Veteran strategy game designer Brian Reynolds has now been with Zynga for a year and a half -- long enough to watch the Facebook game landscape transform from simple word and card games to a polished Flash app experience.
"I think I see a theme in the new releases this year -- it's that maybe this year is the coming of more sophisticated game design practices [in social games]," Reynolds predicted in his GDC Online keynote. "That's cool for players and it's cool for designers, because the games are getting more fun and more interesting to work on."
Early social games were more influenced by best practices from the web, rather than by those from the world of game design. So at Zynga, Reynolds was tasked with applying those traditional practices to social gaming. The result is FrontierVille, launched in June 2009 with the original goal of experimenting with game design in the social space.
Reynolds hoped to figure out ways to make games "more social than they had been," and to use design skill to weave key lessons from Zynga's other successes into a new title. "We weren't trying to launch out in a radical new direction," he points out.
For example, hoping to make the "visiting behavior" that helped popularize FarmVille more social, one of the first things Reynolds wanted to do was give players the ability to click on anything on a neighbor's homestead. This represented an expansion on the far fewer options that FarmVille allowed.
And in FrontierVille, the environment pushes back, requiring players to clear and explore it -- something that "goes back to a classic game design pattern that [was part of] a lot of the games that I worked on," he explains. Refining the environment and receiving a reward for it is part of the Civilization games, for example.
But are these the elements that make FrontierVille fun? Reynolds says that the "Sid Meier school of game design" says that designers have to build, prototype and experience a design to find the fun -- it can't all be conceived on paper.
Finding More Fun
By playing the game himself, Reynolds learned it was more fun to get excitement and visual feedback from the environment in response to clicks. A reputation meter was added to further enforce the social nature of the game, so players can see how they and others score when it comes to helping each other's homesteads.
"The stuff that was great about FrontierVille was stuff that we found along the way," he said.
The invading varmints, the snakes and bears that arrive on the homestead were not part of the original design until "somebody was like, well, shouldn't there be wild animals on the frontier?" So, Reynolds says, they added snakes just in time for user-testing. All the testers who played the game was either surprised or wary when they first found a snake, but felt excited once they defeated one.
When a face for the tutorial was needed, Frontier Jack came along -- and the staff enjoyed him so much that he became a character in the world. "That was the first dawning of the idea that we ought to have some more characters," he says. Another mechanic unique to FrontierVille among social games, the ability to create a family, wasn't originally part of the plan, but once it came to the design table, Reynolds knew it'd stick.
Getting More Social
People use one another's Facebook pages to find information about one another, and people will become similarly curious about the families created by those they know. Will a friend create a spouse that looks like their actual one, or not? Players had a particularly good time designing their spouses, Reynolds observed. "There's lots of special social wiring going on here. We had hit upon, kind of by accident, a really powerful means of social self-expression."
The questing format was also unique to FrontierVille. Reynolds said that once Frontierville incorporated it, other games quickly hurired to implement quests, because they kept players engaged by giving them more to do. And it creates a sense of story, since players want to find out what happens next -- "and you can make them pay to find out what happens at the end of the story," he adds. "Just that very first story arc of getting married on the frontier helped us get players through the tutorial," he says.
The way early players reacted to a particularly funny picture of a sheep in their news feed notification helped the team see that they should continue to create opportunities for players to joke and comment -- in that way, more players feel like they're part of the game, but the game also facilitates real social behavior. The game becomes about genuinely playing with friends, not about spamming them and exploiting the connection for in-game gain.
"Fun monetizes well," said Reynolds. Putting designers into "the retention box" and have them work on retention features is a mistake, because it's the fun parts of social games, made by good game designers, that keep players around. "Game designers design better virtual goods that players want to buy," he says.
"What I really want to do when I'm on Facebook is learn about my friends. The game might be fun, but what I really want is ways to share stuff about me with my friends." Giving players opportunities to laugh together, talk about things they actually care about, makes games more genuinely social and thus more successful.
He emphasized getting a prototype running well and early, since that's when designers will really be able to see what's fun. But then, the development and polish needs to take a lot of time -- good games take longer than one predicts. "Maybe that's another talk," he jokes. "Sneaky ways to get more time... how not to ship your game, 101." But a bad launch can be particularly devastating for a social game, as players will never try a game again if they have a poor experience.
"If you have a bad launch, then you're just done; you're doomed," he says.
It's important to respect the lessons of prior designs, too. Just as Civilization synthesized mechanics from other successful titles into something new, so did FrontierVille mash up Mafia Wars and FarmVille, and add all the new twists Reynolds describes.
"These things really are games," he says. "People play these games for entertainment. All of the kinds of progress and leveling up and self-expression, all the things that we play AAA games for are in here, we're just talking to a different audience. Which does mean we're talking, in some ways, in a different language."
"When you're making entertainment you can't just keep making the same things over and over again. Innovation doesn't mean inventing a kind of game that nobody's ever ever seen before. You'll be lucky if you make one or two things like that in a whole career of game design," he adds. "But innovation also means combining the things we know in new and innovative ways."
"Remember that in social games, the social innovations are the most important of all."