Purpose: This study aimed to investigate effects of a flexible multi-component workplace intervention program for improving physical activity among Japanese white-collar employees in a cluster randomized controlled trial design. Methods: A total of 208 worksites and nested employees were approached. Any worksite interested in a 3-month physical activity promotion intervention and white-collar workers aged 18 years or older were eligible. The worksites were randomly assigned to an intervention or a control, stratified by worksite size. The intervention worksites were offered an intervention program that allows these worksites to select intervention components from a set of options, including environmental changes. The control worksites were offered feedback and basic occupational health services. Physical activity level was measured using a self-reported questionnaire at baseline and at 3 and 6 months of follow-up. Results: Three worksites (92 employees) were allocated to the intervention and five worksites (98 employees) to the control worksites. The overall physical activity level in the intervention worksites significantly increased compared with the control worksites (Coeff = 0.45, SE = 0.19, p = 0.018). For subgroup analyses, the intervention effect on the overall physical activity level was significant among medium- and large-sized worksites but not among small-sized worksites. Conclusion: This trial showed a significant and positive effect of the intervention program on physical activity. The program is unique because of its flexibility and feasibility. However, small worksites might receive less benefit from the program, indicating a need for further support and/or new technologies. Trial Registration: UMIN Clinical Trials Registry (ID=UMIN000024069).