Navigation and the Power of Purity

In the early days of the web we were so excited. Hyper-links were sexy and we would spend the day blanketing our pages with them. Those wonderful blue symbols underlined (also in blue). Spectacular. The job was to find offline content and convert it into HTML and connect it to one of these little blue bugs. Gray was a popular background choice, but mercifully web masters gave in and, for the most part, switched to white. Below is Yahoo’s home page on May 8, 1999.

As time went by content creators became more attuned to writing web copy, not traditional analog copy. This began to differentiate the internet content consuming experience, as writers put importance on the uniqueness of the medium at the forefront.

Copy flourished and with it graphics. Web safe colors limited designers so we saw a primary color palette that brought back grade school art class memories. Eventually browsers began to see more colors and monitors advanced, freeing creative types to polish and refine. Most images were hard edged and PowerPoint like inserts. Not very pleasing.

While designers designed, writers wrote. And of course functionality exploded. Self service, research and commerce grew at a mind boggling pace, completely overwhelming the navigation. Amazon was adding thousands of new products a week. Then lines of businesses. How in the world would one be able to find all this stuff on their site? The answer was of course “The Tab.”

These images are from Amazon’s site in 1999 and 2000. Tab mania infected the world wide web almost overnight. It was the answer we were all looking for and believe me, we jumped on it. This is about the time web site operators began to understand that labeling was one of the most important things they needed to get right. Sites were peppered with corporate tribal language and consumers were confused. Cute names were coined because you couldn’t just call something by its name could you. How pedestrian.

The birth of web Personas and user-centered design helped immensely with navigation and labeling. There were two problems that Personas were great at solving. One was what to call something so a visitor to your site could recognize it, and secondly the density of content to be displayed on a page. It was a challenge to marketing and set up an epic battle between selling and goal accomplishment. Segments tell you how to sell, but Personas tell you how to satisfy. Of course there are many other benefits to using Personas, but these are the two watershed moments in making design better for the user.

Google came on the scene and made search work. I mean really work. The “White Box” began to show up in the upper right hand corner of sites that could organize their unstructured data and catalog content. That was, and remains today very powerful. So much so that on today’s Amazon’s home page not a tab can be found. Search runs across the top of the page and that’s how most people find what they’re looking.

Navigation became nested and included fly-outs, all of which are fine ways to solve for how you make multiple choices available to users without page clutter. But sites are overrun with content and suffer from the weight of an organization’s natural ability to “pile on.” I see this everyday and on so many sites. My advice to you if you control the customer experience on your site is to guard against content creep at all costs. Once something gets on a page it’s very difficult to get it off because high trafficked sites will find that all their links garner clicks. We just don’t know the motivation behind each click.

My favorite property is the property of subtraction. Addition is too easy and is fueled by likes or worse, an org chart. But subtraction. Now that’s an art form. A skill that requires constant honing fed by web analytics, data and real insights. So as you say, “Scalpel please,” many will scream that we can’t get rid of that content or drop that banner. We get x number of clicks from those links. When you look at the number of clicks and put that into the reality of percent of clicks on your entire site, the numbers have quite a few zeros after the decimal point. Not enough to move the stock price, and certainly no one would increase their business goals as a way to justify their existence. Now I will be the first to want to drive business and make money. So bring the data.

I have a theory that I believe would hold true on most if not all often visited web properties. If you take a high trafficked page and put 10 links on it you will get traffic to all 10 links in a given month. If ¬†you expand that to 20 links you’ll get the same outcome. All 20 links will be clicked. But this is not a justification for putting 30 or even more links on the page. I posit that if you reduce the number of links (excluding primary navigation elements) you will drive up traffic to the remaining links. Ten links in total will get as many clicks as the 20 in total, so you are distributing the traffic more evenly across the content. Put your business driving content behind those ten links and call it a day (exaggerating for effect).

Which brings me to the point of this post. Aim for purity in design and navigation. If you can do only one thing with something you will most likely get more people to do it and spontaneously return to do it again and again. Take Pinterest as an example. It made it to 11 million registered users and one billion page views a day in about a year. The fastest growth pace to date. Why? In a word purity. It’s all about one thing, images. when you click the main action button you can do one of three things. That’s right 3 things. Not 5, not 7, not 15. Add a pin, upload a pin or create a board.

That’s it! Simple, crystalline, pure. Anyone can pin and it’s a near perfect metaphor with the cork board and scrapbooking. Genius. Behind that simplicity lurks a lot of other things that will drive toward monetization and potentially commerce. The question is can they keep that behind a clean interface. If they can, then success is inevitable.

Keep it pure. As pure as the driven snow.