Don't Be Bland

Putting more stock in focus groups of users than designers is not a recipe for visually enthralling interfaces. Jason Arber strikes a balance.

A recent article in The New York Times caught my eye. It was a feature about Douglas Bowman, currently creative director at Twitter and previously a senior visual designer at Google. In a feature entitled 'Data, Not Design, is King in the Age of Google', The New York Times' Miguel Helft writes about Bowman being required to run every Google design decision - even "whether a line on a web page should be three, four or five pixels wide" - as test versions past user groups, whose responses would be a significant factor in the final outcome.

You could certainly argue that this kind of engineering-led decision-making has done Google no harm. In a little over 10 years it's become one of the most powerful and influential companies in the world, with a 2008 revenue of nearly $21.8 billion (according to Google Investor Relations). Google pretty much owns the internet, kicking its competitors in the seat of the pants when it comes to online search and advertising. While Apple might be a thorn in Microsoft's side, Google is a chainsaw in its jockstrap.

I've always been an advocate of the 'form follows function' school of design, especially in user interface and application development, where acting on whims and disregarding functional common sense can lead to fanciful but highly impractical interfaces. Of course, it takes unconventional creative minds to push our notions of what interface design can be, but it requires another set of smart people to take those alien interface concepts and channel their essence into something useable.

Apple is very good at this. It took Cover Flow - a technology invented by Andrew Coulter Enright and implemented by Jonathan del Strother for flipping through mp3 album artwork - and applied it to an operating system. This became a rather useful way of quickly finding the right document based on what it looks like rather than its title. In this way, Apple seems the opposite of Google; there's a sense that Apple ploughs its own furrow, confident that it knows what you want even before you realise it. Some find this attitude condescending, but at least there's no pandering to focus groups. Obviously, there's an important place for user testing, but when it becomes the design rationale for everything, the result can be bland and uninspiring.

Google Video, in my opinion, has such an insipid interface it makes supermarket own-brand rice pudding look like a rock star about to drive a Rolls Royce into a swimming pool. Sure, it does the job but, for me, using it is a vaguely unsatisfying process. Look at how sexy Vimeo is by comparison. Hell, even the business-like YouTube does a better job. Here's the rub: Vimeo is still very functional, but has applied a silky varnish of great design that makes watching and interacting with videos a pleasure. It's possible to mix great design and functionality, and end up with something beautiful and unique (the great iPhone applications from Tapbots spring to mind).

Imagine if wonderful movies such as David Lynch's Mulholland Drive had been changed because a test screening audience wanted a more linear plot. I'm not suggesting that Google should start churning out Lynchian interfaces, simply that a user group will invariably funnel them into a very safe middle ground - and if Google really wants to shine, I suggest it puts trust in its designers over focus groups. Maybe the brand whose motto is 'Don't be evil' should look at adopting an updated version: 'Don't be bland'?