THINK PIECES

Who Knows What Pages Work Best? - Part 1 

By Charles Hofacker, Professor of Marketing, College of Business, and
Jamie Murphy, a student at the College of Communications, Florida
State University. Hofacker and Murphy frequently publish research and commentary on
internet marketing in scholarly journals and the New York Times Online.
They operate New South Network Services, an ISP and Web Creative
Firm in Tallahassee, FL

At one time or another, every Internet marketer must have wondered about the best way to set up a web page. Normally, a web site has certain communications, marketing or selling goals. How should a certain page, like the primary home page, be set up so as to facilitate achieving these goals?

Artists generally go with their gut instincts in terms of what looks good or what looks cool. Using their "right-brain" and their creative imaginations, web page designers generally do good work, although clearly some are better than others. Marketing types on the other hand, contemplate the bottom line. What home page format or look would keep visitors on the site longer? How can we get a higher percentage of visitors to click on a certain key link? These kinds of "left-brain" questions often go unanswered and surely represent some of the true mysteries of web life. As practiced web observers, we have our answer to these questions: We don't have a darn clue!

We would like to note here, though, that no one else does either. Now this reminds us of the young novitiate in the middle ages. He was waiting tables on the more established brethren of the order when he came upon a rather vehement argument. Brother Bartholemew was insisting that a horse had 36 teeth as that corresponded to thrice the number of disciples. Brother Eggbert was just as adamant that the answer was 39 as St. Mary must be factored in to the argument. As the conversation paused while both took a sip of soup, the novice asked, "why don't you go into the stable and count them?"

While that young man was quickly thrown out of the order either due to his empirical bent, or perhaps his temerity, we think he had the right idea. To pick a simple example, suppose you are trying to decide which colors should be used for a site. Why not run an experiment and find out the color which seems to do better with respect to the goals of the site? Direct marketing specialists do these sorts of studies all the time. It frequently turns out that minor changes in format or wording can change the response rate to a mailout by a very large amount. Despite decades of experience, direct marketers are still learning how to improve their format and their copy. Our knowledge of why people click is still in its infancy. Perhaps we should borrow a page from direct marketing and experiment.

There are several ways to run web experiments, but the key thing is to insure that as visitors arrive on the site, only random chance dictates which experimental version of the web page they see. At New South Network Services, we have been running experiments like these for a variety of clients and we are constantly surprised by the subtleties of why and how people click. Next time: How to set up and run a web experiment.