Jesse Schell

[Amy Jo Kim] Thanks for joining us today, Jesse, for the Getting2Alpha Podcast. Tell us a bit about your background.

[Jesse Schell] I started programming games when I was maybe about twelve years old, but I’d started making board and card games when I was even younger than that. It had always been something of a hobby for me. I started getting interested in the idea of becoming a real life software engineer as I got into high school. The idea of actually becoming a game programmer, game designer had never, that didn’t even seem like it was a real possibility. I didn’t know anybody had a job like that or anybody who knew anybody who had a job like that. It would have been like saying, “I’m going to become a movie star,” or something. I was focused on just straight up software engineering and I focused on artificial intelligence for a while, but I was always making games on the side.

Then, when I went to Grad school, virtual reality was starting to come out, like ’93. I thought that was really interesting so I started getting into that. That ended up leading me to the Disney virtual reality studio where I started out as a programmer, but gradually took on more and more design responsibilities. That was where I got my real education, in terms of entertainment and in terms of game development and game design. I was there for about seven years doing both interactive theme park attractions as well as massively multi-player games for Disney. I worked on Toontown Online. I was lead designer on that.

Then I moved to Pittsburgh where I started teaching at Carnegie Mellon University at the Entertainment Technology Center, teaching game design there. That’s what led me to write the book, The Art of Game Design. In parallel with that I started my own studio, Schell Games, which I guess it’s been ten, eleven years now, something like that, that I’ve been doing that. We’ve grown it from, I guess we started with four or five people, now we have a hundred people. Mostly focused on educational and transformational games. We do about two-thirds of those and one-third entertainment games, so a lot of variety in the sorts of projects.

Tell us what you mean by transformational games.

Transformational games is the phrase that we like to use that’s meant to encompass both educational games, but also games that are designed to change the player in other ways. Some people use the phrase serious games, I’ve never been a fan of that because it makes it sounds like the games, like if someone accidentally had fun playing your game that would be bad, and of course that’s not what anybody means. The idea of a transformational game is a game designed to change the player. That might be I’m trying to change their knowledge, I want them to know certain things so I’ll have them play the game. It also might be I’m trying to change a habit that they have. That would be a kind of transformation. Or maybe I’m trying to change their attitude about something. All of that falls within the realm of transformational, so that’s a phrase we like to use.

You also mentioned The Art of Game Design, your book, which is now in it’s second edition. What’s the creation story of that? What sparked you to actually sit down and put that into the world?

Boy, that’s kind of a long, that book was a long time in coming. I’ve always been interested in entertainment, what makes entertainment work, the psychology of entertainment. An aspect of my background that I didn’t talk about was I used to be a juggler and a street performer. I worked my way through high school and college doing that stuff. It actually helped me a great deal as I moved into the realm of Disney and working in their theme parks, and even just in games because those were experiences, for me, making real entertainment and dealing with real audiences on a day-to-day basis.

A breakthrough book for me was a book called Magic and Showmanship by a guy named Henning Nelms, written probably in the ’50s where even though it’s extensively about magic, it’s really about what is entertainment and how does entertainment work. That book really got me thinking and when I was at Disney I started writing various articles and things about the nature of entertainment, what are interest curves and how do those work? How does interactive entertainment work and how do you tell a great story but still keep things interactive? I started writing bits and pieces and I’ve been working on this project that I’ve called the Understanding Entertainment Project, kind of riffing off of the Understanding Comics text by Scott McCloud.

I ended up talking to a publisher about it and he’s like, “Nobody’s going to want to buy a book called Understanding Entertainment. No one cares about that, but game design is really hot and you teach a course in game design. Do you think you could write a book about game design?” The more I thought about it, I’m like, “You know it’s all the same stuff.” That’s what got it rolling. The Art of Game Design is very much based off of a lot of these things that I’ve learned over this long period, but a lot of it really congealed when I started teaching a game design class at Carnegie Mellon so some of what’s in the book comes out of there. The central premise of the book is that good game design happens by looking at your game from many points of view, which in the book I call lenses. Lenses are just simple questions that you ask yourself about your game and in the book I enumerate over a hundred lenses that are different ways you can look at your game in order to make it great.

