This article discusses the challenges and emerging directions in untargeted metabolomics, focusing on the identification process in mammalian systems. Metabolomics aims to study low molecular weight molecules in organisms, providing insights into genotype-phenotype relationships. Unlike proteomics, metabolomics faces unique challenges in identification due to the complexity and diversity of metabolites. LC-MS-based metabolomics is a leading technology for analyzing small molecules, but it requires careful sample preparation and data processing. The article highlights the importance of accurate metabolite identification for biological interpretation and discusses the limitations of current metabolomics databases. It also addresses the challenges of false positive identifications and the need for improved methods to assess confidence levels in metabolite assignments. The article emphasizes the importance of integrating multiple data types, such as ion mobility and mass spectrometry, to enhance metabolite identification. It also discusses the role of bioinformatics tools in metabolomics data analysis and the need for standardized reporting standards. The article concludes with the potential of metabolomics to better understand the mechanisms underlying health and disease.This article discusses the challenges and emerging directions in untargeted metabolomics, focusing on the identification process in mammalian systems. Metabolomics aims to study low molecular weight molecules in organisms, providing insights into genotype-phenotype relationships. Unlike proteomics, metabolomics faces unique challenges in identification due to the complexity and diversity of metabolites. LC-MS-based metabolomics is a leading technology for analyzing small molecules, but it requires careful sample preparation and data processing. The article highlights the importance of accurate metabolite identification for biological interpretation and discusses the limitations of current metabolomics databases. It also addresses the challenges of false positive identifications and the need for improved methods to assess confidence levels in metabolite assignments. The article emphasizes the importance of integrating multiple data types, such as ion mobility and mass spectrometry, to enhance metabolite identification. It also discusses the role of bioinformatics tools in metabolomics data analysis and the need for standardized reporting standards. The article concludes with the potential of metabolomics to better understand the mechanisms underlying health and disease.