henhaobio
Sign in

bioRxivBioengineeringPreprint

Open-source benchmarking of dairy and dairy-free products

Aeneas O. KoosisiD, Caroline E. CottoiD, Ellen Kuhl AiD

PDFDOI

Abstract

Dairy-free products offer opportunities to reduce the environmental impact of food. However, they must reproduce the diverse sensory and functional properties of conventional dairy, from foam stabilization in a latte to acid-gel structure in yogurt and melt--stretch behavior in mozzarella. Yet, large-scale, open-source sensory benchmarks across dairy product categories remain scarce. Here we show that the sensory gap between dairy and dairy-free products varies substantially across ten categories: barista milk, butter, cheddar, cream cheese, creamer, ice cream, fluid milk, mozzarella, sour cream, and Greek-style yogurt. We analyze the NECTAR Taste of the Industry 2026 dataset, which includes 1,437 respondents, 3,220 overall-liking ratings, and 79,027 recorded survey responses, including 35,762 product-linked responses. Leading dairy-free beverages, barista milk ($+0.17$), fluid milk ($+0.02$), and creamer ($-0.06$), closely approach their dairy benchmarks in overall liking on a seven-point scale. In contrast, mozzarella shows the largest deficit ($-1.46$), followed by Greek-style yogurt ($-0.68$) and butter ($-0.54$). All three deficits remain statistically significant after correction for multiple comparisons. Across categories, descriptive penalty analysis identifies off-flavor (1.85 points), stale or cardboard notes (1.81 points), and artificial or chemical notes (1.80 points) as the sensory attributes associated with the largest reductions in overall liking. Together, these results establish the first open sensory benchmark for dairy and dairy-free products, identify beverage categories that approach conventional dairy in overall liking, and reveal the sensory attributes and product categories with the greatest opportunities for future innovation.

Citations
0

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Similarity Map
SelectedSame fieldCross-field
THISReed '19Koosis '26Mandel '21Pierre '26Pierre '25Dalcq '19
60% semantic matchNeuroscience
Danielle R. Reed, Joel D. Mainland, Charles J. Arayata·Physiology & Behavior·Jun 12, 2019· 48 citations