Fantastic. You’ve taught, since then, many, many students and you’ve worked with training up young game designers who’ve joined your company, as you mentioned it’s grown from just a handful to now a hundred strong. What are some of the most common mistakes that you see people, particularly first-time entrepreneurs, or first-time game designers making when they’re creating and testing the very early versions of their games?

Oh boy, oh boy, oh boy. There’s a lot of common mistakes, definitely. Wow. It’s sort of different, it’s interesting. The mistakes students make end up being really different than the mistakes entrepreneurs make. I think that’s because they’re often approaching things very differently from each other. When it comes to entrepreneurs, I think a lot of times, one of the most common mistakes is this sort of notion that game development is simple, that there’s some simple formula that you follow and that you’re going to have a guaranteed fun, guaranteed successful game. If only they had that formula then they’d be all set. That’s definitely one.

Another huge one for entrepreneurs is the notion of, “all I need is a great game design and then someone will code it and then I’m in great shape.” When the truth is development must be very iterative. Things must be played into existence and a lot of people don’t, they’re not ready to plan for that, they don’t understand the realities of that. The worst part is when they start to run into that, they feel like they must be doing everything wrong because it wasn’t good right out of the gate when really nothing ever is. That’s just how these things go. You have a lot of, there’s a lot of mistakes and iterations and changes that are going to happen.

Following on that point, what do you do in your classes and in your studio to implement the powerful side of that, the habit of early quick iteration, failing fast, learning a lot from it, et cetera. How do you actually make that happen?

A big part of it is just the process that you use when you plan things out. You need to plan for that iteration. We very intentionally plan pre-production phases, which are all about getting as many of the early experiments out of the way as possible. We very much use agile development processes which allow you to change gears mid-stream and make big changes. We’re big believers in what I call the 50% rule, which is that if you have a fixed schedule as many of us often must have, you’ve got eight months or you’ve got four hundred thousand dollars or whatever you’ve got, halfway through your schedule you should have all of the main game features functioning so that you, in your second half you can spend time making them great. We try and be disciplined about those sorts of things. That really helps you get to the right place, and of course constant play-testing.

That’s fascinating. You’re really talking about the journey from pre-production, early experiments, what some people might call and MVP (Minimum Viable Product) even, to leaving a large chunk of time for polish and integration.

I would make a, not to harp on details, but I would make a very strong distinction between pre-production and an MVP.

Great. Let’s make that distinction really clear.

Yeah because typically MVP you’re talking about minimum viable product and, for me, that refers to the minimum viable product you could actually put out in the world to have people play and get meaningful feedback about that. Pre-production, the way we use that phrase is very different. Pre-production is you’re doing internal tests in order to prove to yourself that this concept could actually maybe work. The idea is that the end of pre-production, you have a pretty good idea of how long it will actually take to make this game. You’ve answered a lot of the early, early questions. Whereas MVP would be, from my point of view, the first fieldable version of the game.

Right, so that’s purely terminology. Some people use MVP to mean a fake landing page. People use that word in really different ways. The reality is it’s a continuum right? It’s not like this clear thing, but just to dig a little deeper into what you’re talking about, during pre-production, during these experiments, do you do any play-testing of small pieces or ideas of the game on your target audience?

Oh yeah, absolutely.

Let’s talk about that because I think I may be using, I talk about early MVPs that are, they can even be sketches or interactive prototypes that you specifically test, not just on your internal team and your investors, but also on your early adopter target audience. That’s one of the things that I find people have a hard time getting their minds around because they don’t want to show early stuff to who they think of as their customers. How do you do that with your focus on educational, transformational games, what practices and hacks have you evolved to get some of that crucial early feedback from your audience?

