Palate-pleasing Wines Stem from IoT and AI Solutions

Artificial intelligence is helping retailers sell wine. New technologies demystify the wine-purchasing experience, giving customers detailed product information and personalized recommendations at the point of purchase.


Article Key

Image credit: Koscar

When the pandemic hit in March 2020, our buying patterns changed. People began panic-buying toilet paper, meat, bottled water—and alcoholic beverages. Wine purchases alone jumped 60 percent initially, but now wine sales are on par with pre-pandemic levels. With tasting rooms closed and social gatherings limited, wine drinkers can use some help figuring out which wines wet their whistles.

New technologies are helping to inform customers about the wines they choose. California-based Tastry is using artificial intelligence to analyze the flavor profile of wines to inform customers and producers about the liquid in their bottles. On the retail end, Koscar has created an intelligent shelf that provides information about the wines for sale.

Better Wine Through Chemistry and AI

Selecting a bottle of wine can be an intimidating experience. Multiple factors influence the taste, such as country or region of origin, varietal, producer, and alcohol level. Even forest fires can impact the flavor of California wines. While taste is subjective, knowing what consumers want can help drive sales for winemakers.

Tastry’s solution analyzes the chemical compounds of wines and feeds that data into the AI engine. The algorithms assess the ratios of each of the compounds in the sample and how they interact with each other to determine a flavor profile. It compares the wine’s flavor profile, mouthfeel, aroma, and color to other wines in the Tastry database and predicts how customers will respond to that wine.

Winemakers can pay Tastry to analyze their wines; in return they get access to a dashboard that predicts how customers will view their wine in a given store or region. Tastry’s “computational blending” technology helps winemakers adjust flavor profiles to make the wine more appealing to consumers, which can increase the wine’s retail value and drive profits.

Stats about wine chemistry

Image credit: Tastry

Tastry tapped its database to develop BottleBird, a wine-recommendation app. Retailers can install the app on tablets alongside shelves of wine to inform buyers. Consumers take a short quiz to establish their flavor preferences, and the app matches wines to their personal tastes. Users can also download the app to a smartphone. Tastry claims that stores using the BottleBird app report a 20 percent increase in sales.

Top Shelf Solution

Another retail solution comes from Koscar, which has created an AI-based smart shelf, powered by Intel® Core™ processors. The two-shelf system has additional wine storage below and a 43-inch LED touchscreen above.

Integrated 3D sensors detect when a bottle of wine is picked up from the shelf, and a corresponding image pops up on the screen. The screen displays an image of the selected wine, along with the price, and notes about its taste, vintage, and production. It also displays a QR code that users can scan to get more information or purchase the win.

Alternatively, users can select a wine by tapping its on-screen image. Multiple floating balls on the screen also contain images of wine bottles. When a customer taps a ball, information about that product is displayed on the screen. Buyers also can scroll or swipe through screen images until they find one they like.

More than 80 percent of consumers purchase wine based on the appearance of the label. That’s great for graphic designers, but it’s not a reliable gauge of taste. With AI, customers can learn more about the wines they select at the point of purchase, boosting retail profits.

Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries.