Pulsatilla Adans (PSA) herbs (Ranunculaceae) have been widely used in traditional medicine in China and other countries. However, the authentication and quality control of PSA herbs have always been a challenging task due to their similar morphological characteristics and the diversity of the multiple components that exist in the complicated matrix. Herein, a novel integrated strategy combining UHPLC/Q-Orbitrap-MS techniques with chemometrics analysis is proposed for the discrimination of PSA materials. We developed a comprehensive method integrating a nontargeted bidirectionally screened (NTBDS) MS data set and a targeted extraction peak area analysis for the characterization of triterpenoid saponins of PSA from different species. After that, partial least-squares discriminant analysis (PLS-DA) was performed on the obtained MS data set and the parameter variable importance for the projection (VIP) value and P value were employed to screen the valuable MS features to discriminate PSA from different species. In addition, the receiver operating characteristic (ROC) curve is used to verify the reliability of MS features. Finally, heatmap visualization was employed to clarify the distribution of the identified triterpenoid saponins, and four medicinal species of PSA were successfully differentiated. Additionally, 34 constituents were reported in PSAs for the first time, 81 triterpenoid saponins were identified as differential components, and 12 chemical ingredients were characterized as potential chemical markers to differentiate the four officinal PSA herbs. This is the first time that the differences in different PSA herbs have been observed systematically at the chemical level. The results suggested that using the identified characteristic components as chemical markers to identify different PSA herbs was effective and viable. This method provides promising perspectives in the analysis and identification of the ingredients of Chinese herbal medicines, and the identification of similar herbs from the same species.