An electronic panel has been used to characterise the organoleptic characteristics of twenty-five extra virgin olive oils from varieties Hojiblanca, Picual and Arbequina, with different degree of bitterness. The method consists in the combination of three systems: electronic nose, electronic tongue and electronic eye. The Principal Component Analysis (PCA), where PC1, PC2 and PC3 explained 59% of the total variance between the samples, has demonstrated that the capability of discrimination of the combined system is superior to that obtained with the three instruments separately. This improvement is due to the increased information extracted from each sample. Partial Least Squares-Discriminant Analysis (PLS-DA) has allowed separation of the groups in function of olive variety with a root mean square error of prediction (RMSEP) lower than 0.099. Using PLS1 and PLS2 regression models, good correlations have been found between the signals obtained from the electronic tongue and the polyphenolic content (measured by chromatographic methods) or the bitterness index (scored by a panel of experts) with correlation coefficients higher than 0.9 in calibration and validation. These preliminary results indicate that the combination of an e-nose, an e-tongue and an e-eye can be a useful tool for the analysis of olive oil bitterness.