A targeted metabolomics LC-ESI-QqQMS application for geographical origin discrimination based on 20 nonpolar key metabolites was developed, validated according to accepted guidelines and used for quantitation-via stable isotope labeled internal standards in 202 raw authentic hazelnut samples from six countries (Turkey, Italy, Georgia, Spain, France, and Germany) of harvest years 2014 and 2015. Multivariate statistics were used for detection of significant variations in metabolite levels between countries and, moreover, a prediction model using support vector machine classification (SVM) was calculated yielding 100\% training accuracy and 97\% cross-validation accuracy, which was subsequently applied to SS hazelnut samples for the confectionary industry gaining up to 80\% correct classifications compared to declared origin. The present method demonstrates the great suitability for targeted metabolomics applications in the geographical origin determination of hazelnuts and their applicability in routine analytics.