Forecasting the customer development of a publishing company with decision trees. (English) Zbl 05280212

Schader, Martin (ed.) et al., Between data science and applied data analysis. Proceedings of the 26th annual conference of the Gesellschaft für Klassifikation e. V., Mannheim, Germany, July 22–24, 2002. Berlin: Springer. Stud. Classification Data Anal. Knowl. Organ., 538-545 (2003).
Summary: Many companies recognize that it is important to concentrate their marketing budget on customers with good prospects in future. This paper proposes a new approach which was applied to a company selling loose-leaf notebooks consisting of a basic volume and several updating volumes. In recent years the customers were valued with respect to their actual transaction volume. The validation has some drawbacks. Therefore, a new type of valuation described in this paper was introduced: now, the customers are valued due to their short term as well as their long term potential. The short term potential is the probability of ordering a basic volume. The number of updating volumes represents the long term potential.
For the entire collection see [Zbl 1023.00021].


62-XX Statistics