Many of Kabam‘s social mobile gamers play for three hours a day, seven days a week. But sometimes, they can complain for just as long.

That’s the blessing and the curse of Kabam’s success in the strategy-gaming niche, where it has published titles like The Hobbit: Kingdoms of Middle-earth. These games have landed Kabam in the top-10 grossing games on mobile platforms, and they’re just part of a larger portfolio of titles that are spreading the company’s risks and creating opportunities for breakout successes.

But success comes at a price. In Kabam’s free-to-play games, only a small percentage of players pay for things. And those folk are very demanding, particularly when they see changes in the games that they don’t like. Andrew Sheppard, the president of Kabam Studios, has to think about this design challenge as he takes the games of the $300 million company to the next level.

I interviewed Sheppard onstage at the Mobile Gaming USA event this week in San Francisco. Here’s an edited transcript of our interview.

GamesBeat: Andrew, since you’ve been there four years, can you tell us what it was like when Kabam got started?

Andrew Sheppard: I joined Kabam about four years ago, as you say. At that time Kabam was about 30 folks. The kind of legend that we talk about is, the original Kabam was a company called Water Cooler. They built social and community apps for Facebook. They had significant success early on, when the platform was a very open space, getting up to 60 million uniques off just community apps that went viral.

Then the ad markets hit a difficult patch, and Kevin, our CEO, realized the opportunity to produce a kind of content where consumers would pay you directly to participate in the experience, and ideally – in the context of free-to-play – pay only when they felt the content was good enough to warrant paying for it. He felt that was a path for the company that would align it more with the consumer, and give the company much more durability.

In 2009 they started building Kingdoms of Camelot. I joined in December of 2009, when the game was still in beta. It was a small office on Castro Street in Mountain View, right above a dim sum shop, which you couldn’t avoid. You had to take the elevator up and be locked in the car with dim sum. But over the next four years, we grew substantially, from a 30-person shop in Mountain View to a 100-person team in Redwood City, and then to a 350-person team in San Francisco and Redwood City, and then today an 800-person company.

GamesBeat: How quickly did you know that Kingdoms of Camelot was your big hit?

Sheppard: In all honesty, when Kevin and I were talking about the role—I took the role more on instinct than I would have historically. But I had a feeling, when I first looked at the game and Kevin and I discussed the early high-level metrics, that it would be really successful. But at that time we didn’t really know.

I could see right away that the community was alive. Not only were there lots of blogs about the game, but inside all of our games, we have—We were the first company to push what would be typically classified as MMO features – alliances, chat, group messaging – into a browser and now a mobile game. The confluence of, this is a game built by a group of passionate developers who love these games, and the users inside the game are telling me it’s great, and it’s on this Facebook platform that doesn’t have any core games currently and is growing like a juggernaut, it felt like the right intersection of forces. We had the platform, the right alignment with the consumer, and developers who love this type of strategy game.

When I joined, I remember my revenue chart for the year was $2 million. We beat that by 16 times the next year.

Andrew Sheppard, president of Kabam Studios

Above: Andrew Sheppard, the president of Kabam Studios.

Image Credit: Dean Takahashi

GamesBeat: You had an interesting dynamic in Kingdoms of Camelot, where you had these 100-person clans back then. Somebody would cause trouble, and then everybody would attack that one person. All 100 would pile in at the exact same moment and you’d have a giant battle. But there was no animation. It’d be over in an instant. You’d get these spreadsheet results back. And people loved it.

Sheppard: My arc through gaming has gone from working on publications like Gamespot, then moving to traditional triple-A companies like Electronic Arts, and then ending up in web and now mobile. I remember a lot of my friends who were still working at traditional game companies fell into the camp of just being opposed to social games. Our game, although a core game, was classified as such.

Certainly web technology wasn’t at the point where you could visualize combat to the degree that we wanted, or battle resolutions, or even the way that players made consequential choices in their universe. But speaking to the power of the core gameplay mechanic, especially for strategy gamers—Strategy gamers care enormously about being able to set up a number of configurations for how they want to participate in the world and then seeing how those actions play out, either in a turn-based fashion or an incidence-based fashion. People were able to develop quite complex emergent gameplay patterns with their alliances to take down competitors or, in the case you described, encourage people to act in socially constructive ways.

