The 21st-century Supply Chain

The best supply chains aren’t just fast and cost-effective. They are also agile and adaptable, and they ensure that all their companies’ interests stay aligned. by Hau L. Lee

Triple-A Supply Chain

D uring the past decade and a half, I’ve studied from the inside I more than 60 leading companies that focused on building and re- ‘ building supply chains to deliver goods and services to consumers as

quickly and inexpensively as possible. Those firms invested in state-of- the-art technologies, and when that proved to be inadequate, they hired top-notch talent to boost supply chain performance. Many companies also teamed up to streamline processes, lay down technical standards, and invest in infrastructure they could share. For instance, in the early 1990s, American apparel companies started a Quick Response initiative, grocery companies in Europe and the United States touted a program called Efficient Consumer Response, and the U.S. food ser- vice industry embarked on an Efficient Foodservice Response program.

All those companies and initiatives persistently aimed at greater speed and cost- effectiveness-the popular grails of supply chain management. Of course, compa- nies’quests changed with the industrial cycle: When business was booming, exec- utives concentrated on maximizing speed, and when the economy headed south, firms desperately tried to minimize supply costs.

As time went by, however, I observed one fundamental problem that most com- panies and experts seemed to ignore: Ceteris paribus, companies whose supply chains became more efficient and cost-effective didn’t gain a sustainable advantage over their rivals. In fact, the performance of those supply chains steadily deterio- rated. For instance, despite the increased efficiency of many companies’ supply

102 HARVARD BUSINESS REVIEW

n

The 21st-Ceiitury Supply

chains, the percentage of products that were marked down in the United States rose from less than 10% in 1980 to more than 30% in 2000, and surveys show that con- sumer satisfaction with product availability fell sharply during the same period.

Evidently, it isn’t hy becoming more efficient that the supply chains of Wal-Mart, Dell, and Amazon have given those companies an edge over their competitors. Accord- ing to my research, top-performing supply chains possess three very different qualities. First, great supply chains are agile. They react speedily to sudden changes in demand or supply. Second,they adapt overtime as market structures and strategies evolve. Third, they align the interests of all the firms in the supply network so that companies opti- mize the chain’s performance when they maximize their interests. Only supply chains that are agile, adaptable, and aligned provide companies with sustainable competitive advantage.

The Perils of Efficiency

W hy haven’t efficient supply chains been able to deliver the goods? For several reasons. High- speed, low-cost supply chains are unable to re-

spond to unexpected changes in demand or supply. Many companies have centralized manufacturing and distribu- tion facilities to generate scale economies, and they de- liver only container loads of products to customers to minimize transportation time, freight costs, and the num- ber of deliveries. When demand for a particular brand, pack size, or assortment rises without warning, these or- ganizations are unable to react even if they have the items in stock. According to two studies I helped conduct in the 1990s, the required merchandise was often already in fac- tory stockyards, packed and ready to ship, but it couldn’t be moved until each container was full. That “best” prac- tice delayed shipments by a week or more,forcing stocked- out stores to turn away consumers. No wonder then that, according to another recent research report, when com- panies announce product promotions, stock outs rise to 15%, on average, even when executives have primed sup- ply chains to handle demand fluctuations.

When manufacturers eventually deliver additional merchandise, it results in excess inventory because most distributors don’t need a container load to satisfy the in- creased demand. To get rid of the stockpile, companies mark down those products sooner than they had planned

to. That’s partly why department stores sell as much as a third of their merchandise at discounted prices. Those markdowns not only reduce companies’ profits but also erode brand equity and anger loyal customers who bought the items at full price in the recent past (sound familiar?).

Companies’obsession with speed and costs also causes supply chains to break down during the launch of new products. Some years ago, I studied a well-known con- sumer electronics firm that decided not to create a buffer stock before launching an innovative new product. It wanted to keep inventory costs low, particularly since it hadn’t been able to generate an accurate demand fore- cast. When demand rose soon after the gizmo’s launch and fell sharply thereafter, the company pressured ven- dors to boost production and then to slash output. When demand shot up again a few weeks later, executives en- thusiastically told vendors to step up production once more. Five days later, supplies of the new product dried up as if someone had turned off a tap.

