steve kwan user experience designer

Five things I want you to know about usability

(December 10th, 2007 - 8:35PM)

Ideas fly fast and furious at Vitrium, and we're always torn about which features should make the cut into our next version. Most other software companies are in the same boat. So how, then, do you determine how new features will affect your application's usability?

Here are some of my usability philosophies. Applying these heuristics has helped me make the right decisions with docmetrics. Before you add a feature to your application, I encourage you to ask yourself if it's in line with these principles.

1. Usability and simplicity are one and the same.

   Keeping your application simple almost guarantees a high degree of usability. I'm working on a theory that usability is basically just another way of looking at simplicity.

   So what is usability's worst enemy? Complexity. The more complex your application is, the worse its usability.

   Many times you'll be pressured by clients or internal forces to add a new feature to your application. Most of the time, this is because a small faction - sometimes even a single person - wants or needs that feature. Sure, adding that feature may placate that one faction, but it increases your application's complexity, which is a disadvantage to everyone else.

   If a feature would add significant complexity to your system, you might want to omit it. It's almost always better to reduce complexity, even if it means fewer features. Most users would prefer to use a simple system that does 80% of what they want, as opposed to a complex system that does 100% of what they want.

   If you absolutely need more complexity, you may want to consider using tiered complexity.

2. Every screen element you add takes attention away from the others.

   This is an important rule that applies to all aspects of interface design. Every element you add makes the others less noticeable. If you're trying to guide your users down a particular path (which you usually are), you'd do well to avoid cluttering your screens with elements.

   Your most important action on any screen should always be the most visually obvious. The easiest way to make an item visually obvious is to remove all the others. So any item that isn't necessary on a page should either be removed or downplayed.

3. There's usually no such thing as an "advanced user."

   All users are inherently simple. That doesn't mean they're stupid; it just means they have better things to do than reverse engineer your interface.

   If you're adding a feature "for the advanced user," you should consider axing it. Very rarely are there legitimate "advanced users" for a product. There's the target market, and then there's everybody else. Everything your target market needs to achieve should be handled simply. If a particular task is very complex for your application and requires an extra interface for the "advanced user," there's a very good chance that those users just aren't part of your target market.

   Rather than bloating the application with features that only a fraction of your users would need, you'd be better leaving them out.

4. Help text is unnecessary.

   That's right. Not undesirable, but unnecessary. Help text is usually used as a band-aid fix to poor interface design. Help text is a crutch that designers rely on when their interface is too complex to speak for itself.

   The truth is that nobody reads the help, so why even bother? As a rule, I try to avoid writing help text as much as possible. I do this not out of laziness, but out of discipline: if my interfaces are sufficiently well designed, they shouldn't require extensive instruction.

5. Usability is not the most important thing to your users.

   Don't get too caught up in usability. Superior service is the most important thing any company can provide to its users. Sometimes talking someone through a problem on the phone is more helpful than a good interface.

   Whenever possible, use your application to solicit feedback from your users. Make them part of the development process. A good interface may lead to user satisfaction, but good support leads to user commitment. If your users feel actively involved in the development of your product, you'll both be happier.