Why We Launch a Game When We're Embarrassed By It

We have a relatively unconventional approach to developing and releasing games.

We launched Give Me Fuel in April. It wasn't a "soft launch", or a "beta". It was a launch. We put the game out there for anyone to play. We didn't hype it up, we didn't push for a featuring on the iTunes store. We knew that the game was confusing to new users. It had almost no tutorial. It had little feedback about what you were doing. The animations were repetitive. There was barely any sound. It was often slow and stuttery. It lacked content.

By the standards of most game companies, it was an alpha, at best, and a pre-alpha in some peoples' eyes. And for us, it was the right time to release the game.


On mobile, conventional wisdom is that you have exactly one chance at success, and that's getting a massive feature spot from Apple on iTunes. You can only do that at launch, and if you blow it, and don't get a feature, it costs you hundreds of thousands of downloads at a value of millions of dollars in free exposure. And all of that is true. If you don't get that, you pass on a massive marketing exposure that's a critical part of most successful games' success. But we consciously passed on that, and instead put out a largely unfinished, extremely rough game we knew had significant problems.


Because we had a question we needed to answer.

Our process has always been about trying to answer some sort of question. With Give Me Fuel, we have a "big idea" behind the game that you can't see yet. But for us to get to that "big idea", we needed a core game loop that does certain things. It has to be something you can play in short bursts. It has to be something you can play for months without getting bored. It has to be something that can withstand a significant amount of evolution without collapsing. That is, it has to have "hooks" for things we're going to be adding later, and it has to be flexible enough to be able to accommodate that.

And there are a number of ways you can answer that question. The most common is that you build an amazing game, polish the hell out of it, make it perfect, and see how it turns out. Obviously, that's not our approach.

The problem is that making a game "perfect" takes a lot of time. Building the game - all the mechanics, the content - that stuff takes a good long while, and making it all understandable and polished? That stuff can take just as much time. And the problem is, if you're doing that in secret, you've only got yourselves to test on (and potentially limited usability testing and the like). If you've got the resources to bolster your own capacity, great. But it's often an expensive, time-consuming, highly iterative process, and a good portion of the time, the difference between how players in user tests play and how real players play is different enough that you can still release with fatal mistakes in your game.

Mobile development is interesting, because it's different than console or PC development. Instead of taking a month to patch something & test & release, we can release patches to certain things multiple times a day. We can then see how players are playing - both by interacting with them, but also because the game tells us what's happening. Our game's a multiplayer game, and a lot of the game takes place on our servers. So we know how many turns a person has taken. We know when they bail out of the tutorial without finishing it. We know when someone bounces back and forth between screens, confused as to what to do.

The availability of data & the ability to iterate quickly totally changes how you can and should develop a game. Instead of polishing something to perfection, you can release it early, get feedback from players as to what they like or don't, look at where people are having problems, and fix that stuff. Instead of "trusting your (the developer's) vision and judgment", you can instead try something, see how it works, and iterate with useful information backing up your choices. And by doing that, you can make progress radically faster and better than you could if you relied on the judgment and intuition of even extremely experienced individuals.

We believe so much in this process that not only do we release our games early, often in embarrassing states, in order to get that iteration & information, it's been part of our logo from the start:

There are two things at work, here. The first is that our games are a combination of things that are familiar, and something that's explicitly new. The game is the star. Where it comes from is the collision of two paths. The path of the familiar is the bold yellow, circular stroke. It's well known, and we can draw on it with confidence and certainty. The W - our identity - is the new. The uncertain. The thing that we're going to get wrong multiple times, changing course, until we get it right and it merges with the familiar to form something that you've never seen before.

We release this way not because we believe the first iteration of our games are a reflection of our ultimate capability. We release this way because we know it's the fastest way to get to the point where those two concepts finally merge into something great. It's a painful process to go through. You release something to the public that you know isn't the best you can do. You know some people will hate it. You know some people will judge you for it. But you also know that you have to do it that way because its the best way to get the best results. It comes often at great cost - both to your ego, and to things like Apple Featuring, media reviews, etc. - all of which are structured to deal with games that "launch big".

But we are trying to do something different. We're trying to build an audience slowly, and make the game better quickly. We start knowing we'll lose a lot of people, but that the folks who can see through the painful bits to find the "core" of the game - if they love it, that answers the question we need to get answered. And then with their help (and the help of everyone that tried the game & quit), we make it better. And we start to ask a different question. Can we make this game great?

We're going to be answering that question, along with a different one, over the next few months. I hope you'll join us for the journey. It should be interesting, I think - because that second question we're trying to answer is whether we can make an entirely new kind of game from this foundation. Something that, to the best of our knowledge, has never existed before.