Wine recommendations from our proprietary algorithm beats sommelier recommendations by 12.3%

As part of our NSERC Engage research project on wine recommendations, we conducted a series of experiments in collaboration with Dr. Cooperstock and his team where we invited participants to blind taste wines paired with food.  Each participant sampled wines recommended by our proprietary algorithm and by a sommelier with WSET Level 3 Award in Wines qualification.  Participants rated wine recommendations from our proprietary algorithm 12.3% better than the wines recommended from a sommelier.

To the best of our knowledge, there are currently no wine and food pairing algorithms to facilitate the online wine purchase experience. Retailers can potentially benefit by applying our proprietary algorithm to recommend customized wines pairings.

A research study, from Cornell University and University of Houston, shows that wine and food pairing recommendations can increase sales of wine by as much as 44.5%. In addition, 35% of purchases on Amazon are due to algorithm product recommendations.

Our algorithm is currently patent pending.