This paper presents a process evaluation that assessed the fidelity and quality of implementation, as well as the acceptability and subjective evaluations of a HIV/AIDS intervention among students and teachers. Methods: The process evaluation was conducted as part of a cluster randomized controlled trial of a theory- and evidence-based school HIV/AIDS intervention in Cape Town. The intervention was designed for grade 8 high school students and delivered by teachers over a six-month period. Twenty-six schools participated in the trial, 13 in the intervention group and 13 in the control group. Results: The success of implementation was varied within and across the schools, with some teachers implementing the intervention with more fidelity than others. This was influenced by a combination of individual characteristics and institutional factors. The factors that aided implementation included compliance with the current outcomes-based education approach; provision of teacher training; provision of teacher manuals with detailed information and instructions about the lessons and activities; continued monitoring and support for teachers; and student enthusiasm for the lessons. Proper implementation was hindered by large class sizes; too many activities in the intervention; teacher resistance to and inexperience in using participatory methods; teacher turnover; the low status of life orientation compared to other subjects; and a general disregard for life orientation among students. Conclusions: These findings are important for improving the intervention and contextualizing the results of the outcome evaluation; and to better plan for further large scale dissemination of school-based HIV/AIDS intervention programmes.