Due to the complexity of typical metabolomics samples and the many steps required to obtain quantitative data in GC × GC–MS consisting of deconvolution, peak picking, peak merging, and integration, the unbiased non-target quantification of GC × GC–MS data still poses a major challenge in metabolomics analysis. The feasibility of using commercially available software for non-target processing of GC × GC–MS data was assessed. For this purpose a set of mouse liver samples (24 study samples and five quality control (QC) samples prepared from the study samples) were measured with GC × GC–MS and GC–MS to study the development and progression of insulin resistance, a primary characteristic of diabetes type 2. A total of 170 and 691 peaks were quantified in, respectively, the GC–MS and GC × GC–MS data for all study and QC samples. The quantitative results for the QC samples were compared to assess the quality of semi-automated GC × GC–MS processing compared to targeted GC–MS processing which involved time-consuming manual correction of all wrongly integrated metabolites and was considered as golden standard. The relative standard deviations (RSDs) obtained with GC × GC–MS were somewhat higher than with GC–MS, due to less accurate processing. Still, the biological information in the study samples was preserved and the added value of GC × GC–MS was demonstrated; many additional candidate biomarkers were found with GC × GC–MS compared to GC–MS. Electronic supplementary material The online version of this article (doi:10.1007/s11306-010-0219-6) contains supplementary material, which is available to authorized users.