Data warehousing is largely directed towards “what to do next” type of decisions that essentially address operational decision-making needs. We argue for developing data warehouse support for deciding on organizational policies: policies evolve and therefore need continual decision-making support. We propose an RE approach for discovering information contents of a data warehouse for policy decision-making. Each policy is represented in an extended first order logic that can be converted into a policy hierarchy. Each node in this hierarchy can be selected, rejected, or modified. In order to take this decision, the relevant information is determined by using Ends Information Elicitation and Critical Success Factor Elicitation techniques for information elicitation. The elicited information is converted into an ER diagram from which star schemas are obtained.