Some investors continue to believe that on-line business-to-business (B2B) exchanges could improve supply chain management dramatically despite the punishment they took in financial markets last year. Many companies are still eager to jump in: during the year 2000, the total investment in B2B infrastructure exceeded $200 billion—an estimated $10 billion of it for public consortia-backed e-marketplaces alone. Still, the financial markets appear to have caught on to something that experienced supply-chain-management practitioners have suspected for a while. Although B2B exchanges (otherwise known as e-marketplaces) can help companies realize certain purchasing and transaction-processing benefits in the short term, broader improvements—particularly reductions in inventory, improved service levels, faster time to market—are harder to achieve.

The promise of real benefits rests on the potential for seamlessly integrating data flows and work processes across entire enterprises and even industries. To realize this opportunity, the power of B2B exchanges would have to be combined with the enterprise-resource-planning and decision support systems that many companies have adopted in recent years. But this kind of integration is hard to pull off, and for industries with complex, segmented supply chains, it is simply inappropriate.

True improvement will emerge only from the understanding that exchanges can’t affect every element of the supply chain equally. The primary benefits will come from their ability to speed up the flow of information and to make it available more widely—which alone can produce gains large enough to justify the big investments that unshaken believers in the future of B2B exchanges continue to make.

Elusive benefits

Products reach customers through a chain of retailers, distributors, wholesalers, manufacturers, and component suppliers. Supply chain management is intended to accelerate the flow of goods, information, and capital in both directions, along the chain’s entire length, and to help companies monitor that flow. Because the costs of managing the supply chain—inventory, the warehouse and distribution center, and freight—can represent 10 to 15 percent of sales in most industries, the savings that B2B exchanges promise could have a genuine impact. Indeed, programs to improve supply chains might raise margins by 1 to 2 percent of sales and improve customer service dramatically.

Yet most exchanges haven’t delivered these benefits. First, an exchange can’t wring huge efficiencies out of all elements of the supply chain; in fact, it can have no impact at all on some of them, such as the physical flow of goods. An electronics manufacturer in California, for example, must maintain surplus inventory because otherwise the company wouldn’t be able to fulfill unanticipated orders until the components for them arrived; those from Taiwan, for instance, may take several weeks to cross the Pacific and clear customs. At best, the improved information flow or collaboration that an exchange offers may eliminate the three to five days ordinarily spent planning, negotiating, and documenting transactions.

The second reason for the failure of B2B exchanges to function as promised is that they themselves have perpetuated certain inefficiencies by failing to recognize that the same supply chain segment in different industries, and different supply chains (or segments thereof) in the same industry, may require different improvement levers. In men’s apparel, for instance, a retailer could have a number of supply chains. One might replenish perennials such as undershirts, white dress shirts, and size-40 regular navy blazers, while a second might stock fashion items, for which demand varies according to the season, the effectiveness of efforts to promote them, and their inherent appeal. A grocery retailer, meanwhile, must manage the flow of perishable produce (such as lettuce and apples), for which demand tends to be fairly predictable, and of nonperishable products (such as soft drinks), for which it can be influenced by heavy promotion. Instead of developing services based on different segments of the supply chain, retail B2B exchanges have so far tried to serve all of these needs at once.

Third, many companies that own information think it gives them a crucial competitive advantage and therefore fear sharing it freely, though companies up and down the supply chain would benefit if they did. Companies know that their business processes and decision support systems have a direct impact on their costs and revenue. The level of mutual cooperation and trust that participants in a B2B exchange must have before aggregating their purchases of, say, copy paper is trivial compared with what would be needed to get them to share information—for example, about forecasts, product life cycles, and bills of material. The idea of confiding financial data to an exchange generates even greater skepticism. Precisely because Dell Computer and Wal-Mart, for example, derive a competitive advantage from their exclusive collaborations and from the proprietary sharing of information with their suppliers, they have avoided public B2B marketplaces and exchanges. Moreover, the tools and techniques needed to optimize and integrate financial flows are just coming into broad use.

Even if it were possible to allay basic fears about the sharing of information, thorny challenges for B2B exchanges would remain: providing security, imposing formats for conveying information, and ensuring that members share information fairly.

Defining the real opportunity

To date, e-marketplaces have focused mainly on procurement—and dealt with it reasonably well. But they have encountered problems in seeking to streamline tasks (such as production planning, inventory control, and scheduling) that lie closer to the heart of supply chain management. To devise solutions, it will be necessary to analyze what exchanges can and can’t do. They will never reduce the time it takes to deliver goods physically, for example. But since the information flow in supply chains is typically linear, fragmented, and inaccurate, they can make a vast difference in this area.

Usually, information is shared only by two players, buyer and seller, at each stage of the supply chain

Retailers, distributors, wholesalers, manufacturers, and suppliers all participate in a typical supply chain. But only two adjacent players—the buyer and the seller—usually share information at each stage, and the nature and amount of what they share depends on the quality of their relationship. After all, this kind of information mostly concerns the actual transaction between them; they rarely communicate their general understanding of market trends or changes within the industry. As a result, the information that each participant uses to make its decisions doesn’t reflect conditions in the industry as a whole, and perspectives diverge. When forecasts of demand, for instance, aren’t reliable, companies must act defensively by accumulating excessive inventory, and they must also pay overtime and incur the expense of expedited service when an unexpected order arrives. If they can’t make last-minute adjustments, they lose sales.

The successful B2B exchanges of the future will replace this linear, bilateral structure with one organized as a hub and spoke—the exchange at the center of the information flow and the individual trading partners arrayed along the circumference. Successful exchanges will also be tailored to coherent segments of the supply chain. When exchanges have established transaction standards and common platforms, the hubs will be able to gather information spanning all levels of the chain. The dispersal of this information by hubs will increase the speed with which it is shared, its accuracy and quantity, and the transparency of the whole chain. As the lead times and search costs of early adopters shrink and forecasts become more dependable, companies will be increasingly prepared to surrender their closely guarded secrets to the exchanges.

If an improved information flow is what B2B exchanges really have to offer companies that want to sharpen their management of the supply chain, two questions should determine where they put their greatest effort and investment. First, what are the characteristics of the different supply chains in which companies participate? Second, what parts of the chain are most affected by better information in the short and long term, and what parts are most relevant for long-term improvement?

Before a company allows its fear of being left behind to push it into investing in and joining a public exchange, it should determine if the supply chain services offered by the exchange comport well with its supply chain segments. It is understandable that companies want to cut their inventories by improving forecasts, for example, but no statistical model can predict the day-to-day demand for a product with a three-month life cycle and thus what levels of safety stock to maintain. In industries such as fashion apparel or personal computers, shorter lead times and response cycles are therefore more likely to generate real improvements in the supply chain.

Companies should decide which elements of the chain could produce the greatest efficiencies and then choose the exchange most likely to promote them. Identifying ways to capture true supply chain benefits from exchanges thus comes down to basic supply chain management. Companies shouldn’t let the hype and excitement of the exchange phenomenon make them neglect their off-line supply chain operations—from demand-planning algorithms to logistics management. If they focus their attention on these basics, the benefits will surely follow.