Background: For a large trial of the effect of group antenatal care on perinatal outcomes in Rwanda, a Technical Working Group customized the group care model for implementation in this context. This process analysis aimed to understand the degree of fidelity with which the group antenatal care model was implemented during the trial period. Methods: We used two discreet questionnaires to collect data from two groups about the fidelity with which the group antenatal care model was implemented during this trial period. Group care facilitators recorded descriptive data about each visit and self-assessed process fidelity with a series of yes/no checkboxes. Master Trainers assessed process fidelity with an 11-item tool using a 5-point scale of 0 (worst) to 4 (best). Results: We analyzed 2763 questionnaires completed by group care facilitators that documented discreet group visits among pregnant and postnatal women and 140 questionnaires completed by Master Trainers during supervision visits. Data recorded by both groups was available for 84 group care visits, and we compared these assessments by visit. Approximately 80% of all group visits were provided as intended, with respect to both objective measures (e.g. group size) and process fidelity. We did not find reliable correlations between conceptually-related items scored by Master Trainers and self-assessment data reported by group visit facilitators. Conclusions: We recommend both the continued participation of expert observers at new and existing group care sites and ongoing self-assessment by group care facilitators. Finally, we present two abbreviated assessment tools developed by a Rwanda-specific Technical Working Group that reviewed these research results.