The shocked electronics giant discovered that vendors had been so busy ramping production up and down that they hadn’t found time to fix bugs in both the compo- nents’ manufacturing and the product’s assembly pro- cesses. When the suppliers tried to boost output a second time, product defects rose to unacceptable levels, and some vendors, including the main assembler, had to shut down production lines for more than a week. By the time the suppliers could fix the glitches and restart produc- tion, the innovation was all but dead. If the electronics company had given suppliers a steady, higher-than-needed manufacturing schedule until both the line and demand had stabilized, it would have initially had higher inven- tory costs, but the product would still be around.

Efficient supply chains often become uncompetitive because they don’t adapt to changes in the structures of markets. Consider Lucent’s Electronic Switching Systems division, which set up a fast and cost-effective supply chain in the late 1980s by centralizing component pro- curement, assembly and testing, and order fulfillment in Oklahoma City. The supply chain worked brilliantly as long as most of the demand for digital switches emanated from the Americas and as long as Lucent’s vendors were mostly in the United States. However, in the 1990s, when Asia became the world’s fastest-growing market, Lucent’s response times increased because it hadn’t set up a plant in the Far East. Furthermore, the company couldn’t cus- tomize switches or carry out modifications because of the

Hau L Lee (tiaulee(p)stanford.edu) is the Thoma Professor of Operations, Information, and Technology at the Stanford Grad- uate School of Business in Stanford, California, and the codirector of tbe Stanford Global Supply Chain Management Forum. He is the coeditor (with Terry P. Harrison andfohn }. Neale) of The Practice of Supply Chain Management: Where Theory and Application Converge (Kluwer Academic Publishers, 2003).

104 HARVARD BUSINESS REVIEW

-The Triple-A Supply Chain

amount of time and money it took the supply chain to do those things across continents.

Lucent’s troubles deepened when vendors shifted man- ufacturing facilities from the United States to Asia to take advantage of the lower labor costs there. “We had to fly components from Asia to Oklahoma City and fly them back again to Asia as finished products. That was costly and time consuming,” Lucent’s then head of manufacturing told me. With tongue firmly in cheek, he added,”Neither components nor products earned frequent-flyer miles.” When Lucent redesigned its supply chain in 1996 by set- ting up joint ventures in Taiwan and China to manufac- ture digital switches, it did manage to gain ground in Asia.

In this and many other cases, the conclusion would be the same: Supply chain efficiency is necessary, but it isn’t enough to ensure that firms will do better than their ri- vals. Only those companies that build agile, adaptable, and aligned supply chains get ahead of the competition, as I pointed out earlier. In the following pages, I’ll expand on each of those qualities and explain how companies can build them into supply chains without having to make trade-offs. In fact, I’ll show that any two of these dimen- sions alone aren’t enough. Only companies that build all three into supply chains become better faster than their rivals. I’ll conclude by describing how Seven-Eleven Japan has become one of the world’s most profitable retailers by building a truly “triple-A” supply chain.

Fostering Agility

G reat companies create supply chains that respond to sudden and unexpected changes in markets. Agility is critical, because in most industries, both

demand and supply fluctuate more rapidly and widely than they used to. Most supply chains cope by playing speed against costs, but agile ones respond both quickly and cost-efficiently.

Most companies continue to focus on the speed and costs of their supply chains without realizing that they pay a big price for disregarding agility. (See the sidebar “The Importance of Being Agile.”) In the 1990s, whenever Intel unveiled new microprocessors, Compaq took more time than its rivals to launch the next generation of PCs because of a long design cycle. The company lost mind share because it could never count early adopters, who create the buzz around high-tech products, among its con- sumers. Worse, it was unable to compete on price. Be- cause its products stayed in the pipeline for a long time, the company had a large inventory of raw materials. That meant Compaq didn’t reap much benefit when compo- nent prices fell, and it couldn’t cut PC prices as much as its rivals were able to. When vendors announced changes

Agility Ottjectives: Respond to short-term changes in demand or supply

quickly; handle external disruptions smoothly.

Methods: > Promote flow of information with suppliers and customers.

>Develop collaborative relationships with suppliers.

>Design for postponement.

>Buiid inventory buffers by maintaining a stoci<pile

of inexpensive but key components.

>Have a dependable logistics system or partner.

>Draw up contingency plans and deveiop crisis

management teams.

Adaptability Olijectives: Adjust supply chain’s design to meet structural shifts in

markets; modify supply network to strategies, products, and

technologies.

