Lessons from a game guru: Candy Crush Saga creator once survived six months without pay

I’ve talked before about the charmed life of Tommy Palm, a key figure at King Digital Entertainment in the creation of the hit mobile game Candy Crush Saga. But in looking at Palm, it’s clear that the line between winners and losers is a thin one.

Palm was part of a team of four people who took a so-so game on the Web and converted it into one of the biggest mobile sensations ever. Last year, King generated more than $1.8 billion in revenues, most of it from one game: Candy Crush Saga.

That was enough to enable the company to go public at a valuation of $6 billion, making Palm a rich man. His story is like many entrepreneurs. He was once so broke that he considered giving up on the game business. He endured six months without a paycheck, and he almost took a job at McDonald’s to pay the bills.

King missed its targets for its most recent earnings, and the company’s stock price tanked. But for someone with humble beginnings like Palm, King’s current troubles aren’t all that bad. He has hit low points in his life before, as we learned during the Casual Connect conference in San Francisco. At that industry event, fellow game developer Arseny Lebedev (the founder of casual game house Signus Labs) interviewed Palm about his life. Here’s an edited transcript of their talk.

Arseny Lebedev: You’re an inspiration to me. You’ve been in the business forever. You have a nickname — Tommy “Danger” Palm. Tell us more about yourself. Where are you from? What’s your background? Why are you in games?

Tommy Palm: I’m from Stockholm, Sweden. I’ve been passionate about game development ever since the age of 12, when I had a Commodore 64 and learned programming with some friends. We made games together. I continued doing that on a hobbyist level until the end of the 1990s, when I started my first game development company, Jadestone. In 2009, I founded Fabrication Games, which King acquired in 2012, together with Alex, our technical director.

Lebedev: What did you work on at Jadestone?

Palm: Our first game was a war strategy game called Kodo, but the first big game that we made was a football manager game with online multiplayer. Mostly for web, but we had some mobile extensions to it. At the time, it was impossible to get those kinds of things working, though, especially with a business model around it.

Lebedev: What about Fabrication Games? That was a mobile company, right?

Palm: Jadestone eventually went more over to gambling, so Fabrication was where we took the mobile content we’d developed and continued with that. We focused solely on making cross-platform games. That’s what we figured would be big in the future.

Lebedev: This was over a period of 15 years or 20 years?

Palm: Yeah, basically, from the end of the 1990s.

Lebedev: What were some of the high points and low points in that long stretch?

Palm: They were often tightly connected. I started Jadestone straight out of school. There was no money backing it whatsoever. It was just me and some guys early on who were able to live on very little money.

We were blown away originally when we saw mobile games on the Nokia 6110. The multiplayer Snakes version was fantastic. We figured this was going to be huge. We just didn’t know it would take 10 or 15 years for it to get there. It was a lot of struggling with great ideas and the limits of technology — plus, nobody really had a clue about business at the time. But we figured out we needed that when we ran out of money.

I remember this vivid moment in early 2000 when I’d been living without any income for the past six months. I’d started to evaluate whether I was going to take a job at McDonald’s to support myself, or if I was going to continue doing this. We were at a conference in Gotland, and it had been raining. Everything was very gloomy. But we’d sent a proposal to Nokia, and we sat down at a bar and got the deal. The contract was huge by our standards at the time, and at that moment, the sun came out. I was almost in tears. That would save the company for another two years, and we started working very closely with Nokia. That’s one of my absolute highlights.

Another of those moments was later, with King, in November 2012, when the mobile version of Candy Crush was released and we saw the numbers for the first weekend. That was one of those moments you never forget. We had a bet going in the team as far as what kind of numbers we were expecting, and I was ten times under that number.

Lebedev: We’ll talk about that soon. But what about a low point? As an indie guy, sometimes the low points just seem so low, and you wonder why you’re even doing this.

