Coaching is a key driver supporting implementation of evidence-based instructional and behavioral practices (EBPs). The purposes of this study were to propose a model of coaching and examine the effects of two coaching mechanisms (prompting and performance feedback) on (a) teacher use of classroom management EBPs and (b) student classroom disruption. Using two counterbalanced, concurrent multiple-baseline designs across seven teachers, modest functional relations were found between highly efficient delivery of coaching using prompting alone and performance feedback alone on (a) an increase in teacher use of EBPs and (b) a decrease in classroom disruption. No consistent differences were observed in the order with which prompting versus performance feedback were introduced, nor were consistent benefits observed when adding combined versus individual use of prompting or performance feedback. Findings and implications for coaching research and practice are discussed.