The paper applies the choice model incorporating attribute cut-offs proposed by [Swait, J.D., 2001. A non-compensatory choice model incorporating attribute cutoffs. Transportation Research: Part B 35 (10), 903–928] to evaluate shippers’ preferences for freight service attributes. A stated preference experiment on a sample of Italian manufacturing firms shows that the marginal utilities of attributes are not constant over the whole range of values presented, and that significant improvements can be made by introducing non-linearities at points specified by individual respondents. It is found that the conventional model overestimates the importance of the attributes, though their relative ranking is maintained. The proposed choice model provides useful information on the difficulty of achieving a modal transfer of freight from road to truck-rail intermodal transport.