Palm: When you’re an entrepreneur, you invest so much — not only money, but your time and your passion for something. The low points can be terrible. I’ve heard a couple of great leaders talk about how depressed they’ve become when their companies are doing badly. I’m amazed, when I meet entrepreneurs, that they can keep doing this. We need those kinds of people to create great companies.

Lebedev: King bought Fabrication. What was King then, and what is it now?

Palm: King has been around for 11 years now. A lot of the original founders are still working very actively in the company. They originally came from a Swedish startup called Spray, where they were working with a lot of cool web ideas.

Lebedev: What was the vision with King when it started?

Palm: It’s changed a couple of times. But it’s been focused on games. Casual, skill-based games were something King was doing for quite some time and leading in that small segment.

Lebedev: It was a website, right?

Palm: It was a platform for games that was shared with a lot of partners. These skill games were games where you would bet a dollar on the result and you’d be matched with someone of equal skills. If you won you’d get the dollar or whatever it was. But there’s a lot of regulation in that area now. It’s different in different regions. It was fairly big at the time.

Lebedev: How many employees are there at King now?

Palm: Currently we have more than 800 in 11 different offices. We have seven studios where we have game development going. We have a big studio in Stockholm, and then Malmo as well.

Lebedev: Why Sweden? It’s so expensive. It’s like 7 euros for a Coke.

Palm: There was a really good session at the Nordic Games Conference about why Sweden is such a good location. One of the things people said was that it’s very expensive, so we need to develop better tools to automate a lot of the really boring stuff. I found that interesting. Maybe the fact that it’s so expensive to hire people makes it important to be streamlined, to have just a few people doing creative work.

Lebedev: How were King’s earlier games conceived? They had hundreds of games, right?

Palm: Yeah. The website now has more than 200 games. They’re built by very small teams. The concepts can come from many different directions. We have sessions where we essentially do hackathons and brainstorm concepts. It’s just two or three people working for a short period of time to take something from an idea to a game concept that we can have players test.

That’s very important to the success of King. We’ve been able to eliminate a little bit of the risk in taking half of the company to work on somebody’s idea. Instead, we can have a small, agile team do something we can try out and get some numbers on as far as how it works.

Lebedev: I wanted to talk more about King and what’s going on there. One of the things I’ve noticed is that sometimes King’s artwork is really out there. It has a unique look. How do you think the visual aspects of your games affect them?

Palm: If you look at those 200 games we have there, they vary in quality because they’ve been produced over a long time. The games that come out of King now have a more consistent type of polish. One important thing to note is that we try to make our game teams own their product. If they want to go in a certain direction, they should be allowed to do that. That’s an important strength. It’s not so hierarchical. This is typically Scandinavian, I would say. It’s a very flat organization.

Lebedev: I ask the question because the original Candy Crush game, the Flash game, it’s not the best-looking game. But that’s interesting, because it was King’s most successful game on the site. Or it was in the top five.

Palm: The original art style was a little bit more gritty, I would say? It’s like French 1920 type of candy shop look and feel to it. When we did the Facebook version, we didn’t have a lot of that.

Lebedev: Let’s talk more about that. King traditionally has games on its site. Now the site is called RoyalGames. There’s all these games on that site. Those games moved to Facebook and then to mobile. But the Facebook games and the mobile games are tied together. Can you talk more about the idea behind that? It seems obvious, but it’s probably hard to execute, having a good sync between two different platforms.

Palm: It’s one thing to do that in theory. Getting it to work in practice is a whole different bag. That’s how Fabrication Games and King met originally. It was at a conference. We were presenting the same basic thoughts about what’s going to be big in the future. One of those things was games you could take from one platform and continue on another one.

In the early discussions, when we were looking at designing games like Candy Crush, we all wanted it to be the same game, but with input from PC and mobile. The user behavior there is also quite different. That was definitely a challenge. We took some decisions on the design there to, for instance, make the mobile game available offline. That was really important, but it delayed the original mobile release a bit.

