Balsamic vinegar of Modena (BVM) and traditional balsamic vinegar of Modena (TBVM) are highly appreciated typical Italian products. The quality control and authentication assurance of both these balsamic vinegars are very important topics. In the recent years, the interest to develop new and standardized analytical procedures, able to further enhance the quality and commercial value of these typical and unique products and to preserve them from possible sophistications and adulterations, is increased. In this work, 76 samples of both BVM and TBVM were analyzed by 1H NMR spectroscopy coupled with multivariate data analysis. The spectral data were analyzed by principal component analysis (PCA), general discriminant analysis (GDA) and classification tree analysis (CTA). The best and very promising model was obtained by a GDA which shows 98.6% of total variance explained by the first canonical function and a predictive capacity of 98.4% with a good separation between clusters. The signals of 5-HMF, α-glucopyranose, malic acid, succinic and acetic acids and the signal at 3.3 ppm were found to be the most statistically significant variables. •Balsamic and Traditional Balsamic vinegars of Modena were analyzed by 1H NMR.•Spectral data were analyzed by chemometric techniques.•NMR and statistics are powerful tools in balsamic vinegars characterization.•A model able to classify the samples was obtained.