind of for us since engaging the whole shooting match, we don't look at it in very simple way. We kind of have a pretty rich view of it.

00;07;44;11 - 00;07;59;20
James Gwertzman
So I could see some of the game designers listening to this right now. Starting to tear their hair out and saying there's no formula for games is no mathematical equation that's going to create a really fun game. And yet you seem to be breaking it down and you're saying sort of maybe there is. Can you talk a little bit more about that?

00;07;59;20 - 00;08;01;05
James Gwertzman
That tension that that exists?

00;08;01;05 - 00;08;27;09
Troy Skinner
Yeah, that's a that's a really great question. So the game designers keep your hair on. I'm on your side. There is no formula for making games. I guess the way I would say it, an analogy would be something like if you think about food, like Thai food or or pizza, you wouldn't say there was one perfect pizza. And we all eat the same exact pizza.

00;08;27;09 - 00;08;45;04
Troy Skinner
Like we have thin crust and that crust and different toppings. But if you were someone that studied pizzas a lot, you would come up with a framework for things that people tend to value in a pizza. You know, they like it, they like the crust on it, they want a sauce, they want some toppings, maybe cheese, maybe not.

00;08;45;16 - 00;09;16;07
Troy Skinner
And it's kind of like that when we're looking at different genres and so we're not trying to describe what a genre is. We're trying to measure why people engage in different games and genres. And so when we look at something, you'll see patterns. When you test 500 games where, you know, like as an example, if you have a shooter, then high up on the motivational chain is going to be people's desire to get better at the game. Investment, skill growth, whether experiencing skill growth

00;09;16;07 - 00;09;53;18
Troy Skinner
And then as we've gotten into more of a live environment, novelty coming through seasons, people making some RPG progress through account leveling or battle passes have come up the chain, whereas those things didn't exist in old Counter-Strike or in old Halo games. So it's not proscribing what design's supposed to be. It's more measuring successful games and saying these are the components of motivation that are firing in this game and then trying to have a conversation with designers about why that works, what features are bringing that value, and whether we have an idea as designers to do something that fulfills the requirements of pizza, but in a new and interesting way.

00;09;54;08 - 00;10;14;22
Crystin Cox
You know, I think that, you know, James is sort of hitting that theme we've touched on a couple of times, which is this tension that sometimes exists more classically between creative and data and sometimes telemetry. But I think you have an interesting perspective on it when we talk about research, because there can also be a very similar tension that exists in research.

00;10;15;04 - 00;10;38;18
Crystin Cox
I'm curious from your perspective, you engage with a lot of different teams over a wide swath of the industry. How have you seen developers' relationships with data change? Do you find yourselves still trying to, you know, say, like, guys, you know, this you got this is what data is and this is why you should care? Or are people more savvy now?

00;10;38;26 - 00;10;45;23
Crystin Cox
Or maybe somewhere in between or maybe even new things are developing. But yeah. How have you seen the relationship with data change?

00;10;45;26 - 00;11;16;29
Troy Skinner
Yeah, that's a great question for me. If you look at a ten or 15 year period of time, the developers relationship to data has changed massively, like back when I was in the publisher developer game. What data meant was some vague notion of meta critics score being predictive of what your retail product would do. And there was no, you know, data was like publicly available stuff like, you know, going to, you know, Gamasutra and hearing someone like cite one number without any context.

00;11;16;29 - 00;11;37;06
Troy Skinner
And so there was an absence of data. And in that environment and I participated in this as a developer, it was just basically we all were arguing for our own personal perspective with no data framework to kind of interact with. So it was, it was more of like, you know, we were just arguing about what a good game was and we didn't have the same language.

00;11;37;08 - 00;12;02;02
Troy Skinner
Some people thought it was supposed to be fine and some people thought it was supposed to be innovative. And some people thought, it was supposed to be high quality. But we were all making up what each of those words meant and so there just... data wasn't really a thing. And then, you know, you kind of fast forward to now and I think, you know, most developers I work with rely upon data much more extensively and have become more used to what data does well on what data does poorly.

00;12;02;08 - 00;12;26;26
Troy Skinner
So I think the starting point with data is what data does well, especially from a telemetry standpoint, is it tells you what is happening and it doesn't necessarily tell you why it's happening. And so I think there's still responsibility for a designer or anyone running a product to have a framework for what makes a good game. And then use data to validate hypotheses.