Methods: > Monitor economies all over the world to spot new supply

bases and markets,

>Use intermediaries to develop fresh suppliers and logistics

infrastructure.

>Evaluate needs of ultimate consumers-not just immediate

customers.

>Create flexible product designs.

>Determine where companies’products stand in terms of

technology cycles and product life cycles.

Alignment Objective: Create incentives for better performance.

Methods: >Exchange information and itnowledge freely with vendors

and customers.

> Lay down roles, tasks, and responsibilities clearly for sup-

pliers and customers.

>Equitably share risks, costs, and gains of improvement

initiatives.

OCIOHVR 3004 105

st-Contury Supply Chain.

in engineering specifications, Compaq incurred more re- working costs than other manufacturers because of its larger work-in-progress inventory. The lack of an agile supply chain caused Compaq to lose PC market share throughout the decade.

By contrast, smart companies use agile supply chains to differentiate themselves from rivals. For instance, H&M, Mango, and Zara have become Europe’s most profitable apparel brands by building agility into every link of their supply chains. At one end of their product pipelines, the three companies have created agile design processes. As soon as designers spot possible trends, they create sketches and order fabrics. That gives them a head start over com- petitors because fabric suppliers require the longest lead times. However, the companies finalize designs and man- ufacture garments only after they get reliable data from stores. That allows them to make products that meet con- sumer tastes and reduces the number of items they must sell at a discount. At the other end of the pipeline, all three companies have superefficient distribution centers. They use state-of-the-art sorting and material-handling tech- nologies to ensure that distribution doesn’t become a bot- tleneck when they must respond to demand fluctuations. H&M, Mango, and Zara have all grown at more than 20% annually since 1990, and their double-digit net profit mar- gins are the envy of the industry.

Agility has become more critical in the past few years because sudden shocks to supply chains have become frequent. The terrorist attack in New York in 2001, the dockworkers’ strike in California in 2002, and the SARS epidemic in Asia in 2003, for instance, disrupted many companies’ supply chains. While the threat from natural disasters, terrorism, wars, epidemics, and computer viruses has intensified in recent years, partly because supply lines now traverse the globe, my research shows that most sup- ply chains are incapable of coping with emergencies. Only three years have passed since 9/11, but U.S. companies have all but forgotten the importance of drawing up con- tingency plans for times of crisis.

Without a doubt, agile supply chains recover quickly from sudden setbacks. In September 1999, an earthquake in Taiwan delayed shipments of computer components to the United States by weeks and, in some cases, by months. Most PC manufacturers, such as Compaq, Apple, and Gate- way, couldn’t deliver products to customers on time and incurred their wrath. One exception was Dell, which changed the prices of PC configurations overnight. That allowed the company to steer consumer demand away from hardware built with components that weren’t avail- able toward machines that didn’t use those parts. Dell could do that because it got data on the earthquake dam- age early, sized up the extent of vendors’ problems quickly.

Most companies overlook the idea that supply chains

should be agile. That’s understandable; adaptability and

alignment are more novel concepts than agility is. How-

ever, even ifyour supply chain is both adaptable and

aligned, it’s dangerous to disregard agility.

!n 1995, Hewlett-Packard teamed up with Canon to

design and launch ink-jet printers. At tbe outset, the

American company aligned its interests with those of

its Japanese partner. While HP took on the responsibil-

ity of producing printed circuit boards (or”formaters”).

Canon agreed to manufacture engines for the LaserJet

series. That was an equitable division of responsibili-

ties, and the two R&D teams learned to work togetber

closely. After launching tbe LaserJet, HP and Canon

quickly adapted the supply network to the product’s

markets. HP used its manufacturing facilities in Idaho

and Italy to support tbe LaserJet, and Canon used

plants in West Virginia and Tokyo.

But HP and Canon failed to anticipate one problem.

To keep costs down. Canon agreed to alter the number

of engines it produced, but only if HP communicated

changes well in advance-say, six or more months be-

fore printers entered tbe market. However, HP could es-

timate demand accurately only three or fewer montbs

before printers hit the market. At that stage. Canon

could modify its manufacturing schedule by just a few

percentage points. As a result, the supply chain couldn’t

cope with sudden fluctuations in demand. So when

tbere was an unexpected drop in demand for the Laser-

Jet III toward tbe end of its life cycle, HP was stuck witb

a huge and expensive surplus of printer engines: the

infamous LaserJet mountain. Having an adaptable and

aligned supply chain didn’t help HP overcome its lack

of agility.

and implemented the plans it had drawn up to cope with such eventualities immediately. Not surprisingly. Dell gained market share in the earthquake’s aftermath.

