Data Merge: Stores Can Learn A Lot From Their Web SitesData Merge: Stores Can Learn A Lot From Their Web Sites

When it comes to business intelligence, online retailers have an advantage over brick-and-mortar stores. Online retailers often know who their customers are, since they have to ship them their purchases, while traditional retailers typically don't. (Supermarkets with customer-discount cards and retailers with their own credit cards are exceptions.)

Because online retailers can track customers through Web sites--and even what sites they came from--they often better understand the cause and effect behind a sale than stores do. Brick-and-mortar store operators know little more than what was sold and when.

But retailers with storefront and online-sales channels can use information collected from the Web to further their customer-relationship management efforts. For example, retailers that detect changes in customer buying patterns through a Web site, such as affinity buying (goods purchased together) or increased demand for particular products, can apply that knowledge by changing store displays or stocking up on sought-after items.

"You can think of it as an early-warning indicator of what's hot and what's not," says Usama Fayyad, CEO at DigiMine Inc., a maker of data-mining software that retailers such as J. Crew Group and Nordstrom Inc. use to analyze online sales.

Using DigiMine, for example, Nordstrom analysts in 2001 noticed that a lot of online shoppers were searching for navel rings like the one worn by a model pictured on Nordstrom's Web site. Nordstrom didn't carry the rings, but it quickly added them to its inventory both online and in its stores after seeing the sudden demand.