The personal computer (PC) industry provides a unique and compelling case for creating value through supply chain integration. The industry's dramatic history of outstanding growth, its sheer pace and magnitude of change, and increasingly complex and global supply chains have created substantial challenges for participants. Traditional channel structures and behaviors in this sector's supply chains have not kept pace with the changes.

Unlike industries such as food service, consumer packaged goods, automobiles, and apparel, the PC industry has not yet matured to high levels of integration in its supply chain. Inefficiencies have emerged that now are preventing potential growth. These supply chain inefficiencies have led to chronically delayed new products, demand distortions, scarcity and allocation problems, inventory obsolescence risks, and unpredictable service levels.

With such a turbulent pace of change, participants in the industry are under increasing pressure to identify and exploit new value sources. Relationships among many supply chain participants (suppliers, assemblers, distributors/resellers, and retailers) currently are insufficient to provide solutions. Instead of seeking greater cooperation and interactivity, these four groups are displaying a degree of gamesmanship toward each other, which serves to undermine rather than improve the supply chain's efficiency.

To assess ways to improve this industry's supply chain, Andersen Consulting initiated a study in conjunction with Stanford University and Northwestern University. The research team undertook an extensive survey of more than 200 companies, comprehensive site visits to 10 leading companies, and secondary research on supply chain performance and financial analysis—providing a comprehensive picture of each segment of the industry.

Our goals were to understand how companies were integrating supply chains, to identify opportunities for improving performance and value creation, and to define critical factors for successful implementation. This article highlights the findings of that research and outlines the most promising strategies for supply chain integration in the dynamic PC industry. Many of the strategic principles offered hold relevance for a wide range of industries.

The PC industry is an ideal subject for this type of investigation. For one thing, while the industry's supply chains have become increasingly global and complex, they are not as mature as those in many other industries, which show higher levels of integration. Furthermore, little research has been done to date on supply chain integration in this industry. Finally, as the discussion below underscores, the industry's supply chain is an intensely dynamic one.

Changing Product and Profit Dynamics

The pressures and tensions among PC supply chain participants stem from several sources. Most of these are byproducts of rapid growth, which surged with the introduction of higher-density transistors in the early 1980s. The industry grew dramatically from $2.7 billion in 1981 to $150 billion in 1996, with compound annual growth levels of 13.7 percent. This robust growth is expected to continue through the year 2000, according to Dataquest, a market-analysis firm.

The explosive growth has been fueled by advances in technology. Yet these very same advances have wreaked havoc on product and profit dynamics. For example, technology breakthroughs have drastically shortened the life cycles of microprocessors, components, and desktop units from an average of seven years in the 1980s to a forecast four years by the year 2000.

Furthermore, consumers are increasingly demanding products with unique configurations, which has resulted in a proliferation of product offerings. To illustrate, the number of models tripled from about 400 in 1990 to more than 1,400 by 1995. And yet all of this costly differentiation has not stopped sweeping commoditization, as competitors have become increasingly able to emulate one another in ever-shorter time frames. The result of these trends has been a rapid decline in average selling prices and margins. In fact, profit margins for the top assemblers such as Compaq, Apple, Acer, and Dell have fallen by about half between 1990 and 1995.

Supply and Demand Misalignment

With companies seeking to capture and maintain market share in a world of price cutting and falling margins, there is little room for error in getting products to market. If the product is not available to customers—particularly during the crucial days of introduction and ramp-up, when margins are at a premium—the impact on market share and profit can be devastating. Exhibit 1 depicts the strong correlation between market share and product availability, using Compaq as an illustration.

As shortages steadily intensify, market share flattens as early as three months out and then begins to decline precipitously only one month later. Despite this tremendous pressure on availability, the PC industry has had difficulty achieving adequate supply, particularly of the hottest new products. In fact, chronic shortages of hit products have plagued many industry participants. The graph in Exhibit 2 depicts the typical supply-and-demand misalignments that occur throughout a PC's product life cycle.

As the graphic shows, demand is typically highest at or shortly after launch time, when supplies are most precarious. During this period of scarcity, distributors and resellers often over-order as a hedge against continuing shortages, creating 'phantom demand.' As planning systems typically use previous-period demand to forecast the future, the phantom demand gets factored into forecasts, often triggering overproduction. In the meantime, true demand tends to taper off when competing products come on the market.

Evolving Supply Chain Structure

The PC industry's changing supply chain structure has itself created further challenges for supply chain value enrichment. Until the mid-1980s, leading PC companies performed the majority of activities in the chain. By the second half of the 1980s, however, the industry had evolved to a networked structure, where many independent companies joined forces to bring a product to market.

