PUB - Publikationen an der Universität Bielefeld

This paper presents a new model based on Poisson regression for analysing consumer preferences on the Internet. Our approach bears resemblance to conjoint analysis and delivers insights into the preference structures underlying the usage or non-usage of alternative product and service offerings. In contrast to conjoint analysis, the model uses Internet transaction data as it automatically results when a request is launched. In doing so, the model provides useful information with respect to potential improvements of the current range of products or services. The basic idea of our approach is demonstrated by means of usage data provided by a commercial web portal and online booking data provided by a well-known tour operator.