These online quizzes that match you with your ideal wine are worse than horoscopes

Maybe you joined an online wine club at some point and were asked to take a “palate quiz” – a quiz that asks you if you drink your coffee black or with cream. , whether you are attracted to dark chocolate or milk chocolate, etc. a way to predict whether you would like to drink Merlot or Sauvignon Blanc. These sorting mechanisms have become a popular tool in the sphere of wine e-commerce, mainly aimed at beginners. It’s a good idea: a non-intimidating way to help people figure out which wines they might like.

I just wish they actually worked.

The palate quizzes seem grounded in the idea that everyone has an ideal taste they seek – that I, a bitter black coffee drinker and sharp dark chocolate eater, would want to drink bitter-tasting wines. But it just doesn’t match the way I or anyone else I know eats or drinks. We are all inconsistent, liking certain sweet foods and despising others; some of us are quite omnivorous, with an appetite for everything. A person may like to butter a piece of toast, but not a toasty, buttery Chardonnay.

Wine clubs like Winc and Bright Cellars have been giving customers palate quizzes for years. Now, a California startup called Tastry has launched technology that promises to be more advanced, using artificial intelligence to map a wine’s flavor “matrix” and tailor it to a person’s palate.

“I had a hypothesis that you could understand the human palate by creating new technology designed not to look at the quality of wine, but to understand how consumers perceive wine,” Tastry CEO Katerina told me. Axelsson, during a telephone conversation. earlier this summer.

His company analyzes individual wines down to the molecular level, then compares them with other wines in his database. This data-driven approach should theoretically eliminate the errors of human subjectivity: two different people can describe the same wine in completely different ways, but the molecules don’t lie.

When you answer Tastry’s palate quiz – available on its app, BottleBird – the system predicts which wines will suit your particular palate. These recommendations are “80-93% accurate,” Axelsson said, meaning most of the time people say they really like the wine suggestions. As a person continues to use the app and offers more data points, she says, that accuracy rate increases.

I downloaded the app and answered the survey. He presented me with a series of flavors and aromas and asked me if I was pro, neutral or anti. I answered honestly (I’m pro-black olive, anti-sour candy, neutral on cigar tobacco smell) and waited impatiently for the app to calculate my results.

The first wine listed, matching 99% my idiosyncratic palate, was Barefoot Sweet Red ($5.82). My heart sank. It’s not a wine I would like to buy.

Part of the reason I don’t want to buy the Barefoot Sweet Red is the sweetness. It’s not that I categorically hate sweet wines: I like a Riesling Auslese, a semi-dry Vouvray, a fortified Madeira. But I’ve tasted a lot of wines under $10 that have “sweet” on their labels over the years, and they’re usually made to taste jammy, overripe, and sweet. That seems like a nuanced distinction that might be difficult for an app to navigate.

Sweetness isn’t the only reason I don’t want to buy the Barefoot Sweet Red. We all buy wines for many reasons other than just the taste: we love the label, we love the story of the winemaker, we’ve been to where the grapes were grown, the company matches our personal values. Barefoot, owned by Gallo, the nation’s largest wine conglomerate, may not tick those boxes for some people.

Maybe a better way to put it is this: people want to drink wines that feel cool to them. I don’t know how an algorithm can solve this.

There are other uses for a technology like Tastry’s besides an application for customers. Some top wine companies rely on data collected by Tastry to figure out how to blend their wines into a concoction with widespread appeal. (Tastry doesn’t release individual user data, Axelsson said, but it creates heatmaps that show which flavor combinations are most popular in certain geographies.) Offering its technology to wineries and retail is the company’s primary revenue channel; he says it has been used by 150 wineries or stores so far.

Even more exciting is the possibility of this technology being used to lessen the effects of the smell of smoke, which Tastry announced last week that it will begin offering. Wineries still do not fully understand how wildfire smoke affects the taste of a wine, and there are few remedies available. Tastry’s analysis of a wine could provide a deeper insight into how smoky-smelling compounds interact with other flavor compounds. It certainly seems worth it.

I love the concept of a palate quiz – it’s fun and might make wine more accessible to newcomers. But every time I pick one up, indicating my preference for M&Ms over Skittles, I feel like transporting myself from my computer to a wine store, where I can ask a real human being: what do you recommend?

Melvin G. Rodriguez