Limited variation in consumer choices observed in the market may restrict marketing managers’ ability to determine what drives their customers’ decisions. For example, shopping centers typically are fixed over time with little variation in their attributes and any given customer will consider only few alternatives. In this paper, we introduce a novel approach to overcome this challenge by drawing on consumers’ mental representations of decisions. The approach bridges choice modelling and means-end chain analysis and is rooted in utility theory. It relies on the fact that there are gains and costs to including additional components in the mental representation of a decision. The gains are that with every decision component, the individual is better able to discriminate between choice alternatives and the probability of making the right choice increases. The costs are the mental efforts of evaluating each additional component, which involve memory retrieval, inference and judgment tasks. We propose that akin to information search theory, additional decision components are activated in a mental representation only if the expected gains of evaluating alternatives on the decision component exceed the cognitive costs of doing so. This theoretical proposition is formalized in a model of individuals’ latent utility for the alternatives in a decision. The utility function drives the activation of attribute and benefit components in the mental representation such that they are included only if variation in their level has a strong enough impact on the decision (i.e., above the individual’s mental cost threshold). This formalization enables us to simultaneously model decision utility and the cognitive activation of attributes and benefits, and hence decompose the utility of fixed alternatives from mental representations of decisions. We illustrate the model using data collected in a tailored online study of individuals’ mental representations of a hypothetical shopping decision problem.