Lebedev: So that was the idea from the start, that you were going to put games on Facebook and mobile and they were going to be synched. Bubble Saga was the first big success on Facebook, but that wasn’t the first Facebook game from King. There were a couple.

Palm: When we looked at the reason why Candy Crush, the first big game, had worked—There were a few game elements there. Your fingers weren’t ever in the way. If you looked at the first version of Bubble Witch Saga we did, it gets a little too small for mobile phones. We updated Bubble Witch Saga 2 so it would be much easier to play.

Lebedev: How did you pick which games moved to mobile? Was it just performance?

Palm: Today we have this vision of cross-platform for most of the games we want to do. They work on both platforms. But looking at it originally, it was a process of us transitioning from online games to Facebook games to cross-platform games.

Lebedev: Was Bubble Witch a big success? Like you said, the mobile game wasn’t the best. It was a little bit small.

Palm: The original mobile game was successful, but then we had Candy Crush coming after it, and that was even more successful of course. It was a good iPad game, the first Bubble Witch, but very small and low in detail on iPhone. Also, if you look at the differences for Bubble Witch Saga 2, you’ll see that we updated the colors quite a lot.

Lebedev: Does King employ witches?

Palm: That depends on who you mean. But no.

Lebedev: Let’s talk about Candy Crush. When Candy Crush came out, as you said, it was a big success. Was it featured by Apple when it came out?

Palm: Not originally, to my knowledge. There are many different reasons why it didn’t get a big feature at the time. But since then, it’s gotten a lot of features. We obviously work closely with Apple and Google.

Lebedev: I don’t want to press the subject too much, because there’s a lot about why Candy Crush is successful. But I just thought it was interesting that when it came out, its success wasn’t driven by Apple immediately. I think its success was driven by a lot of other factors.

Palm: Yeah. One thing that was a huge factor was that there was already a big audience on Facebook at the time. They were actively — if you looked at the Facebook app page, you’d see that they were already asking about a mobile version of the game every day, in big numbers. When we released the mobile version there was instantly a big buzz there. People downloaded it straight away.

Lebedev: Can you talk about some of the tricks you guys use between the Facebook and mobile version? My girlfriend would always switch devices to get more lives and so on. These are all design aspects that were thought through.

Palm: It’s a combination. It’s something we talked about originally, how we wanted it. Now we see that people really do that. They run out of lives on one platform and go right over to the other platform, which isn’t bad for us at all. It’s something we encourage.

Lebedev: I noticed, when I was playing Farm Heroes Saga, there used to be a loop where you could get infinite money. I brought it up to one of your guys and he said that was a feature. I thought that was pretty cool. Let’s talk more about some controversial things. Why isn’t Candy Crush a copy of Bejeweled? Because you get a lot of negative press about that. I don’t think it is. I think it’s two separate games.

Palm: I especially see that a lot from the hardcore community, the forums people. One thing that they’re not looking at, perhaps, is that match-three games is really the genre. We’ve made several match-three games with Candy Crush already. There were also many match-three games before Bejeweled. But these games definitely look at each other and see what’s been done well and try to add new elements to it. With Candy Crush, there’s the fact that you can combine power-up candies together, which is something new.

Lebedev: Is there a bit of a fear of change? My philosophy is that there’s never really cloning. It’s all just iterations. Maybe you can talk about your own opinion in that area. Some of the Fabrication games, they were simple games. It’s not like you copied a solitaire game.

Palm: The game development community thrives quite a lot when it’s not so much about patents on games. You can look at concepts and try to improve them if it’s something worthwhile for players to play and spend their time on.

Lebedev: You said the word “patent.” Why did King cause itself all these troubles with these public copyrights on common words?

Palm: I’m not an expert in that field at all. But there’s a great letter from our CEO on the website talking about those issues. One thing he says is that we have a policy of protect and respect. We’ve seen a lot of identical games, basically, that are trying to mimic a famous brand of ours and pull people in. We’ve been trying to stop that, and one efficient way of doing so is by having a trademark.