Games do go through, over the life cycle of a server, these different states of being in complete chaos – almost like Wolf Run, or an early Automata type of thing – to being in a steady state. It’s fun to see that play out and change over time.

GamesBeat: And you’re able to reskin these games – the same game – with different themes like Rome or Atlantis.

Sheppard: Early on, we went down the path—on the Facebook web, we had a couple games that were very similar. We had Kingdoms of Camelot, Glory of Rome, and Global Warfare. What we learned by doing that was, there wasn’t enough difference in the core gameplay to justify these latter games existing at scale. What you’ve seen with Dragons of Atlantis and Kingdoms of Camelot for mobile and The Hobbit is, we’ve gone much more heavily into adding differentiating gameplay. Although it’s very subtle, we’re changing the configuration of how the battles work and the way the universe coexists to drive a different type of emergent behavior that’s targeted toward the audience.

The Hobbit is a strategy game, and we classify it in the subgenre of empire strategy game. But it has factions, and it increasingly has a feature set that’s more appropriate to a mass market audience. Kingdoms of Camelot is much more hardcore. It’s kind of a ren faire audience. They’re looking for a lot more lore about Camelot or Avalon or Arthur. We try to invoke that in that universe.

Dragons of Atlantis

Above: Dragons of Atlantis

Image Credit: Kabam

GamesBeat: And then what you get out of this is loyal users. They’re playing six days a week sometimes.

Sheppard: Absolutely. Across Web and mobile, although the stats are somewhat different, we find that our average paying user plays three hours a day, six days a week. It’s astounding, because when you think about the battery life on some phones, they don’t get three hours if you’re using them constantly.

Pirates: Tides of Fortune

Above: Pirates: Tides of Fortune

Image Credit: Kabam

GamesBeat: Monetization also followed. How did that part happen? How did you guys figure that out?

Sheppard: I don’t know how many people in the room worked back in the social days, when virality was the only thing people talked about. But very early on, Kabam, consistent with our approach to market entry and company strategy, chose to differentiate from the crowd that was doing virality. It felt that virality was, one, not really great for the user experience, and two, something that you couldn’t ultimately control.

There’s always going to be a bad actor in a viral ecosystem that sends so many notifications that the platform has to clamp down. Just by virtue of creating that ecosystem, it’s a race to the bottom as far as platform behavior. In contrast, with monetization, you’re leading into something we believe very heavily at Kabam, which is that free-to-play is the most democratic form of game development. People will only pay to participate in your game if they see value in what you’re selling. Because of that relationship, you immediately are going to build your game in a way that paying users care about. Ideally, too, in a way that encourages more people to pay. It’s easy to listen to a vocal minority and end up building something small. We generally aim not to do that.

GamesBeat: What have you learned about the motivations people have for buying things? I’ll just put mine out there, which is—I’m in a 100-person clan in The Hobbit. If I fall behind, and I’m like No. 100 in the clan, I’m going to worry that I’ll be kicked out, because I’m too small a player. They want to grow. So the motivation is there to just start spending like crazy. I haven’t done that, but the fear of being kicked out of my clan is a good reason for monetizing.

Sheppard: That’s definitely a driver. The way that we look at product development, we actually start by thinking—this is pulled more from traditional gaming, but we think about the primary and secondary audience for every game. We map it against the genre preferences for those audiences.

Some players are very much driven by the social engagement component, whether it’s keeping up with folks or helping other people. Some people are driven by competition. Some of our competitors from the web talked about revenge or dominance being a motivator for spend. We’ve actually found that that form of monetization is less durable. It’s also more toxic to the community. When you have someone who has the ability to spend a large amount in these games, and you give him the power to decimate everyone that’s casual, it creates a bullying mechanic. It encourages people to not come back. It’s not fun to play with that person.

There are also people who care about progression. This is more common in RPGs. The RPG user cares a lot about advancing through something, whether it’s advancing through a character tree or advancing through a universe or a storyline. They’ll often pay to advance through that storyline.

