Microsoft teams up with Starbucks on predictive drive-thru ordering and bean-to-cup blockchain

Microsoft teams up with Starbucks on predictive drive-thru ordering and bean-to-cup blockchain

4:41pm, 6th May, 2019
Microsoft CEO Satya Nadella discusses the company’s work with Starbucks. (GeekWire Photo / Kevin Lisota) Starbucks is as a coffee company, but that hasn’t stopped its leadership from turning to a familiar Seattle ally to up its digital bonafides. Today at Microsoft’s Build developers conference, the two companies ran down a series of initiatives they are working on together. The new projects include everything from using machine learning to gather customer preferences, to connecting coffeemakers in stores to the internet to blockchain services for tracing coffee. Microsoft and Starbucks have , and they have an established connection at the highest levels. Starbucks CEO Kevin Johnson is a former Microsoft executive who ran the worldwide sales and the Windows divisions before In introducing the latest fruits of the alliance, Microsoft CEO Satya Nadella applauded the collaboration between Starbucks’ software engineers and its business side. Nadella cited Starbucks as one of many examples of a consistent theme: increased hiring of software engineers at companies outside the industry that are embracing high tech solutions. “They are coming together to completely take what is that iconic experience that is Starbucks and incorporating digital throughout,” Nadella said. Starbucks CEO Kevin Johnson(Starbucks Photo) Here are a few of the innovations the two companies are working on: Predictive drive-thru ordering: The Starbucks app already gives plenty of recommendations based order history. The company is extending that technology to the drive-thru, with digital boards that will make order recommendations. Starbucks’ drive-thru technology, which is being tested in the company’s Tryer Center innovation hub in Seattle, doesn’t have the same luxury as the app’s detailed customer order history. But it will rely on store transaction histories and more than 400 other store-level criteria, such as inventory, time of day and weather. Will this technology someday lead to facial recognition software or license plate readers on the order boards that will be able to see customers and their cars when they pull up and recommend their favorite order? It’s unclear, but Microsoft did say that customers will eventually be able to opt in to more personalized drive-thru recommendations. Connected coffee-makers: Starbucks is using several Azure cloud products to connect and secure the more than a dozen pieces of equipment in each of Starbucks 30,000 stores, from coffee machines to grinders and blenders. Starbucks’ IoT-enabled machines collect more than a dozen data points for every shot of espresso pulled, including type of beans, the temperature of a cup of coffee and water quality. Starbucks hopes that connecting its devices to the cloud will make it easier to be proactive about maintenance instead of reactive. The machines 16 hours a day, every day, at high volume, and glitches or breakdowns can cripple stores’ operations. Connecting all of its machines to the cloud allows Starbucks to more rapidly update its menu. Before, Starbucks would have to send thousands of USB drives and upload them manually, a process that would take weeks. Now, the company just sends out updated recipes through the cloud. Bean to cup blockchain: Microsoft just announced its new last week, and the tech giant is already putting it to good use in the partnership with Starbucks. Microsoft is working with the coffee giant Starbucks teased last year that is getting closer to reality. The idea is to ultimately connect coffee drinkers with coffee farmers, who can potentially then take advantage of new financial opportunities. And customers will be able to use the Starbucks mobile app to trace the journey of their coffee from the farm where the beans originate all the way to their cup. Starbucks said it worked with more than 380,000 coffee bean farms last year. The company promised to open source the pilot program and share what it learns.