Lebedev: So it’s not that you hate independent developers. There are just SEO tricks. If I call a game something close to Candy Crush Saga, everyone will come to me.

Palm: We’ve seen quite a lot of that. There was an article in Business Insider that had a long list of games that were aiming to trick people who are unfamiliar with our games.

Lebedev: Do you think that independent games and companies matter in a concurrent ecosystem?

Palm: Yeah. We had this period in the console and PC space where people complained a lot that there was no innovation. Everything had stagnated a little bit because nobody wanted to take any risks. That’s something indies always do. I love to go to the indie showcases and look at the game concepts they come up with. A lot of innovations come from those types of games. I love a thriving indie community. In Sweden, we have a really strong indie scene. King takes every opportunity we can to try to sponsor events.

Lebedev: After the IPO, would you say King has turned evil?

Palm: Ah, no. Well, it’s obviously — King is not a person. It’s a big company. Many people, especially if you follow the forums, you can get that feeling, though.

Lebedev: I mean, I think King is a nice company. I like your games. But I don’t know what you have against raccoons. The villain in Farm Hero — you guys just hate raccoons, or what?

Palm: No, no. We like raccoons. But they make for an excellent villain.

Lebedev: I wanted to ask more about working at King. You guys have, what, 100 trillion offices?

Palm: More like 11.

Lebedev: You mentioned the agile process and wanted to talk more about that. I’ve never heard anyone at King talk about what happens inside. Could you talk more about the office culture, maybe, in Sweden and other places?

Palm: Sweden is where I’m most familiar with the culture. We typically use scrum, two-week sprints. People come together and talk as a team about what they’re going to do. We release new updates and content for games every second week. Typically a new episode with 50 new levels. We also take a look at old levels, when we see that people get stuck there a lot.

Lebedev: What language do people speak there?

Palm: Typically English.

Lebedev: So everyone is — because you guys have a Spanish office, a German office, but it’s usually English?

Palm: We also recruit a lot of non-Swedes into the Swedish office. That’s one great thing about Stockholm, for instance. It’s a good place to live, so people don’t mind relocating. It’s very hard to find somewhere nicer to live.

Lebedev: Let’s talk about the agile process. At Signus, my company, we tried to do agile, and I always tell everyone that you do, but I think that every time you hear about agile, it’s not fully agile, not a full process. It’s very difficult to get everyone to work like this. One of the pieces of agile is continuous improvement. So how do you guys do it? Do you physically have boards and cards?

Palm: Yeah, we do a lot of that. But it’s also — the teams can have their own versions of this, something that works really well for them. It’s important to be flexible in your methods.

Lebedev: What happens when two studios work together? Or does that not happen often?

Palm: We have a platform functionality in Sweden, for instance, that needs to be used by the other offices, so we come together several times a year to work on that. We travel a lot to different studios to update the team on changes.

Lebedev: What are some weaknesses that you’re seeing personally?

Palm: I don’t know about weaknesses, but we definitely have bottlenecks around recruiting engineers. We’ve been focusing a lot on recruiting and getting the studios up to strength. That’s gone fairly well, though, I would say. It’s impressive to see that the London studio was able to do such a great game with Farm Hero Saga, the first game that was released from there.

Lebedev: How do you work? What’s your personal management style like? How do you keep yourself organized?

Palm: If you work with me, you know that I’m a man of chaos. I don’t have a team at the moment. I work a little bit with design still, but I work pretty independently in that area. When I was the CEO of Fabrication Games, though, I personally liked to work with people who take a lot of responsibility themselves. Management by not managing was my favorite way to work. I think that’s pretty Swedish, to make sure that people have the tools, that they can fail once and hopefully they’ll learn from that mistake.

Lebedev: Let’s talk more about the market. It’s important to look at the market, because it’s really changed. A handful of companies are at the top. King is one of them. Who are King’s competitors? Or does King have any competitors?

