In today's economy, companies around the world must deal with demanding customers, create competitive advantage, and continually generate new revenue or cost savings. These realities have driven them to implement technology to "link up" and share information with one another. To date, these links have ranged from electronic data interchange (EDI) and value-added networks to extensible markup language (XML) and the Web.

The next phase of this sharing process is a more complex and committed form of collaboration. It entails companies' sharing formerly proprietary information with other companies that are members of what's called "collaborative business communities" (CBCs). The proprietary information would include data on such things as inventory levels, pricing, and manufacturing capacity.

Within a CBC, business information is shared in a form that can be processed directly by enterprises that are a part of a common marketplace or a common value chain. CBCs are not about enabling a "virtual team" of people spread across multiple enterprises to work on the same document. Instead, they're about developing a shared demand forecast for a group of companies with similar requirements. Changes made to this forecast would be reflected immediately to the community of suppliers.

CBCs offer value to enterprises in a number of ways. They can enable secure process-driven collaboration with customers, distributors, suppliers, contractors, transportation carriers, and other trading partners. CBC participation can improve the visibility of critical operational data and accelerate the velocity of products and services moving among the community members. Participants also benefit from performance metrics that are designed for the business community as a whole, rather than metrics that optimize only their own operations at the expense of other community members. Additionally, community members will benefit from the sharing of process and performance analysis among CBC members. Each member of the supply chain will receive real-time alerts over the Internet as plans are updated or operational events deviate from plan. For example, via a Web site posting, a manufacturer could share with its suppliers quality data for a particular manufacturing run.

One of the key values of the CBC is that the members share a common infrastructure that allows them to implement a collaborative operations planning (COP) process. A COP process provides members with the ability to examine the entire supply chain holistically and respond to demand change optimally by considering overall business objectives. For example, this collaborative approach might show that it is more cost effective to incur higher shipping costs and keep the production schedule on track than to make more costly changes to the schedule.

The COP process recognizes the strategic value of inventory, not just the cost. It is based on an understanding of the time and service value of the warehouse in managing the flow of product to meet market demand. By dynamically rebalancing inventory within the CBC and taking advantage of deployed aggregate stock across the member companies, the members can respond quickly to demand and buffer themselves against forecast error at distributed warehouse locations. Most importantly, CBC members can use inventory, materials handling, and production capacity across the entire enterprise more efficiently, thereby minimizing additional capital cost. The net result: improved profitability at a faster rate of return. Obviously, it's much more cost effective to buffer uncertainty with information than with inventory. For companies trying to maximize their competitiveness, the COP process presents potentially sustainable competitive advantage to CBC members.

Collaborative business communities can achieve this sustainable competitive advantage by capitalizing on three key advantages: visibility, velocity, and variability.

Visibility expresses the need for all of the members to understand the availability and location of inventory, shipments, and other resources throughout a network that stretches from the suppliers' supplier to the customers' customer. Poor visibility creates uncertainty across the supply chain, which causes enterprises to use time buffers (for example, increasing cycle times) and material buffers (larger inventory piles). Visibility-enhancing systems fall into two categories: those that enhance an enterprise's visibility of supply chain information (including past performance; current conditions; and trading-partner plans, capabilities, and abilities), and those that enhance trading-partner visibility of the enterprise's information (including order-status information and inventory availability). Both types of systems must be implemented for effective supply chain performance. Enterprises must develop a data-sharing strategy that effectively segments data and securely reveals it to selected supply chain partners. CBCs provide the framework in which such strategies can be formulated.

Poor velocity of information also leads to what is called the "bullwhip" effect, where suppliers overcompensate for lack of accurate forecasts by building up extra inventory to avoid out-of-stock situations. Collaboration-enabling technologies, however, now are robust enough to take the sting out of the bullwhip.

The most widely adopted collaboration-enabling technology is the Internet, which enables cost-effective, close-to-real-time communication regardless of platform. Because the Internet is basically a transport pipe, however, it alone will not get the job done because it is blind to data semantics. Companies seeking to enable collaboration with trading partners through application-to-application communication will need to develop core competencies in new integration methods, such as message brokers and processware. They will seek to make their collaborative business community more responsive than competing CBCs by leveraging the Internet and integration technologies to create interenterprise applications that increase the velocity of supply chain activities. This is already occurring in some high-tech segments, where customers can obtain access to a company's sales configurator via the Web and configure the desired product. The configurator then automatically checks material and manufacturing capacity availability not only for the selling company but also for that company's key component suppliers. It then gives a firm delivery date—which often reduces the order-processing cycle by weeks.

Although many trading partners and customers are asking for increased visibility in a number of ways, what they really want to know is, "When are events deviating from my expectations?" In other words, they want to be alerted to unexpected variability so that they can: (1) take corrective action before their own operations are negatively affected, say, by a material shortage (that is, managing variability to reduce costs) or (2) reconfigure their activities to take advantage of changing conditions, say, a new product selling faster than expected (managing variability to enhance revenue).

