Byline: Denise Power

NEW YORK — Using analytical software to guide direct mailings, Fingerhut has lifted its profit rate 10 percent among customers who shop infrequently. Now the catalog and Internet firm is seeking to expand that process to other shopper segments, such as new catalog customers and potential customers.
The direct marketing company, a subsidiary of Federated Department Stores also may apply its internally developed solution to optimize customer contact at Federated’s stores. Randy Erdahl, director of business intelligence at Fingerhut, said the system is not being used in other Federated businesses yet, but it’s the next logical step.
“Even if you’re not in the catalog business, this can be applied for any kind of customer contacts,” he said.
The system helps target Fingerhut’s 120 catalogs to the households in the firm’s database that are most likely to purchase products. Fingerhut has a total customer base of six million.
“The system takes into account many variables, including the mix of retail price points, mix of merchandise categories and the mix of new merchandise — 30 different metrics in all — to evaluate saturation and similarity among catalogs” mailed to a particular customer, Erdahl explained.
By analyzing historical purchasing data collected over time, the direct marketer is currently targeting mailings most effectively to its repeat customers. Going forward, Fingerhut plans to use the system to target first-time customers, those shoppers on whom only a small amount of shopper-specific purchasing data has been collected. In addition, according to Erdahl, the system may also be used for prospecting new customers.
The system is called “mail stream optimization,” and Fingerhut developed it in cooperation with consultants from IBM, Armonk, N.Y. Powerful data analysis software applications from SAS Institute in Cary, N.C., and Torrent Systems in Cambridge, Mass., are used to analyze customer purchasing behavior to predict which consumers are most likely to respond to which of Fingerhut’s catalogs.
“It’s essentially a strategy for streamlining high-volume, direct mail to targeted customers and to reduce advertising costs,” Erdahl said.
Many times, one Fingerhut catalog has merchandise overlapping another catalog’s offerings, and both get sent to the same address. The company is using the system to identify and eliminate redundant mailings.
By determining which customers are most likely to respond to a particular catalog’s offering, and eliminating those that are redundant, Fingerhut is able to scale back on the size of each mailing, reducing costs and bolstering response rates and profits.
Erdahl cited one example that resulted in a 10 percent increase in profits generated from a group of shoppers characterized as “inactive,” meaning those whose most recent purchase was made a year or two ago. Fingerhut conducted mailings to a control group of these shoppers, using conventional direct-mail methods, and to a test group, using the new mail optimization system.
In the test group that targeted customers precisely, Fingerhut reduced direct mail costs by 8 percent and lost only 0.7 percent in revenue, Erdahl said.