Who’s Mining the Store?

7-Eleven stores have always been packed with all kinds of stuff, from the mundane (cigarettes and beer) to the delightfully bizarre (Pina Colada Slurpees?!). But until recently, the stores were bare of IT systems.

By the mid-’90s, in an industry that thrives on local decision making, 7-Eleven had punted many of its stocking decisions to vendors such as Coke and Frito-Lay. These suppliers knew better than headquarters what was selling and what wasn’t, according to James Keyes, 7-Eleven’s president and CEO. “We have lost our way a bit,” he says, using the present tense because he believes the challenge of pinpointing what customers want—and when they want it—is far from over. Since Keyes took over in 2000, he has put a heavy emphasis on IT, not only to improve stocking decisions but also to deliver new services to customers.

Keyes’ eyes were opened to the strategic importance of IT during a trip he made in 1990 to visit 7-Eleven’s licensees in Japan. (At the time, he was an executive with 7-Eleven’s retail gas business.) He was amazed by the way the stores customized their offerings to local demand and by the assortment of fresh foods they offered, from sushi to sandwiches. The Japanese did it with scanning data, rudimentary data warehouses and a nascent in-store ordering system. When he became CEO, Keyes knew that the U.S. stores had to do the same. At a time when 7-Eleven’s traditional niche is being invaded by everyone from drug stores to Wal-Mart, the survival of the business may depend on it. We asked Keyes to explain what he does to get the people in his IT department to think like retailers.

CIO: How has 7-Eleven’s use of technology changed over time?

James Keyes: Back in the ’20s when people started buying refrigerators and stopped buying ice, we were small enough that the people in the stores were able to talk to customers one by one and ask, “You don’t need ice, so what do you want instead?” In the next 50 years, the business became so complex that we lost touch with the customer; we couldn’t just talk to them any more.

[Until 1995], we had virtually no technology in the stores at all. We didn’t know what we sold by store. It was total guesswork. It would take us almost 20 days to collect sales data from the stores because it was all manual. Vendors were stocking our shelves themselves. The first step in 1995 was to automate the basic accounting functions. Then in 1997, we began layering other capabilities on top of that with the intention of being able to put decision making back into the stores.

We discovered that, with technology, we can track every item literally every hour and use that point-of-sale data to drive revenue. In effect, we have come full circle in our ability to respond to the needs of the customer. Because now, we can use technology as a surrogate for being able to talk to every customer who walks in the door. We don’t just have to rely on items we know will sell quickly, like 12-packs of beer; we can actually stock something that we’re taking a chance on. And in short order, we’ll understand how the customer responds.

Before we had much technology in the stores, we used the approach of “space selling.” To build sales and traffic, you’d use big displays and make customers literally trip over the product. But we discovered that you don’t need a whole door full of Coca-Cola Classic for it to sell well, you can sell just as much with one shelf. That’s freed up space to try something new—let’s say Red Bull. We partnered with Red Bull and said, Let’s give it a couple of shelves. When customers went to buy their Coca-Cola, they saw an equivalent display of Red Bull. And pretty quickly, we discovered that using space to drive new opportunities, rather than using it to [push known products harder], allowed us to increase our sales-per-square-foot in the stores.

What convinced you that technology could help generate sales?

This got its start at our stores in Japan. Here in the United States, we had recognized in the ’80s that the future of convenience stores was going to be to sell proprietary products and fresh foods. They are unique, have higher margins and drive more store visits. Our answer at the time was the hot dog grill. Hot dogs were about the only thing that we could manufacture in the store. We were wildly unsuccessful in using [outside providers] or any more efficient means of producing high-quality food.

I went to Japan in 1990, and I saw that there were customers lined up to buy sushi at 7-Eleven stores. And I’m scratching my head, saying, You know, we can’t sell a decent hot dog, and they’re over there selling high-quality sushi; how are they doing that? The answer was this: We had licensed two 7-Elevens in Japan 30 years ago, just as scanning technology became available. They grew up with scanning in the stores. They discovered that they could manage a fresh-food manufacturing capability using that data. They created commissaries, bakeries, places where they could make high-quality, fresh rice products.

They did it by taking the scanning data, converting it into a friendlier format and pushing it back to the store managers. But they stopped short of having the system make the buying decisions for the managers. Most companies assume that data and a good algorithm can make better demand decisions than a human being. But there isn’t an algorithm in the world that can predict what is going to [sell] in the local neighborhood that evening. Only a store manager can do that. Fresh food is now 40 percent of sales in Japan, with only 10 percent [waste].

When I saw what they did in Japan, I became a believer in the power of technology to make every 7-Eleven store better at retailing, and I also became a believer in our ability to satisfy customers’ desire for portable, high-quality fast food.

7-Eleven’s CIO, Keith Morrow, reports to you. How is that relationship structured?

Many companies look at the IT department as a service organization, and they satisfy the needs of their internal clients. We look at the technology department as a key stakeholder in our ability to satisfy the customer, so IT is as important a player as merchandising in their ability to help us respond more quickly to customers. Technology is a key part of our strategy, and our CIO is a key member of our executive committee and involved in our strategic planning process.