Palm: It’s a question that the CEO could probably answer better. I would see it pretty much as — we’re aiming to do these bite-sized games that you can pick up and play at any time. The competition is whatever people use to spend their time. We see, at least in Sweden, that traditional broadcast TV is losing a lot of its audience. One of the reasons is that people are using their smartphones more. There are a lot of things competing for people’s time.

Gaming is a great entertainment form that way. You can take it out and play it for a little while and it’s still meaningful, especially in a social context.

Lebedev: You’ve been talking about a lot of derivative projects. You tried licensed goods a little while ago. Can you talk about how that’s gone? I have the Candy Crush socks. My mother bought them for me. Has that been difficult?

Palm: It’s super interesting work. I’ve been involved myself with the socks, for instance. We say at King that we’re primarily a games company. That’s what we excel at and what we’re continuing to do. But we want to do things that feel meaningful for players and fans, things that will put smiles on their faces. We’ve been doing things that feel like fun products to work with. We now have a licensing team in London, so we’re taking that more and more seriously.

Lebedev: King is great at doing these bite-sized games, like you said, but what about more mid-core games? What do you think of that area?

Palm: I’m a big consumer of games myself, ranging from hardcore to mid-core. But I think that when it comes to King, we’re aiming at doing more casual, social games. That’s where we’re seeing a big appetite in the market – games that you can play with your family members. Women have been really important to us.

Lebedev: King is very slow to move to new platforms. I don’t think you guys had any Kindle games for a while. It’s only been in the last year. Is there a reason for that? Why aren’t you on Blackberry, for instance?

Palm: Again, it’s very much a question about where we focus. Our teams are small. We want to be everywhere, but we can’t. We tend to do things that a lot of players are asking for. We haven’t seen a lot of fans on Blackberry asking for us to bring our games there. The challenge of going to that platform is too big for it to make sense.

The focus this last year has been the Asian market. We opened up offices in Japan and Korea, and we’re working with Tencent in China. That’s been one of our focus areas.

Lebedev: My last big question is, what would be your advice after 15-plus years as a developer? What might be three points of advice?

Palm: My first would be to be very passionate about what you do. You should go for something you’re interested in. With King, with my companies, it takes a lot longer than you would think. If you want to waste 15 years of your life, you’d better do it with something you really love. That would be my strong advice. If you’re passionate about making games, that’s a good area to work in.

Also, with indies now, I think that the mobile games market is in many ways very democratic. You have the opportunity to compete with big companies. But to do that, you need to grasp analytics and marketing, which are two big areas on one side of making great games.

Lebedev: I’ve never interviewed anybody, and I’ve always wanted to ask these famous questions. I’m going to ask you these questions and you have to answer with your first thought. What’s your favorite word?

Palm: That would be a Swedish word, Web Server Application Programming. That’s a course I created at the Royal Institute of Technology. I met the guys I still work with at King there. We were studying how to make big scalable server systems.

Lebedev: And what’s your least favorite word?

Palm: “No.” I’m a very positive person. I tend to forget everything that’s negative. I don’t like to work with people who are.

Lebedev: What turns you on? Candy?

Palm: My fiancé?

Lebedev: What turns you off?

Palm: God, I don’t know.

Lebedev: Clash of Clans?

Palm: No, I can’t think of anything there.

Lebedev: I don’t know how old you are, but — what sound or noise do you like?

Palm: In my spare time, I’m a nature person right now. I really enjoy the forests, bird songs and things like that.

Lebedev: What sound or noise do you hate?

Palm: That’s tough. Airplane noises? I spend a lot of time on airplanes.

Lebedev: You don’t have to translate for us. What profession, other than your current one, would you like to try?

Palm: When I was young, once upon a time, I took a career decision where I had to go for programming or for art. I’ve always wondered how I would work out as an artist. I actually did art for a few of our early game concepts, and it looked terrible, so maybe it’s good that I didn’t go into that.

