There is an increasing interest in the development and implementation of digital therapeutics (apps) in individuals with serious mental illness (SMI). However, there is limited understanding of the role of neurocognition and social cognition on engagement with apps. The present study is a secondary analysis of a pilot randomized controlled trial ( = 62) comparing a tailored digital intervention to treat tobacco use disorder in individuals with SMI to a standard of care digital intervention for the general population. The purpose of this study was to examine the impact of neurocognition, social cognition, and clinical characteristics on indices of app engagement in users of the tailored app compared to users of the standard of care app. Correlational analyses demonstrated that individuals with low levels of neurocognition and social cognition engaged more often and for longer duration with the tailored app compared to the standard of care app. In a series of multilevel zero-inflated negative binomial models, assignment to the tailored app remained the most robust predictor of app interactions (Risk Ratio [RR] = 1.72;  < .01), duration of app use (RR = 6.47; p < .01), and average length of interaction (RR = 2.70; p < .01), after adjusting for key demographic and clinical characteristics, and two measures of cognition. This is one of the first studies to demonstrate that digital therapeutics can be designed to mitigate the impact of neurocognition and social cognition on device engagement in SMI populations. Recommendations are made to advance the use of new analytic models to uncover patterns of engagement with digital therapeutics.