GDC Online: Koster Shares Lessons Learned In Social Game Design

At a GDC Online talk in Austin today, Raph Koster, VP of creative design at Disney-owned social game developer Playdom (Sorority Life), challenged myths about the social game audience and gave triple-A developers tips on making the leap from big-budget packaged games to social game development. "The biggest thing involved in making this shift, making this leap is mental adjustments," Koster said, adding that to go into social gaming, you have to be more than just a "bad-ass game developer." Filling in at the session for scheduled speaker and co-founder of Playdom subsidiary Metaplace John Donham, Koster offered a series of tips to help with those important mental adjustments: Almost All Console Games Are Niche Koster showed a list of PlayStation 3 games that have sold over 1 million units. With games like Uncharted 2 and God of War 3 on the list, it represented a powerful lineup of well-made games that reached a large audience. Then he showed a list of social games that had over 1 million users – 219 games for the month of August alone. "That makes console gaming look like a very small marble in the toybox," said Koster. While Koster conceded, "it's undeniable that triple-A titles make piles and piles of money… [with social games], monetization is on a whole different kind of scale." Audience It's not true that people just play Facebook games on the side while doing "primary" Facebook activities like uploading pictures or checking on friends, Koster said. He also said it's inaccurate to say that people only play Facebook games for five minutes, or that social game players are unintelligent. People play these games for hours, he said, adding that 35 percent of the audience will purchase virtual goods at some point. "These people aren't casual. … [but] this isn't just a plain old hardcore gamer market," said Koster. "These are the people that watch Dancing with the Stars, not [The] Big Bang Theory." And they're willing to pay for the experience, he said. Get Independent Feedback Koster stressed the importance of tracking metrics and making necessary changes to a social game as rapidly as possible. He said the process of using metrics instead of drawing on his own experience and supposed know-how was initially counterintuitive for him, given his more traditional, core-focused design background (he was previously creator on Star Wars Galaxies). But by taking advantage of access to metrics, triple-A developers moving into social games no longer have to wait after years of development "to find out if something works," Koster said. Social game makers can find out almost instantly. It's better to "have your dream crushed after 48 hours rather than five years," said Koster. "…Establish a process that gets you feedback at every stage, including before you even start." The creative director also pointed to an important resource for fast play-testing and feedback, UserTesting.com. There, developers can spend $39 for a video of a visitor using the developer's social game or website, and also receive a written summary describing their experience with the product. Does Retention Matter? Koster told developers their "key metrics are virality, retention and ARPU [average revenue per user]." Without those elements, "your business goes away" he said. He added that despite the industry's constant emphasis on social games' virality, "It's not the be-all end-all anymore." "…What makes the game viral has to be inside the core loop of the game mechanic," said Koster. But he said today, virality is more about drastically reducing user acquisition costs and, in his opinion, "ARPU is the knob that's easiest to turn." Specialization Is Bad Another adjustment Koster said traditional core game developers need to make is to throw out the concept that specialization is good. A social game company needs to be structured considerably different from a core game developer, he argued. "Think of roles as hats," said Koster." You have to structure your team in such a way that all of these hats are worn by [a lot less] people." Koster said that in social game development a "bad-ass" level designer isn't as valuable as a level designer who was also a retail manager that knows about selling products at a store, for example. "What you need is people who can wear more than one hat." I Don't Care What You Think... Prove It "Not every feature makes your game better," Koster said. But with core games, it can be extremely difficult to prove that. Again, metrics can instantly show the effect of a new feature in a social game. Koster recalls a situation where Playdom put achievements in a game, for example, and the veteran team was confident that it knew how to implement them. But the metrics revealed a surprise: "[Achievements'] chief effect was that people recommended the game slightly less," said Koster. In triple-A game development, Playdom would've left the feature as it was, because they'd have no idea that it wasn't good for the game. Now, instead, Playdom is on version seven of the achievements system, Koster pointed out. "When we started I thought we'd be right 90 percent of the time," said Koster. But in reality, "our hit rate was 10 percent. … We're better now -– it just takes us three tries till we get something to work." Koster also encouraged developers to realize that "every user is a place to learn, and there are always more of them." He said it was hard to convince himself that it's worth annoying some players for the sake of testing and learning, but in the end it is to benefit the wider base and helps gain more users. Everyone Has To Know The Metrics And What Their Goal Is And those highly-important metrics should not be kept a secret from the team. Instead, project them on the wall, and use certain metrics to motivate the team and drive them to goals. "What it does is focus the team very, very powerfully," said Koster. However, "metrics are not a cure-all," he warned. "They are for optimizing. … Metrics are awesome, but you have to be just as creative, just as awesome a designer as you ever have been." There Are New Winners Today Koster also said not to listen to detractors who say that the social game bubble is destined to soon burst. While double-digit percentage user base declines for leading games like FarmVille make headlines, "it is not true that this market is over and done with. … The business challenges have just shifted," said Koster. Aside from declines of major social games, the smaller developers are coming up and making significant strides in the category, he said. It's Not Spam Notifications on Facebook are one of the more contentious topics between social game detractors and proponents. "Wall post spam is not spam to people who are playing the game," Koster argued. Koster said that the notion behind sending those notifications about a social game is the same thing that drives people to send birthday cards to friends and relatives. He argued that there's real emotion and social aspects that go into notifications. "'Virality' is in many ways a terrible name for what this really is. … It's denigrating to call that 'spam,'" Koster said. It's As Much Of A Business As It Is A Game "You can no longer be a designer that doesn't understand money," said Koster. "…You have to be constantly on your toes." With social games, tiny changes, such as a link placement or the color of an invite could boost numbers or drag them down. "You have to constantly treat this like a living business at every moment." Try And Level The Playing Field Finally, Koster encouraged the use of tools and techniques that can level the playing field with competitors. He suggested using cross-product marketing networks like Applifier and leveraging other games in your studio to bring them to other products. "Find other ways to broadcast your message," he said.