The way Starbucks uses data for competitive advantage is instructive for all businesses, including SMEs. Here are some reasons to look beyond the coffee next time you see the ubiquitous green logo.

Loyalty Cards Are Customer Data Devices

Loyalty cards at supermarkets make sense because of the huge variety of possible products customers could buy, and the importance of discounting. In sectors where pricing is critical, loyalty card data is crucial to getting promotions right. But in a coffee shop, most people buy the same drink in relatively small quantities, so the value of loyalty card data doesn't seem quite so obvious. The same applies to many other small businesses with different types of product or service, but a similar type of relationship with customers.

Nevertheless, Starbucks pioneered digital loyalty cards in coffee shops, first in a credit card style format, then as a mobile phone app. By linking the loyalty card to phones, it was able to improve its relationship with customers, especially in terms of convenience. Equally important was the data it captured about customers.

Starbucks now has details of what customers buy, when and where they buy it, and even whether they eat in or take away. This allows the company to offer rewards geared to a customer's needs, and offer them new products and services they might like. It can also personalise more conventional promotions such as offering cold drinks on hot days,messaging customers' phones at different times of day based on what's likely to suit.

If you're still stamping your paper loyalty cards and giving away free fifth or tenth drinks - or the equivalent in your business - you might want to think about what you'd gain by knowing more about your customers, and being able to keep in touch with them when they're not in your store.

Data-inspired Products and Menus                   

Personalised promotions are important to Starbucks, but an equally valuable use of customer data is developing new products and menus.

When you're serving hundreds of millions of drinks, cakes and sandwiches, it can be very expensive to find out customers don't like a new product. So customer data is a source of ideas for new Starbucks products. For example, data about non-dairy milk usage across stores might suggest new variants of drinks to launch, and could also lead to separate menus for customers with allergies or food intolerances.

It's also possible to turn getting customer feedback into a customer benefit, for example by offering exclusive events to loyal or targeted customers, where they can try out new recipes ahead of others. Again, any business could do this, but data about customers and products makes this kind of idea easier and cheaper to put into practice, with greater, more systematic business benefits.

Few SMEs have the sophistication or scale of Starbuck when it comes to product development, but some of their concepts can be repurposed for a smaller environment. One way of doing this is to explore how you could use data to inform and validate product hunches and inspirations. 

Data to Keep Costs Down

One of Starbuck's biggest outgoings is the cost of real estate. Before opening a new branch, it uses sophisticated data models and analysis - including artificial intelligence - to predict likely customer levels and forecast revenues. Most SMEs don't have access to this level of data, nor the tools to use it even if they did. However, the principles Starbucks applies to model the possible future performance of a store can be applied by any entrepreneur with a laptop. The starting point is the same straightforward spreadsheet of revenue sources and costs most business owners are familiar with. What Starbucks reminds us is how to use data to create different models and scenarios for different situations, and how far the technology to help us do that has come. The underlying lesson is that the better we understand and use all the available data about our business, the more effectively we can plan and manage anything the future holds.

Another area where Starbucks uses data to keep down costs is equipment like coffee machines and ovens. These are connected to the internet, and send data to central locations for monitoring and analysis. Some maintenance and repairs can be delivered over the internet - such as new machine settings. Others can be predicted and addressed before machines break down, minimising store disruption. Most of this monitoring and analysis is done automatically by computers and AI.

The very smallest businesses may have limited opportunity to apply either of these lessons, but if you're already slightly larger, or expect to grow, look for opportunities to use data to model costs better than you might otherwise do, and see if data technologies can help reduce costs of modern, internet-connected equipment.

Few small business can grow or succeed today without using and understanding data.

Starbucks demonstrates some of the most sophisticated ways of doing this, although many of these are not relevant for SMEs and entrepreneurs. However, some of what they do with data provides valuable lessons and reminders for businesses of any size. One of the most important is to treat data with as much care, respect and skill as you would any valuable business resource or asset.

Was Rahman is an expert in the ethics of artificial intelligence, the CEO of AI Prescience and the author of AI and Machine Learning. See more at www.wasrahman.com