Background: Cognitive mapping is a participatory research methodology that documents, in visual form, a construct of the local environment in which people live and work. We adapted this method to provide detailed data about study locales to inform recruitment and retention strategies for HIV prevention community based clinical trials. Methods: Four cognitive mapping studies were undertaken between 2005 and 2010, in and around Johannesburg in Orange Farm, and Hillbrow. Participants included members of clinical trial Community Advisory Boards (CAB), young people recruited from schools in Hillbrow and an organization for out of school youth, and men recruited from a database of men in the community interested in taking part in research. These groups participated in reconnaissance walks and in depth interviews, and drew detailed sketch maps. Results: The cognitive maps defined the physical boundaries of the research locales, provided insights into their social histories, and identified important characteristics of the population such as movement, social and sexual networks, ethnic and other divisions. Important differences between the official cartographic maps and the cognitive maps were raised. The mapping data was applied by identifying key areas for recruitment that recruitment staff members were less familiar with and that may otherwise have been overlooked. Conclusions: Cognitive mapping is an effective, rapid and low cost method that can be used to inform recruitment and retention strategies for community-based clinical trial research. The method also provides a means for clinical trial researchers to involve the local community in research and to familiarise them with the social setting.