Less Is More

Those gadget-obsessed, control-freak programmers love to fill products with gizmos and features, but that tendency is contrary to a fundamental insight about good design. Less is more.

When an interaction designer has done a particularly good job, the user will be quite unaware of the designer's presence. Like service in a world-class restaurant, it should be inconspicuous. When the interaction designer has accomplished something really good, users won't even notice it. In an industry that promotes "coolness" as a design objective, it really gets tiresome to find my way so often obscured by interaction artifacts that have obviously taken some poor programmer lots of time and work. Too bad his efforts didn't go into something effective. Many visual designers think that good design is cool, and occasionally it is, but no matter how cool your interface is, less of it would be better.^[4] Again, the point is that the less the user sees, the better a job the designer has done. Imagine watching a movie and seeing klieg lights in the corners of the frame or hearing the director yell "Cut!" at the end of a scene. Imagine how intrusive that would be and how it would break the viewer's spell.

^[4] In my book, About Face, I introduce over 50 powerful design axioms. This is one of them.

Super programmer and software designer Kai Krause is famous for his unique interfaces. Kai has created some of the most powerful and interesting graphical-manipulation software. His products always have breathtakingly beautiful interfaces. They also tend to be inscrutable, kind of like a game. In addition to his programming ability, Kai is a visual designer, and his products reflect the visual designer's willingness to make things obscure like modern art for the sake of effect. This works because his user base is other visual designers and hobbyists. It doesn't go over very well outside that world.

In programming, there is always an infinite variety of ways to solve any given problem. Experienced programmers, as they explore their options searching for the optimum solution, occasionally stumble on a technique that allows them to throw out hundreds or even thousands of lines of code. This only happens when the programmer has made a valuable conceptual leap forward. When she can toss out lots of code, her program is getting better. Less code means less complexity, fewer bugs, fewer opportunities for invalid interactions, and easier maintainability.

Interaction designers share this sensation. As they explore their options, they discover places where they can dispense with entire screens or discard large and complex dialog boxes. The designer knows that each element of the user interface is a burden on the user. Each button and icon is one more thing that the user must know about, and must work around, to get to what she really wants. Doing more with less is always better.

If the designer is doing well, she is removing interface from a product. She is not designing screen after screen of buttons and gizmos. A product manager from a large software company visited us one day, inquiring about having us redesign a product for them. He told us that he expected the interface to have about a dozen dialog boxes. We explained to him our process and then quoted a price for their design. It was about $60,000, if I remember correctly. The manager then exclaimed, "But that's outrageous! It's $5,000 per screen!" I didn't have the heart to tell him that we would probably reduce the dialog box count down to one or two and that the price when calculated on a per-screen basis would be a lot higher. He just didn't get it. Paying for design on a per-screen basis is like paying a waiter by the number of trips he makes to each table. A better waiter makes fewer trips, and a better designer always creates lots less interface.

Sometimes being an interaction designer can be so frustrating! If, as a designer, you do something really, fundamentally, blockbuster correct, everybody looks at it and says, "Of course! What other way would there be?" This is true even if the client has been staring, empty-handed and idea-free, at the problem for months or even years without a clue about solving it. It's also true even if our solution generates millions of dollars for the company. Most really breakthrough conceptual advances are opaque in foresight and transparent in hindsight. It is incredibly hard to see breakthroughs in design. You can be trained and prepared, spend hours studying the problem, and still not see the answer. Then someone else comes along and points out a key insight, and the vision clicks into place with the natural obviousness of the wheel. If you shout the solution from the rooftops, others will say, "Of course the wheel is round! What other shape could it possibly be?" This makes it frustratingly hard to show off good design work.

Computer scientist Alan Karp says, "Almost every patent application I have submitted has been rejected as 'obvious.'"

When I say less interface, I don't mean less functionality although that can sometimes be the case. I mean that the user doesn't have to interact with the program any more than is absolutely necessary to get any particular task accomplished.

In this chapter and in Chapters 9 and 10, I presented a brief look at our most widely used design tools. They have proven to be very effective in designing products and services ranging from industrial control to enterprise planning to consumer products. In the next chapter, I will examine some other available tools that claim to help create better-designed products.