Given the importance of education and the growing public demand for improving education quality under tight budget constraints, there has been an emerging movement to call for research-informed decisions in educational resource allocation. Despite the abundance of rigorous studies on the effectiveness, cost, and implementation of educational interventions, school- and district-level practitioners find that the applicability of these studies to their decision-making problems in the real educational context can be limited. The purpose of this paper is to address three methodological challenges that hamper the influence of educational research on educational decision-making. These three methodological issues include how to generalize or extrapolate effectiveness and cost information from the evaluation site(s) to a specific context, how to incorporate information from multiple sources, and how to aggregate multiple consequences of an intervention into one framework. In particular, the author introduces Bayesian decision theory, a method that processes research findings and subjective judgments using a transparent mechanism, to address these three challenges so as to provide clear, localized direction for decision makers.