Even with increased collaboration and visibility of information, companies still will have to manage performance variability. CBCs provide a structure in which this can be done more effectively. Managing variability involves detecting variability, understanding its context, recognizing the supply chain effect of variability, and understanding actions required by variability (for example, whom to alert and whether to trigger a new manufacturing plan). Community members must be able to react quickly to changes in demand plans, inventory levels, and market forces. They also must be able to operate with shorter cycle times and lower inventory levels. The capability to exchange information quickly and comprehensively within a CBC better enables members to accomplish these activities.

To be effective, a collaborative community needs a management structure. It can work like a trade association or like local government, with elected leaders. Or, the CBC can be driven by a subset of the membership who are willing to fund leadership positions out of their corporate budgets. Under another alternative, a third party (say, a venture capital firm) would fund a community manager out of the advertising revenue, transaction fees, or other revenue generated by interactions among community members and with nonmembers.

Whatever the community's structure, the duties and metrics of the community manager must be decided up front. Among other duties, the CBC manager will be responsible for converting enterprise "sell side" requirements to the "buy side." The manager also will ensure that the multi-enterprise production and fulfillment processes are profitable for the CBC members—whether the demand and the orders are generated within or outside of the community.

Any manager needs tools to do his or her job, and the CBC manager is no exception. A CBC manager needs management software and services, such as supply chain planning and supply chain management packages, to enable the real-time optimization of manufacturing constraints such as process constraints, resource constraints, manufacturing capacity, and technology constraints. Furthermore, the software and services must optimize these constraints not only within the enterprise but also among trading partners.

Just as the manager's duties and metrics must be spelled out, so too must the requirements for these community management tools. These requirements include developing a CBC management application that facilitates collaborative decision-making by capitalizing on the data from the different enterprise systems, by communicating among all of the functions across the channel, and by developing key performance indicators. The application must have decision-support logic to analyze millions of bits of data, optimize those activities, and determine the nature and timing of necessary actions. Finally, the CBC management application must be able to determine the actual transaction instructions, support decisions to manage the flow, and then measure the operational execution throughout the entire supply chain. With this type of management infrastructure, CBC member companies can begin to match information and product flows to actual consumption.

Increasingly, enterprises will need direct, real-time business-to-business integration that must be supported by the CBC management software and services. This integration will require data consistency and integrity across applications as well as integration with enterprise resource planning (ERP) and supply chain execution (SCE) applications. The management software and services also must integrate information from outside the enterprise. It must provide secure, Internet-enabled, business-to-business integration with other members of the CBC and even nonmembers. The interface needs to provide for easy integration with internal enterprise applications as well as external systems used by trading partners.

To enable this integration, collaborative business communities need an infrastructure that includes both XML (a common language for developing interactive content) and common semantics in developing applications. Both are critical for developing truly collaborative communities. A collaborative community must have a grammar that conveys the meaning and context of the shared forecasts and plans as well as the assumptions upon which they are based. CBCs will use XML to realize more fine-grained interactions, with richer and more flexible data structures.

One of XML's main benefits is that it provides a common language for structuring documents that flow between business partners in a supply chain. For instance, it is possible today for a SAP Financials application to "view" data in a Baan manufacturing application, even when they reside in different enterprises. However, without a common data dictionary, the application logic of one application cannot "understand" the data held by another without human intervention to interpret the data. XML and the semantic reconciliation tools based on it effectively address this issue.

XML standards by themselves, however, will not eliminate the need for data transformation. All of the CBC members must use the same semantics in developing their applications. Common semantics almost never happen when integrating pre-existing applications, unless there is an agreed-upon technical infrastructure. For example, a message that explicitly associates the inventory-level data element with the value "5" is not useful until the contextual meaning of "5" can be coded and "understood" by the recipient application. XML as a syntax specification cannot create messages that are fully understood by heterogeneous applications with different data models, any more than the alphabet by itself generates meaningful words or phrases.

In short, when developing the information infrastructure for their collaborative communities, members will have to keep the following guidelines in mind:

• The programs must use the same semantics.
• Integration vendors will have to be selected by the CBC, rather than by each enterprise.
• Enterprise developers will have to use standard semantics in their "home-grown" applications.
• Older applications will have to be either rewritten to follow the agreed-upon semantic standards or interfaced through "interpreter" systems (or service bureaus).

Because of these complex integration issues, we believe that CBCs will be relatively small groups of enterprises, rather than entire industries. Furthermore, there will not be universal compliance with these semantic standards because each enterprise will be a member of multiple CBCs. This means we will continue to see a need for application integration tools and services through 2002—at least.

