The patterns of growth and development of the therapeutic alliance over the course of therapy have been of continued interest to psychotherapy researchers. The purpose of this study was to investigate whether a simple institutional metacommunication intervention with clients had an effect on the development of the alliance. This adjunctive instruction involved inviting therapy clients to take a proactive role in their treatment by encouraging feedback to their therapist about various aspects of the therapy process. In this randomized controlled study (N = 94), clients were assigned to 1 of 2 conditions: (a) an institutional "adjunctive instruction" condition in which patients were contacted by clinic personnel at the beginning of the remediation phase (Session 5) and encouraged to take a proactive role in their treatment and (b) a "control condition" that contained no institutional adjunctive instruction. Between-condition differences in the alliance were tested, controlling for baseline influences and the early therapeutic alliance. Clients' postsession reports from Sessions 1 to 24 indicated that the adjunctive instruction increased the alliance over the course of therapy vis-a-vis the control condition. The adjunctive instruction appeared to have fostered clients' evaluation of their therapists' interest in their welfare. The results indicate that interventions, even brief or subtle, can produce lasting benefits in the alliance when targeted at specific psychological processes. Systematic metacommunication from the institutional level appeared to reinforce clients' therapeutic alliance with their therapists in individual treatment. (Contains 1 figure, 2 tables, and 1 footnote.)