00;12;27;06 - 00;13;09;10
Troy Skinner
Whereas when someone just looks at data first and says, I'm going to have data explain to me what to make, that's not what it does well. You have to come with the framework. You have to be asking questions and then using data to validate or invalidate the hypotheses that you're carrying. So I think many more developers, even if they don't use fancy science talk in their relationship to data, I think a lot of people have internalized where, you know, data supports them in making decisions in tough, confusing environments as opposed to thinking about, you know, James, you kind of set this rabbit to run earlier, but instead of thinking about data telling you what game to

00;13;09;10 - 00;13;13;29
Troy Skinner
make, data is much better at helping you answer questions that you know how to frame.

00;13;15;22 - 00;13;34;10
Crystin Cox
You know, I'm interested to hear your thoughts, because I think we've asked some data scientists and analysts this question, but I'm curious how your answer might be different, because you have a very, as a researcher you have a bit of a different point of view. For teams, and I hear this a lot, like teams are a little bit intimidated sometimes by data.

00;13;34;10 - 00;13;49;13
Crystin Cox
They're like, it seems like I have to have a lot and I don't know exactly what to do with it. What would your advice be? Like what are the, what are the first things like that team should start caring about? What are the things that you sort of see as the most important to measure?

00;13;49;18 - 00;14;14;09
Troy Skinner
Yeah, it's a good question. So, you know, this goes back to the themes we've already hit on, which is for me, data is for hypothesis validation or invalidation, right? You're using the scientific method to iterate towards better engagement, better monetization, greater value for players. And so you need to before you can build the data framework, you need to have a theory of how your game engages.

00;14;14;09 - 00;14;43;10
Troy Skinner
And obviously when we design anything new, we don't always know how it works. But that's kind of why you need to work from my perspective. You need to have a sense for why people do things long term, why they engage. And then you look at your game and say, What is our theory of this? If we're a new kind of builder or we're a new kind of MMO or a new kind of battle royale, do we think that we're going to engage just like something that exists in our genre?

00;14;43;19 - 00;15;07;14
Troy Skinner
Or do we have a new premise based upon denser event patterns or new modes, or some leveling system, etc. etc., all these features have an implicit hypothesis written inside them of how they're supposed to engage players. And so what So then you once you have a hypothesis, then you say, Okay, what data could I collect to tell me

00;15;07;14 - 00;15;27;18
Troy Skinner
if not only is the game engaging the way I want to but what aspects of the game are driving engagement? And so my starting point is like essentially, what's our plan? What's our hypothesis? But then thinking through how we would answer that question and that's the same thing you do when you kick off research. Someone has a question about like, whatever, why is League of Legends so successful?

00;15;27;28 - 00;15;47;14
Troy Skinner
And the starting point is, well, why could it what why do we think that might be the case? Like, is it because there's so many different characters? Is it because it's so intense? Is it because, you know, blah, blah, blah, blah, blah? And you write all the possibilities down and then you collect information to validate or invalidate each of those possibilities.

00;15;47;14 - 00;15;49;09
Troy Skinner
And that's kind of what I do with the data framework. [ad break]

00;15;52;00 - 00;16;13;09
Crystin Cox
The Art of LiveOps is presented by Azure PlayFab. PlayFab is a full featured back end solution that powers LiveOps for games of all sizes. Whether you need multiplayer services, analytics or LiveOps tools, PlayFab can help you reach your retention, engagement and monetization goals. Learn more and you can get started for free at

00;16;13;09 - 00;16;21;20
Crystin Cox
PlayFab.com That's played PlayFab.com P-L-A-Y-F-A-B dot com and you can get started for free. [end ad break]

00;16;24;06 - 00;16;27;03
James Gwertzman
Hey, welcome back to the Art of LiveOps. Today we're talking to Troy Skinner

00;16;27;03 - 00;16;35;04
Crystin Cox
He's a researcher who looks into the motivation behind why people play games and what keeps people engaged. We're having a great conversation. Let's dive right back in.

00;16;36;08 - 00;16;53;13
James Gwertzman
So you mentioned I think you said like 130 different metrics that you, you or you know, you've kind of broken this intrinsic motivation down to these all these different data points. Yeah. Based on what you're saying, and it's actually an important point, you can't just take all 130 and try to get them all to go to ten, you know, like that's not the point

00;16;53;13 - 00;17;01;27
James Gwertzman
The point is to sort of understand for my game what's, what's the engagement in my game and then which of those would then be appropriate for that particular experience I'm creating.

00;17;02;03 - 00;17;02;28
Troy Skinner
Exactly. Yeah.

00;17;03;04 - 00;17;10;03
James Gwertzman
What are some examples of those metrics? I mean, I'm assuming we're not talking about like, you know, DAU we're talking about something, you know, much more fine grained.

