Three data sets profiling the organization, financing, and services of state mental health agencies, compiled routinely for all states, were examined to assess their potential for use in mental health policy analyses. Strengths included standardized protocols for compiling data at the state level, a high degree of completeness in two of the data sets, consistency in data definitions over time, and built-in data integrity and editing procedures. Data are aggregated at the state level and lack user-friendly features that facilitate downloading to statistical software. However, potential use of these data as contextual factors in multilevel models makes them an important resource for consideration.