This study aimed to employ a comprehensive data screening strategy to identify the chromones and coumarins present in Saposhnikovia divaricata (SD). Initially, the five-point mass defect filter (MDF) method was utilized to screen the respective three subclasses of chromones and coumarins for MS1. In comparison to the traditional MDF method, the number of interference peaks was reduced from 3462 to 1053, representing a decrease of 69.58 %. Then, diagnostic fragment ion filtering (DFIF) was used to screen product ions selected by MDF method, which was based on the fragmentation rules of each subclass to screen whether it met the conditions. Finally, we characterized 94 compounds from SD, including 40 chromones and 54 coumarins, among them, 82 chromones and coumarins were identified by combining the above two methods, 12 coumarins were identified by combining MDF and references, they were classified into other coumarins. Twenty one compounds were identified for the first time from SD, and 3 chromones and 7 coumarins were unknown compounds. Through untargeted metabolomics analysis of SD samples from 12 different regions, significant differences were found in SD samples from different areas, and 20 differential metabolites were distinguished, including 13 chromones and 7 coumarins. This study established for the first time a comprehensive strategy combining MDF, DFIF, and untargeted metabolomics to evaluate SD quality. The results indicated that this method is an efficient, accurate, and promising approach for classifying and exploring compounds in complex natural product systems, providing a basis for evaluating the quality of SD from different sources.