00;17;10;24 - 00;17;32;07
Troy Skinner
Yeah. So these are psychometrics. So they're self-reported experiences from the player because I research games and ask people what their experiences are. So an example would be in the goals category, long term goals, short term goals, goal density, which means I don't have one goal, but I have lots of different goals. Then you can start going to progress, which is are you making progress

00;17;32;07 - 00;17;55;14
Troy Skinner
against those goals? Are you making progress toward your current goal, toward your long term goal? You know, you would get into different kinds of growth, so are you experiencing skill growth. Are you experiencing engagement contingent growth, which is RPG systems? Are you experiencing a lot of novelty surprise discovery? Do you feel there's different kinds of autonomy? Do you feel like a game like Sims has a lot of narrative autonomy because people are authoring stories.

00;17;55;24 - 00;18;14;20
Troy Skinner
A game like Minecraft or weirdly enough, Magic, The Gathering have lots of experimental autonomy or Path of Exile have more experimental autonomy because you're tinkering withstuff. So it's basically if you just think about the same as a food, you wouldn't say the best food is the one that has all the spices in it. You would say What makes you know, what spices does Thai food use?

00;18;14;29 - 00;18;33;11
Troy Skinner
It's the same this and motivation in games, which is it's it's basically coming up with more refined language about what's working about that game. And so that's a quick example of some like goals and growth items. But there's autonomy and there's social items and relatedness and competence items control mastery, blah, blah, blah.

00;18;33;12 - 00;18;56;01
Crystin Cox
Yeah, I think that's a great analogy because I know when we've worked together, it's been so valuable because it really can feel as a developer, like there's this intense pressure to add all the spices. Like there's a real sense sometimes when we do very sort of shallow investigations where we go, what do players want? And the answer is everything, right?

00;18;56;01 - 00;19;22;27
Crystin Cox
They want at all. Like they want all the features and all the things. I think it like that perspective that you bring has I know has been really valuable for us in trying to go, how do we dig a little deeper than that and ask, what's a good combination for this game that's like going to work for us, for our community as well, not just for, you know, some generic idea of what the what a gamer is?

00;19;22;27 - 00;19;43;05
Troy Skinner
Yeah. I mean, I think you put your finger on one of the hardest parts of being a developer is like everyone basically knows in the industry now is that engagement is what it's about. And so over the last ten years, we've all worked our hardest in our own way to hold our players longer, which means if you're coming with a new game to the market, there are no players on the sidelines.

00;19;43;19 - 00;20;10;12
Troy Skinner
You need to pull players away from a successful game. And in most cases, if you're doing that, that game's already made tons of money and they've been adding more features. So you're coming in with less money with no established audience, and you're trying to pull players away from something else that already is satisfying their players. And so a lot of the challenge you get is the developers really how do I make something more compelling with less money or less development time or less features?

00;20;10;20 - 00;20;34;20
Troy Skinner
And, you know, one strategy that is more precision about where the action is going to happen for your game so you don't pile every feature or every spice into the game. You say there is something that people haven't experienced before that I think will be super compelling that will differentiate us. And that's what we're going to focus on and these other things which are valuable in certain contexts we don't think we need right away.

00;20;34;20 - 00;20;45;13
Troy Skinner
And so I think I think having clarity around what your plan is for you know, bringing value to players is is an important way to attack an entrenched competitor.

00;20;45;13 - 00;21;03;05
James Gwertzman
I have not never heard a more clear explanation or a succinct description of the challenge facing every game designer right now, which is exactly right. Trying to pull players away from existing successful games with less money, less resources, less time. And yeah, and that's exactly what every I mean, when I talk to game developers and say, what are your biggest pain points?

00;21;03;13 - 00;21;07;28
James Gwertzman
Number one is always audience. How do we get an audience for my game? It's a single thing they stress about the most.

00;21;07;28 - 00;21;08;08
Troy Skinner
Yeah.

00;21;08;08 - 00;21;17;08
James Gwertzman
And I think what you're essentially saying here is this is a framework for, for how to do that. And I love your description of precision, almost like a surgeon, precisely understanding what you're trying to create.

00;21;17;17 - 00;21;35;03
James Gwertzman
So let me change it. So for the people who are listening to this right now and they're thinking what I'm thinking, it is, Oh my God, I want to know what these 130 metrics are. And I want to start understanding how to get more precise about my own because I because I think as a discipline, we starting to have gam desing programs and people starting to approach it with a lot more rigor than they used to.

00;21;35;11 - 00;21;58;16
James Gwertzman
But think a lot of people in industry are still coming at it very just from a gut. There's not a lot of frameworks going on. And I think I think I love what you're describing because this is a framework you can think about it sort of craft the engagement model you're looking for. How would I start learning about your theory or how if I wanted to listen to this, take a more rigorous approach to crafting my my engagement model, where would I start?

