Consumers are thought to be rational in their choices. But sometimes their choices comprise intransitivi-ties. Current paper presents a mathematically supported experimental study of consumers' intransitivepreferences concerning product attributes. We used a personalized experimental design developed on thebasis of Q methodology to maintain subjectivity. In our experimental study we are investigating intran-sitivity occurring in participants' preferences during selection between simple, medium complex, andcomplex products. By the Q method we gain the same number of less important (middle level) attributedimensions for all participants. The participants' task is to make pairwise comparisons of preferencebetween specific realizations of each product group. Since conjoint analysis is based on the measu-rement and testing of consumer preferences, reassessment of the phenomenon may have impact onmarketing research methodology. This research is to test an experimental design which-with someadjustment-could be able to measure revealed preferences in a more precise way.