This presentation will explore how we can better understand and leverage engagement in digital mental health interventions—not just as a usage metric, but as a meaningful indicator of clinical effectiveness, patient needs, and long-term outcomes. Drawing from over a decade of work in the field, I will present insights from real-world deployment and clinical studies to illustrate the importance of examining how, when, and for whom engagement drives impact. A central focus will be on new findings from my current work at Bright Therapeutics, where we lead the development and research behind Recovery Path, an app for individuals receiving treatment for substance use disorders, including opioid use disorder (OUD). I will present results from a recent multisite study showing how implementation of Recovery Path in opioid treatment programs was associated with significant improvements in patient retention. Clinics where counselors were trained to use the app achieved a significant increase in 30-day retention, a key predictor of long-term treatment success. These findings reinforce the idea that digital tools, when meaningfully integrated and used to support clinician-patient relationships, can have a real and measurable impact on care. The talk will also draw from my prior work at SilverCloud Health, an online platform for depression and anxiety with wide-scale deployment across health systems in the US and UK. In partnership with Microsoft Research, we identified patterns of engagement and developed a machine learning model that predicted treatment success. These insights were then embedded into the platform as a clinical decision support tool to help guide human coaches toward personalized, data-driven support. This translational effort—from exploratory data analysis to real-world deployment—offers a blueprint for how digital mental health services can move beyond basic adherence metrics to achieve precision engagement. Together, these projects offer a cross-condition, cross-platform perspective on how digital interventions can fulfill their potential through thoughtful measurement, design, and integration of engagement strategies. The presentation will examine questions such as: What types of engagement actually matter for outcomes? How can we distinguish meaningful engagement from noise in real-world data? What is the role of clinicians in sustaining engagement, particularly in high-risk populations such as those with OUD? I will close by inviting the audience to reframe the conversation around digital mental health—from a narrow focus on downloads and logins to a richer, more clinical understanding of engagement as a key ingredient in treatment success. This session will be of interest to researchers, clinicians, digital health developers, and policymakers seeking to optimize the effectiveness and equity of digital mental health solutions.