00;21;58;16 - 00;22;19;19
Troy Skinner
yeah. There's a couple of publicly available resources, you know we...we do 50 projects a year, so we're pretty booked up with stuff that's behind a curtain. But my partner Scott Rigby did about six or seven years of GDC talks where he talked that he's the one that made the initial framework within video games that I've been collaborating for eight or ten years.

00;22;19;28 - 00;22;34;28
Troy Skinner
He's a brilliant speaker, very charming, much more entertaining than I am. And so, you know, if you have access to the GDC vault or a lot of times his stuff will get featured on Gamasutra. It's probably on YouTube as well. And I think there's I probably done three or four GDC talks as well. So there's some more detailed information there.

00;22;35;04 - 00;22;56;00
Troy Skinner
And then Scott and one of the founders of Self-determination Theory, Dr. Rich Ryan, wrote a book called Glued to Games that also, you know, it's now eight years old, but it is the starting point that we've been elaborating. So those are that, you know, the easiest ways. And then, of course, you know, I test video games for a living and do workshops for a living.

00;22;56;02 - 00;22;58;13
Troy Skinner
And so if you ever want to talk to me, then you reach out to me.

00;22;58;13 - 00;23;18;16
Crystin Cox
Yeah. And I can definitely highly recommend looking at Scott Rigby's GDC talks and Troy's. They're really great. And a lot of them have been published out onto YouTube, off the vault. So even if you don't have login access, I think there are several of them that you can take a look at there that are really quite good.

00;23;19;08 - 00;23;51;15
Crystin Cox
So it's really worth looking into. I'm curious to talk a little bit. You and I have actually talked about this before. But there's a lot that happens in Liveops that has to take into consideration this very long amount of time that passes, where we go...we're really looking for success that last over five years, ten years. And I know that you there's been times where you've been able to go back and study games at multiple times in their in their life cycles.

00;23;51;15 - 00;24;16;24
Crystin Cox
I'm curious to hear you talk a little bit about how some of the relationship with motivation in games changes as games get older and as player communities spend longer and longer with them, because I know early on it's natural. There's a lot of novelty associated with games or innovation, but that five, ten years in is almost never going to be sustained.

00;24;17;03 - 00;24;25;22
Crystin Cox
So I'm sort of curious how the priorities change and what you see as trends happening for games that are starting to emerge out into really long tails.

00;24;27;22 - 00;24;51;10
Troy Skinner
So it's such a good question, but a hard question to answer because the way in which formulas change is different genre, but you put your finger there's a couple of quick points I can say that are broadly applicable. So you made the first one which is the biggest one, which is any new game. One of the first components of motivation is people are like, this is a new game.

00;24;51;10 - 00;25;11;13
Troy Skinner
I've never seen it before. I'm interested to see what it does differently than any other game I've played before. And so novelty tends to be a really big part of early funnel experiences for people. And if you think about novelty primarily it's not always true. There's different formulas, but a lot of novelty comes from exposure to new content.

00;25;11;23 - 00;25;30;04
Troy Skinner
And so it tends to be if you imagine like something simple like an MMO with ten zones, well, novelty's going to be a big part of the formula through the ten zones. And then when you get through that and you get into elder game, novelty is going to tank and other things are going to have to replace it or players are going to wash.

00;25;30;18 - 00;25;54;10
Troy Skinner
And so what usually replaces something in a very, very long tale is people start to form motivation around their social identity within the game. So humans are always narrativizing everything that they're doing. You know, we're telling the story all the time about what a great researcher I am and how I get better over time and how people are in love with the things I write, blah, blah, blah, blah, blah.

00;25;54;10 - 00;26;16;02
Troy Skinner
I have some, you know, fantastical narrative, which is probably not true running in my head that, you know, I'm grabbing whatever evidence I can to support it, you know, keep myself smiling every day. The same thing happens in video games where people are basically they have goals, they're accomplishing things, they're making choices with their character or the way they play or how they approach the game that differentiate them from other players.

00;26;16;13 - 00;26;48;24
Troy Skinner
And they narritivize that and really good games do a good job of essentially recognizing things that differentiate other players and publishers that not other players and then get feedback from their peers about what differentiates them and what makes them special. And so if you just think about early games tend to be content based and novelty based and late games tend to be about I have an identity that I want to support and maybe a live event being successful in it supports it, or having a cosmetic that no one else has supports it, or being the first person on my server to do X thing supports it.

00;26;49;01 - 00;27;14;07
Troy Skinner
People are putting together a narrative that's less content based and more about something that differentiates from other players, that gets them recognition and in broad strokes, that's the difference between early game and late game. That formula, though, is very different in a shooter or a MOBA than it is in an MMO or, you know, looter shooter because they just have different conventions for how they're going to hold people's attention.

