The Brunauer−Emmett−Teller (BET) method has been widely used to characterize surface area of various porous materials. As a result of some oversimplified assumptions, the performance of BET method can be compromised in micro-and mesopores. Thus, the consistency criteria were introduced to improve its performance. However, the applicability of the consistency criteria for kerogen mesopores has not yet been assessed. In this work, N 2 adsorption in graphite and kerogen mesopores is conducted by grand canonical Monte Carlo (GCMC) simulations at 77 K, while the accuracy of the standard BET method (BET_STD) and the BET considering consistency criteria (BET_CC) for surface area estimation is assessed. For graphite slit mesopores, BET_CC can better describe N 2 adsorption behaviors in graphite mesopores than BET_STD. For kerogen mesopores, the surface area from BET_CC has a better linear correlation with CH 4 excess/total adsorption under in situ conditions than that from BET_STD. In other words, the surface area from BET_CC is a better indicator than that from BET_STD to predict CH 4 adsorption behaviors. Our work provides important insights into BET surface area characterization in kerogen mesopores and optimization of characterization processes to better assess gas-in-place estimation in shale media.