Zynga’s chief designer shares tips learned from social games

Zynga has become the envy of the video game industry because it has so many users for its Facebook games. So Brian Reynolds, chief designer at Zynga and a veteran of the traditional video game industry, got a lot of attention for his talk about designing hit social games.
The talk at the Dice Summit was also an interesting contrast to the morning speech by Jesse Schell about how traditional game developers fear Facebook. Reynolds said that Zynga has six of the top 10 games on Facebook with more than 239 million monthly active users playing games that don’t cost anything to start playing. FarmVille, Zynga’s most popular game, has more than 79 million monthly active users. About 3 percent to 5 percent of those players actually pay real money to buy virtual goods inside the games. That small percentage is actually generating lots of revenues that has driven up Zynga’s valuation and enabled it to raise $180 million in its last funding.
“This is why this industry is being taken seriously these days,” Reynolds said.
Typically, the Facebook games are built with $100,000 to $300,000 budgets over four weeks to 12 weeks with a team of 10 or so people. By contrast, console games cost $10 million to $40 million over two years with teams that can hit 100 people. The Facebook games are more like services, updated every week with new features for users and responses to user feedback, whereas traditional games rarely offer such updates. FarmVille itself was built in just five weeks.
“Welcome to the web,” Reynolds said. “The whole game is sitting on servers in a room. You have control of the entire game at all times. The metrics are available in real time.”
Zynga pays attention to the reach of a game, its retention rate, and revenue generated. Any new features that are introduces should be simple and align closely with one of those measured goals. In FarmVille, for instance, one of the most effective new features introduced had to do with the withering of crops. If you didn’t pay enough attention to your crops, they would wither. Then it would be embarrassing for you if your friends visited your farm. So, Reynolds said, shame is a very good motivator to get people to keep coming back to tend their crops.
The pillars of success are play (having something fun to do), express (a way to share your success with others), and invest (to do things in the game that will pay off in the future). If a player is successful with a game, they gain social capital because their real friends will see how well they are doing.
If developers aren’t sure what effect a feature will have, they measure it. They can test a feature with just five percent of the users to see if they like it. And they can run parallel tests, such as seeing whether users respond to a text message with pink lettering or red (pink works better).
Carrying features over from traditional games, such as difficult fights with a boss character, make intuitive sense. But they won’t work unless they inspire some kind of social behavior (like requiring you to ask a friend to help you fight the boss) or the purchase of a virtual good. Reynolds ran out of time while giving his presentation. A few folks in the crowd groaned when he skipped the screen on “how to make money.”
In closing, Reynolds said don’t mindlessly port a game from old platforms to the new, forget all that you know about how design should be done, remember that fun game designs drive users to return, and be humble enough to change your plans if user data suggests you need to do so.
Reynolds will speak on a panel on Next Generation Social Games at our GamesBeat@GDC executive game conference at the Game Developers Conference on March 10. And if you’re a game entrepreneur, consider entering the Who’s Got Game contest for best game startup. Finalists will go up on stage at GamesBeat@GDC.