Then there’s also people paying to participate in something that’s of limited duration. Whether it’s a sale or a promo, which is the more commonly understood concept, or an event like a boss raid in some of our newer titles–Phoenix Age, with Underworld Empire, is really strong at this.


Above: Wartune, a Kabam mobile game

Image Credit: Kabam

GamesBeat: You can gift currency or a really valuable thing over to other players. You can buy something with your own real money and then give it away. That makes you the most popular person in your clan.

Sheppard: Absolutely. Helping others is a big one.

GamesBeat: Did all of this translate into mobile fairly easily?

Sheppard: It was a big transition, moving from web to mobile. For those of you that started off in web and are either approaching mobile or in mobile now, it’s such a different ecosystem. In many ways, especially as the fidelity of the devices improves, it’s becoming more like console or, if it’s Android, more like PC. The release cycles are longer. The updates take more time than web. With web you can make realtime updates and see it in the game right away. The games are much higher quality. Our newer games are substantially more expensive to build – not just the visual quality, but also just the depth of the features and gameplay that people expect is much higher.

GamesBeat: What’s the team size or the usual budget these days? Is a million bucks going to get you very far?

Sheppard: Depending on geo – we have studios in Beijing, Vancouver, Seattle, and San Francisco – we have teams ranging from 20 on a very light casual-core game all the way up to 75 on a live product that’s at scale. The Hobbit is about 75 folks. Dev costs can range between $2 million and $6 million, depending on which project you’re talking about. In some cases we’re anticipating higher dev costs.

The functional mix on mobile is very different from web. Not only for our games, where you generally have a persistent meta layer, which is either server-based or global, there’s more backend complexity. Relative to console, you have more backend engineers. Relative to web, you have much more frontend and UI engineers. A lot more QA as well.

GamesBeat: You guys crossed $300 million last year and doubled revenues. That was basically being successful in mobile?

Sheppard: Yeah. Mobile, as a backdrop, is an incredible platform. You’re looking at 35 percent growth on mobile revenues over last year, in gaming alone. $12.3 billion global market. 45 percent of that is Asia, 24 percent North America, 4 percent Latin America, the rest of it Europe and the Middle East. Gaming apps are 40 percent of all app downloads. Games constitute 75 percent of revenue.

For us as a company, we’re a bit unique in that all of our games are long tail games. We focus on building games that run more like television shows. We want to produce, with a beta test, a pilot television program. If the market deems it’s good and people see value in it and want to pay money for it, it becomes the first season of a game. Through expansion packs we drop new episodic content – effectively new series of content that, at each step, tries to align with where the consumer wants the game to go. That’s the democratic element of free-to-play.

We’re literally talking to users, both inside the game and out — where do you want to see this thing arc? Both paying and non-paying users.

GamesBeat: There’s some sophisticated content that you can launch, sort of like a DLC in consoles almost, but some of it is also very simple, like just holding a tournament. Tournaments are things that the clans go crazy over. They’re all trying to get first, so they can get more free stuff. That seems like it’s simple to administer, right?

Sheppard: Much simpler to execute, and to some degree you can automate it. We’ve modeled our operations to be more off of an Asian developer. We have what we call live operations, but at some other companies it would be called service operations. The function would be game masters. They’re charged with, in many ways, being the party planner for each server, where they jump in every day and they’re running a schedule of events. Sometimes just purely engagement, sometimes monetization-related. Puzzle & Dragons does this very well. I’d say, though, as a western developer, we’ve done the best job of pulling that over and adapting it.

Part of the reason why we generally bias away from automating it is we find that the preferences for types of events – not just the mix of events, but the payouts, the duration, the challenge – vary substantially by country. Perhaps better described as by language and by time of day? You want somebody that’s attuned to the audience and knows the gameplay super deep to be in there with the users making active decisions about what programming is happening at any given time.

Kabam has published Fast & Furious titles under a license.

Above: Kabam has published Fast & Furious titles under a license.

