This article examines the convergence of two popular school improvement policies: instructional coaching and data-driven decision making (DDDM). Drawing on a mixed methods study of a statewide reading coach program in Florida middle schools, the article examines how coaches support DDDM and how this support relates to student and teacher outcomes. Authors find that although the majority of coaches spent time helping teachers analyze student data to guide instruction, data support was one among many coach activities. Estimates from models indicate that data analysis support, nevertheless, has a significant association with both perceived improvements in teaching and higher student achievement. (Contains 4 tables, 1 figure and 12 notes.)