00;27;14;17 - 00;27;37;28
Crystin Cox
Yeah, and as you said, like people's identity around those genres is different. You know, like it's, it's very. Yeah, the story I'm going to tell myself about why I'm still playing, you know, Counter-Strike after 15 years is really different than the story I might tell myself about firing up, you know, League of Legends, even though it's a competitive game as well.

00;27;37;28 - 00;27;44;22
Crystin Cox
But it's a, it's a team based game and it's got its a different elements to it or even wildly to, you know, really different than something like an MMO, right?

00;27;44;25 - 00;27;47;25
James Gwertzman
Or Candy Crush or a Bejeweled or wants to play in those games after five years.

00;27;48;17 - 00;27;49;19
Troy Skinner
Totally

00;27;49;19 - 00;27;53;15
James Gwertzman
it's a different social structure. I think you're still social structure. I think you're right. But it's very, very different.

00;27;53;20 - 00;27;54;02
Crystin Cox
Hmm.

00;27;55;06 - 00;28;18;02
Troy Skinner
Well, that's different again from Animal Crossing or Skyrim, which, you know, Skyrim's more identity based. You know, Animal Crossing or Sims is much more about your, your creative persona, about who you are as an artist within that tool box, you know, so they all have social identities, but they have different things that support them because the genres are, you know, essentially allowing people to differentiate themselves.

00;28;18;07 - 00;28;18;16
Troy Skinner
Totally.

00;28;18;19 - 00;28;40;02
Crystin Cox
And I love that framework because as someone who worked in MMOs for many years, you know, for a long time, there was this sort of general idea especially outside of the MMO development space that's like, yeah, MMOs or there's something about guilds, I guess, that kind of makes people stick, or maybe people feel just like really obligated to meet some sort of like rating, you know, calendar.

00;28;40;02 - 00;28;57;00
Crystin Cox
And, you know, it's somewhat frustrating because when you work in that space, I think you do get a greater sense for like now there's a lot of...there's a lot going on in there around. Yeah. Player identity, shared community history. And the way that those things come together is really different for each game.

00;28;57;27 - 00;29;14;21
Troy Skinner
Yeah. You know, a really good example of that. I don't play the game because it'd be too hard for me and take too much time. But all the stories that pop out of Eve Online, I mean, all those and we talk aboutplayers narrativizing their experience and the community's narrativizing their experience like that game generates the craziest stories.

00;29;14;28 - 00;29;27;09
Troy Skinner
And people are like mining rock for like three years to eventually tell this like peak story. Wow. That's a really good example of that, that in effect, we're not really content based. It's some narrative, shared narrative and the player base has had.

00;29;27;10 - 00;29;46;21
James Gwertzman
So, so I've got a pair of questions that are related. The first one is we're talking a lot of intrinsic motivations here. What about extrinsic motivations? Like leaderboards or, you know, stats, or you're just talking about, you know, my narrative why I'm playing with my internal narrative. But then it's related to external factors like a uinque piece of clothing I have or a unique, you know, cosmetic item.

00;29;46;29 - 00;30;27;04
Troy Skinner
Yeah, great question. So intrinsic and extrinsic motivation are a little bit squirrely. So if you do something like pay your kid to do homework, that's going to be interpreted in most cases as controlling. And what you'll find is over time, the kid will transfer motivation from whatever intrinsic value there is from doing well in school. And the joy of learning to I do this for five bucks and it'll actually devalue their, you know, the reasons why they do homework and their motivation to do it, especially in the absence of the $5 will paint So intrinsic motivation in contrast is I just want to do it for my own purposes.

00;30;27;04 - 00;30;50;10
Troy Skinner
So there are all these complex places where it's subjective whether a stimulus is intrinsic or extrinsic. So a perfect example of that would be something like a reward in a game for finishing a quest or hitting a certain place on the leaderboard. And what it has to do with how it will profile for an individual is basically how they frame that experience.

00;30;50;19 - 00;31;12;12
Troy Skinner
So if I have a long term goal to hit max level and finishing this quest will give me a sword that makes me level twice as fast, even though I was in theory incentivized to do the quest for the sword I feel like I have ownership. I'm owning that relationship because that reward is instrumental to my personal goals.

00;31;12;20 - 00;31;31;24
Troy Skinner
So, even though you could theoretically say that's an extrinsic reward, that's not how it's experienced by someone who has essentially made a plan to get this sword because it would be the best way to level and it's the same way with leaderboards. If you're only value for playing a game is being number one on the leaderboard, that's going to just going to be purely extrinsic motivation.

