Read the Original Article at http://www.informationweek.com/news/showArticle.jhtml?articleID=6502604
Every retailer Web site has its lookers and its buyers. Victoria's Secret may have more than its share of the former. To help turn more browsers into buyers, the lingerie and beauty products retailer is using online marketing analysis services from Coremetrics Inc., the companies disclosed this week.
Victoria's Secret launched its Web site in December 1998 and is frequently ranked among the top 10 online retailers. But collecting data to understand why shoppers buy what they do hasn't been easy. For a couple of years, the company collected Web-log data for analysis, but finding insight in the overwhelming volume of information was difficult, says Ken Weil, new media VP.
Late last year, the company enlisted Coremetrics to collect and store clickstream data from its Web site. The data is organized and made available via the Web to Victoria's Secret merchandising, marketing, and Web-site managers for analysis. "Collecting the data is only half the battle. Organizing it is the other half," Weil says.
Managers analyze the data to determine how users navigate through the Web site before making a purchase, tracking shoppers' steps backwards from checkout. On average, only 2% to 6% of visitors to retail Web sites make a purchase. (Victoria's Secret's conversion rate is, well, a secret, Weil says.) The company experiments to see what drives sales, such as running different pictures of the same pair of shoes to see which sells more. Competing "impulse-buy" items are also offered to shoppers at checkout to gauge their popularity.
Analytical results can lead to Web-site redesigns to provide shoppers with the shortest route to what they're looking for. "If we can get people to where they want to go in three or four clicks instead of five or six, they're less likely to drop off the site," Weil says. Speed is also critical: Managers get meaningful data analysis in six to 12 hours with Coremetrics, Weil says, compared with days with Web-log data.