Objective This study investigated the cost‐effectiveness of a community‐based colorectal cancer‐screening program (C‐CRCSP) in Shanghai, China, among the residents in the urban, suburban and rural areas. Methods A Markov model was constructed to evaluate the cost‐effectiveness of a 25‐year annual C‐CRCSP including 100 000 populations. Cost‐effectiveness was determined by the incremental cost‐effectiveness ratio (ICER); referring to either life‐years gained, or quality‐adjusted life‐years (QALYs) gained. The threshold was gross domestic product per capita. Univariate and multivariate sensitivity analyses were performed to investigate the influence of compliance, prevalence, technological performance, medical cost and annual cost discount rate (3.5%) on ICER. A probabilistic sensitivity analysis evaluated the probability of the cost‐effectiveness of C‐CRCSP at different maximum acceptable ceiling ratios. Results Compared with no screening, the C‐CRCSP resulted in total gains of 7840 QALYs and 2210 life‐years (LY), at a total cost of CNY 58.54 million; that is, the ICER were CNY 7460/QALYs and CNY 26650/LY. Stratifying by residency, the cumulative gains in QALYs and LY were estimated to be the lowest in the urban populations compared with the rural and suburban populations. The cost for the urban population was 3‐fold and 6‐fold that of the suburban and rural populations. The ICER for QALYs ranged from 2180 (rural) to 16 840 (urban). Conclusion The cost‐effectiveness of a C‐CRCSP in Shanghai was most favorable for the rural population, while the urban population benefits less in terms of QALYs. ICER could be enhanced by measures that increase compliance.