Testing conventional wisdom

For the most part, organizations involved in KM assume that the information they classify as knowledge is indeed factual and accurate, thereby providing a realistic glimpse into what's going on. But what if that so-called "knowledge" isn't what it's made out to be? In other words, what if an organization's purported knowledge is simply conventional wisdom that has never been questioned?

At A&E Television Network, verifying conventional wisdom--or conversely, poking holes in it--is something that So Young Park, the director of e-commerce, has started to do recently. As head of the broadcast company's online e-commerce efforts, Park is always keeping an eye on the latest promotions and dynamic online content, and how that information affects conversion rates, or the number of site visitors who make a purchase while online.

With software from Art Technology Group Inc., Park and her small e-commerce group constantly tweak the online content--in terms of product offerings, banner advertising, graphics and copy--in the pursuit of higher sales. "We do different tests for all sorts of creatives to increase our knowledge about the areas of the site that perform well and why they do so," she says. For example, the group often creates campaigns featuring two versions of copy, and then can see in real-time which version gets the better response. The concept, called A/B testing, is a tried-and-true technique for direct marketers who are seeking to fine tune their promotions. The advantage for online marketers is that the results of the tests become available when there's still time to do something about them.

In one such example at A&E, Park fiddled with the copy of a promotion during this year's Easter season. Essentially, Park flip-flopped the copy to see which version of a promotion would elicit the better response. One version said something to the effect of: "Spend US$50 and you'll get a free gift." While the other said: "Get a free gift when you spend $50." The creative graphics in both versions were identical, and Park assumed--based on her own conventional wisdom--that the ad touting the free gift first would garner the most takers. Yet a test of promotions proved otherwise: The ad mentioning the money first attracted more buyers, to the tune of 50 percent more. "It may sound like a little thing, making such a small change, but the results were really surprising," Park says. Surprising not only because Park was incorrect in her initial assumption but also because such a seemingly small change resulted in a fairly significant performance.

With the ability to quickly test promotions, A&E is compiling a paper-based knowledge management system of sorts. In a binder dubbed "Marketing Death Match," Park keeps a running score of all the tests, complete with copy, text and results. The binder is available to the eight- person e-commerce group as a resource for learning what works in promotions and what doesn't.

So with a combination of software and low-tech paper, A&E e-commerce group is able to validate what it suspects, discount what it assumes, and share what it knows.

Join the newsletter!

Error: Please check your email address.
Show Comments
[]