Löffler, S.S.LöfflerBaier, D.D.Baier2022-03-132022-03-132016https://publica.fraunhofer.de/handle/publica/40167310.1007/978-3-319-25226-1_262-s2.0-84981549535Conjoint analysis is a widespread method for modeling and measuring preferences for multi-attributed products in marketing: A sample of customers are asked to evaluate (fictive) offers (attribute-level-combinations). From these individual responses partworths are estimated and used to design and price offers that maximize, e.g., market share, sales, or profit. However, it can be theoretically and empirically argued that partworths not only vary across individuals but also within them. In this paper, we discuss an approach that respects these variations. Partworths are situation-specific modeled at the individual level. The empirical partworth distributions are estimated using Bayesian procedures. The approach is applied to waterpark design and pricing using simulated and real data. It is shown that taking these variations into account influences the maximization.en004005006conference paper