The antitussive effect of Platycodonis Radix is closely related to the components in saponins fraction of Platycodonis Radix extract (SFPRE); however, these active components and their holistic mechanism remain unknown. Hence, a new method by integrating spectrum-effect relationship analysis with metabolomics analysis was applied to study the active components and their holistic mechanism simultaneously. For spectrum-effect relationship analysis, chemical fingerprints of ten batches of SFPRE were developed using UHPLC-LTQ-Orbitrap MSn; antitussive effect were evaluated using a classic mice-cough model induced by ammonia liquor. Spectrum-effect relationship was analyzed by partial least squares regression (PLSR) analysis. For metabolomics analysis, the altered metabolites related to cough in serum were identified by UHPLC-Q-TOF/MS and orthogonal partial least squares-discriminant analysis (OPLS-DA); metabolic pathway analysis was depended on MetaboAnalyst 4.0, KEGG database, METLIN database and HMDB database. Our findings showed that 10 identified components of Polygalacin D (peak 26), Deapio-platycodin D (peak 21), Platycodin D (peak 23), β-Gentiotriosyl platycodigenin (peak 37), Platycoside G3 (peak 17), Platycoside C (peak 25), Platycodin D3 (peak 16), 3-O-β-D-glucopyranosyl platycodigenin (peak 33), Platycoside F (peak 19) and 3″-O-acetyl platycodin D3 (peak 15), and 2 unidentified components (peak 45 and 44) possessed antitussive effects. The metabolomics analysis result showed that 19 metabolites were potential biomarkers related to the cough, 16 of which could be restored to normal levels by SFPRE. These biomarkers were involved in arachidonic acid metabolism, linoleic acid metabolism and glycerophospholipid metabolism. The current study may facilitate the development of antitussive medicines with fewer side-effects based on Platycodonis Radix.