In many metabolomics studies, metabolite identification by mass spectrometry (MS) often is hampered by the lack of good reference compounds, and hence, NMR information is essential for structural elucidation, especially for the very large group of secondary metabolites. The classical approach for compound identification is to perform time-consuming and laborious HPLC fractionations and purifications, before (re)dissolving the molecules in deuterated solvents for NMR measurements. Hence, a more direct and easy purification protocol would save time and efforts. Here, we propose an automated MS-guided HPLC-MS-solid phase extraction-NMR approach, which was used to fully characterize flavonoid structures present in crude tomato plant extracts. NMR spectra of plant metabolites, automatically trapped and purified from LC-MS traces, were successfully obtained, leading to the structural elucidation of the metabolites. The MS-based trapping enabled a direct link between the mass signals and NMR peaks derived from the selected LC-MS peaks, thereby decreasing the time needed for elucidation of the metabolite structures. In addition, automated 1H NMR spectrum fitting further speeded up the candidate rejection process. Our approach facilitates the more rapid unraveling of yet unknown metabolite structures and can therefore make untargeted metabolomics approaches more powerful.