Routine outcome monitoring (ROM) systems that identify clients at-risk for treatment failure using outcome and therapeutic process measures are a recognized evidence-based practice. However, only 3 empirical studies have tested ROMs in group therapy, producing mixed results. This randomized clinical trial tested the Outcome Questionnaire System, the ROM system with the most empirical support for individual therapy patients, with 430 group therapy patients who were randomly assigned to 2 experimental arms (Group Questionnaire [GQ] + Outcome Questionnaire-45 [OQ-45] vs. OQ-45). Given the strong evidence for progress feedback, the primary purpose of this study was to ascertain if therapeutic relationship feedback using the GQ reduced rates of relationship deterioration and failure when progress feedback was held constant. Group leaders simultaneously ran pairs of groups that were randomly assigned to the 2 conditions. Of the 430 patients enrolled in 58 groups, 374 attended more than 4 sessions. Results showed that therapeutic relationship predicted improvement in outcome and that feedback reversed the course of relationship deterioration and reduced rates of relationship failure. Although there were no effects on attendance and dropout for feedback, the 2 experimental arms produced mixed results for the OQ-45 not-on-track cases. The combined relationship and progress feedback (GQ/OQ-45) was associated with fewer outcome deterioration cases, while the progress feedback condition (OQ-45) showed higher outcome improvement cases. Findings are discussed with respect to previous group ROM studies, clinical implications, and future research. (PsycINFO Database Record