Features are your Enemy

“All other things being equal, the simplest solution is the best.” – Occam’s Razor

Feature Creep, the result of giving in to the attitude that “this feature would be nice” in your product, inevitably results in increased design and development costs, bloating of your product, and a reduction in simplicity.  Here, I propose a strategy for the discovery and brutal elimination of unnecessary features.

Ideas reproduce.

Those driving ideas and products tend to be idea generators – people who bubble with enthusiasm and want to leverage every possibility in their upcoming product or version.  As creatives involved in the process, designers also think up options and features that may provide usefulness.

During the brainstorming phase of a project, the team ideally produces a wealth of directions that the product can take, and explores all avenues.  Ideas flow freely, bounce off each other, and breed like rabbits.  They should be remembered, captured, tagged, identified, explored, and discussed.

Then, they should be decimated until only the best remain.

Natural selection.

After the optimistic, criticism free generation phase of the project, it’s time to take the gloves off.  Set expectations with the team and ground rules for discussion.

1. Ensure you have defined your problem.

   Without knowing what you’re solving, the solution will be meaningless.  Write down the problem as concisely as possible.  This should have been done before the brainstorming stage, but sometimes the problem can shift as brainstorming continues.  Do not leave this step until your problem is identified.

2. Collect your solutions.

   Your brainstorm session has ideally generated a wealth of interesting and insightful solutions.  However you have decided to manage your solutions – sticky notes, bullet points, twitter posts -collect all of these solutions in a single place.

3. Evaluate each solution relative to the problem.

   How well does the solution address the problem?  Which is simplest?  Examine each solution with the highest suspicion.  It should stand up to attacks from many angles, supported by how well it addresses your original problem.

4. Sidebar all solutions that don’t directly relate to the probem.

   Some of your solutions will be crazy, head-in-the-clouds, never buildable, frilly junk that were fun at the time but can probably be tossed now.  Others are great ideas that could be included in a “version 2″ of the interface you’re building, and these can be kept on the sidelines until later.

5. Design and enjoy your clean, simple, realistic interface.

   With this now-limited solution to your problem, you will have a manageable design and development project, in which you have the resources to focus on quality interaction, simple visuals, and standards based development.