The data-driven coffee – analyzing Starbucks’ data strategy

Posted by

We often go through life without pausing to think about the behind-the-scenes of events or things that are part of our daily lives. While ordering yet another “Tall Latte, please”, I remembered reading that Starbucks had been investing in being data-driven. But did I know how? Did I know the story of that Latte? We take such everyday things for granted, without ever realizing that what might be “just a cup of coffee” for us, in reality, it’s a cup that carries in its biography many decisions and strategies unbeknownst to us. So I came home, and started to dig into the details of that tech-infused cup of coffee. The idea behind this post is to share those insights with you so that you can also get inspired to make data-driven decisions.

While the coffee beans are used to brew addictive coffees, Starbucks uses the “data beans” primarily for these purposes: Expansion, Personalization and Innovation.

Expansion

Learning from failures is essential. In 2008, when Starbucks’ CEO, Howard Schultz, was faced with the tough decision of closing hundreds of stores, he insisted that they take a more analytical approach going forward — choosing the right location is essential for success of businesses like Starbucks. So today the company uses a smart system called Atlas, developed by a location-analytics company called Esri. This data driven approach takes into account features like population density, average incomes, and traffic patterns to estimate the profitability and economic viability of the new store. Also, it predicts impact to already existing Starbucks in the area — compete with competitors and don’t bring your own brand down. This means that now the decisions are driven by data and not just human opinions and gut feelings. Of course, this does not imply that humans are completely out of the loop. At least not yet!

Personalization

With 90 million transactions a week in 27,000 stores worldwide, Starbucks is on the right footing for making marketing and business decisions based on data. But where is this data exactly collected from?

Starbucks launched a rewards program and a mobile app so that they could get “users in and data out”. Data collected via the reward programs (more than 13 million active users) and the mobile app (more than 17 million active users) allows the coffee giant to learn about exact user behaviors — “Do you like to indulge more on a rainy day? Do you like muffins over croissants with your pumpkin flavored tea? Do you order a different type of coffee on workdays than on the weekend?”

Here are some of the ways in which the data collected is being used for creating personalized experiences.

No matter which Starbucks store in the world you visit, your data collected via the app allows the smart system at the new store to know about your order preferences — your usual drink for “that kinda day”. Also, Starbucks has been trying to harness the power of AI to provide recommendations. They have enabled a program called Digital Flywheel for the rewards members’ accounts. Deciding what to eat is not always an easy decision, so AI is being used to lure customers into buying more by providing suggestions based on factors like purchase history, weather, time of the day and birthdays. The same AI technology is also being used to send consumers personalized offers and to re-engage them in case it had been a while since they had last given their bucks to Starbucks.

Starbucks is trying to maximize their AI adoption by deploying voice and text recognition AI. My Starbucks Barista is a virtual barista in the mobile app for easily placing orders via voice or text messages as reduced friction in placing orders means potentially more sales.

Today, U.S. customers use the mobile app to pay for 34% of orders. 12% of orders are placed in the app rather than at the counter; at 3,600 locations, that figure is 20% or above during peak business hours. This translates into a significant data footprint and reflects the importance of investing in customer-facing technologies. I believe the mobile app and their rewards program are a key to their continued success in the future. As the amount of data keeps accumulating, AI recommendations will keep getting better and better, and satisfied customers will keep coming back.

One important thing to be noted here is that adding a mobile ordering service is more than just creating a mobile app. The entire workflow needs to be reexamined and adjusted to complement such new technologies. Starbucks had to add new roles and resources to prevent customers (who had ordered on their app) walk away because there was a “mobile queue” at the store upon arrival — we are living in the era of instant gratification.

Innovation

Product offerings

When Starbucks decided to expand their business and serve their products from grocery stores, they let the data do the talking in deciding the products to offer.

For example, the analysis that about 43% of tea drinkers don’t add sugar in their tea and about 25% don’t add milk to their iced coffee when drinking the beverage at home, led Starbucks to create two unsweetened ice tea K-Cups — Mango Green Iced Tea and Peachy Black Tea. Pumpkin spice caffe latte and iced coffee without milk or added flavors are some other fruits of such data crunching.

From offering alcohols in certain locations to coming up with special Frappuccino promotions to take advantage of heat waves, Starbucks is making business decisions led by data.

Starbucks of the Future

Starbucks_Reserve_SODO_(27)

The coffee giant has recently upped its game. Starbucks Reserve SODO (Seattle) is a showcase of  “Starbucks of the future”. Not only Reserve SODO’s concept is theatrical from a design perspective, but it’s also driven by IOT and cloud services.

In 2018, Starbucks has further strengthened and deepened its relationship with Microsoft. It has started using more than 50 Azure cloud services — along with Power BI analytics — and it also works closely with Microsoft on planning new initiatives.

A cutting-edge piece of machinery used at the futuristic store is the smart Clover X coffee-maker. The smart coffee maker is an example of how all the equipment in the future-stores will be smart and connected to the cloud for data collection and remote maintenance. While today employees need to record temperatures of food storage cases manually, in the coming years it will be all automated via sensors.

It is important that the almond croissant in Oslo tastes the same as in New York. To ensure this, the warming ovens in the stores are programmed to heat items for a specific amount of time, as instructed by Starbucks’ food-quality experts. Today whenever the menu evolves, the updates for these ovens are shipped to all the stores via USB drives. This is a painful  and slow process — not so futuristic. Here Azure Sphere is going to come to aid so that the “Starbucks of future” can be powered by internet-of-things — commercial devices that will be connected to internet, but with some local computational capabilities so that they can keep chugging even in case of connectivity loss (Microsoft’s mantra of Intelligent Edge).

download

 

End Notes

I hope you enjoyed learning about Starbucks’ data strategy. Next time you sip your coffee or savor your delicacies, stop for a moment and think about the intricate smart show running behind-the-scenes!

                                                                           .  .  .

Thanks for reading! If you want to be notified when I write something new, you can follow this blog. 🙂

 

References

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s