How Promotions Work: SCAN*PRO-Based Evolutionary Model Building
Harald J. van Heerde/Peter S. H. Leeflang/Dick R. Wittink*
We provide a rationale for evolutionary model building. The basic idea is that to enhance user acceptance it is important that one begins with a relatively simple model. Simplicity is desired so that managers understand models. As a manager uses the model and builds up experience with this decision aid, she will realize its shortcomings. The model will then be expanded and will lead to the increase of complexity.
Evolutionary model building also stimulates the generalization of marketing knowledge. We illustrate this by discussing different extensions of the SCAN*PRO model. The purpose of published model extensions is to increase the knowledge about "how promotions work" and to provide support for more complex decisions. We summarize the generated knowledge about how promotions work, based on this process.
JEL-Classification: M31.

Withholding of Knowledge in Organizations**
Matthias Kräkel*
This paper examines a principal’s trade-off when he decides whether to transfer knowledge to other members of the organization. Although knowledge makes an agent more productive (productivity effect), knowledge transfer could cause the agent to become self-employed. The agent would then become a strong competitor of the principal (competition effect). I show that there is also an effort effect, which determines the principal’s optimal knowledge transfer and his preference for either a principal-agent relationship or a duopolistic competition with the agent. The principal’s decision depends crucially on whether knowledge transfer leads only to a relative competitive advantage for the agent, or additionally to an absolute advantage when the agent becomes self-employed. I also show that teamwork makes a principal-agent relationship more attractive for the principal and that such effort sharing leads to lower costs.
JEL-Classification: L2, M2.

Capturing Customer Heterogeneity using a Finite Mixture PLS Approach**
Carsten Hahn/Michael D. Johnson/Andreas Herrmann/Frank Huber*
An approach for capturing unobserved customer heterogeneity in structural equation modeling is proposed based on partial least squares. The method uses a modified finite-mixture distribution approach. An empirical analysis using quality, customer satisfaction and loyalty data for convenience stores illustrates the advantages of the new method vis-à-vis a traditional market segmentation scheme based on well known grouping variables. The results confirm the assumption of heterogeneity in the individuals’ perception of the antecedents and consequences of satisfaction and their relationships. The results also illustrate how the finite-mixture approach complements and provides insights over and above a traditional segmentation scheme.
JEL-Classification: M31, C11.