Nokia and Ericsson provided a study in contrasts when in March 2000, a Philips facility in Albuquerque, New Mexico, went up in flames. The plant made radio fre- quency (RE) chips, key components for mobile tele- phones, for both Scandinavian companies. When the fire damaged the plant, Nokia’s managers quickly carried out

106 HARVARD BUSINESS REVIEW

The Triple-A Supply Chain

design changes so that other companies could manufac- ture similar RF chips and contacted backup sources. Two suppliers, one in Japan and another in the United States, asked for just five days’ lead time to respond to Nokia. Ericsson, meanwhile, had been weeding out backup sup- pliers because it wanted to trim costs. It didn’t have a plan B in place and was unable to find new chip suppliers. Not only did Ericsson have to scale back production for months after the fire, but it also had to delay the launch of a major new product. The bottom line: Nokia stole market share from Ericsson because it had a more agile supply chain.

Companies can build agility into supply chains by ad- hering to six rules of thumb:

• Provide data on changes in supply and demand to part- ners continuously so they can respond quickly. For in- stance, Cisco recently created an e-hub, which connects suppliers and the company via the Internet. This allows all the firms to have the same demand and supply data at the same time, to spot changes in demand or supply problems immediately, and to respond in a concerted fashion. Ensuring that there are no information delays is the first step in creating an agile supply chain.

• Develop collaborative relationships with suppliers and customers so that companies work together to design or redesign processes, components, and products as well as to prepare backup plans. For instance, Taiwan Semiconduc- tor Manufacturing Company (TSMC),the world’s largest semiconductor foundry, gives suppliers and customers proprietary tools, data, and models so they can execute design and engineering changes quickly and accurately.

• Design products so that they share common parts and processes initially and differ substantially only by the end ofthe production process. I call this strategy “postpone- ment.” (See the 1997 HBR article I coauthored with Ed- ward Feitzinger,”Mass Customization at Hewlett-Packard: The Power of Postponement”; This is often the best way to respond quickly to demand fluctuations because it al- lows firms to finish products only when they have accurate information on consumer preferences. Xilinx, the world’s largest maker of programmable logic chips, has perfected the art of postponement. Customers can program the com- pany’s integrated circuits via the Internet for different ap- plications after purchasing the basic product. Xilinx rarely runs into inventory problems as a result.

• Keep a small inventory of inexpensive, nonbulky com- ponents that are often the cause of bottlenecks. For ex- ample, apparel manufacturers H&M, Mango, and Zara maintain supplies of accessories such as decorative but- tons, zippers, hooks, and snaps so that they can finish clothes even if supply chains break down.

• Build a dependable logistics system that can enable your company to regroup quickly in response to unex-

pected needs. Companies don’t need to invest in logistics systems themselves to reap this benefit; they can strike alliances with third-party logistics providers.

• Put together a team that knows how to invoke backup plans. Of course, that’s only possible only if companies have trained managers and prepared contingency plans to tackle crises, as Dell and Nokia demonstrated.

Adapting Your Supply Chain

reat companies don’t stick to the same supply net- iworks when markets or strategies change. Rather, such organizations keep adapting their supply

chains so they can adjust to changing needs. Adaptation can be tough, but it’s critical in developing a supply chain that delivers a sustainable advantage.

Most companies don’t realize that in addition to unex- pected changes in supply and demand, supply chains also face near-permanent changes in markets. Those struc- tural shifts usually occur because of economic progress, political and social change, demographic trends, and technological advances. Unless companies adapt their supply chains, they won’t stay competitive for very long. Lucent twice woke up late to industry shifts, first to the rise of the Asian market and later to the advantages of outsourced manufacturing. (See the sidebar “Adaptation of the Fittest.”) Lucent recovered the first time, but the second time around, the company lost its leadership of

OCTOBER 2004 107

The 2JLst^^iitury Supply Chain.

f the rittest

Many executives ask me, with a twinkle in their eye,

if companies must really keep adapting supply chains.

