A two-dimensional framework for Decision Support Systems (DSS) cost-benefit analysis is proposed. One dimension reflects the degree to which the supported decision is structured, the level of managerial activity, the level of uncertainty and the source of information used, while the second dimension classifies DSS based on the phase of the supported decision according to Simon's model for the decision-making process. The review of the current literature reconciles some of the contradictory findings in the DSS cost-benefit literature, and shows that the adequacy of the valuation method depends on the attributes of the system. The main DSS valuation issues are identified, and the paper discusses the potential gap between price (determined by estimated costs or the market) and the value (which is subjective). The magnitude of this gap is found to be correlated with the extent to which the supported decisions are structured.