It’s a great question. It’s a difficult thing that everybody has to wrestle with. It happens in different ways. Partly I think it happens through, you can do it through simple conversations with your target audience. I remember when we created Pixie Hollow, which was a massively multi-player game for girls based on Tinkerbell, we had a whole design worked out that we felt pretty good about and we said, “Yeah before we get into any production, we should, let’s spend some time talking to girls and see how they feel about it.” We didn’t start by just saying, “Here’s what we’re thinking.” We started out by just asking them really simple questions, questions like, “Do you think you would like to be a fairy?” And we got an enthusiastic, “Yes!” Okay, that’s what you like to hear. “If you were a fairy, what would you do?” We thought we knew what the answers were going to be and we were wrong. The answer was, “Fly.”

We hadn’t been thinking about that because we’d been really focused on the Tinkerbell movie where Tinkerbell does a lot of things. She does some flying, but flying isn’t prominently featured, so we hadn’t been thinking about flying. Flying was really central with the girls we talked to. We started talking to more and more girls and like, “Yeah, wow. They really expect that.” That wasn’t in our plan at all. We immediately changed our plan. We made it so that you’re flying every second of the game. Early conversations can be really important and those can go a long way.

Not just early conversations, early conversations that are testing in a real way, looking for real answers, you’re fundamental premise.

Yeah, exactly.

Not testing like “side” things. The fundamental premise, right?

Yeah, absolutely. You’re fundamental premise is what are you doing.

Right. All that detail and all that beautiful design around the detail is for naught.

Yeah, because the stuff on the edges, you can change that later, but the central activity, the main loop of the game, if you get that wrong, stop, like what are you doing? Early conversations are one way, but then another way is you just, you learn to couch your early, unfinished prototypes in a way that you can get useful feedback.

Tell us more.

Well, it’s a tricky bit of business because people, artwork really influences people and if I give you something with really rough unfinished artwork I’m not going to get the best feedback. You as a designer, you know that. One of the rules that we know is if my game’s really ugly and it’s still fun, I’m in really good shape. It’s okay to test things that don’t look very good but have the right play mechanics. One of the advantages, you actually have a lot of benefits when you work with kids because kids have a good imagination and adults have a bad one. You can ask kids to imagine that things were different than they are. Adults have a harder time doing that.

The other good things is if you bring people really rough prototypes that are unfinished, they’re more likely to give you honest feedback. If you give somebody something really polished, they’re less likely to tell you the truth about how they feel about it because they feel like it’s already done and so that their feedback’s not going to matter. If you bring a thing that’s all kind of “janky” and broken and a lot of rough edges, they’ll tell you the truth about what they like and don’t like about it.

That’s a golden nugget. Let’s talk tools for a minute and process. Let’s say somebody is eager to follow through on that advice and they want to do it. What are some of your favorite prototyping tools? Do you just sketch stuff up and put it in a slideshow? Do you use Unity? What do you guys do for really rough mock ups that you want to get that early feedback on?

Everything is different, every project is different, it totally depends. For 2D games, we will sometimes use PowerPoint a little bit, you can kind of mock up what the game is going to look and feel like in PowerPoint. We’ll do a little bit of that, but usually we do like to just code the thing up in whatever system it’s ultimately going to be implemented in so that we can get traction moving that direction. A lot of that comes down to getting artists and programmers who are used to that kind of rapid prototyping because not everybody is optimized for rapid prototyping. It’s a very different way to work where you have to be okay with certain things being sloppy because a lot of it may get thrown away. The idea of prototyping is every prototype is designed to answer a question, the faster it can answer that question, the better a prototype it is, even if it’s a really ugly prototype. If it answers your question, then it’s the best one possible.

Fantastic. That’s a really concrete tip for entrepreneurs who want to do smart prototyping is make sure your prototype is designed to answer a specific question as quickly as possible. Do you have any other tips on, iteration is really a theme throughout, and I’m thrilled to hear it because that’s been my experience as well, both as a game and product designer, is that it might be 5% design up front, but iteration is where all the magic happens. What are some of your other tips for people that want to do this right, eager young entrepreneurs?

