20bits

Creating Appaholic has given me the opportunity to see what apps succeed and why. Here are some rules of thumb to consider when writing your Facebook app.

1. The Complexity Ceiling

   Facebook is simple. The features on Facebook are simple, even compared to similar features on other sites, e.g., Flickr vs. Photos. My hypothesis is that no application which is more complex than the most complex feature on Facebook will succeed. You can compare three similar apps: Bookshelf, Bookshare, and Visual Bookshelf. Of these Visual Bookshelf started the latest, is the most simple, and also has two orders of magnitude more users than the other applications. The more simple you make your app the better it will do.

2. Try to Be Social

   Just because you've built it doesn't mean the users will come. Facebook is still first and foremost a social platform and that's why people are using it. If your application has no social component it's just going to flop. Not only will it be difficult to spread virally but there's little compelling reason for people to install it. For example, let's say you have a blog and want to promote its content with a Facebook app. You should not just create a Facebook app which displays your blog posts in people's profiles; rather, try to add a social component. What are my favorite articles? What have I commented on? It's then easy to see what my friends like and what they have commented on.

3. If You Can't Be Social, Be Viral

   Writing good social applications is hard. If it weren't Facebook wouldn't be worth what it is. You can, however, write an application which does nothing more than spread itself. If your idea is funny enough, like Vampires or Zombies, people will use it. It's still too early to tell whether these apps have staying power, but you'll at least get your fifteen minutes of Facebook fame.

4. Don't Be Too Weird

   People are used to how Facebook works. If your application is totally foreign they just won't understand it, even if it's the most usable, straight-forward piece of software. The top applications tend to fall into one of two broad categories (again, iLike is the exception). In the first category the applications are viral for virality's sake. Why this works is self-evident, since the applications exist for no other reason than to spread themselves. In the second the applications augment or complement an existing Facebook feature. Top Friends, Graffiti, and X Me all fall into this category, for example, which behave like Facebook's friends list, wall, and poke features, respectively.

5. Fads Exist

   Even though the Facebook platform is only two months old fads have already come and gone. If you're just looking to get a respectable number of users in a short period of time then it's worth paying attention to these fads. The "quotes application" genre is an example. There are applications which add quotes to your profile from everything from Star Wars and Family Guy to Friends and Scrubs. Don't blame me if the fad becomes unpopular and your app fizzles, though.

6. Quality Matters

   This might be obvious, but quality matters. Apps like iLike and Causes provide a set of very high-quality, social features, so even though they're not the most simple apps they're still compelling. You also have to be prepared to deal with your growth. You can model your growth using Appaholic to predict how many users you'll have in a few days, weeks, or whatever. If it's more than you anticipated make sure you have the hardware to handle it — users will uninstall apps that are slow or broken, even if it's not the apps' fault.

These rules aren't hard-and-fast, and many successful applications break some or all of them. iLike and Causes, for example, are relatively complex compared to most other applications and even Facebook itself, but they're still wildly successful. They get to bend the rules because they're extremely social and have high-quality features. If you can match that level of quality then go for it, but I personally think it's easier to create a simple application and add features as it grows than to create a complex application and hope users don't get lost. Users are more important than features.