Screening and identification of active components from traditional Chinese medicines is rather challenging due to the diversity and complexity of chemical components. Herein, a comprehensive strategy based on a spectrum-effect relationship model and LC-MS analysis was developed to screen active components from Terminalia chebula fruits. The water extract of T. chebula fruits was subjected to macroporous resin column and then eluted successively with water and 30%, 50%, 70%, and 95% ethanol. The 30% ethanol eluate fractions of eighteen batches from T. chebula fruits were used for the spectrum-effect relationship study. The IC 50 values for acetylcholinesterase inhibitory and 2,2diphenyl-1-picrylhydrazyl scavenging activities were measured, LC fingerprints were established, and 15 common peaks were specified. The spectrum-effect relationship between common peaks and IC 50 values was investigated by principal component analysis, gray relational analysis, partial least square and multiple linear regression. The 30% ethanol eluate fraction was further characterized by LC-MS analysis. The chromatographic peaks (Peaks 1,2,3,5,12,14,15) making great contributions to the efficacy were screened through a spectrum-effect relationship model, and sixteen components were further identified. The results suggested that the proposed strategy is simple and effective for acquiring active components from a complex matrix.