In the design and evolution of software intensive systems, it is desirable to make informed decisions as early as possible in the life cycle. To do this, it is both necessary to be able to predict properties of these future systems and to know how one would like to prioritize among those properties. This paper addresses the latter problem of how to make trade-offs between non-functional properties of software intensive systems. An approach based on the elicitation of utility functions from stake-holders and subsequent checks for consistency among these functions is proposed. A sample GUI is presented, along with some examples. Limitations are discussed and several avenues for future work, including empirical validation, are proposed.