Companies may find it tough to accept the idea that

they have to keep changing, but they really have no

choice.

Just ask Lucent. In the mid-1990s, when the Ameri-

can telecommunications giant realized that it could

make inroads in Asia oniy if had local manufacturing

facilities, it overhauled its supply chain. Lucent set up

plants in Taiwan and China, which allowed the com-

pany to customize switches as inexpensively and

quickly as rivals Siemens and Alcatel could. To align

the interests of parent and subsidiaries, Lucent execu-

tives stopped charging the Asian ventures inflated

prices for modules that the company shipped from the

United States. By the late 1990s, Lucent had recaptured

market share in China, Taiwan, India, and Indonesia.

Unhappily, the story doesn’t end there, because

Lucent stopped adapting its supply chain. The com-

pany didn’t realize that many medium-sized manufac-

turers had developed the technology and expertise to

produce components and suhassemblies for digital

switches and that because of economies of scale, they

could do so at a fraction ofthe integrated manufactur-

ers’costs. Realizing where the future lay, competitors

aggressively outsourced the manufacture of switching

systems. Because ofthe resulting cost savings, they

were able to quote lower prices than Lucent. Mean-

while, Lucent was reluctant to outsource its manufac-

turing because it had Invested in its own factories.

Ultimately, however, Lucent had no option but to shut

down its Taiwan factory in 2002 and create an out-

sourced supply chain. The company’s adaptation came

too late for Lucent to regain control of the global mar-

ket, even though the supply chain was agile and

aligned.

the global telecommunications market because it didn’t adapt quickly enough.

The best supply chains identify structural shifts, some- times before they occur, by capturing the latest data, fil- tering out noise, and tracking key patterns. They then re- locate facilities, change sources of supplies, and, if possible, outsource manufacturing. For instance, when Hewlett- Packard started making ink-jet printers in the 1980s, it set up both its R&D and manufacturing divisions in Vancou-

ver, Washington. HP wanted the product development and production teams to work together because ink-jet technology was in its infancy, and the biggest printer mar- ket was in the United States. When demand grew in other parts of the world, HP set up manufacturing facilities in Spain and Singapore to cater to Europe and Asia. Al- though Vancouver remained the site where HP developed new printers, Singapore became the largest production facility because the company needed economies of scale to survive. By the mid-1990s, HP realized that printer- manufacturing technologies had matured and that it could outsource production to vendors completely. By doing so, HP was able to reduce costs and remain the leader in a highly competitive market.

Adaptation needn’t be just a defensive tactic. Compa- nies that adapt supply chains when they modify strategies often succeed in launching new products or breaking into new markets. Three years ago, when Microsoft decided to enter the video game market, it chose to outsource hard- ware production to Singapore-based Flextronics. In early 2001, the vendor learned that the Xbox had to be in stores before December because Microsoft wanted to target Christmas shoppers. Flextronics reckoned that speed to market and technical support would be crucial for en- suring the product’s successful launch. So it decided to make the Xbox at facilities in Mexico and Hungary. The sites were relatively expensive, but they boasted engi- neers who could help Microsoft make design changes and modify engineering specs quickly. Mexico and Hungary were also close to the Xbox’s biggest target markets, the United States and Europe. Microsoft was able to launch the product in record time and mounted a stiff challenge to market leader Sony’s PlayStation 2. Sony fought back by offering deep discounts on the product Realizing that speed would not be as critical for medium-term survival as costs would be, Flextronics shifted the Xbox’s supply chain to China. The resulting cost savings allowed Micro- soft to match Sony’s discounts and gave it a fighting chance. By 2003, the Xbox had wrested a 20% share of the video game market from PlayStation 2.

Smart companies tailor supply chains to the nature of markets for products. They usually end up with more than one supply chain, which can be expensive, but they also get the best manufacturing and distribution capabil- ities for each offering. For instance, Cisco caters to the de- mand for standard, high-volume networking products by commissioning contract manufacturers in low-cost coun- tries such as China. For its wide variety of mid-value items, Cisco uses vendors in low-cost countries to build core products but customizes those products itself in major markets such as the United States and Europe. For highly customized, low-volume products, Cisco uses ven-

108 HARVARD BUSINESS REVIEW

dors close to main markets, such as Mexico for the United States and Eastern European countries for Europe. De- spite the fact that it uses three different supply chains at the same time, the company is careful not to become less agile. Because it uses flexible designs and standardized processes, Cisco can switch the manufacture of products from one supply network to another when necessary.