Today, many key supply chain activities in the PC industry have evolved into myriad industries of their own. The industry's supply chain was further altered as some components, such as microprocessors and disk drives, became concentrated among fewer suppliers, both geographically and commercially. Meanwhile, sources for other types of components began to proliferate rapidly at sites scattered all over the world.

Industry View of Integration

Through the research work and the team's direct experience in the PC industry, it was apparent that the industry, as a whole, did not view itself as being highly integrated. The survey results reveal a low level of perceived integration with upstream suppliers and downstream customers.

Ironically, these relatively low scores belie the extensive attention and activity related to improving supply chain integration that actually is occurring in the industry. The flurry of current and planned activities appears to indicate considerable interest among industry participants. As depicted in Exhibit 3, software developers, hardware suppliers, assemblers, and distributors/retailers are reporting substantial integration activity—both current and planned.

Despite all this movement, there appears to be considerable confusion among participants about the best ways to integrate. In many cases, significant value is being overlooked or simply not being captured. Many supply chain participants are still focused exclusively on products, neglecting aggressive management of the supply chain itself as a strategic source of value. Some are ignoring the chain's structure and dynamics as a source of added value. And others are limiting their integration efforts to quick, tactical solutions, rather than launching more far-reaching strategic initiatives. In some instances, implementation efforts have not been successful, mainly because the support relationships are not yet developed.

Although the PC industry is experimenting with supply chain integration in many positive ways, some initiatives have backfired by alienating the very participants whose cooperation was sought. The study team's site visits revealed a wide spectrum of cooperation among supply chain partners. Some companies have done particularly well in achieving harmony. In other cases, however, there is discord among partners who have "burned out" on trying various integration concepts that have been ill-conceived or poorly executed.

Analysis of the research reveals substantial opportunities for supply chain integration in the PC industry. These vary in their suitability for different participants as well as in the benefits they offer. We believe that companies should vigorously pursue those integration opportunities that hold the greatest value potential for their unique supply chain situations. More specifically, they should:

• Pursue compression of the supply chain's structure by understanding all the options and making choices strategically. Companies miss out on substantial value by failing to grasp the possibility of modifying the fundamental supply chain structure, by rejecting it as being too risky, by executing poorly, or by attempting to do too much at once, even though some options may conflict with one another.
• Implement serious collaborative planning with supply chain partners—not just because it enhances value, but because it has become a requirement for survival. Our research indicates that the risks for this particular initiative are relatively low and the rewards are high. We believe that striving for collaboration will always deliver some value; not trying is no longer an option.
• Design products for maximum market responsiveness. This effort requires viewing product planning as the greatest single source of integration value for an entire supply chain, even though this perspective breaks with traditional engineering practices. This is the most overlooked and underutilized value-enhancement opportunity for supply chains; most integration activities involving design have been limited to product assembly and specification issues.

To properly assess and illustrate the value of these integration activities, the research team developed an economic model using performance indicators, such as inventory turns and gross margins, as well as economic value measurements. As a starting point, we used Stern Stewart & Co.'s published value measure known as Economic Value Added™, or EVA.

EVA asserts that value is created when a company earns operating profits in excess of the cost of capital employed, where the cost of capital includes the cost of both debt and equity financing. A business with a zero, or break-even, EVA is just meeting its operating and shareholder-investment requirements.

EVA is driven by four levers: revenues, operating margins, the amount of capital employed, and the cost of capital employed. Any action that leads to a change in one of these levers will, in turn, change a firm's EVA. We thus can measure the impact of a specific supply chain initiative on the value a business creates by measuring the effect of that initiative on these four levers. For our study of the PC industry, we incorporated the EVA model into a larger model that we specifically designed to illustrate the impact of supply chain integration initiatives on value creation. Exhibit 4 displays that model.

Based on our primary and secondary research of the best practices in the PC industry, we developed benchmark percentage improvements in inventory turns and margins made possible by implementing each of the three major integration initiatives discussed in this article. We then analyzed the impact these improvements would have on EVA.

To illustrate the magnitude of value creation possible, we developed a representative financial model for the PC industry. (See Exhibit 5.) This financial model constitutes the baseline supply chain employed in our study and yielded the value numbers used in this article.

The baseline "companies" used in the model are not actual entities. Instead, the research team created model companies based on the typical financial performance shown by actual industry-segment participants. Our analysis indicates that the three initiatives we propose can create substantial value for supply chain participants.

