[In this in-depth Gamasutra postmortem, the team at console and PC developer Demiurge Studios (Mass Effect for PC) explain what went right and wrong in creating popular iPhone word game WordFu, from swift iteration to localization complexity and beyond.]

WordFu is a puzzle word game developed by ngmoco and Demiurge Studios for the iPhone and iPod touch. Its initial release was in late February of 2009, and had three updates spanning the following months.

The project started as a simple dice-based word game, and quickly grew to be a much bigger game than we expected. We learned a lot along the way, and had a great time making and playing WordFu.

As with all games on this platform, climbing to the top is a mysterious and exciting prospect -- one we saw briefly before falling back into the ever-growing jumbled conglomeration that is the Apple App Store. But this isn't about reaching the top 10; it's about our journey there.

What Went Right

1. ...And I Want It Yesterday

As the first iPhone game Demiurge had ever developed, there were many things that were completely new to us. We didn't expect to be able to implement and iterate new features so quickly -- a pleasant surprise. With this new, fast development process, adding a new feature took two days to implement and three days to iterate and turn it around to the publisher.

This sort of turnaround was completely foreign to us, since we're used to developing much larger-scale games. Normally a new feature (of proportionally the same magnitude of impact that the features we added to WordFu took) can take up to a week to implement and weeks to iterate upon. Additionally, we use Scrum development methods in most of our projects, so new features aren't even tasked until the next sprint.

Features like the power-up dice were added quickly and easily thanks to the rapid turnaround process. Within a week, we were able to have dice that froze time, temporarily doubled word scores, and allowed a die to be flicked. This didn't include the art changes needed for the feature, but normally a feature that changes the scope of the game takes longer than this to successfully integrate.

2. Easy Updates

Since features are relatively easy to implement, we could take feedback and quickly put it into the game. The Apple App Store made this process easy and fast, so when customers and critics requested features, we could turn around and easily update WordFu.

Some people didn't like the stress (or what we like to call "challenge") of a short starting time, so our first update allowed people to choose a longer starting time. We also allowed people to play their own music while playing WordFu. Multiplayer was limited to playing against people in your local Wi-Fi network, so we added the ability to challenge your friends via email. This feature was expanded in the next update.

In WordFu version 1.2, challenging your friends to beat your score was augmented to include Facebook and Twitter posting. These are two venues that the iPhone and iPod touch can uniquely interact with, whereas other gaming platforms have more traditional multiplayer networks. It was exciting to see what new ways people could share and experience WordFu.

The third and final update included localization for German, French, and Spanish versions of WordFu. These posed some interesting challenges, since we had to translate many of the art assets in game, which I discuss later.

Title screens in German and Spanish

Being able to listen to your fans and critiques and actually make changes is a satisfying experience, especially when they realize that you're actively improving the game. It also helps when the process of implementing, sending off, and updating content is relatively fast and easy.