00;31;32;01 - 00;31;52;01
Troy Skinner
But if instead what someone is saying is, I play this game and my goal is to get better at the game, and I want feedback that validates that I'm getting better whether that is something on a leaderboard or a cosmetic or someone in my guild noticing that I have a certain title. And essentially my goal is to be successful and to be recognized.

00;31;52;09 - 00;32;05;06
Troy Skinner
Then these theoretically extrinsic things like achievements, etc., etc. are now things that are interpreted by an individual, a player, as supporting their own personal goals. And so it profiles as intrinsic. Does that make sense? James It's a little bit squirrely.

00;32;05;24 - 00;32;21;18
James Gwertzman
That sounds, that sounds great. I get that. That makes total sense. I'm actually part of me is thinking about that in a context of relationships. I'm thinking a lot of you're describing, I'm thinking of like, you know, dating someone or, you know, the arc of a, of a long term marriage or something, which is sort of yeah, but that's a whole different topic going back really.

00;32;21;18 - 00;32;52;19
Troy Skinner
It's really similar actually. I'm glad you said that. One of the reasons people have midlife crisis is because they essentially accomplish all the goals they had in mind for relationship or for their life. And what essentially happens is people need to be growing or they run into problems and so what happens in an arc of a relationship is you're meeting someone, you're having lots of new experiences, high novelty, you're feeling like you're growing your relationship together and then the relationship plateaus and then you're in the elder game. And then you're going to get much more into social identity.

00;32;52;25 - 00;32;58;21
Troy Skinner
So your instincts are dead on. Has the same structure because it's all human motivation. So it applies and well.

00;32;59;26 - 00;33;15;12
James Gwertzman
That's exactly what's going on. It was, you know, your notion of elder game in a relationship was exactly what, you know, you're going to go in or else you are going to stagnate and get in a rut. Okay. Different topic, though. Totally. Switching gears, let's talk about monetization because, you know, we talk about engagement and boosting engagement and how that's, you know, primary goal.

00;33;15;25 - 00;33;33;14
James Gwertzman
But obviously to run a successful business and to be successful as a developer, you have to ultimately monetize that engagement or else you're not going to be able to keep growing the game and paying the bills. So and at the same time I know a lot of times, again, developers often struggle a little bit of monetization. It feels a little bit like, yes, they have to and it's important, but at the same time.

00;33;34;02 - 00;33;46;07
James Gwertzman
I know for some people it's sort of like a dirty word almost seems like I don't want to worry about money even though I know I should. How do you talk about the role of, of, of, of, you know, getting players open their wallets and what the motivations look like for that and how that ties into your framework?

00;33;46;14 - 00;34;04;29
Troy Skinner
Yeah. So I think, you I know we didn't plan this out, but you set this chain of discussion up perfectly. So if we just go back to the discussion we had about rewards, how they can be interpreted as extrinsic or intrinsic depending on people's goal states and how they're oriented towards it, it's the same way with monetization.

00;34;05;12 - 00;34;29;29
Troy Skinner
So if for example, in the game you have a goal to accomplish to be a very successful player that's very important to your guild and helps you and, and, you know, like you have a central role that you want to fulfill. That's your goal state. And there's something that you can purchase that does not ruin someone else's experience, that supports that.

00;34;30;14 - 00;34;49;20
Troy Skinner
Then people are going to have very high value for that monetization. Or a simpler example is in something like a battle pass, if people's goals are to stand out in their peer group and there's a way in which you can buy a paddle pass and then work hard, play skillfully to earn a cosmetic before the other people in your cohort does.

00;34;49;27 - 00;35;17;20
Troy Skinner
So you can be the first one to have a certain skin on, then what what monetization is in this content is a way to support people's intrinsic motivation. They have goals that they want to accomplish and paying helps support that. It becomes way more complex in certain environments because if you're in a competitive multiplayer game, that's a zero sum win loss situation and you pay for gameplay advantage.

00;35;17;26 - 00;35;43;00
Troy Skinner
Then when some person spends, it obviously essentially disadvantages everyone you're playing against. So that's where it gets very tricky but if everyone's essentially running their own story and people are mostly playing together or playing side by side, then it's, you basically just need to understand their goal states and basically sell people something that they've that essentially helps them accomplish their goals.

00;35;43;09 - 00;36;13;21
Troy Skinner
An extreme example which, you know, goes both ways. If you test something like a 4x game, like Game of War or Game of Thrones, essentially you can think about the metaphor for monetization there kind of like poker, high stakes poker, where to play at the top tables, to be in the guild and be some of the head of a guild that fights for the throne there's a huge table stakes, thousands of dollars in terms of upgrades to play at that VIP table.