The values listed in Exhibit 6 represent the gap, or value-creation opportunity possible through implementing each of the three strategic initiatives identified. The results clearly indicate that the companies studied are not realizing substantial value. The specific assumptions related to inventory or margin-performance improvement that drive this value enhancement are discussed below.

The supply chain-compression strategy focuses on the supply chain's overall structure as a critical source of hidden value to the participants who form its various links. Much of that dormant value is tied up in finished-goods inventory. The more complex a chain's structure, the more likely that stock (at various stages of completion) is languishing on the shelves of several, if not all, participants. In addition, the more finished-goods inventory is dispersed throughout the chain, the more difficult and costly it is for the chain, as a unit, to switch gears when consumer demand shifts. Customer needs go unsatisfied, while obsolete goods accumulate.

For the PC industry, the problems caused by supply chain complexity are most evident in the commercial channel, where there can be as many as four or five different entities that hold various stages of finished-goods inventories under their roofs—PC assemblers, retailers, distributors, resellers, and value-added resellers.

To calculate the impact on value, the research team measured inventory reduction and then factored in the margin improvement made possible through reduced obsolescence—1 percent for every 10 days of inventory. The research indicates that compressed supply chain structures can deliver inventory reductions ranging from 10 to 65 percent for assemblers and 10 to 100 percent for distributors, depending on the level of compression achieved.

The research confirms that numerous unfulfilled compression opportunities suitable for PC companies exist. It specifically identified four basic options for supply chain compression available to commercial PC market participants. Each option can deliver progressively greater value, yet each also is progressively more challenging to undertake. These basic options, which can include any number of variations, include intra-company postponement, inter-company postponement, the sales agent model, and the direct model.

Intra-Company Postponement

This type of compression is probably the most manageable to achieve because it occurs within a single company's operations. In this model, final product configuration is delayed, shifting those activities from manufacturing locations to distribution facilities. By doing this, a company can minimize its stock of finished goods and always have a ready supply of semi-configured product that can be customized just in time, according to shifts in demand. The research indicates that intra-company postponement can reduce the total inventory investment required by a PC assembler from 10 to 30 percent.

Inter-Company Postponement

As with the intra-company model, this form of postponement has PC assemblers ship semi-configured product to distributors who add components, such as disk drives and memory, only when orders are received. The major difference is that final configuration is shifted from manufacturing to an outside provider, such as a distributor, reseller, or retailer. Though this one difference makes this strategy significantly more challenging, it eliminates even more finished-goods inventories. Under the value-creation model, both assemblers and distributors achieve 10- to 30-percent inventory reductions.

Agent Model

The PC industry has yet to embrace a true "sales agent" model, although this strategy can enhance supply chain value even more than postponement. This strategy further compresses activities by drastically changing the roles of the supply chain partners. Instead of holding inventory or managing order fulfillment, distributors become sales agents, focusing on sales, service, training, and network configuration. The chain's assemblers ship orders directly to customers. This eliminates virtually all inventory at distributor sites without increasing inventory at other locations. The sales agent model can deliver an increase in inventory turns of 10 to 60 percent for assemblers and distributors.

Direct Model

The direct model creates the most compressed PC supply chain of all the models described here by eliminating all of the supply chain intermediaries between the assembler and the end customer. Within this model, assemblers handle all sales and customer-service functions in addition to final configuration and order fulfillment. Distributors and retailers disappear, as does all channel inventory. The direct model provides the clearest demand signal to the assembler. Compared to a traditional supply chain structure, it can reduce inventory investment by 50 to 70 percent.

From these results, it might seem obvious that all assemblers should immediately move to a direct model. However, the company's current supply chain structure is a critical consideration in choosing a compression strategy. Traditional, multi-tier manufacturers cannot immediately switch to a direct model without seriously jeopardizing their strong channel relationships and putting their business at considerable risk. Furthermore, differences in customer knowledge of products, varying degrees of product complexity, and the ease of use and reliability of products must be factored into the analysis. Ultimately, each firm will determine the appropriate mix of compression strategies based on its current position and overall company objectives.

A critically underutilized key to lean inventories and increased market agility is the supply chain partners' ability to collaboratively plan and execute the combination of tasks that bring a product from raw materials to market. The PC industry has barely scratched the surface in exploiting the opportunities and technologies available for this type of integration. Yet, these opportunities and technologies are accessible to virtually any company—and the risks are relatively low.

Rapid technological change, ever-shorter product life cycles, and increased supply chain complexity in the PC industry all have compounded the challenge of matching supply to demand. Close collaboration among supply chain partners can better align the two and thus enhance the value of the network's combined activities.

