Rapid Response: Inside the Retailing Revolution
A simple bar code scan at your local department store today launches a whirlwind of action: data is transmitted about the color, the size, and the style of the item to forecasters and production planners; distributors and suppliers are informed of the demand and the possible need to restock. All in the blink of an electronic eye. It wasn’t always this way, though. HBS Professor Janice Hammond has focused her recent research on the transformation of the apparel and textile industries from the classic, limited model to the new lean inventories and flexible manufacturing capabilities.
Once upon a time, suppliers held all the cards. Henry Ford's dictum that consumers could have any color car they wanted as long as it was black proved wrong in the extreme, but for years manufacturers in this country kept their hands firmly on the spigot of supply and determined when people could get what they wanted. Choices were more limited, delivery time was measured in months, and warehouses were typically piled high with mountains of expensive inventory—often comprised of too many unpopular products and too few hot sellers. Today, this scenario is ancient history. The customer is king, trucks unload their just-in-time goods from automated distribution centers on a weekly basis, and the variety of products available is seemingly infinite. A typical department store now stocks some 800,000 items, with that number climbing to one to two million in flagship operations.
Spurred on by advances in information technology, a revolution has been taking place in retailing during the past two decades, and nowhere more so than in the apparel and textile sectors. The first to chronicle these changes and their implications in detail are HBS professor Janice Hammond, Frederick H. Abernathy and John T. Dunlop of the Harvard University Faculty of Arts and Sciences, and David Weil of Boston University. Their recent book, A Stitch in Time: Lean Retailing and the Transformation of Manufacturing—Lessons from the Apparel and Textile Industries (Oxford University Press), is based on eight years of extensive study, including a comprehensive database compiled under the aegis of the Harvard Center for Textile and Apparel Research (HCTAR).
The keys to success in an age of product proliferation, the authors found, are no longer economies of scale and cheap labor but an up-to-the-minute knowledge of what sells and what doesn't, flexible manufacturing capabilities that can respond appropriately to demand, lean rather than fat and costly inventories, and the rapid replenishment of stock. "The old way was to gear planning and production decisions to forecasts and guesses made months in advance of a selling season," write Hammond and her coauthors. "Now firms receive periodic ongoing orders based on actual consumer expenditures. Lean retailing allows department stores, mass merchandisers, and other retail outlets to capitalize on information, allowing them to minimize their exposure to demand uncertainty."
To accommodate real-time demand and diminish the risks inherent in "perishable" inventories, retailers depend on timely information gleaned from technologies such as product bar codes, point-of-sale scanners at the checkout register, and electronic data interchange (edi), which facilitates the accurate and instantaneous exchange of sales figures between the manufacturer and the retailer. In the mid-1980s, these advances began to chip away at the barriers commonly found throughout the various stages of production and distribution.
"That's when Sam Walton set the wheels of change in motion," explains Hammond, "by pulling suppliers closer to him, insisting that they implement the information technology necessary for exchanging sales data, adopt standards for product and shipping container labeling, and use modern methods of material handling, thus assuring a variety of products at low prices." As a number of other leading retailers followed suit and as data collection and computing methods became more sophisticated, once autonomous units found themselves part of an "information-integrated" channel extending from the acquisition of raw materials to production and sales.
With the resulting improvements in demand forecasting and production planning and practices, inventory management for both the retailer and the manufacturer has become much more of a science. Since stores can track consumers' preferences and update their orders from day to day, lean retailers require a rapid response from their suppliers. Indeed, a replenishment delivery time of no more than one to two weeks is now regarded as standard operating procedure.
In their book, Hammond and her colleagues observe that apparel makers have several options for meeting this considerable challenge. Most still concentrate on holding in inventory finished merchandise for which there is a relatively stable demand—products described in the study as "basics" (that is, those with a long shelf life, such as men's white dress shirts) and "fashion basics" (khaki pants with pleats, for instance), which together accounted for 72 percent of total dollar volume in a hctar sample of the apparel market.
But in the long run, the researchers see this as an inefficient strategy for satisfying the requirements of lean retailers. Going forward, they predict, the preferable alternative is to add short-cycle production processes based on flexible planning. "A manufacturer can pay somewhat more," they write, "to make certain units—those with a high weekly variation in sales—in quick production lines and still reap a better return than the supplier would by making all of the product in a slower plant" that benefits from economies of scale. It's up to management, they add, to determine the right combination of short- and longer-cycle lines in domestic and foreign sources. "Balancing these assembly lines by establishing for each [stockkeeping unit] the precise pattern of expected variability in demand provides the means for maximizing profits. Lean retailing practices will continue to push suppliers in this direction."
In a fast-paced environment where time-to-market and short-cycle production are powerful levers of competitive advantage, proximity has taken on much greater significance in all but "fashion" items, where once-a-season orders still prevail. As a result, the People's Republic of China, Hong Kong, Taiwan, and Korea—long the biggest sources of U.S. apparel imports and paragons of the cheap-labor model—have seen their shipments to this country decline dramatically, from 38 percent in 1991 to 16 percent in 1997.
Replacing much of the Asian connection has been a region-based realignment that includes the United States, Mexico, and the Caribbean Basin. According to Hammond, chances are good, therefore, that "the skilled processes involved in the manufacture of garments amenable to rapid replenishment will remain in this country for the near term, while the lower labor costs in nearby Mexico and the Caribbean will favor these areas for the assembly phases of production."
Given this scenario and the changes in technology, management, and manufacturing practices documented in their book, the authors are optimistic about the future of U.S. apparel and textile firms, even as many other commentators see danger ahead as a result of liberalized global trade agreements and the phasing out of the protective measures offered by the Multi-Fiber Arrangement. "This important sector of our economy is more advanced and productive than ever," concludes Hammond. "As Mark Twain might have put it, any reports of its impending death are greatly exaggerated."
From Working Knowledge: A Report on Research at Harvard Business School, Winter 1999
Research by Janice H. Hammond