Many schools are moving to develop benchmark tests to monitor their students' progress toward state standards throughout the academic year. Benchmark tests can provide the ongoing information that schools need to guide instructional programs and to address student learning problems. The authors discuss six criteria that educators can use to develop, select, or purchase benchmark tests: alignment, diagnostic value, fairness, technical quality, utility, and feasibility. Together, these criteria determine benchmark tests' ability to provide both accurate information about students' progress and useful feedback to improve instruction. (Contains 2 Figures.)