Lebedev: What profession would you not like to try?

Palm: Like I say, I seriously considered working as a cashier at McDonald’s.

Lebedev: When you die, what do you want your epitaph to say?

Palm: “Try again”?

Question: Did Fabrication Games have an early iteration of Candy Crush in the works? If not, why did King buy you?

Palm: Candy Crush came about at King. It was a web game first, in 2011. The Facebook version was beginning development when we joined King in early 2012. We had the game engine for cross-platform mobile and tablet functionality, and we had an expert team for developing mobile games. We were two very good puzzle pieces to fit together. We felt that we didn’t have the skill set at all on the social side. But we had a small team that was very focused on mobile technology.

Question: What kind of features have you tried in Candy Crush that you wouldn’t like to repeat again?

Palm: I’ve gotten this question before. Basically, the stars were very much aligned for Candy Crush Saga. There’s not a whole lot of things that went wrong, looking back and seeing that it became the world’s biggest game. We’re a bit humbled by how everything connected so well over time.

I did go back to the team and asked them, “What would you say were our biggest mistakes?” One thing that came up from the original Flash version, which was made by one programmer in 2011 — that code base was actually continued and integrated into the bigger Facebook game. There was an inheritance of old code. The guy who made that code wasn’t even involved in the project anymore. It would have been much faster to throw that code away and rebuild it.

Another thing was the fact that there were two different teams, between the Facebook version and the mobile version. That meant, for instance, that we had the game calculating scores differently between platforms at one point. What we did then is move the teams together and corrected the same bugs in the mobile version that were in the Facebook version, so they’d calculate in the same way

Question: Have you ever played a Zynga game? Do you have any opinions on them?

Palm: A little bit? I played FarmVille, the first one. There’s obviously a lot of talented people at Zynga. Some of them are now working at King. I don’t have a strong opinion there, though. It’s tricky to be one company talking about another. But there’s a lot of passionate people there that I know and I’ve met. But it’s a big company. It’s hard to say something general.

Lebedev: I worked on some Zynga games.

Palm: Oh, yeah? Do you have any opinions?

Lebedev: I can’t say which ones, though, because I’d get a cease-and-desist for speaking about it. But I liked it. It was nice. Great people.

Question: Once you’ve created magic in one game like Candy Crush, what do you to re-create that kind of magic again in a completely different game?

Palm: We’ve been giving our game teams a lot of space to work on their concepts, making sure they make the best possible games they can. We’re seeing that working very well. If you look at the charts, it’s not only Candy Crush. Bubble Witch 2 is a game I’m playing a lot of myself.

Our idea is to make sure we come up with a lot of new game concepts in an early phase. We’re trying them out and trying to make them into casual social games. We see that there’s a lot of interest from people around the world. Today we have 352 million unique users, which is a humbling number to say the least. These people are hungry for new content. It’s not like they just play the game once and don’t want to try it again. They love having something to spend their time with in between the bus ride home or — well, in the U.S. you don’t ride the bus so much. Hopefully you’re not playing games while you’re driving your car.

Question: Apart from a great game concept, user acquisition is another aspect of making a great game. What do you think plays a bigger part for user acquisition — PR, ads, Facebook, viral, or cross-promotion through your own games?

Palm: The biggest channel for Candy Crush is people showing it to their friends and persuading them that this is a really good game they should download. Other than that, we’re obviously doing some advertising and PR. We feel that a broad approach has worked best for us. We work with Facebook, with PR, with advertising, and all these things come together nicely.

One difficult thing about marketing is that it’s not a linear equation. You don’t start with $100 and expect to have double the result if you go in with $200. The stakes are pretty high. But once you get to that point where have this ability—A lot of people have become aware of the game and the brand. Now we see that the cross-platform features work very well for us. When we have a popular concept, people know about King as a brand and want to play similar games. That’s an important factor now, though it wasn’t as important for Candy Crush Saga in the early days, because that was our first really big game.