         |
 |
 |
     |
|
Retail Systems Alert recently spoke with Marshall Fisher, UPS professor of operations and information management at the University of Pennsylvania’s Wharton School. Fisher is a cofounder of the Consortium for Operational Excellence in Retailing (COER) with Harvard Business School professor Ananth Raman. Fisher and Raman are also cofounders of 4R Systems, a company that develops supply chain planning software for short lifecycle product management. Retail Systems Alert: How did COER originate? Fisher: “Professor Raman and I started the three-year research project in early 1997 and we recruited 32 retailers to work with us. The Sloan Foundation-funded research project was supposed to end in 1999. Walter Salmon [professor emeritus, Harvard Business School] helped us a lot in securing retailers and being our mentor. He said, ‘Nobody knows what the frontier is in terms of best practices in forecasting and supply chain, let alone who is on the forefront.’ He viewed the project as taking stock of the state of the art in retail supply chain planning. Our mantra became a quote from one retailer: ‘We are awash in data and starved for information.’ COER worked with the retailers to survey what each was doing in their approach to forecasting and supply chain planning, to identify who was good at various aspects of planning, and to share these best practices within the group. We have also conducted projects with some of these retailers to develop new tools for choosing which stores are best for product tests, for making initial buys and generally making supply decisions over the life of a product to maximize life cycle profitability.” Retail Systems Alert: What are COER’s challenges today? Fisher: “Today, five years later, through our research and other academics’ research, as well as through 4R and other companies, there are lots of analytic tools. So, while math might have been the bottle-neck five years ago, it’s not the bottle-neck today. We have developed the framework – planning success is the result of good data, good math, and good execution. It seems that good data and good execution are the biggest challenges today.” Retail Systems Alert: Could you define those, please? Fisher: “Good data is what’s going on, what we have sold and not sold. Good math is what we shall buy and ship to the stores, and good execution is making this happen to 500 stores with 50,000 SKUs routinely.”
Fisher: “We have found that the quality of store/SKU-level inventory data is very low. Raman and his students have done a lot of research in this area. [See Ananth Raman, Nicole DeHoratius & Zeynep Ton, “Execution: The Missing Link in Retail Operations,” California Management Review 2001 for some of the results of this research.] Roughly, what we have seen is that if the computer record says that an item is in stock 95 percent of the time, customers indicate that they see it in stock only about 75 percent of the time. What is interesting today with the Internet, is that many retailers are making their inventory records available to customers. Customers can order appliances over the Internet and pick them up in a store. Well, as retailers have begun to offer consumers access to their inventory records, it has revealed the state of accuracy of that data. Typically, twenty percent of the time, when the customer checks on the product availability and it says it’s in stock, they can’t find it in the store. Of course, this can happen for all kinds of reasons, scanning errors, exchanges not recorded correctly, shipment errors or misplaced products in the store.” Retail Systems Alert: What challenges exist on the execution site? Fisher: “If you start with good data, a key component of which is sales and inventory data, and then good math, which is how to turn that data into forecasts and buying decisions, good execution then should result in products being shipped to a store and onto the shelves in a timely manner according to the plan. But this often happens very haphazardly. Products get left in the backroom or get put in the wrong place, and stock outs at vendors delay receipt. We worked with a lot of retailers to design and conduct product tests where we put a product in 20 stores to see how it would sell. A huge source of error and variation in those tests occurred because different stores displayed the products with different degrees of prominence. And it’s very hard for the corporate office to control what happens in stores in terms of how a product is displayed. In e-commerce they talk of ‘the last mile,’ the difficulty of getting a product delivered to a customer’s house. Well, in a store it’s ‘the last 100 yards.’ That means getting products from the loading dock to the shelf. And this is particularly difficult with new products.” Retail Systems Alert: Are there any success stories out there? Fisher: “It always seems to me that retailers could learn a lot from the automobile industry. If you look at an automobile plant with its assembly line, it is largely an exercise of getting the right part, to the right worker, at the right time. Toyota, who has been very successful, believes that much of its success rests on the ability to control and execute parts logistics better than other companies. They found that if they delivered fewer parts more frequently, and on time, the result was a more reliable system than if they held large qualities on inventory all over the plant. The same thing happens for retailers. Often a lot of things get lost in the back room of the store.” Retail Systems Alert: How could retailers learn from the previous example? Fisher: Well, if you play out the Toyota solution, it would lead you to perhaps get rid of the back room. Actually, Belgian shoe retailer Brantano, who spoke November 5 at our more recent COER conference, got rid of its backroom and now keeps all inventory in-store. They stock one pair of everything for every style and every shoe and every size, and at the end of the day, they simply re-order where there are holes. They claim that they can get products from the DC to the shelf faster than from the back room to the shelf. Retail
Systems Alert: Is this a common solution? By shipping in smaller quantities and producing in smaller quantities, you would appear to have a less economical system. But Toyota discovered that by using the flow system – where you ship today what you are going to sell tomorrow – the benefits of more accurate execution outweighed the apparently higher production costs. I think the ability to know what is going on in the store and accurately executing is a huge challenge for retailers today.” Retail Systems Alert: In what other ways could retailers solve their in-store problems? Fisher: “The other thing Toyota did was involve their workers. In our work, we have seen that product placement in a store really determines its sales. This is an area that store people could manage more because they are right there on the floor. Most retailers view the store employee payroll as something to be minimized. Perhaps this is something retailers have to rethink.”
|
