BACKGROUND: A growing number of the population are using the Internet for health information, such as physical activity (PA). The aim of this study was to examine the effectiveness of delivery modes for a behavior change program targeting PA. METHODS: A randomized trial was conducted with 192 subjects randomly allocated to either a face-to-face, Internet-mediated, or Internet-only arm of a 12-wk intervention. Subjects included inactive adults with Internet access. The primary outcome variable was self-reported PA, assessed at four time points. RESULTS: The results showed no group x time interaction for PA F(6, 567) = 1.64, p > 0.05, and no main effect for group F(2, 189) = 1.58, p > 0.05. However, a main effect for time F(3, 567) = 75.7, p < 0.01 was observed for each group. All groups were statistically equivalent immediately post-intervention (p < 0.05), but not at the follow-up time points (p > 0.05). The Internet-mediated and Internet-only groups showed similar increases in PA to the face-to-face group immediately post-intervention. CONCLUSIONS: This study provides evidence in support of the Internet in the delivery of PA interventions and highlights avenues for future research.