In addition to deciding upon the CBC's management and technological infrastructure, members must define strategies and tactics for collaboration. Such definitions require an understanding of the power dynamics of one's supply chains as well as an understanding of the capabilities and abilities of one's trading partners. Collaboration is more likely to succeed in transforming relationships if it: (1) is relatively unintrusive to the trading partner's organization, (2) avoids a quantum leap in technology or business process, and (3) creates a true win-win partnership among the CBC members.

The CBC management team should not attempt to define a single strategy for collaboration. Enterprises will need to adopt different strategies for different distribution and supplier channels and relationships. Two traditional strategic approaches are using coercion when dealing with suppliers and being submissive when dealing with powerful customers or distributors. Participating in multiple CBCs will somewhat reduce these tendencies. The CBC manager will establish "rules of engagement," and the incentive to define strategy based on power relationships will be reduced.

Yet, collaboration does not mean the end of coercion. In fact, collaboration can lead to the emergence of powerful channel masters. For example, powerful players in any community, marketplace, or supply chain will insist on sharing production and distribution plans "in their way." This means they'll want their suppliers to implement the collaborative software and processes that they have designed. In such cases, it will be up to the management team of each enterprise to ensure that the collaboration is in its best interest. Within a CBC, enterprises faced with win/lose propositions from channel masters could appeal to the CBC management to develop a more win-win proposition that would optimize value to the community overall. Nevertheless, coercion, in one form or another, will persist as a strategic consideration.

Enterprises, therefore, will have to be very explicit in their requirements for collaborative activities with their trading partners. Sometimes companies fall into the trap of idealistically believing that forming a CBC will automatically result in across-the-board collaboration from all of the trading partners. Our experience with Fortune 500 companies, however, suggests that collaboration succeeds when driven and enforced by a small number of channel-master enterprises. These channel masters have a common goal of reducing shared or "community" costs but still wish to maintain their competitive advantage in other process or operational areas.

Once the CBC is established among a relatively small group of supply chain partners, it can be expanded geographically. Collaboration across international borders is a significant challenge. Although it is true that the Internet has created a global marketplace for even the smallest companies, it also has presented some global-sized problems for companies attempting to sell their goods and services in multiple countries. For example, if a purchasing manager or consumer in another country enters an order on your Web site, you may be able to confirm the order in real time and price it in the customer's local currency. You may even be able to allocate your inventory to this order and commit to a delivery date while the customer is still online. That order, however, is still not complete because it must be screened for compliance with export regulations and terms and conditions of sale. Global fulfillment has its own set of challenges. These include various import and export restrictions, regulatory issues, and differences in the technical and process interface requirements found in countries around the world.

Implementing a global collaborative business community requires that the CBC manager bridge the technical and administrative gap between the business community and the numerous government agencies. The CBC manager must keep up to date on all regulations, procedures, and documentation requirements in the various countries. Governments often will require full and immediate compliance with even the most recently changed regulations. CBCs will help integrate the financial and technical requirements of a company's global suppliers and customers with the documentation and compliance requirements of host governments—even those requirements that are only weeks or days old.

Collaborative business communities are emerging in a number of business sectors. Two of the most promising come from the computer industry and the grocery industry.

Computer manufacturers today use complex processes and a multitude of expensive applications. Yet they often wind up guessing at demand and roughly estimating inventory levels at locations around the world. This can result from a lack of agreement on something as simple as how a part number is defined or how inventory queries can be made through a standard interface. The negative impacts are felt on production planning, channel allocation, and the cost of returns.

The computer industry has addressed the situation through RosettaNet, a global business consortium formed in 1998 to standardize and streamline these fulfillment activities. RosettaNet has created a master dictionary of industry-specific products and other data elements, an implementation framework to guide computer companies seeking to comply with the standards, and a series of Partner Interface Processes (PIPs), which define how different systems can be interfaced to each other. The consortium membership of RosettaNet includes most of the leading IT companies, such as IBM, Microsoft, Intel, SAP, Oracle, Hewlett-Packard, Cisco, CompUSA, Compaq, EDS, AOL/Sun/Netscape, and NEC.

RosettaNet demonstrates what all industries, trade groups, and collaborative business communities must do to adapt existing systems, processes, and transaction standards to do business in real time via the Internet. Although RosettaNet is by no means complete, it is an instructive model. Companies looking to develop Internet-ready business processes should carefully examine the RosettaNet documentation for clues about how to organize their own supply chain or CBC.

The retail grocery industry's UCCnet is another successful example of data-driven collaboration among enterprises. Currently in its pilot phase, UCCnet is being tested by six companies: three retailers (The Kroger Company, SUPERVALU Inc., and Wegmans Food Markets Inc.) and three manufacturers (Frito-Lay Inc., Procter & Gamble Co., and Ralston Purina Co.). Once the pilot is complete at the end of first quarter 2000, UCCnet will immediately begin the rollout to additional companies in the industry. As the capabilities expand, it ultimately will move beyond the grocery industry.