Gap, too, uses a three-pronged strategy. It aims the Old Navy brand at cost-conscious consumers, the Gap line at trendy buyers, and the Banana Republic collection at con- sumers who want clothing of higher quality. Rather than using the same supply chain for all three brands. Gap set up Old Navy’s manufacturing and sourcing in China to ensure cost efficiency. Gap’s chain in Central America to guarantee speed and flexibility, and Banana Republic’s supply network in Italy to maintain quality. The com- pany consequently incurs higher overheads, lower scale economies in purchasing and manufacturing, and larger transportation costs than it would if it used just one supply chain. However, since its brands cater to different

jrha_Trlpla-A Supply

fancy. The Japanese automobile maker had expertise in tracking U.S. trends and geographical preferences, but it felt that it would be difficult to predict consumer re- sponse to a hybrid car. Besides, the Prius might appeal to particular consumer segments, such as technophiles and conservationists, which Toyota didn’t know much about. Convinced that the uncertainties were too great to allo- cate the Prius to dealers based on past trends, Toyota de- cided to keep inventory in central stockyards. Dealers took orders from consumers and communicated them via the Internet. Toyota shipped cars from stockyards, and dealers delivered them to buyers.

Although Toyota’s transportation costs rose, it cus- tomized products to demand and managed inventory flawlessly. In 2002, for example, the number of Toyotas on the road in Northern California and the Southeast were 7% and 20%, respectively. However, Toyota sold 25% of its Prius output in Northern California and only 6% in the Southeast. Had Toyota not adapted its distribution sys- tem to the product, it would have faced stock outs in

The best supply chains identify structural shifts, sometimes before they occur, by capturing the latest data, filtering out noise, and tracking key patterns.

consumer segments. Gap uses different kinds of supply networks to maintain distinctive positions. The adapta- tion has worked. Many consumers don’t realize that Gap owns all three brands, and the three chains serve as back- ups in case of emergency.

Sometimes it’s difficult for companies to define the ap- propriate markets, especially when they are launching innovative new products. The trick is to remember that products embody different levels of technology. For in- stance, after records came cassettes and then CDs. Video- tapes were followed by DVDs, and almost anything ana- log is now or will soon become digital. Also, every product is at a certain stage of its life cycle, whether it’s at the infant, ramp-up, mature, or end-of-life stage. By mapping either or both of those characteristics to supply chain partners, manufacturing network, and distribution sys- tem, companies can develop optimal supply chains for every product or service they offer.

For example, Toyota was convinced that the market for the Prius, the hybrid car it launched in the United States in 2000, would be different from that of other models because it embodied new technologies and was in its in-

Northern California and been saddled with excess inven- tory in the Southeast, which may well have resulted in the product’s failure.

Building an adaptable supply chain requires two key components: the ability to spot trends and the capability to change supply networks. To identify future patterns, it’s necessary to follow some guidelines:

• Track economic changes, especially in developing countries, because as nations open up their economies to global competition, the costs, skills, and risks of global supply chain operations change. This liberalization results in the rise of specialized firms, and companies must peri- odically check to see if they can outsource more stages of operation. Before doing so, however, they must make sure that the infrastructure to link them with vendors and cus- tomers is in place. Global electronics vendors, such as Flextronics, Solectron, and Foxcom, have become adept at gathering data and adapting supply networks.

• Decipher the needs of your ultimate consumers – not just your immediate customers. Otherwise, you may fall victim to the “bullwhip effect,” which amplifies and distorts demand fluctuations. For years, semiconductor

OCTOBER 2004 109

HBR Spotlight

The 21st-C?entury Simply CShai

manufacturers responded to customer forecasts and cre- ated gluts in markets. But when they started tracking de- mand for chip-based products, the manufacturers over- came the problem. For instance, in 2003, there were neither big inventory buildups nor shortages of semiconductors.

At the same time, companies must retain the option to alter supply chains. To do that, they must do two things:

• They must develop new suppliers that complement existing ones. When smart firms work in relatively un- known parts of the world, they use intermediaries like Li & Fung, the Hong Kong-based supply chain architects, to find reliable vendors.