Image Credit: Kabam

GamesBeat: I’m curious about the portfolio strategy versus the super major hit strategy. Supercell has just three games. King is really built on one major game. EA has 900 games, but it can’t catch up with those guys. You guys have the portfolio strategy, focused on a particular genre. Is this going to be a winning strategy?

Sheppard: There are pros and cons to both approaches. The portfolio strategy gives you the ability to diversify risk a bit. Interestingly enough, all of the companies, relative to their revenue, probably have about the same percentage of dollars running into R&D. It’s much more about whether you allow the market to see the games that you think are not going to be hits. As a company, we’ve chosen to put more games out than not, because we feel that free-to-play revenue experience, which in traditional games would be called shipping experience, is the most important experience for the teams to have.

For those of you that have lived through shipping a game, whether it’s traditional or mobile, whether it’s service-based or a packaged product, including the 99-cent games as well, there’s nothing better than putting it out there and getting immediate feedback from the audience on whether it’s working or not. It needs to have one or two cycles to try to steer the game toward what that audience really wants.

Our ethos as a company is, we believe in our experience to date, especially in the categories we’ve had experience in, but we’re incredibly humble about categories we don’t know. We assume the best way to learn is to begin that dialogue with consumers. When we pushed from strategy into racing games with Fast & Furious, and when we pushed into RPGs with Heroes of Camelot and now Underworld Empire and Castle Age, in all of those situations we went out knowing that the games weren’t perfect. We were tabling a lot of internal debate as to modify the games and letting the audience tell us what we should do.

GamesBeat: The loyal fans, did they move over from Facebook into mobile? Or are you mining new users?

Sheppard: The approach we took with Kingdoms of Camelot, which was our first mobile title—It was a great confluence of factors. It was original IP, a franchise that had reached 20 million users on the web. When we brought it to mobile it ended up being the top-grossing app of 2012 on iOS. Surprisingly, there’s a low degree of overlap with the web. Part of that was because we didn’t connect the backends. We also adhered by the platform guidance of not doing any user migration. We didn’t feel that was necessary.

The thesis for that was, the taxes are equivalent. There’s not really a huge benefit to that. Users will find it one way or another. What we were trying to find out is if mobile reached a different audience, and what we found is that at that time, it did reach a different, but overlapping audience from a user standpoint, but a very different audience from an engagement standpoint.

A good way to describe that is to think about the way that you live your lives today. If you’re anything like me, you’ll use your mobile phone at a conference. You’ll use your laptop at the office. You’ll use your tablet at home in bed. Each one of those use cases effectively makes you a different consumer. When you have a game that’s restricted to Facebook and web—we were able to tap into a share of time that we couldn’t otherwise. It ended up working out well.

There was a huge benefit for us separating the two games, also, because we made subtle changes to the way the session designs were architected. We steered into shorter session times, and we also removed some of the more core elements of Kingdoms of Camelot web. It’s a more punitive game design, where players can hurt each other more.

The Hobbit

Above: The Hobbit

Image Credit: Kabam

GamesBeat: You have this benefit of having loyal users with a large percentage of them who will monetize. The double-edged sword is that some of these whales are very demanding. There are maybe a few incidents that spell this out. I think there was a policy change in Dragons of Atlantis that closed a loophole, basically, and sort of shut down some activity that was not really desirable. But you had a huge protest, to the point where the whales are putting out petitions to Kabam.

And there was another recent virtual goods hack, where some veteran players took advantage and used that. There’s this paranoia in tournaments, where players saw others who were really fast and winning in tournaments, and they were suspecting that Kabam was trying to boost the tournaments so that everyone would spend money to try to beat this unbeatable player. You have these very loyal users who are really annoying as well.

Sheppard: [Laughs] It’s definitely a balance. The passion that people have for the games—we have people in Kingdoms of Camelot web, which we’ve invested in for four years straight, but if you look at the product it doesn’t seem to have changed substantially. In part because the user doesn’t want it to change. There’s this concept that we talk about internally called “don’t move the cheese.” If people are used to the button being in a certain location and you move it for no good reason, they get upset. It changes their pattern of engagement.