00;36;14;04 - 00;36;40;17
Troy Skinner
The people that spend that money have extremely high satisfaction for both the game and for the spend. In the same way that someone likes to play, high stakes poker would have high satisfaction for. The people that aren't satisfied, that are sad about that monetization scheme are the people that have the goal to play at the VIP table and don't want to spend the money and so there's no simple answer to how monetization motivation works.

00;36;40;17 - 00;37;01;02
Troy Skinner
What you need to basically think about is two different dimensions, which is what are people trying to accomplish in the game? And is there something that allows them to accomplish that more efficiently in a way that feels fair to both them and to the rest of the community? And so if you can knock off those two pieces within genre conventions and you tend to have at least a pretty successful foundation to go forward.

00;37;02;24 - 00;37;21;11
Crystin Cox
Yeah, I like that. I mean, it definitely tracks with a lot of my experience of in some ways it's very simple, just makes something very valuable that people feel great about buying that, you know, serves their goals. And in some ways it couldn't be more complicated because it's so specific to every game. And there's no easy answer because.

00;37;21;24 - 00;37;47;00
Troy Skinner
That's the problem. People want it to be tools like I see a lot of PMs feel like there are techniques that apply in all contexts, and it doesn't work that way because you have to understand where identity's coming from. What people are trying to accomplish in a particular game. And then essentially, for lack of a better term with the economy in that game is where value comes from, what rewards are, and then what the interactions between rewarding one player is and the rest of the group and their competitors.

00;37;47;07 - 00;37;54;08
Troy Skinner
And so that is not a one size fits all problem. That's a have to work through it in your context problem.

00;37;55;16 - 00;38;21;03
Crystin Cox
And it's really very similar to what you're talking about before as far as rewards and this question of like is it extrinsic or is it intrinsic, it's the same problem I had so many discussions with people about daily log in rewards and this sort of constant debate when you have to deliver like some design criticism where you're like, this is going to this is not going to be effective for very long because this design for this daily rewards system is going to feel really extrinsic.

00;38;21;19 - 00;38;27;08
Crystin Cox
And the people like but but other people do it and it's great and it's like, yes, because it doesn't feel extrinsic.

00;38;27;12 - 00;38;27;21
Troy Skinner
Yeah.

00;38;27;23 - 00;38;29;13
Crystin Cox
It's you have to actually design it well.

00;38;29;15 - 00;38;44;22
James Gwertzman
Well, and what it means is you end up it's so interesting because it's like you look at a pile of things for sale in a store and what you're saying is behind every one of the in a well-designed game and everyone is objects is a different set of motivations and reasons for that object to be there in that store.

00;38;44;24 - 00;38;59;27
James Gwertzman
Yeah. That it's not just like, oh yeah. And this is the armor buffing the field buffing the this buffing that up off. It's like no, this is for the player who wants this and this is for the player wants this and this player wants this and it's there. They all fit together. They reinforce each other. But ultimately in a well-designed game, you know, sort of true.

00;38;59;27 - 00;39;09;12
James Gwertzman
You really do have something for everyone. But it's not just because I like red and you like blue, it's because I want to feel important and I want to feel mastery. And each player is going to need a different thing.

00;39;09;18 - 00;39;27;28
Troy Skinner
That's exactly right. And if you another way to say it, which is exactly what you just said, but just in different words, which is people that people have different skill levels. People are in different parts of the game. And basically what people are going to do is try to write a narrative that feels feasible for them given their skill level and experience level.

00;39;28;11 - 00;39;55;00
Troy Skinner
That's going to give them social validation. And if someone is brand new to the game and not good, then the things that they can get validation for are different than something that's been playing for four years that's the best in the world. And you need, if you want to sell to all those people, then you need to understand what each of those player profiles need and have merchandise or experience or you know, acceleration opportunities that are appropriate to each.

00;39;56;00 - 00;40;17;22
Crystin Cox
This is all really good stuff. I know we're almost out of time, so I want to make sure we ask our favorite question that we ask everybody, which is can you share a LiveOps disaster that you've experienced, just any, anything that you've that you've seen where things did not go right? We always love to get these from people.

00;40;17;22 - 00;40;19;18
Crystin Cox
I think it helps keep us all humble.

00;40;19;18 - 00;40;20;01
Troy Skinner
Yes.

00;40;20;01 - 00;40;21;17
Crystin Cox
And they're always interesting.

00;40;21;21 - 00;40;46;12
Troy Skinner
Well, I have tons of humbling stories, luckily, so I haven't done a lot of LiveOps, but I did do it once where I came in maybe in the last four months of The Matrix online. And I was a developer with no LiveOps background and the job available in Seattle that day was someone to run tech ops community and customer service.