• They must ensure that product design teams are aware of the supply chain implications of their designs. Designers must also be familiar with the three design-for- supply principles: commonality, which ensures that prod- ucts share components; postponement, which delays the step at which products become different; and standard- ization, which ensures that components and processes for different products are the same. These principles allow firms to execute engineering changes whenever they adapt supply chains.

Creating the Right Alignment

G reat companies take care to align the interests of all the firms in their supply chain with their own. That’s critical, because every firm – be it a sup-

plier, an assembler, a distributor, or a retailer – tries to maximize only its own interests. (See the sidebar “The Confinement of Nonalignment”) If any company’s inter- ests differ from those of the other organizations in the supply chain, its actions will not maximize the chain’s performance.

Misaligned interests can cause havoc even if supply chain partners are divisions of the same company, as HP discovered. In the late 1980s, HP’s integrated circuit (IC) division tried to carry as little inventory as possible, partly because that was one of its key success factors. Those low inventory levels often resulted in long lead times in the supply of lCs to HP’s ink-jet printer division. Since the di- vision couldn’t afford to keep customers waiting, it cre- ated a large inventory of printers to cope with the lead times in supplies. Both divisions were content, but from HP’s viewpoint, it would have been far less expensive to have a greater inventory of lower-cost ICs and fewer stocks of expensive primers. That didn’t happen, simply because HP’s supply chain didn’t align the interests of the divisions with those of the company.

Lack of alignment causes the failure of many supply chain practices. For example, several high-tech compa- nies, including Flextronics, Solectron, Cisco, and 3Com,

The Confinement of Nonalignment

It’s not easy for executives to accept that different firms

in the same supply chain can have different interests,

or that interest nonalignment can lead to inventory

problems as dire as those that may arise through a lack

of agility or a lack of adaptability. But the story of

Cisco’s supply chain clinches the argument.

All through the 1990s, everyone regarded Cisco’s

supply chain as almost infallible. The company was

among the first to make use of the Internet to commu-

nicate with suppliers and customers, automate work

fiows among trading partners, and use solutions such

as remote product testing, which allowed suppliers to

deliver quality results with a minimum of manual

input. Cisco outsourced the manufacturing of most of

its networking products and worked closely with con-

tract manufacturers to select the right locations to sup-

port its needs. Ifever there were a supply chain that

was agile and adaptable, Cisco’s was it.

Why then did Cisco have to write off $2.25 billion of

inventory in 2001? There were several factors at play,

but the main culprit was the misalignment of Cisco’s

interests with those of its contract manufacturers. The

contractors accumulated a large amount of inventory

for months without factoring in the demand for Cisco’s

products. Even when the growth of the U.S. economy

slowed down,the contractors continued to produce

and store inventory at the same pace. Finally, Cisco

found it couldn’t use most of the inventory of raw ma-

terials because demand had fallen sharply. The com-

pany had to sell the raw materials off as scrap.

have set up supplier hubs close to their assembly plants. Vendors maintain just enough stock at the hubs to sup- port manufacturers’ needs, and they replenish the hubs without waiting for orders. Such vendor-managed inven- tory (VMI) systems allow suppliers to track the consump- tion of components, reduce transportation costs, and, since vendors can use the same hub to support several manufacturers, derive scale benefits. When VMI offers so many advantages, why hasn’t it aiways reduced costs?

The problem starts with the fact that suppliers own components until they physically enter the manufactur- ers’ assembly plants and therefore bear the costs of in- ventories for longer periods than they used to. Many sup-

110 HARVARD BUSINESS REVIEW

pliers are small and medium-sized companies that must borrow money to finance inventories at higher interest rates than large manufacturers pay. Thus, manufacturers have reduced costs by shifting the ownership of invento- ries to vendors, but supply chains bear higher costs be- cause vendors’ costs have risen. In fact, some VMI systems have generated friction because manufacturers have re- fused to share costs with vendors.

One way companies align their partners’ interests with their own is by redefining the terms of their relationships so that firms share risks, costs, and rewards equitably. For instance, the world’s largest printer, RR Donnelley (which prints this magazine) recognized in the late 1990s that its supply chain performance relied heavily on paper-and-ink suppliers. If the quality and reliability of supplies im- proved, the company could reduce waste and make deliv- eries to customers on time. Like many other firms, RR Donnelley encouraged suppliers to come up with sugges- tions for improving processes and products. To align their interests with its own, however, the company also offered to split any resulting savings with suppliers. Not surpris- ingly, supplier-initiated improvements have helped en- hance RR Donnelley’s supply chain ever since.