Right. Well, I definitely believe in, what we call the rule of the loop, the more iterations you do the better it’s going to be. It’s as simple as that. Anything you can do to optimize, you know, shorten the duration of those iterations, the better your thing’s going to be. That isn’t the way people usually think. When you choose your tools, you want to choose tools that let you change things a lot and change things frequently. If you could do multiple iterations in a day or in an hour, that’s much better.

The other idea that people often forget about, they get so focused on we’re going to build this “thing”, they forget about the idea that they can do multiple prototypes in parallel. You could perhaps do, maybe you’re going to do an art prototype purely as a PowerPoint, while you do a performance prototype that’s maybe coded up in Unity. Then you have a totally separate game play prototype that maybe somebody coded up in Flash and you could have three different people working on those three different things totally in parallel, in three totally different bases, but they’re answering parallel questions so that later you can put it all together. Parallelize your prototyping is a definitely good tip.

If your pre-production goal is to know how long it’s going to take to build the game, that seems like a really good thing to do. Your specialty, as you said earlier, is transformational games, particularly in education. Can you share with us some of what you’ve learned about doing educational games, particularly ones for kids, versus just the more general world of game design.

In a lot of ways, transformational games are not that different than regular games, it’s just much harder because when you do entertainment games, it’s important that they be engaging and that they be fun and that they get people coming back and they get people excited. Then, when you get into transformational games, you have to do all of those things and you have to really change the way a person looks or thinks or what they know. That’s pretty tough. One of the big things with transformational games is subject matter experts. It’s really, the business of educating somebody is really deep and complicated.

Even if you want to make something as simple as, “Hey we’re going to make a game about algebra.” It’s not enough for you to know some algebra, you need to have someone who’s an expert on algebra, and even that’s not enough, you need someone who is an expert on the best ways for people to learn algebra. What are the hangups that people have because when you’re creating an educational game, the idea is the game is an alternative. There’s already a way to learn algebra, but there must be something wrong with it because we are, people are asking for better and more engaging ways.

It’s really important to get yourself back to the question of what problem am I really solving? You need the help of the subject matter experts in order to understand that, so that you know that you’re solving a real, legitimate problem. So many entrepreneurs, they get all excited about some concept they have and it’s an exciting idea to them, but it’s not actually solving a real problem and as a result the product doesn’t really get much of a reception. You have to find a real problem, focus on that real problem and then use best-in-class expertise about getting that problem solved. That’s the part I like most about transformational games is game designers, game developers, they can’t do it alone. Subject matter experts can’t do it alone. They get to come together and create this thing that none of them could have made alone. It really makes a difference in people’s lives.

What is your experience with or thoughts about adaptive learning techniques within that world?

When you say adaptive learning, tell me more about what you mean?

It’s very popular among educational gaming startups, at least the ones I’m talking to, and adaptive learning means, like say Khan Academy does it in a light way. It means that the system, you can interact with the system and it will adapt the challenges that it feeds to you based on what you’ve done so far. On the one hand, you can say, “Well every good learning game would do that, of course.” On the other hand there’s very sophisticated algorithms for doing this across subject matter that certain companies out there offer as platforms. I’m curious if you’ve had experience either with those platforms or with building some of that tech into your own games, or into your in-house libraries.

My experience with that so far has been that the fancier the algorithm is, the less effective it generally is. I think there are some simple things that you can do in that regard, that end up making a really big difference. What I found is when people have these really deep, complicated neural net algorithms that are supposed to do all this and try and provide that, my experience has been as of yet, for all that effort, the difference it makes ends up sometimes being small and then the problem you run into is you end up being constrained by the system. You start to say, “Hey, I’ve got a creative idea about a different way we can do this,” and someone there who’s in charge of this big, convoluted adaptive learning system says, “Wait, wait, wait. Our system doesn’t handle that, so you can’t do it that way.” My experience, so far, hasn’t been great. I’ve found them not to be especially effective and that they can be kind of constraining. When you keep them simple then I think, I’ve seen some good results. There’s still, I think a lot of room for improvement.

