Aquifer storage and recovery (ASR) is an effective strategy for water resources management and has been widely used in many contaminated and saline aquifers. However, its recovery efficiency (RE) may be significantly affected by mass transfer limitations. A numerical model is developed to simulate ASR performance by combining the convergent and divergent dispersion models with a first-order mass transfer model. By analyzing the concentration history at the pumping well, we obtain simple and effective relationships for investigating ASR efficiency under various mass transfer parameters, including capacity ratio and mass transfer timescale, and operational parameters, including injection durations and well-pumping rates. On the basis of such relationships, one can conveniently determine whether a site with mass transfer limitations is appropriate or not for ASR and how many ASR cycles are required for achieving a positive RE. Results indicate that the immobile domain may function as a contaminant source or sink or both during the recovery phase and RE usually improves with well-flow rate, the decrease of capacity ratio, and the ASR cycles. However, RE is a nonmonotonic function of the mass transfer timescale and the injection duration. A critical timescale is given for quantifying this nonmonotonic behavior. When the injection period is greater than such a critical value, increasing injection period results in a higher RE. Contrarily, when the injection period is less than the critical value, increasing the injection period may even yield a lower RE.