00;40;46;12 - 00;41;14;01
Troy Skinner
And so that's the job that I got. And the game was just about the launch and we didn't have enough content. And so one of the brilliant ideas that someone had maybe I was involved, I can't remember, was we were going to basically get GMs on our team to impersonate big character's, roleplay them in the game. So people like Morpheus, Trinity and so we were writing narratives outside the game.

00;41;14;01 - 00;41;32;13
Troy Skinner
We were contacting players. We're kind of like, you know, doing this like live action role playing in an MMO. And then, you know, after doing this for like three weeks a month, something like that, and kind of having a very loose handle on why this was interesting, what you should do with it, what the rules were, that of it were how it worked, etc..

00;41;32;20 - 00;41;56;21
Troy Skinner
I got a call in my office from an executive from Warner Brothers, and they told me they had Laurence Fishburne on the phone and he'd heard that people were playing Morpheus in the game and he wanted to see it. And I had 5 minutes to set something up, so I had no plan. I had no idea, you know, we had no script or, you know, there was nothing to go with.

00;41;57;01 - 00;42;29;28
Troy Skinner
I had no experience, no plan, no no preparation, no nothing. And so I remember going into a room where we have lots of people that kind of did this job and someone was tasked to play Morpheus, and we got some players in this faction together, and he was giving a speech. And of course, because, you know, it's an MMO, it's the Internet, the first thing that everyone all these are students trolling Morpheus doing lots of lewd animations in front of him, talking hash, all while Mr. Fishburne is, you know, watching on screen and listening to it on the phone.

00;42;30;06 - 00;42;42;23
Troy Skinner
He was not pleased. The executive of Warner Brothers was not pleased. You know, we use some console commands to affect the situation but it was not it was not a time when I covered myself in glory.

00;42;43;24 - 00;42;45;29
Crystin Cox
Wow. That's a great story.

00;42;46;06 - 00;42;49;10
Troy Skinner
Is it? It felt pretty horrible, actually. But there you go. Yeah.

00;42;49;15 - 00;43;10;09
Crystin Cox
Well, now you can look back, though, and we can all we take it all go, wow, that sounds rough. I will say, though, that that kind of behavior, the GMs coming into like occupy characters and do things live and in a game is something that we used to do all the time and early MMOs And we don't do it very much anymore.

00;43;10;17 - 00;43;15;18
Crystin Cox
Some of you've just highlighted some of why. Not just because we're all...

00;43;15;18 - 00;43;16;23
James Gwertzman
Players' unpredictability, you mean?

00;43;16;24 - 00;43;34;26
Crystin Cox
Yes, it does not. Yeah. Not because we're worried that movie stars will watch us and get unhappy. But but because players are pretty unpredictable and it can be really challenging to do that stuff live without a great plan. But it's something that's kind of an untapped area these days. There's some interesting stuff you could still do with it.

00;43;35;14 - 00;43;41;24
Crystin Cox
Maybe not as much with a like beloved iconic character...

00;43;41;24 - 00;43;42;10
Troy Skinner
Yeah, it was tough.

00;43;42;10 - 00;43;43;29
Crystin Cox
...that's like owned by a giant corporation.

00;43;44;12 - 00;43;46;24
Troy Skinner
It's tough, tough. It was a tough hand.

00;43;48;11 - 00;44;02;17
James Gwertzman
Alright, well thank you so much for your time today. This has been a phenomenal episode to kick off our third season with because it gives us a framework for so much of what we talk about. Here on the show that we've never actually really had before. So I'm grateful for your time today.

00;44;02;24 - 00;44;03;24
Troy Skinner
Thanks for having me. It's fun.

00;44;04;19 - 00;44;09;25
Crystin Cox
Yeah. And thank you so much again, Troy, is there anything you want to shout out before we're done?

00;44;10;06 - 00;44;27;21
Troy Skinner
Just very briefly. Obviously, I'm a researcher and I spend all my time thinking about why games work. And I work at a company called Immersyve, which is IMMERSYVE. And so if you like to contact us and talk about video games, that's great.

00;44;28;06 - 00;44;29;28
Crystin Cox
Awesome. Great. Thank you so much.

00;44;32;01 - 00;44;34;08
Crystin Cox
Thanks for listening to The Art of LiveOps Podcast.

00;44;34;14 - 00;44;39;14
James Gwertzman
If you liked what you heard, remember to rate, review, and subscribe so others can find us.

00;44;39;22 - 00;44;44;09
Crystin Cox
And visit PlayFab.com for more information on solutions for all your LiveOps needs.

00;44;44;19 - 00;44;45;14
James Gwertzman
Thanks for tuning in.