Those people, they love the game. They’re in there for social reasons, to compete on whatever vector is important for them. They go through cycles of spending and not spending, loving and hating the game, loving and hating Kabam.

Outside of things that our outside of our control, like policy changes or malicious attacks by outside parties, everything that we do internally is governed by, what’s the right thing for the user in the long term? For us, one of the strategic pillars of the company is franchise durability. We want all of our games to be, ideally, decade-long franchises. That’s a difficult balance when you have not only a large player base, but an incredibly passionate player base, with very different views on how the games should run.

We try to learn from people who’ve done this well in the past. Blizzard always does an exceptional job. CCP, the way that the engage the community and the governance and the roadmap design for Eve Online is exceptional. We’ve had situations where both non-paying and paying users can vote for features to develop in our games. That’s been very effective.

Dragon of Atlantis

Above: Dragon of Atlantis.

Image Credit: Kabam

GamesBeat: The distinction versus those companies, though, in free-to-play you have two to 10 percent of the players paying for something. Those are the ones who are doing the complaining. Normally, in game design, you can ignore the extreme minority and design to make it appealing to the vast majority of players. In this case, those are the guys paying you. You can’t ignore this small group that wants to steer you in a certain direction.

Sheppard: Our paying users are classified into different tiers of VIPs. Yes, it is difficult to separate the noise from what’s really important to users. Sometimes we’ll have situations where we’ll have an event—I’ll take an example where we make a mistake. We run an event incorrectly, maybe the payouts aren’t done precisely, and we’ll have a paying user come in and send a ticket to customer support and say, “I’m so mad at Kabam! I’m never going to spend on this game again!” And you’ll literally see on the support ticket that he just made a purchase.

In those situations, we always engage with the customer, especially if it’s a top level VIP. In some situations we’ll even call the users and talk to them about what happened and try to understand. Can we abstract this into a thing that can be fixed? A lot of the team investments we’ve talked about in the past are purposed with getting this kind of operational stability for our users.

Now, with respect to design changes, that’s a much more nuanced engagement method. There’s always a significant percentage of passionate players that are just going to dislike change initially. The overall arc is, if we do make a design change, which we generally avoid in a loop that’s already working, we do wait to see how things stabilize. Both on a metrics basis, so it’ll be all of the standard KPIs that pretty much every conference talks about, but also on a qualitative basis, through customer support tickets, through internal dev chats with users, and also with our chat. We’ll have our game masters give us a sense of how people feel inside the games.

GamesBeat: Do you communicate in advance with players about this, before you roll it out?

Sheppard: If it’s a really big feature, we’ll do a single-server rollout, very similar to the ways a lot of companies do a Canadian beta launch for a game. We’ll usually split test the design there to make sure it’s not game-breaking. Once that’s known good, we’ll roll it out worldwide. Or if it’s a feature that we’ve already deployed in one game and we know that the users love it—let’s say alliance raids, something like that. Then we’ll have a little more confidence and go live with it out of the gate.

Generally speaking, some degree of tension with the user base is good. If your players are not unhappy with changes you’re making, then either you’re not making enough changes, or you’re not making aggressive enough changes to keep them engaged. Dropping back to the TV metaphor, we’ve all had series we watched where we disagree with the way the writers took the storyline, but by virtue of diverting from what we expected, it becomes a form of entertainment unto yourself – something you can debate with your friends. We think that’s okay.

GamesBeat: If you have to do something like sunset a game, you’re probably often faced with the opportunity cost—if I take these 10 or 20 people and put them to work on a potential blockbuster, that payoff may be a lot more than whatever revenue is coming in from an older game.

Sheppard: One of the biggest challenges for us in particular—if we were building games that launched, made money, and then just went to zero, it’s a very different calculus from when you build a game like Kingdoms of Camelot web, and you’re literally running it for four years, with aspirations to run it for 10 years, profitably and with a team on it.