That’s great to hear. Where’s your focus these days? What’s on your horizon? Where are you and your team spending your time?

We definitely have a mix of things and that’s always been part of the strategy for our studio is to have a diverse portfolio so that as the market shifts around, we can shift around with it. Some of the things that are kind of hot for us, definitely tablets replacing text books we see as a big trend, so we’ve been getting good educational games designed for tablets, designed to integrate with classroom experience. That’s been a big one for us. We’ve been getting into virtual reality. I have a long history with it personally and now that it’s coming to the market, I think it’s going to be very powerful. We’re doing some development in that space.

Related to that, we really like physical-digital hybrid experiences. We’ve got this game called Happy Atoms that we’re very excited about. It’s a actual physical modeling toy that models the way atoms bond into molecules, but then it works with an augmented reality app for your phone or for your tablet that does a visual scan of what the player has built, what molecule they’ve built. Then it can identify it and that all feeds into the game. Similarly we’re doing physical, digital museum exhibit hybrids and theme parks. Those are all big trends for us.

Do you need beta testers for your molecule game?

We absolutely do and if you go to Happyatoms.com you can sign up to be one.

Great. I’ll make sure to share that on the webpage along with this podcast.

Great.

Where can folks go if they want to know more about you and get your book and just learn more?

The easiest thing is go to Jesseschell.com and my information, information about Schell Games, et cetera, it’s all there.

Great. Just a couple more quick questions. One is, if you had a super power as a game designer, what is your super power? What’s your sweet spot, the thing that you most just dearly love to do?

Oh boy, oh boy. What do I dearly love to do the most? Boy that’s tough because I like so many things.

I feel like I’m pretty good at seeing experiences in my head. Being able to get ahead of things and… the thing I actively, and I guess this is the thing. I actively train on this is the ability to observe my own feelings about a thing that I’m interacting with. I talk about this a bit in my book. It sounds sort of weird, and I partly, like I do daily meditation which I feel very helpful in this regard because it teaches you to observe your own thought process while it’s happening without interrupting it. You can be a really, really efficient designer if you can interact with a thing, and observe yourself interacting with it, and observe your feelings and emotions, and then you can look at that and say, “Wait a minute, why was I… I felt frustrated there, or I felt a lack of interest there, or I felt overwhelmed there. Then, why was that?” That seems to be a thing I’m pretty good at and I love doing. Absolutely.

The other thing, of course, that I’ve had to get good at in more recent years is team building and team maintenance. That’s, I’ve been slower to get good at that, but lately that seems to be where I end up having to spend a lot of my time. It seems like it’s going pretty well.

That’s great to hear. What internal tools do your teams use for communication? Do you use Slack or HipChat or Atlassian?

Good question. One of our big beliefs is that we want teams to be able to choose their own tools. We don’t like mandated tools. Different teams offer best practices. Certainly wise use of email is very important. Some groups are experimenting with Slack. Since agile is really important to us, different teams are using different things to keep track of burn down charts and keep track of schedules. We’ve tried a number of different tools, but it seems like a lot of our teams are finding JIRA in the long run ends up being their preferred tool for project tracking. Really we use a wide variety. It depends if you’re working with outside clients, sometimes that changes the tools you want to use. Bigger projects want to use different tools than smaller projects. Mostly we keep it fairly simple.

I’ll tell you a very simple thing that we do, we make sure everybody sits next to each other. We do six to eight projects at a time and every time a new project starts, desks are moved so that the people working on something are all right next to each other. That face-to-face communication is the most important communication.

Thank you so much for your time and for sharing your perspective and wisdom, Jesse.

Sure thing. Great talking to you.