
Combining trained sensory panels with analytical chemistry to build comprehensive flavor maps that guide recipe iteration and quality control.
Flavor is ultimately a human experience, and no amount of analytical chemistry can replace trained sensory evaluation. At the same time, sensory panels alone can't identify the specific compounds driving perception. The most powerful approach combines both.
Trained panel scores for a multi-strain Baco Noir vs standard single-strain control.
Our sensory analysis program pairs trained tasting panels with GC-MS (gas chromatography–mass spectrometry) and HPLC (high-performance liquid chromatography) to build comprehensive flavor maps for every recipe iteration.
These maps correlate specific metabolite concentrations with sensory descriptors — linking, for example, a particular ester concentration to perceived fruitiness, or a phenolic compound to perceived body and mouthfeel. Over time, this data feeds back into our AI models, improving predictive accuracy.

For our partners, this means every recipe comes with a full sensory profile and the analytical data to reproduce it consistently at scale. Quality control becomes data-driven rather than subjective.