Zynga: The Journey From Farm to City

Zynga VP of product development Mark Skaggs used his session at GDC today to discuss what his team learned between shipping monster hits Farmville and Cityville on Facebook - and his main advice is to always try and achieve the impossible.

Farmville wasn't originally Skaggs' idea. At the time, he'd been working on a mediaeval RTS game for Facebook, but technical restrictions meant that development had stalled. It was during a development meeting centred around how to apply Skaggs' extensive experience working in RTS development (he'd been project lead on C&C Red Alert 2 and C&C Generals) that fellow executive Bing Gordon threw an idea into the area.

"Why don't you guys just make a farm game?"

The idea had merit. Farming, he reasoned, was essentially resource management, much like the core of any RTS title. However, the slow pace and rhythm of that management meant that people would play at their own pace, coming back daily to keep tabs on progress. Skaggs took the idea on board and decided to run with it, but could such a parochial concept ever challenge the current star in Zynga's crown - Mafia Wars?

Skaggs' strategy was simple. Get things done quickly. If you can't make something outstanding quickly, then don't worry. Make it acceptable and put it in anyway. If you can, all the better, but don't get stuck trying to turn every aspect of your project to gold. Getting to market quickly was the most important factor. This concept of 'minimum value product' was the core of Farmville's development.

Skaggs took control and responsibility - becoming the "single strong voice" who made all decisions on what would go into the project, avoiding any prevarication on the development process.

"We adopted something I called five minute design. If we couldn't figure out how to make the feature really fun, we got it out of the way and right into the game."

That policy was taken as gospel. The reason Farmville's characters look so similar to Yoville's? They're exactly the same code - ripped from the previous title and dropped in wholesale.

As the project grew, people were pulled in from across the company to work on it. Human resources weren't in short supply, but tech resources were less easy to come by. Hosting and serving the content was going to be more difficult. Skaggs was told that the release of Farmville would have to wait until the tech used to power Mafia Wars was upgraded, at which point the new game could use those second hand resources.

But there was a second option. Use the as yet unproven cloud technology to power and distribute the game.

Taking that option, Skaggs said was a pivotal decision. Having the ability to add new servers simply and quickly meant that rapid growth could be easily accommodated. Adding that hardware via the traditional model would have taken weeks - stalling Farmville's growth considerably.

When Farmville launched it had no way of making money. There was no monetisation option, no in-game items for sale. There were no metrics. As a business model, it didn't look great. As Skaggs puts it - "there was no parity with competitors, but we offered the better experience."

But Farmville prevailed. Jokingly a target of 400,000 daily users was set for 11 days. Farmville achieved that number in just five days. Skaggs was given the received wisdom that Farmville would peak at eight, perhaps twelve million DAUs. At the height of its success, Farmville touched 32 million daily users.

The market was clear - Mafia Wars ads, which featured a smoking gangster and scantily clad women, were pulled from Farmville because Skaggs wanted "a mom, sat playing Farmville with a baby on her lap, to not see anything which could upset her." The formula was set. Simple core gameplay + Zynga best practices + moms = success.

Those same moms would go on to spread the word among friends, marketing virally for Zynga without even realising it.

So Farmville prospered, massively. But what next?

Facebook had changed. People were saying that Farmville had been a once in a lifetime opportunity, that nothing else could ever grow so fast or so far. A ceiling expectation of 10 million DAUs was set.

Skaggs changed his process. The idea of minimum value product was out. Instead the mantra became 'get it good or get it out'. Quality became the new focus.

However, two thirds of the development team were fresh graduates. The title was nearly handed off to the newly acquired offices in Austin, thinking that a new perspective was needed. Skaggs' new focus on polish over punctuality meant that a globally organised launch date, with television and online media focused and aligned and waiting, was missed. The game wasn't ready and wasn't good enough, Zynga's audience would have to wait.

Within two weeks of launching, Cityville hit eight million DAUs. After 30 days that number was 14 million. Farmville took four and a half months to reach 20 million. Cityville took just ten weeks.

The process paid off. Zynga might come in for criticism for its rapid release schedule, the impression that each product is a reskinned version of the last, but in reality, at least according to Skaggs, the development process has changed completely. Polish, quality and finesse are the new keywords.

That's not to say speed isn't important. Getting to market quickly is still an incredibly important factor, particularly when there are plenty of other companies who appreciate one of Skaggs other points - if someone else is doing something well for an extended period of time, you'd be stupid and naive not to copy it and incorporate it into your next product.

That said, ambition is clearly at the heart of Skaggs' manifesto. He began the conclusion to his speech with a quote attributed to another hugely successful entertainment mogul who often came in for criticism over his methodology and dominance: Walt Disney.

"Sometimes," said Uncle Walt, "It's kind of fun to try the impossible."