Tesco’s New Strategy to Improve Customer Satisfaction Hinges on Innovative BI Tools
By Joe Mullich
Business intelligence has historically leveraged structured data from relational databases-that is, rows and columns of numbers. However, huge amounts of insight can be found in unstructured data. This data resides in, among other sources, email messages, reports, press releases, blogs, online discussions, surveys, Web pages, maps, notes from call centers, and other day-to-day mechanisms of information sharing. Until recently, such unstructured data has been difficult for analytics tools to incorporate.
With the growing importance of this data, new techniques recently emerged to integrate the typically siloed and distributed unstructured data into BI systems. These techniques unlock this data’s power to improve decision-making, customer service, and up-sell and cross-sell potential.
Improving customer satisfaction-and the overall customer experience-is one of the top three priorities for half of the companies whose executives responded to a December 2010 survey by Bloomberg Businessweek Research Services. In fact, almost three out of four of the 307 executives who responded said that improving customer experience is among their top five priorities for 2011.
A leader in this emerging area is Tesco PLC. The fourth largest retailer in the world, Tesco has more than 5,000 stores worldwide, about half of which are in the United Kingdom. (It also owns the Fresh & Easy grocery chain in the United States.) In fiscal 2009-2010, Tesco tallied £62.5 billion in sales of food, clothing, electronics, and other items and services in its wide-ranging portfolio.
“Customer data is in the fabric of every decision we make in the company,” says Mike McNamara, Tesco’s CIO. With its 16-million-member loyalty card program, every single basket transaction tells the company something about its shoppers. “We can probably tell a recession is coming before the government can,” he says.
Leveraging its vast databases, Tesco does four coupon mailings a year to its loyalty cardholders. These coupons are split equally between rewards (getting customers to return and purchase an item they bought before) and incentives (getting customers to sample something new). “If we need to do a price cut of 100 million pounds across the business and want to target a specific customer demographic, we know what products will have the most impact,” McNamara says.
Given the importance of information to every aspect of its operations, Tesco has come to look at data in several different ways. In McNamara’s view, information has three dimensions-timeliness, accuracy, and granularity. “If you are doing flash reporting, the information needs to be timely, but just accurate enough,” he says. “The detailed stuff that the operations analysts look at needs to be very, very accurate, but not so timely.”
Information, McNamara says, must be considered across these different dimensions. “It’s also important to recognize that information comes in structured and unstructured form,” he says.
If Tesco is developing a new product, such as a new set of crockery, the designers in the United Kingdom will use a wide range of information when collaborating with sourcers in China. The patterns will spring from “snippets of things people have seen in magazines, photos, and video clips,” McNamara says. “Everything [comes] together to create a new product, and it’s information of many types, not just numbers.”
Finding ways to share this unstructured data is crucial not only to efficiency but also to top-line and bottom-line growth. For example, Tesco used unstructured data to improve customer satisfaction and retention. In its 2,000 stores in the United Kingdom, Tesco rolled out technology that allows customers to provide instant in-store feedback on their shopping experience in a number of ways: They can fill out a card or use their mobile devices to leave a voice message, send a text, or send an email.
The speed with which Tesco turns this unstructured data into action speaks to the heart of “the new competitive battleground” of the retail industry-“shopper centricity,” says Pierre Lever, managing director at Planet Retail, an analyst firm that covers the retail industry.
For example, the voice mail that a customer leaves while in a Tesco store is instantly translated by language processing software. “A disgruntled customer gets an instant response to try to save their business,” Lever says.
Using business intelligence software, all of that unstructured data is turned into a structured format that can be fed back to the stores, thereby creating continuous improvement. Lever says this capability allows Tesco to move beyond the standard practice of focusing on unique customers and instead think in terms of the unique customer experience, which may be even more important.
By instantly processing unstructured data, Tesco finds out what customers “think when they’re interacting with the store, in the store, and with the staff,” Lever says. “The half life on experience is very short. If you can capture that close to the experience, you can get some really interesting insight” that is more relevant than customer profiles or demographics.
Finding new ways to understand and derive actionable steps from these different types of data is crucial to the Tesco’s ongoing success. “We have an extraordinary amount of data we don’t even use,” McNamara says. “We try to process as much as we can-and use as much as we can-but we end up discarding 98 percent of it. If we can find better ways to turn [that data into action], we’d all be winners.”
He remembers being in an airport lounge and seeing business travelers with laptops staring at spreadsheets of information coded in green, yellow, and red. “There has to be better ways of representing the information and bringing it to life,” McNamara says. “If we can find those ways, we can start using that 98 percent of information we now chuck away.”
For more information insights about Tesco’s strategy, watch this video:
Click here To stay up-to-date with the latest conversations around data and analytics as a member of the BITV Linkedin Group.
Click here To answer the latest BI TV Poll “Which One of the Following Describes Your Companies Use of Analytics?