How ZipZapPlay Saved Itself With a Hit Facebook Game

[Curt Bererton is the CEO and co-founder, with Mathilde Pignol, of social game developer ZipZapPlay. Their company recently came to prominence with a social game called Baking Life, a bakery sim that grew to a peak of 6.7 million monthly active users in just three months.

The below story is an “as told to” monologue, drawn from a recent Inside Social Games interview with Bererton.]

ZipZapPlay started up in late 2007 to build PlayCrafter, which is still active. The short and slightly sad story about PlayCrafter is that it’s one of the better, if not the best game creation platforms that has been made. The idea was to make it as easy to create a game as it is to write a book. If Mozart needed to write a book, code a game and create music all at once, you wouldn’t have these great masterpieces. Game design is like that because you need to know multiple fields.

But the core flaw for PlayCrafter was the business model and the youth of the target audience, which inherently meant they didn’t monetize as well. It broke our hearts a bit that it didn’t totally take off and make us a bunch of money, because that meant we had to leave it behind and move on to another space.

So to some extent that was the origin – starting with something very hard and learning some hard business lessons. In retrospect it was our own naivete to assume we could do PlayCrafter when there wasn’t an industry around that concept. LittleBigPlanet should have been a sign for us to stop sooner. That was supposed to be the platform example, but its sales were not as high as Sony had hoped.

When we started to look at Facebook, the first game we built was called Cat’s Cove, in August 2009. You had these cat pirates, and they would attack each other in turn-based combat. Really the point was, we’re gonna build this Facebook game and get it out the door really fast, because we know we’re gonna screw it up. When we built that game it was about learning technology, what processes worked and what didn’t and doing a lot of split tests.

We did a lot of things wrong there. But the one thing we learned from PlayCrafter after working on it for so long, was if it’s not working out, you need to kill things and move on. And even at that time, the money was getting tighter. Lack of money is a significant motivator.

After Cat’s Cove we started seeing the world in a different way, which now is part of how we think. There’s a spectrum from pure innovation to low-risk products. I’d say PlayCrafter was way over on the hisk risk side, just short of crazy indie games. Zynga is on the other side — they almost exclusively place smart bets now, and they clearly have a recipe that works. Baking Life is more on the conservative end of the spectrum. Part of the reason was that we needed a win, otherwise we were going to sell the company.

But we always need to do something new that hasn’t been done before — it’s part of our DNA. So how do you balance the risk without dying? Baking Life, we felt, was a good compromise. We did a bunch of testing of various themes. Baking did fairly well, and we knew Restaurant City and Cafe World had done fairly well. So we changed the core mechanic to be more of a retail store with cashiers and customers that waited in front of them, and we did our new viral design based on user-generated content. Those things combined to make what we felt was a relatively smart bet.

We also knew that baking was popular outside this space, but hadn’t been done on Facebook (at least to our knowledge). Even if you think about Hollywood blockbusters, if it’s similar but different in some way, people will still like the next blockbuster. To use a hardcore gaming metaphor, people loved Quake, but then also loved Halo and Half Life. We’ll be adding new things based on successful core pieces.

What I hope to see is us moving towards more and more innovation, but as a small company, if we start way over here in crazy town, we might be successful, or we might crash and burn. I believe the best approach is to build revenue with smart bets and then use that revenue to fund the cool innovative ideas that could change the way social games are made and experienced.

Having built several things before, we knew fairly quickly, based on the metrics, that Baking Life would probably do pretty well. We didn’t know that it was going to do as well as it did. I like to believe that when you’re making any game, there’s a combination of skill and luck. So I think there was a fair amount of skill because of what we’d learned from previous games, but there was some luck as well. Who knows, maybe the retail store mechanic could have been boring and people wouldn’t have liked it at all, or creating cupcakes and baking wouldn’t be interesting or fun.

It did work out pretty well, and looking at our metrics, we’d already worked out a process by which we could improve them. Are people finishing the tutorial? If not, where are they dropping off? There’s a general philosophy — we look at metrics a lot, but don’t make decisions solely off them. They should influence design, but not drive the design, because you’ll end only being able to optimize or split test your way up to a certain point, past which you can’t make a big step over to a new idea.

Why is Baking Life successful? The art style, people like to bake, a solid tech stack, tasty cupcakes — there are a lot of pieces that came together to make it popular, and a lot of lessons that came back on what did and didn’t work about the tutorial, architecture, etc. With features like the CupCake Creator, we were trying to bring a little bit of our user-generated content flavor to something we think is important in the social space: the viral aspect, making something more directly meaningful to the person you’re sending it to. We like that idea, that UGC can be powerful, but we learned some hard lessons about how not to do it early on. Now we still like to include that flavor, but you can’t make the whole thing about it.

We could have a 10 hour conversation on what did and didn’t work. One thing people underestimate is that social games look simple, but they’re actually pretty hard. On the technical side alone, if you’re not moving fast you’ll lose users to people like Zynga, who have it down. We think the production values are only going to go up, and that’s where we want to be.

The success was to some extent a pleasant surprise, and we didn’t have a lot of money, so we didn’t have a big team to start chugging away on the next game. The big problem was getting the right people in place, and we set a pretty high bar for who we hired. If we were another company, we’d be at 30 people right now. Instead we’re 10, and may be 12 or 13 soon. We set a high bar not just for skill, but for personality.

Our philosophy is that we aren’t looking for prima donnas or drama queens, just smart and cool people who love games and get things done. Turns out this combination is pretty hard to find. Probably the most significant factor in our success is the people I work with: my co-founder Mathilde and our awesome team of great people. Having fun, highly skilled people without egos working together means you can move quickly without getting bogged down in drama or politics. Speed is the probably the most important factor in the success in almost any business, and this is doubly true in social games. I’m just happy to go to work every day and make great games happen.

Now that we have the revenue, we’re hiring. With social games you can’t just let the ball drop – we’ve been down a bit lately because we’ve focused on hiring, but we’re getting good people on board and getting up to speed quickly and getting to where the next game can pick up more steam.

The other thing is that it’s much easier to recruit technical people because there’s often a right answer. Designers and producers are harder, mostly because there isn’t a “right answer” you can test them for, rather you have to look for great organization and deep understanding of design. But at the end of the day, it has to be a company that I would want to work at too. I didn’t put in three years of startup level effort to hate coming into work.

I also know how much we’re making, so even if you look at some of the acquisition prices in terms of dollars per MAU or DAU, Playdom got a pretty nice number, but to some extent we know that’s based on real revenue. At minimum, a social game should have half a cent per DAU, up to 10 cents per DAU or higher.

If we can show that we aren’t a one-hit wonder, I’d say our story is, we tried one thing and it totally didn’t work, learned from our mistakes, tried another and it worked better and kept on trying and learning until we got something pretty great. We’ve learned a lot that makes me think that the probability of doing it again is much better now than it ever was before. Eric Ries is on our board of advisors and influenced us a lot with his ideas of a lean startup, although we think about a minimum viable game instead of a minimum viable product.

It’s about design know-how and knowing what features you can cut while still leaving something appealing to players. Keeping only the core features allows you to build and test your game with the real market as fast as possible, the beauty of social gaming (and the online space in general) is that you can always change it later via smart design or smart A/B testing.

I think at the end of the day, if you know what you’re doing, you can make something successful on Facebook. It’s still an order of magnitude better than other platforms for getting your game in front of a large audience and its great communication channels.

Maybe we’re a counter-example, but we built and maintained Baking Life with just six people up until the last few months. At least a few small companies have figured out how to do it. Either they got really lucky, or they figured something out.