Despite the growing consensus that consumer preferences are often constructed when decisions are made, we still know very little about the factors that determine the strength and stability of constructed preferences. We propose that the manner in which preferences are formed has a significant effect on their strength and stability. In particular, we propose that option set configuration, asymmetric dominance versus compromise, drives the decision process and whether consumers (consciously) focus on the chosen option or the context. This, in turn, affects the resulting decision process trace, as reflected in choice stability, confidence, and attribution. The results of five studies support this proposition. Thus, the strength of constructed preferences is a function of choice problem characteristics and the construction process.