Multiple Catalog Mail Campaigns: Who Gets Mailed Next, and Which Catalog Should It Be?

In recent years, there is an increasing trend in the numbers of catalogs database marketing managers are mailing. The increased-mailings are complicating the assessment of the basic questions: Which mail campaigns were successful? Which mail campaigns inversely affected other mail campaigns? The task of measuring the success of multiple mail campaigns typically consists of inputting the results into Excel© to create a series of spreadsheet-based analyses among a multitude of customer mailing segments per campaign. The next task is to perform the analyses in greater detail – at the customer level. The latter requires reformatting the results in contingency tables of the multiple mailings, for which Excel and statistical methods are ineffective as the analyses are cumbrous and often lead to inconsistent findings. Lest the most important task is not mentioned, traditional statistical methods and spreadsheet analyses cannot determine: Which customers get mailed the next catalog, and which catalog should it be?

The purpose of this article is to illustrate a new nonstatistical approach to solve the otherwise analyses of multiple catalog mail campaigns. The approach uses the machine learning method of GenIQ Model©, which effectively addresses all three tasks stated above. Moreover, many other questions can be answered, such as: Do customers who only respond to promotional mailings “fall” in recency until the next special promotion? Is the back-end revenue of the promotion offset by the reactivations within a reasonable time period? Are follow-up mailings effective in reactivating customers? Are they kept active and profitable?