I headed the Internet department at computer manufacturer Gateway 2000 for two years and it was quite a busy site selling more than $1 million per day in direct over-the-Web orders in the US alone. Page views were in the millions per day. At one point we had a large cow face on our main page (please don’t ask me to explain the black and white spotted cow fixation that Ted Heads are into). There was also a drop-down list with a button that said "GO!" on the page with links to all the computers currently on offer: such and such a Pentium and such and such laptops. The drop down box links went straight to the page where you could configure and order a computer. One day one of the ponytails came to me and convinced me the drop-down box looked really ugly right in the middle of the cow’s face. She was right so I said go ahead and take it off knowing there were other several ways people could navigate around the site.

A couple of days later the suits called me up and said sales were down by a significant amount over the last couple of days ‘what’s going on?? You been having server problems again?’. Without admitting my stupidity to anyone, I immediately had the drop down box put back in the middle of the cow’s face. Sales went back up immediately. So I said to myself, I said: "self, if putting the drop down list back on the main page made sales go up so much, maybe we should fiddle with it some more and see what happens." So I got my technical guys to make me a tool that allowed me to see a graph on my desktop computer that showed the effect on traffic of changes made to the site in real time. Because we had so much traffic, I could watch the activity graph on my screen go up or down almost immediately in response to changes we made to a page. So I changed "GO!" to "BUY NOW!" Sales instantly plummeted so I changed it back to "GO!" and the cool little activity graph on my computer went right back up. I tried different wording, different colors and different positions on the page and found that "GO!" worked as well as anything else I could think of. Bright colors worked well and the best position for the drop down list was right in the middle of the cow’s face. Upper right corner or upper left corner also worked well.

This was fun so I spent a fair amount of time using my new tool to fine-tune wording and navigational elements on several pages leading up to our shopping cart and check out. I found that adjustments to wording could have a quick and significant effect on sales rates. We managed to increase sales by streamlining our site and eliminating page views that weren’t needed – our goal was to get the buyer the information they needed efficiently and get them to the checkout stand as quickly as possible.

During this process I realised we could use the same techniques to increase page views if we wanted to. Basically it involves fiddling with wording, positioning of the wording and any graphics that are involved in your site navigation and measuring traffic on a particular page before and after making changes to see what effect your changes had. This seems like very basic common sense but few people actually bother to test how things work on their pages in a careful, step-by-step, before and after manner. If you do this carefully you can make changes and quickly see if they cause more hits. Little by little you can find out exactly what wording to use to get people to click on the links you want and go to the pages you would like them to go to.