HIV/AIDS prevalence is still high in Botswana. The main aim of this study was to assess and compare sexual risk behaviours of circumcised and uncircumcised men before and after the launch of the safe male circumcision programme. Data used for analyses were derived from the 2008 and 2013 Botswana AIDS Impact Surveys. Modified Poisson regression analysis was used to obtain prevalence ratios (PR) as measures of association between circumcision status and multiple sexual partners, transactional sex, inconsistent condom use and intergenerational sex. The proportion of circumcised men increased two times between 2008 (12.5%) and 2013 (25.2%). Prevalence of multiple sexual partnerships was high among uncircumcised than circumcised (54.6% vs. 46.4%) men in 2008, but in 2013 after the introduction of the SMC programme it was slightly high among circumcised men than uncircumcised men (23.2 vs. 21.8%). In the adjusted analyses, being circumcised was significantly associated with having multiple sexual partners (2008=adjusted PR=1.31, CI=1.10-1.57; 2013= adjusted PR=1.12, CI=1.01-1.41) and transactional sex (2008=adjusted PR=1.98, CI=1.26-3.11; 2013=adjusted PR=1.60, CI=1.09-1.22) for both survey periods. These results indicate the need to continuously sensitise and encourage men to stop multiple sexual partnerships and transactional sex. Moreover, there is need to encourage all men to use condoms consistently.