These days, it seems that both traditional bricks-and-mortar retailers and online "e-tailers" are drowning in a sea of customer information – knocked by wave after wave of data from online transactions, point-of-service scanners, membership programs and even sensor chips on shopping carts. The question is, with all this sophisticated technology on hand, why have department store markdowns over the last 20 years grown from 8% to a whopping 33% of sales? Why do lost sales today often exceed 10% of revenue? And why is it that customers are still going away dissatisfied because what they want isn’t in stock? In an article entitled "Rocket Science Retailing Is Almost Here – Are You Ready?" in the July-August 2000 issue of the Harvard Business Review, Wharton operations professor
These days, it seems that both traditional bricks-and-mortar retailers and online "e-tailers" are drowning in a sea of customer information – knocked by wave after wave of data from online transactions, point-of-service scanners, membership programs and even sensor chips on shopping carts. The question is, with all this sophisticated technology on hand, why have department store markdowns over the last 20 years grown from 8% to a whopping 33% of sales? Why do lost sales today often exceed 10% of revenue? And why is it that customers are still going away dissatisfied because what they want isn’t in stock?
In an article entitled "Rocket Science Retailing Is Almost Here – Are You Ready?" in the July-August 2000 issue of the Harvard Business Review, Wharton operations professorMarshall Fisher and two co-authors take a hard look at why retailers today may be, as one of their study participants put it, "awash in data but starved for information," and what they can do about it.
Fisher is blunt about the problem: "The big challenge for retailers is achieving what they call the four ‘rights’ – the right product in the right place at the right time and at the right price – and they really are failing miserably." He estimates the costs associated with mismatches between supply and demand, for example, as "about 10% of revenue, in an industry where profit is about 2% to 3% …"
Today’s large retailers, Fisher notes, have powerful information systems that can store customer data in giant warehouses. Some companies keep data for a few weeks, others for a number of years. In both cases, however, many businesses don’t know how to use the information they collect and often don’t make it available to their merchants.
Fisher and his co-authors – professor Ananth Raman of the Harvard Business School and Anna Sheen McClelland, a former research associate at Wharton – conducted a multi-year study in which they worked with 32 leading retailers, including Borders, Bulgari, Nine West, Staples and Tiffany & Co., among others, to find out how they use information technology to understand their customers. The researchers deliberately chose retailers of short life-cycle, innovative products such as fashion apparel, shoes, toys, books, music and consumer electronics, speculating that the unpredictable demand for these products would make them the hardest cases. They surveyed and visited each retailer, working with some of them to improve their capabilities.
While most of the retailers in the study wanted very much to make better use of sales data, they possessed little expertise and few tools to do it. Another factor is at work here also, says Fisher: Some retailers have such long supply chains that even if they discover a particular item has strong early sales, they can’t restock it within the season. "If you look at apparel, it’s the story of chasing low cost through Asia, resulting in long lead times. Then their lead times are so long that they can’t react to sales data, so why bother to try. It’s a vicious cycle."
But Fisher asserts that turning customer data into action is eminently doable and very effective. In addition to his professorship at Wharton, Fisher is the founder of 4R Systems, a provider of software and consulting for supply-chain management of short-lifecycle products. His company helps retailers determine the economically ideal amount of inventory at each point in the supply chain and each time in the life cycle of an item.
Fisher presents his Rocket Science Retailing concept as a roadmap for retailers on getting the most mileage possible out of consumer transaction data. (He traces his inspiration for the "Rocket Science" concept to the movement of the same name that swept Wall Street in the late ’70s, when the arrival of information technology revolutionized the investment world.) Using examples from companies that have done well using data and those that haven’t, Fisher explains how Rocket Science Retailing requires improving capabilities in three basic areas:
• Accurate Forecasting: Rather than relying on the gut feel of a few individuals, companies should update forecasts based on early sales data, track and estimate forecast accuracy, test products and implement a blend of bottom-up and top-down forecasting. Example: Book and music retailer Borders tracks sales by product category at each store, and periodically uses its merchandise planning system to automatically adjust each store’s assortment. A store in Alaska, for instance, might be well stocked with books about small planes.
• Supply Chain Speed: Companies should improve supply chain responsiveness so as to reduce stockouts on hot sellers (popular items that sell out) and also to reduce markdowns because less merchandise needs to be ordered initially. Example: Zara, which ships apparel worldwide from Spain, stores raw materials and reserves production capacity at factories in anticipation of demand.
• Risk-Based Inventory Planning: Companies should track stockouts, use this tracking to estimate lost sales and inventory levels to balance the risks of lost sales and excess inventory. Example: At Bulgari, the Rome-headquartered jewelry manufacturer, the researchers found that stockouts on a single item had lowered company profits by 5% of sales.
In turn, these three capabilities must rest on a foundation of:
• Accurate and Available Data: Companies should ensure accuracy of sales and inventory data, and retain and share this data with merchants. Example: Office superstore Staples practices a "zero balance walk" in which an employee searches 20% of the store daily for stocked-out SKUs (stock-keeping units). The reasons for the stockouts are then traced to uncover possible computer inventory discrepancies.
• Organizational Design: Companies should blend "left-" and "right-brained" capabilities to build better communication between the MIS department and merchandising/planning. Example: In their article the authors quote one prominent retail CEO as saying, "the merchandising-MIS relationship is broken," and another stating that "the only time [the MIS managers] communicate with me is when they ask for a $30 million write-off on some previous project that now has to be abandoned."
As Fisher and his co-authors note in their article, "Retailers can’t continue to suffer growing markdown losses yet disappoint a significant portion of their customers who can’t find what they want. They can’t continue to ignore billions of bytes of unused sales history that could help solve these problems…
"Every decade sees a retailer who innovates so powerfully that it rewrites the rules for other retailers and for all companies in the retail supply chain," they conclude. "In the ’80s it was Wal-Mart. In the ’90s, it was Amazon.com. We believe the next retail innovator will be the one that best combines access to consumer transaction data with the ability to turn that information into action."