Value is increased through a number of levers. As with a compression strategy, one of the most important is improved inventory performance. Study results suggest that collaborative planning can lead to chain-wide inventory reductions ranging from 10 to 50 percent. These rewards are shared by all participating partners, from manufacturers and assemblers to distributors and dealers. (Exhibit 7 shows the improvement in inventory turns and gives the corresponding EVA enhancements.)

Collaborative planning and execution activities send a more accurate demand signal throughout the supply chain, serving to minimize waste and maximize responsiveness. The key activities are collaborative demand planning, synchronized order fulfillment, and joint capacity planning. Each activity requires progressively greater interaction among the partners.

Collaborative Demand Planning

Supply chains are paralyzed by information distortion, a phenomenon sometimes known as the "bullwhip effect." This is the amplification of order variability from the downstream sites to the upstream sites of the supply chain. The bullwhip effect can result in every site using its local demand information to make future forecast projections, infrequent ordering to take advantage of batch-order economics, fluctuating prices, or gaming behavior in shortage situations. Consequently, the demand signals received by a manufacturer can be very different from the actual demand pattern at the consumption level. Such information distortion would result in erroneous capacity decisions, excess inventory, poor customer service, wasted transportation, high overheads, and suboptimal utilization of resources.

One way to mitigate the bullwhip effect is through collaborative demand planning, a technique that can generate a truer demand signal for all the partners in the supply chain. Collaborative planning means sharing customer and operational data among all partners. It also may include joint decision making in forecasting demand. This continuous information loop eliminates, or substantially lessens, the type of incremental distortion that occurs when each participant puts its own spin on demand and supply information. The result is that everyone's understanding of demand remains unified, clear, and truer to reality.

Synchronized Order Fulfillment

Synchronized order fulfillment goes further. It is characterized by negotiated or joint decisions (such as order size and frequency) and by the transfer of management, and possibly ownership, of inventory from the supplier to the supplier's customer. Thus there is less incremental distortion of demand. The greater the synchronization, the greater the value added to the entire supply chain's performance. The most highly synchronized order-fulfillment approaches let customer-demand numbers directly drive orders, instead of basing orders on forecasts. While this form of total synchronization can do more to improve supply chain performance, it makes participants more dependent on one another. And this, in turn, increases their risk.

Joint Capacity Planning

Currently, there is little coordination among supply chain firms on jointly planning medium- to long-term materials and capacities. This lack of collaboration can lead to significant misalignment of capacity and demand over time and can result in individual partners lacking vital information. Furthermore, supply chains that fail to collaborate effectively will have difficulty gearing up or cutting back capacity when fluctuations in customer demand are large and long-lasting. Long-term, joint capacity planning is a critical source of further supply chain value. Accordingly, it should be explored by all PC industry participants.

Use of Collaboration-Support Technology

So much of collaborative planning and execution depends on timely exchange of quality information among supply chain partners. Hence, information technology plays a vital role in supporting such efforts. A variety of highly advanced tools now available are specifically tailored to these applications. The challenge is for management teams to build their technology arsenals to support collaboration—not the other way around. The collaboration among partners is the real solution, whereas the technology is one important enabler among a number of others.

Some of the key technologies used are electronic data interchange (EDI), enterprise resource planning (ERP), decision-support systems, and electronic commerce, or eCommerce. EDI, for instance, is used to electronically transmit a range of data between supply chain partners, including purchase orders, inventory, point-of-sale information, demand forecasts, and advance shipment notices. But it is not a magic bullet when it comes to collaboration. Rather, EDI's power to enhance depends on how information is used, not just on its availability. Without properly leveraging the information EDI makes available, supply chain partners can only expect to benefit from reduced administrative costs through automation.

Designing for responsiveness seeks to maximize supply chain performance at the product's design stage. It takes into account all the implications of a particular design, from raw-materials procurement to delivery at the end-customer's site. The designer seeks to achieve the best possible balance between innovation and product performance on the one hand, and optimal supply chain efficiency and agility on the other. The result is building into every product release both maximum value for all supply chain participants and the best possible run for a new product.

PC assemblers who design for responsiveness can achieve impressive returns. Our research shows that designing for responsiveness can reduce inventory and improve gross margins by 4 to 8 percent. Once again, the higher the level of integration and the greater the resulting supply chain efficiencies, the higher the rewards.

The highest-potential value-enhancement activities are supply chain structural analysis; designing for postponement and product variety; and use of standard, off-the-shelf, and intergenerational components. Every PC company should explore all three approaches.

