Focus groups were formerly associated with market research but have recently gained some measure of social scientific respectability. In this article, I briefly recapitulate the history of focus groups and then examine their role in 4 (2 Dutch, 2 American) policy research projects. In each case, the focus groups provided an understanding of the interests and values of different stakeholder groups and permitted the analysts to predict the groups' reactions to policy alternatives. This served to link the focus groups to the underlying policy problem, to set the policy issues in their appropriate context, to take due account of the technical complexities of the situation, and to orient toward integrating the results of the focus groups with the other tools used in the policy analysis. The four cases shared a "spiral" model approach to focus groups, in which the discussion moves from generic to specific toward the object of focus rather than tackling it directly. This permits both a breadth and depth of perspective and helps avoid posturing. I conclude that focus groups provide a value for policy analysis because they enable participant stakeholders to become part of the process, help uncover misunderstandings that conceal underlying agreements among stakeholders, and uncover potential problems of implementation.