Due to the increasing complexity of Data Warehouses (DW), continuous attention must be paid for evaluation of their quality throughout their design and development. DW quality depends on the quality of all requirements, conceptual, logical and physical models used for DW design. Various authors have proposed metrics to assure the quality of conceptual, logical and physical models for DW. However, there is no significant work in the DW literature to assure the quality of a requirements model. A good quality requirements model may lead to a good quality DW. In this paper, we propose a qualityoriented requirements model for a DW. In the proposed model, the notion of perspective is introduced to capture the intention of the agents (stakeholders) associated with their goals. The agent may view a soft goal from two perspectives: decisional and quality perspective. In the former, the agent may refine a soft goal into the goal having well defined criteria for its achievement and termed as decisional goal, whereas in the later the agent may define the various constraint (timing constraint, budgetary constraint etc.) associated with the decisional goals. The agents suggest the decisions for achieving their decisional goals considering these constraints. Thus, the decisional goals and the constraints specified in the decisional and quality perspective respectively should be maintained as meta-data of the DW. In this way, the quality of DW requirements model will be enhanced, which may lead to enhance the quality of conceptual, logical and physical model of DW.