 This paper investigates the use of nitrogen adsorption experiments to characterize the poor structure of shales. Three models are used to analyze the data, the Barrett-Joyner-Holenda adsorption, BJHD model, the Barrett-Joyner-Holenda desorption, BJHD model, and the density functional theory, DFT model. The BJHD model is found to be less accurate when applied to the shale samples, while the DFT model is more accurate. Additionally, the DFT model is shown to be more consistent in its results than the BJHD model. Finally, the authors propose a workflow for optimizing the BJHD model based on the results of the DFT model. This article is authored by Ming Wang, Zhu Li, Zhikai Liang, and others.