UCCnet is an open standards-based extranet for exchanging important information across the supply chain. It uses a Web-based architecture so that companies of any size will be able to participate. One of the key applications for this extranet over the next two years is likely to be collaborative planning, forecasting, and replenishment (CPFR). We believe CPFR foreshadows how other CBC applications will be developed over the next several years.

To institutionalize the kind of collaborative behavior seen in RosettaNet and UCCnet, enterprises will have to break down a number of barriers. Although browser-accessible applications and the Internet will increase information velocity at less cost than previous technologies, they do not address the issues of variable business processes and disparate data semantics among enterprises. Just as Esperanto failed to become the universal language, universal data semantics for business will never become a reality.

Rather, specific supply chains (for example, the electronics industry or the auto industry) will continue to develop their own semantic standards, and specific CBCs will continue to come up with even more granular semantic standards. Any company doing business across multiple supply chains and multiple CBCs will still need to deal with multiple semantic definitions and business processes. Thus, supply chain management excellence means using emerging technologies to harness the diversity of trading partners, rather than creating a vanilla standard data process and business model.

Another barrier is the tendency among enterprises to resist industrywide collaboration because capitalism does not reward a level playing field. The Demand Activated Manufacturing Architecture (DAMA) project is an illustration. DAMA was attempting to use collaboration to make the U.S. apparel supply chain more competitive with low-cost apparel manufacturers in other parts of the world. Yet the project, which was funded partially by the U.S. government, stalled in the prototype stage after four years of effort. The problem: In large groups, enterprises do not trust one another; they are interested in gaining an edge, not in creating an environment in which the actions taken deliver the most benefit across the supply chain.

The growth of collaborative tools, applications, processes, and business models represents a fundamental shift in how supply chains are managed. Over the next few years, it will change forever the strategies of enterprises that have entered into collaborative agreements. The most progressive companies will move from an internally focused cost-reduction strategy to an externally focused revenue-enhancement one. The success of this strategic shift will rest on management's ability to purge their enterprises of dysfunctional processes and "win/lose" relationships and replace them with more agile and collaborative processes that can leverage community-level strategies and metrics.

In the short term, enterprises would be well advised to carefully select a small group of partners and processes for collaborative activity rather than looking to industrywide initiatives and standards to provide the solution. The vendors that will be able to survive the move to CBCs will be those that solve fundamental problems—such as semantic reconciliation across multiple enterprise planning applications, or those with deep industry domain expertise. And because CBC applications and their infrastructure will be hosted externally, we expect a substantial growth in the sophistication and quality of today's commerce service providers (CSPs) and application service providers (ASPs) to meet this challenge.

As collaboration begins to displace the largely antagonistic relationships among members of the supply chain community, the impact on industries, markets, companies, and business processes will be profound. Some of the key impacts are:

• Services will eclipse products. By exploiting the information potential of CBCs, companies will use services like vendor-managed inventory and direct store delivery to differentiate themselves. This will require on-the-fly business process change, facilitated by CBC technology, as companies modify their logistics operations to meet unique customer needs.
• Demand will drive production. Electronic components currently lose value every day they sit in stock. Outdated products can't be sold and unused production capacity is wasted. All of this costs companies a lot of money. As a result, production is postponed until demand is well understood—moving increasingly toward make-to-order and late-assembly strategies. Companies may even choose to source products from competitors to meet urgent customer needs.
• Prices will match market conditions. CBCs enable better yield management, which allows companies to reduce perishable inventory and maximize profits in commodity or supply-driven markets. The Internet allows virtually every industry to leverage variable pricing across distributed inventory, using new intermediaries like brokers and auction houses to move excess capacity.
• Tens of thousands of e-marketplaces will emerge. Some of these will sell off excess inventory (in the tradition of MetalSite in the steel industry), but most others will focus on bringing buyers and sellers together. Note, however, that many of these marketplaces will have precisely the same limitations that individual Web sellers have: namely, limited integration infrastructure with the individual merchants, poor fulfillment capabilities, and lack of global support should problems arise.
• Industry portals will become obsolete. As CBCs take hold over the next two years, many of the business-to-business e-marketplaces that only offer "portal-like" services for bringing buyers and sellers together will be pushed aside or will drive each other out of business. The fundamental reason is that their interest lies in creating a cheaper (lower margin) market, not in solving any fundamental problems of the buyers and sellers that participate in the markets. If we've learned anything in our e-commerce experience to date, it is this: Vendors that do not address the fundamental business or technical problems of their customers will not survive competitive pressure or an economic downturn.

Combined, these trends point to a new generation of supply chain activity. Success in this environment will go to those organizations that know how to collaborate in a manner that will benefit both themselves and their community partners.