Challenging ‘Factor–Cluster Segmentation

potential for profit increase through more targeted marketing activities with a higher effect on market demand within the targeted segment (Kotler 1997). The burden of responsibility on researchers to identify the optimal segments is different for segmentation studies of different nature. In the case of a priori (Mazanec 2000) or commonsense segmentation (Dolnicar 2004) and extensions thereof (concepts 1, 3, 4, and 5 according to the classifica-tion of segmentation studies proposed by Dolnicar 2004) the crucial decision is the selection of the segmentation criterion or criteria. For instance, a destination might choose to target young tourists using age as the commonsense criterion. On closer evaluation, however, it might turn out that using the stage in the family life cycle would have been a better choice, as the destination’s strength lies in providing optimal services to young families, rather than young singles or groups of young tourists. In the case of post-hoc (Myers and Tauber 1977), a poste-riori (Mazanec 2000), or data-driven segmentation (Dolnicar 2004) and extensions thereof (segmentation concepts 2, 4, 5, and 6), this burden of responsibility lies on the research approach of the data-driven segmentation study undertaken. Because the process of data-driven segmentation consists of numerous components, most of them requiring a decision on the part of the researcher, it is more difficult to avoid potential misinterpretations or suboptimal procedural decisions than in the case of a priori segmentation studies. The present study investigates one particular aspect of the data-driven segmentation process: the way the original answers of respondents are used to derive market segments. Sara Dolnicar, PhD, is a full professor at the School of Management and Marketing (University of Wollongong) and the director of the Marketing Research Innovation Centre (MRIC) i