Abstract: Horizontal well staged fracturing is the key technology to realize the economic development of shale gas. The closure of fracturing fractures in the production process will have adverse effects on the exploitation. Due to the large errors and serious oscillations of dynamic production data, it does not match the internal boundary conditions of the seepage flow mathematical model. Therefore, there are few quantitative methods to evaluate the difference of fracture characteristics between fracturing fluid flowback stage and shale gas production stage based on dynamic production data analysis currently. Based on this concern, a new method of production dynamic data analysis based on deconvolution is proposed to quantitatively evaluate the difference of fracture characteristics between flowback stage and production stage in this paper. Firstly, the seepage flow models and their Laplace solutions corresponding to flowback stage and production stage are given. Secondly, the pressure deconvolution algorithm is used to normalize the dynamic production data of the two stages. Then, the normalization parameter adjustment of deconvolution calculation and the parameter adjustment of theoretical seepage flow model calculation are mutually restricted in the process of typical curve fitting, and the fracture half-length and fracture conductivity of the two stages are interpreted respectively. Finally, the conductivity modulus is introduced to quantitatively evaluate the difference of fracture characteristics between flowback stage and production stage. The established method is used to analyze 10 wells in the field. The results show that this method can effectively quantify the difference of fracturing fracture characteristics between flowback stage and production stage; compared with the flowback stage, the fracture conductivity decreased by about two orders of magnitude in the production stage, and the fracture closed significantly. The analysis method established in this paper has important reference value for the optimization of stimulation measures in the later stage of shale gas reservoir.