• Supply-chain structural analysis. A fully realized design-for-responsiveness effort recognizes that the business case for a new product is made at the product-development stage. It makes sense to fully evaluate, at that early stage, just how the supply chain will perform for the new product. As the development team makes decisions about suppliers, components, and design details, it should be considering the optimal structure for the supply chain to support each new product design.
• Designing for postponement and product variety. In designing for postponement and product variety, the design team aims to minimize the negative impact of variety on time-to-availability and supply chain efficiency. This is accomplished by using modular designs and changing the order and timing of assembly. Postponement is a particularly important leg of a sound product-variety strategy. It calls for carrying modular, generic components to an optimal point in the supply chain, usually further downstream, where differentiating options will be added. By adding the desired variety features at that optimal point, demand can be pooled for the different product options and final-configured inventories kept to a minimum. Designing for postponement and product variety effectively executes the compression strategies and benefits discussed earlier.
• Use of standard, off-the-shelf and intergenerational parts. Another way to streamline the supply chain is by using interchangeable parts. The research found that many companies encourage product designers to use standard and/or off-the-shelf parts whenever possible. Suppliers also are providing incentives to PC assemblers to standardize parts. However, programs that encourage the use of intergenerational parts and intergenerational suppliers are less common and therefore appear to be an important source of untapped value potential for many companies.

Supply chain compression, collaborative planning and execution, and designing for responsiveness can present sizable implementation challenges to many companies. But which initiatives should be attempted? Which firms will make good partners? What are the risks? Since there is no one instant formula, the answers to these questions will vary for every company. To make an informed decision, organizations should consider supply chain integration from three key perspectives: rewards vs. risks, reshaping relationships, and execution as strategy.

Rewards vs. Risks

In considering supply chain integration, a full and realistic assessment of potential value and the potential risks should be undertaken. The comparative risks and costs associated with the three initiatives are shown in Exhibit 8. All three initiatives can reduce inventory assets, which has a further positive net impact by reducing inventory risk, obsolescence costs, and time-to-availability—all of which increases margins. But the complete value picture also must include the positive signal that bottom-line improvements send to stakeholders. In addition, rewards and risks have to be viewed as equitable to all partners.

In addition to assessing value, management teams need to evaluate the risks of supply chain integration for their own companies and their supply chain partners. Many companies, for example, have not considered the full cost and risk implications for their business processes, assets, technologies, and systems. Nor have they thought about the implications for change management. Without this kind of rigorous analysis, integration activities will fall short of desired results.

Reshaping Relationships

All integration initiatives require some reshaping of supply chain relationships—ranging from transactional, to interactive, to interdependent. Transactional relationships do not require much shared information or decision making. Hence, these should not be difficult to implement. A more interactive relationship, however, requires shared information, some joint planning, and some shared decision-making. This requires deeper cooperation, a willingness and capability to learn new skills, and a degree of mutual trust. In the case of the interdependent relationship, the trust factor is most important. As the boundaries between companies become blurred, information is shared extensively, as are decisions, investments, and assets. The goodwill, candor, and skill required in such relationships cannot be overemphasized.

As the above range of variables indicates, whether a company chooses to integrate its supply chain—and how it chooses to do so—must take relationship building heavily into account. Numerous partnership experiences over time have shown that it takes a superlative fit to make even the best integration ideas deliver. Not every supply chain partner will be equally suited to the demands of integration. Companies, therefore, must select their integration partners carefully. What are their capabilities? How flexible are they? Can they be trusted? Can they share your vision of the benefits and risks? These are the kinds of questions that must be answered before considering implementation of any but the most transactional relationships.

Execution as Strategy

As the rate of change in the PC industry continues to accelerate and the availability of competitive information keeps expanding, the leadtimes for strategy formulation, design, absorption, and implementation are shrinking—or actually disappearing. The firms that implement strategies first will be the most likely to gain the advantage and realize the value they seek. Companies that want to begin integrating—or further integrate—their supply chains should consider the following guidelines:

• Regularly implement strategies faster and more effectively than the competition.
• Define and utilize critical capabilities to their fullest, while not overreaching.
• Effectively launch and manage experiments.
• Learn, adapt, and build from repeated successes.

As these guidelines suggest, learning through implementation has the potential to deliver incrementally higher value than deferring all implementation until a "perfect" strategy can be forged. With due consideration of rewards, risks, and relationships, the most important thing is to make the best possible choice for now—and get started as soon as possible.