Objective: This study aimed to evaluate whether coaching features were successfully transmitted via electronic health record (EHR) communication and to evaluate their relationship with weight change in a previously tested EHR-based coaching intervention. Methods: A secondary analysis from the Maintaining Activity and Nutrition through Technology-Assisted Innovation in Primary Care (MAINTAIN-pc) study randomized clinical trial was conducted in nine primary care practices and one specialty practice (endocrinology) affiliated with the University of Pittsburgh Medical Center. Eligibility included age 18 to 75 years, intentional 5% weight loss in the previous 2 years, access to an internet-connected computer, and receipt of care from a University of Pittsburgh Medical Center primary care provider. Survey data were collected during the randomized clinical trial. Results: Participants content with intervention delivery via the EHR and those who felt a strong connection to their coach had significantly less weight regain (p = 0.013 and p < 0.01, respectively). Participants who had needs unmet by the intervention (e.g., "in-person" support in a group setting or individual settings) regained more weight (p < 0.01). Conclusions: The data suggest heterogeneity in the patient population regarding preference for in-person versus EHR-based coaching formats. Such heterogeneity might result in the differential success of EHR-based interventions.