Typical metabolomics experiments produce large amounts of information that need to be transformed into biologically relevant information such as metabolite identities and concentrations. The data preprocessing steps include peak detection, alignment, normalization and quantification. As for analytical quality, the quality of the data processing also needs to be monitored; which is generally a challenging task. Owing to the increasing demand for metabolomics analyses of large sample series, such as in a medical epidemiological setting, there is also a need to correct the potential ‘batch effect,’ i.e., the analytical variation that may occur due to minor changes in laboratory conditions over extended periods in time. Advances have been made in this area, although processing of large‐scale metabolomics data remains an important and emerging aspect of methodological developments.