Myself and other top executives, including Keith, [get together] on Monday mornings to address both pressing tactical issues and any strategic decisions for the week. Then every Tuesday, we have a meeting with the entire management staff—as many as 2,000 to 3,000 people via videoconference. It is best to communicate with everyone directly so everybody hears the same message [about the business strategy]. Then, all the managers cascade that information down to their staffs.

I also have one-on-one meetings with Keith, but the main opportunity is that we have made him part of the central business team.

How do you decide whether to fund a new technology investment?

Many IT investments are made with a strict reliance on cost reduction [to generate ROI]. We do have that element, but we are using revenue enhancement as the primary return on investment criteria for our technology investments—and holding our people accountable for it. It’s harder to measure, and it’s harder to justify, you know, how much of the sales lift came from the technology versus better training or better product? And so therefore, a lot of companies find it very difficult to justify their technology investment on the revenue side. But we’ve had 33 consecutive quarters of improved same-store sales. It’s no coincidence that our introduction of [the decision support system in the stores] was at about the same time that we began to see those sales gains.

What do you like to see in an IT business case?

I do something unfair. [Laughs] I tell IT that they have to generate cost savings from the project that will cover the costs of implementing the technology. Then, I tell the sales- people that they have to have X percent revenue increases after the project is completed. It gives them a target.

Is there an IT project you wish you’d never funded?

OK, I’ll give you one. In 1997, approximately, we had become very successful in the ATM business. We were also selling almost $5 billion worth of money orders per year, and prepaid phone cards were a big piece of our business. We recognized that all of these services could be delivered electronically, in a self-service environment.

So when I became CEO, we were partnered with NCR to develop a new generation of ATM. We were trying to do things that were very sophisticated at the time—like print a money order, cash a check. No one [at NCR] believed enough in the business model that people would do self-service for other things [besides cash]. So what we ended up doing was taking preexisting hardware and software and jamming it all into one 9-foot monstrosity of a box. We were well ahead of the technology curve, but customers didn’t want to use it.

The good part about it was we learned that we could sell a heck of a lot of check-cashing services, money orders and other things if we could do the kiosk right. It gave us enough confidence to [continue development]. We made it Web-enabled, and we shrunk the box down to 3 feet. We now have a device we call “Vcom” that’s in 1,050 7-Eleven stores—and that goes beyond the ATM functions to offer [online shopping]. I’m hoping to introduce products that are only available through 7-Eleven—or that we have a temporary exclusive on. I can see us offering a new record for download through Vcom for 30 days before it hits the stores, for example. We’re still early in terms of consumer acceptance of a wide range of financial services at an [ATM kiosk]. But it’s very promising.

You said a lack of commitment to the project was the main problem. Explain what you mean by that.

It goes to the question of who really understands the customer. NCR said, “We’ve been making ATMs since ATMs were around. We know what the customer wants in an ATM.” So rather than looking at this as an R&D effort, they were focusing their R&D energy in other areas.

So this was not an issue of internal commitment, then?

No. One of the advantages of being chief executive is I can get all the internal commitment that I want. [Laughs] The mistake I made was that I left our IT group out of the project for too long. I let the merchandising department run the project at the beginning because they were behind the product concept. But if I had it to do over again, I would have brought some merchandising people into IT and have the project originate from IT.

If you could change one thing about IT at 7-Eleven today, what would it be?

Here’s the ongoing challenge—it’s the same one I give my merchandisers: The opportunities for change are never ending. The IT department—just like the merchandise department—tends to get comfortable with the project that they have on the table. Achieving the right balance between finishing what we’ve got versus keeping an eye on new and better technology is the challenge.

Right now, RFID is coming down the road. There’s a tendency for IT groups to look at something like RFID in the [obvious] sense: Oh, that’s a supply chain opportunity; we can track inventory. Well, there’s also another element of RFID that makes it much more able to serve the customer. Imagine every food item with RFID technology on the label so that as you bring it to the microwave or convection oven, the product tells the oven how long to cook. These are things that you can do with RFID to be able to satisfy the customer. So getting our technology folks to think like retailers so that they’re looking at emerging technologies in the 7-Eleven way—to better satisfy the customer—that’s probably the greatest challenge.

How have you told them to address that challenge?

We are making them part of the retailing team. In the ideal world, the IT group will not be seen as an IT group, they will be seen as intermixed within our organization so that it’s hard to tell the IT person from the retailer. Now that’s—like I said—in the ideal world. [Laughs] We’re evolving toward that.

The way I’m trying to do it is by rotating people from the business into IT. We tried doing it the other way, with IT people moving into the business, but it didn’t work. Inevitably, the business units wanted to go off and start their own projects, and we would wind up with two different IT departments essentially competing with each other. It works much better to bring the businesspeople into IT and then send them back. It leads to better integration between the two groups.

[When I came in, IT was doing its traditional work] automating back-office functions like accounting. Now we’re stretching them to improve decision making at the store level. The final frontier is to take them into satisfying the end customer. We want them to create systems like Vcom that drive revenue with customers. Getting there will take time. But it’s coming.