A non-targeted analytical approach using gas chromatography-mass spectrometry (GC-MS) is proposed for the analysis of the free and bound volatile fractions of three emblematic indigenous Greek white winegrape varieties belonging to Vidiano, Malagousia, and Savvatiano and establish volatile varietal markers using multivariate chemometrics. A total of 89 free and 103 bound volatile compounds were identified, categorized into alcohols, aldehydes, esters, acids, terpenes, norisoprenoids, C6 compounds, phenols, and ketones. A robust Partial Least Squares Discriminant Analysis (PLS-DA) prediction model was developed and validated, and successfully classified the grape samples according to the variety with 100 % accuracy, demonstrating the potential of volatile profiling as a non-targeted fingerprinting approach for varietal discrimination. Malagousia variety exhibited the highest concentration of terpenes, particularly in the bound fraction, highlighting its intense floral character. The proposed analytical workflow was successfully applied for the investigation of the unique aroma profiles of Greek white winegrape varieties. The findings contribute to the understanding of unique volatile fingerprints of the winegrape varieties and support the adaptation of vineyard and winery practices to enhance their varietal expression.