10 Steps to Designing the Next Killer App

Posted by Miles Evans

Coming off the end of another sleep deprived week long work binge I realized that for 8 months I have been doing nothing but spec’ing out socially delicious AJAX’ed masterpieces, complete with mockups, flowcharts, marketing plans, and brand boosting viral schemes.

The real deal is that designing fluid GUI’s that actually work in a browser has finally arrived somewhat, and everyone is clambering to build the next can't do without web service. Everyone with any vision at all is cashing in - me included.

The secret is designing applications that people actually want to use. Easier said than done right? Actually it isn’t that tough. Here are some points to rip on:

1. Build applications that find compelling and useful ways to filter data. If you take a look at virtually any web 2.0 startup you will see that most of them are built around the idea of filtering or sorting content in unique and novel ways. Think digg not Slashdot.
2. If it’s already been done you’re too late. Quit copying your competition and innovate your way to success. Sit down and think of what type of application you could make if you could do anything you wanted without restrictions. Chances are it can be built.
3. Don’t even think about plastering your new web service with Adsene. In fact set a bandwidth barrier for which you can safely operate without losing money before you need to monetize. Remember, free is always cool, and there is no better trust builder than a useful free service that other chumps charge for.
4. Always try to work with a massive data set if it fits your model. The more data you are collecting the more interesting ways you can filter that data. By collecting reams of data you will also be opening yourself up to capturing your long tail and niching your content.
5. Build your application around a community rather than build a community around your application. The majority of online users are socially active. People go online to hang with other people and share their surfing experience. If you are not providing them clear channels for interaction, collaboration, and discussion, you can bet your competition is.
6. Rapid design frameworks. I’m a bit divided on the use of these as the adoption of an entire framework just to whip out some PHP seems a bit redundant. Having said that most of the AJAX frameworks actually do bring something to the table as javascript is completely retarded to work with even if you know what you're doing.
7. Break old design habits. Ditch the 780 pixel wide 800x600 safe design for fluid layouts that fit 1024x768. Get with the program it’s 2006.
8. Dreamweaver has absolutely no place in your set of design tools. Say this to yourself 10 times as you remove it from your PC and install UltraEdit.
9. Outsource everything! If you are a great programmer but cannot even draw stick figures properly, spec out what you want and hire a freelancer. The time you waste buggering around in Photoshop or Gimp does not justify a $50 logo.
10. Less is more. The reason why the simple layouts are so appealing should be obvious after 10 years of chunder inducing animated gifs and whorish flash intros. Start with a clean layout and focus on extreme functionality over gradients and rounded corners. Remember - In a world of diminished attention spans, confusing your user for 2-3 seconds always means game over.

Happy Spec'ing. Don't forget to digg!

Posted Sep 11, 2006 at 10:39 PM | Permalink | Trackback URL | Del.icio.us | DIGG!