Any evaluation of the benefits of policy-oriented social science research faces fundamental difficulties. These include the uncertainty in determining a causal link between research and the outcome of a policy or the value of a policy outcome. Nonetheless, firm connections can be established between policy research and policy outcomes if there are strong links that bridge the gaps between social science research and the various parts of the policy process. These connections can be established often enough to make it possible to learn about the relationship between research and outcome and the key variables that affect the social profitability of the underlying research. This essay uses the author's experience with agricultural price policies in Asia, Indonesia in particular, to examine these connections. Four issues pervade the analysis of price policy in Asia: How does an analyst know what policy is best? How can an analyst best communicate the results of research to policymakers? Can a new policy be implemented? Does the new policy work? This last issue, the evaluation of policy, is often neglected, but it can provide an important input into the design of policy and should be made an integral part of any policy process. The author's experience in Indonesia suggests four factors that can make policy-oriented research successful. First, the analyst should be involved with the same policymakers or in the same policy setting for the long term. Second, there is a need to find a balance between keeping analysis and advice confidential and the ultimate publication of the key models and results. Third, the analysts should rely on the analytical paradigms of the mainstream of the economic profession even while examining deviations from their underlying assumptions. Lastly, there should be continuing demand from policymakers for problem-oriented analysis.