Preeclampsia is one of the most common severe pregnancy complications in obstetrics, which is considered a placental source disease. However, the mechanisms underlying preeclampsia remain largely unknown. In this study, UPLC-MS/MS-based metabolomic and lipidomic analysis was used to explore the characteristic placental metabolites in preeclampsia. The results revealed that there were significant changes in metabolites between preeclampsia and normotensive placentas. Weighted correlation network analysis (WGCNA) identified the correlation network module of metabolites highly related to preeclampsia and the clinical traits reflecting disease severity. The metabolic perturbations were primarily associated with glycerophospholipid and glutathione metabolism, which might influent membrane structures of organisms and mitochondria function. Using linear models, three metabolites had an area under receiver operating characteristic curves (AUROC) ≥ 0.80 and three lipids had an AUROC ≥ 0.90. Therefore, metabolomics and lipidomics may offer a novel insight for a better understanding of preeclampsia and provide a useful molecular mechanism underlying preeclampsia.