This work aims to study the effect of compound green tea (CGT) on liver lipid metabolism in mice based on metabolomics of liquid chromatography-mass spectrometry (LC-MS), and preliminarily identify potential biomarkers and pathways of action by using a metabonomic network database to explore the lipid-lowering effect of CGT. In this study, forty mice were randomly divided into four groups: compound tea treatment group (DH), high-fat model control group (NK), normal control group (CK) and positive drug group (YK). After a month of different interventions, the mice were weighed and the blood lipid indexes were detected. In addition, differential liver metabolites were monitored by using LC-MS. The results showed that CGT and positive drug treatment were able to decrease body weight, liver coefficient, TC, TG and LDL levels of obese mice, while increasing HDL levels. Among the 110 compounds obtained, 54 metabolites were significantly altered in the four comparisons. More importantly, 15 remarkably downregulated metabolites involved in Lysopc 16:1, Lysopc 18:1, and Lysopc 18:2 were found in the DH group when the mice were treated with CGT; meanwhile, the positive drug Xuezhikang was able to significantly downregulate 14 compounds, including (±)18-HEPE, and 6 keto-PGF1α, compared with the NK group. Besides, KEGG enrichment analysis also revealed the important metabolic pathways, such as linoleic acid metabolism, Biosynthesis of unsaturated fatty acids, and α-linolenic acid metabolism, were related to fatty acid metabolism. These results suggested that CGT could regulate the lipid metabolism in the liver of hyperlipidemia mice, and may regulate 54 potential biomarkers in mice through a related metabolic pathway to make them return to a normal state and improve the disorder of lipid metabolism.