This study extends the classical hedonic price model by taking advantage of the heterogeneity of consumer preferences and testing the existence of different price functions. A finite mixture model identifies consumer segments in the Italian wine market and estimates the hedonic functions within each group. The analysis is extended beyond the mean to the tails of the distribution for each group via quantile regressions. The analysis in classes proves to be more efficient both on average and at the quantiles and presents a more accurate wine market segmentation, unveiling sources of heterogeneity and asymmetry.