Small changes to operations or service delivery can improve the efficiency and effectiveness of public programs. Yet, decision-makers often lack timely and reliable evidence of whether such changes produce the expected results. Rapid-cycle evaluation is a low-cost way to determine, with a high degree of confidence, whether and for whom a program change causes the intended improvements or potentially results in other, unintentional consequences. Mathematica Policy Research is a leader in helping clients to use this rigorous approach as a decision support tool for continuous program and quality improvement.