SAN FRANCISCO, CALIFORNIA – OCTOBER 03: (L-R) Mithril Capital Management Founder & Managing General … [+]
Getty Images for TechCrunch
An age-old marketing need is to identify a consumer base, its preferences, and the strength of a product’s message and presentation. In the past, the focus group was a most widely used method, which may or may not have been near perfect, but it sufficed. Today, the latest digital technology may threaten to kill off the focus group for good.
According to DRINKS, its “proprietary tech platform, PAIR,” powers everything from a wine recommendations engine (identifying consumers) to a wine label design function (product presentation). Similar to Netflix’s targeted content recommendations, PAIR platform’s overall data matrix recommends wines to first-time or repeat consumers and helps those working within the three-tier system understand which wines could perform best in different states and across age groups/genders, while it also provides information to use in label development specific to consumers’ aggregate or personal preferences.
Zac Brandenberg, CEO DRINKS
Courtesy DRINKS
DRINKS CEO, Zac Brandenberg says, ”Wine is one retail category that hasn’t yet had its transformative DTC [Direct to Consumer] moment…PAIR enables us to modernize our approach entirely. The technology helps us optimize packaging and label designs, get the right bottle in front of the right consumer, and curate digital experiences that feel authentic to individual consumer preferences. It’s a platform that can be applied more broadly to grocery e-commerce down the line, and we’re excited to be driving it forward.”
The company already services the retailer, Kroger, wholesaler Boxed and the delivery service, Thrive Market. DRINKS also owns the wine delivery platform, wineinsiders.com. The company claims its platform goes beyond typical sampling and/or rule-based recommendations models like “Customers Also Like or Also Bought”. DRINKS says PAIR’s effectiveness is aligned with its ability to offer “evidentiary data in multiple consumer segments.”
According to Brandenberg, PAIR captures data through a combination of human and machine intelligence; then, the platform personalizes product recommendations by comparing consumer attributes (behavior, demographics, geography) with product attributes (branding, packaging, color, vintage, etc.). The company describes PAIR as “a machine learning deep neural network model that uses all the wine attributes in our product graph.”
By plotting the attributes of two separate wines, PAIR provides “observable differences between how wines are perceived by shoppers.” The claim is that by optimizing wine selection in real-time PAIR’s digital technology removes the one-size-fits-all merchandising of retail shops, and even goes so far as customizing the retailer “storefronts and wine recommendations in real-time for each shopper.”
Brandenberg’s example: “One segment might only be interested in us doing all the work for them. If we were to select a case for them, it would likely be a narrower variety of more popular varietals and labels that are neutral and traditional. Alternatively, a customer segment of image or status-seeking consumers may be presented with a wider range of varieties with bolder labels and packaging that looks great on the dinner table or in Instagram posts.”
The company presents a metrics chart to show PAIR’s recommendations effectiveness against random or sample-based recommendations (if you like this, you may like that). The claim is that over a five-week period, PAIR selections effectiveness outpaced random selection/recommendation by more than 60%.
Before DRINKS acquired Wine Insiders, consumer sales were almost all generated through direct mail, which meant the consumer base was heavily weighted toward older and more affluent (Baby Boomers). Wine Insider’s shift to digital almost immediately lowered the company’s average customer age by more than 20 years (GenXrs and early Millennials), which in turn created the need for addressing the Instagram phenomenon with a “younger, hipper, more vibrant and colorful, higher contrast, and with physically larger labels so they would stand out in product thumbnails on a website.”
DRINKS claims that its digital platform predicts with certainty how different sets of consumers will respond to a collection of wines. In addition, the company says PAIR can predict the opposite with certainty: who would be most likely to purchase a specific group of wines?
Certitude in marketing–is that arrogance or quite a concept?
