CO2 dry fracturing is a promising alternative method to water fracturing in tight gas reservoirs, especially in water-scarce areas such as the Loess Plateau. The CO2 flowback efficiency is a critical factor that affects the final gas production effect. However, there have been few studies focusing on the flowback characteristics after CO2 dry fracturing. In this study, an extensive core-to-field scale study was conducted to investigate CO2 flowback characteristics and CH4 production behavior. Firstly, to investigate the impact of core properties and production conditions on CO2 flowback, a series of laboratory experiments at the core scale were conducted. Then, the key factors affecting the flowback were analyzed using the grey correlation method based on field data. Finally, taking the construction parameters of Well S60 as an example, a dual-permeability model was used to characterize the different seepage fields in the matrix and fracture for tight gas reservoirs. The production parameters after CO2 dry fracturing were then optimized. Experimental results demonstrate that CO2 dry fracturing is more effective than slickwater fracturing, with a 9.2% increase in CH4 recovery. The increase in core permeability plays a positive role in improving CH4 production and CO2 flowback. The soaking process is mainly affected by CO2 diffusion, and the soaking time should be controlled within 12 h. Increasing the flowback pressure gradient results in a significant increase in both CH4 recovery and CO2 flowback efficiency. While, an increase in CO2 injection is not conducive to CH4 production and CO2 flowback. Based on the experimental and field data, the important factors affecting flowback and production were comprehensively and effectively discussed. The results show that permeability is the most important factor, followed by porosity and effective thickness. Considering flowback efficiency and the influence of proppant reflux, the injection volume should be the minimum volume that meets the requirements for generating fractures. The soaking time should be short which is 1 day in this study, and the optimal bottom hole flowback pressure should be set at 10 MPa. This study aims to improve the understanding of CO2 dry fracturing in tight gas reservoirs and provide valuable insights for optimizing the process parameters.