For us the calculus is much more about, what is the level of innovation and investment required to keep that long-term arc on the existing product? We start there because we view franchise durability as the most important foundation of the business. Then, how do we grow the experience and capability of the overall staff at Kabam to always stay linked to where the market is? For anyone that’s working in the space, the pace of change in the space is relentless. Having your best people stay on one title has an opportunity cost to it. It means they’re not being forced to stay contemporary with the market, which means that their skill sets are degrading. Very much like traditional, teams that have worked together have much higher hit rates, especially if they’ve already had a pattern of success, whether it’s in a particular genre or across multiple platforms.

Our general people strategy, and it’s part of why the company has grown so large relative to our competitors, is to always have our best people building forward, launching, and then fortifying games. Then have newer people, to the company or the industry or the platform, backfilling and learning as they go. There’s always enough tension on personal and professional development at all times.

Castle Age

Above: Castle Age

Image Credit: Kabam

GamesBeat: What’s the future of free-to-play?

Sheppard: I personally believe in the power of free-to-play. As a company, we feel the same way. The parallels in other forms of media, I feel, are very clear, where you see almost all forms of paid media moving to models where large majorities of the population can access the content at some quality level for free. Whether it’s streaming video on the web or listening to the radio on your cell phone, all that content is effectively free.

If you look at music as a parallel for gaming, then the example Dean gave about tournaments being a key driver of engagement in our games—these days, with the emergence of the 360-degree contract, people pay for concerts. They pay for the opportunity to participate in a temporal, special experience with an artist.

What that affords us is the ability as an industry to produce content that’s more mass-market than it’s ever been. I’m sure everyone’s had this experience. You can talk to your grandmother or your extended family, and they’ve probably played Candy Crush. It’s kind of mind-blowing. Three or four years ago, smartphones weren’t at that high a penetration rate.

Then, within that, you have percentage points of people within that ecosystem who are actually paying to support the ecosystem. All the developers are aligning their dev efforts against what that broader audience wants, and very specifically against what that narrow audience wants. It ends up being a space that is actually much harder on content developers, where the rising bar of content quality that we’ve seen in traditional gaming ends up being accelerated, because people have gone through it before and they’re anticipating it. But it’s also much more nuanced. You have to convince people not only to pay in your game, but also to take time away from other games they’re playing today.

I do think it’ll cascade outside of mobile. It’s interesting to me. You’ll see microtransactions introduced to a title that’s asking for $50 or $60 up front. Outside of DLC, what I’ll call macrotransactions, there’s a big resistance to microtransactions in those games. But I’ll tell you, if you start giving away those high-quality games for free–A lot of those people are already playing our games, or Supercell’s games. It’s not like they’re not familiar with the model. It’s just that they’re resistant to paying on both sides. I think you’ll see free-to-play expand outside mobile and web.

Question: For free-to-play, what do you think is the biggest untapped opportunity, something that’s going to be big in the future?

Sheppard: My view might be a bit different than most, but one area I’ve been spending most of my time on in the last year is international. I believe that it’s easy, in North America, to just think about the North American market. Most American game companies organize their business operations, their product development, and their product, to be very geocentric. It’s unintended, but it’s simple to think about it, in contrast, by looking at the games that come out in China. The games coming out of China are monoculture. They’re built for that society. Usually when we see a Three Kingdoms game coming to the U.S., we know it’s not going to be successful.

Starting with just the pitch for our games, all the way through launching the game, scaling it through marketing, and then doing live operations and effectively the sales and marketing inside of the game, we’ve built out a global distribution and sales and marketing organization. It’s given us the ability to not only grow our presence in Europe by double-digit amounts over the last six months, but also push into markets with more confidence and more scale than we would have — markets like Russia and Turkey, which are pure growth markets right now.

I feel like there are very few companies — especially in the core category, which is more operationally complex – that can build that out. It takes a commitment up front, and a belief that this is a global market as opposed to a regional market.

If you think about what the opportunity is behind that, Puzzle & Dragons is just a single-country product, right? Supercell and King are much closer, but they’re largely western products. All of those associated games are massive. They’re meaningful relative to 10-year, 20-year triple-A franchises that everyone knows and loves. So what is the potential for a game that actually does go global in appeal?

There’s a lot of reasons to believe that may not happen, or it may be further off. But that’s a big opportunity. I think someone should build for it. That’s what we’ve been focused on.