Sometimes the process of alignment involves the use of intermediaries. In the case of VMI, for instance, some financial institutions now buy components from suppli- ers at hubs and sell them to manufacturers. Everyone benefits because the intermediaries’ financing costs are lower than the vendors’ costs. Although such an arrange- ment requires trust and commitment on the part of suppliers, financial intermediaries, and manufacturers, it is a powerful way to align the interests of companies in supply chains.

Automaker Saturn’s service parts supply chain, one of the best in the industry, is a great example of incentive alignment that has led to outstanding results. Instead of causing heartburn, the system works well because Saturn aligned the interests of everyone in the chain-especially consumers.

Saturn has relieved car dealers of the burden of man- aging service parts inventories. The company uses a cen- tral system to make stocking and replenishment decisions for dealers, who have the right to accept, reject, or modify the company’s suggestions. Saturn doesn’t just monitor its performance in delivering service parts to dealers, even though that is the company’s only responsibility. Instead, Saturn holds its managers and the dealers jointly ac- countable for the quality of service the vehicle owners ex- perience. For example, the company tracks the off-the- shelf availability of parts at the dealers as the relevant metric. Saturn also measures its Service Parts Operation (SPO) division on the profits that dealers make from ser-

_The Triplo-A Supply Chain

vice parts as well as on the number of emergency orders that dealers place. That’s because when a dealer doesn’t have a part, Saturn transfers it from another dealer and bears the shipping costs. The SPO division can’t overstock dealers because Saturn shares the costs of excess inven- tory with them. If no one buys a particular part from a dealer for nine months, Satum will buy it back as obsolete inventory.

That kind of alignment produces two results. First, everyone in the chain has the same objective: to deliver the best service to consumers. While the off-the-shelf availability of service parts in the automobile industry ranges from 70% to 80%, service part availability at Sat- urn’s dealers is 92.5%. After taking transfers from other retailers into account, the same-day availability of spare parts is actually 94%. Second, the right to decide about inventory replenishment rests with Satum, which is in the best position to make those decisions. The company shares the risks of stock outs or overstocks with dealers, so it has an interest in making the best possible decisions. Fittingly, the inventory turnover (a measure of how effi- cient inventory management is, calculated by dividing the annual cost of inventory sold by the average inven- tory) of spare parts at Saturn’s dealers is seven times a year while it is only between one and five times a year for other automobile companies’dealers.

Like Saturn, clever companies create alignment in sup- ply chains in several ways. They start with the alignment of information, so that all the companies in a supply chain have equal access to forecasts, sales data, and plans. Next they align identities; in other words, the manufacturer must define the roles and responsibilities of each partner so that there is no scope for confiict. Then companies must align incentives, so that when companies try to max- imize returns, they also maximize the supply chain’s per- formance. To ensure that happens, companies must try to predict the possible behavior of supply chain partners in the light of their current incentives. Companies often per- form such analyses to predict what competitors would do if they raised prices or entered a new segment; they need to do the same with their supply chain partners. Then they must redesign incentives so partners act in ways that are closer to what’s best for the entire supply chain.

Seven-Eleven Japan’s Three Aces

S even-Eleven Japan (SEJ) is an example of how a company that builds its supply chain on agility, adaptability, and alignment stays ahead of its ri-

vals. The $21 billion convenience store chain has remark- ably low stock out rates and in 2004 had an inventory turnover of 55. With gross profit margins of 30%, SEJ is also

OCTOBER 2004 111

The 2 X s t -Centui^Siipply Chain

one ofthe most profitable retajiers in the world. Just how has the 9,000-store retailer managed to sustain perfor- mance for more than a decade?

The company has designed its supply chain to respond to quick changes in demand-not to focus on fast or cheap deliveries. It has invested in real-time systems to detect changes in customer preference and tracks data on sales and consumers (gender and age) at every store. Well hefore the Internet era began, SEJ used satellite connections and ISDN lines to link all its stores with distribution centers, suppliers, and logistics providers. The data allow the supply chain to detect fluctuations in demand between stores, to alert suppliers to potential shifts in requirements, to help reallocate inventory among stores, and to ensure that the company restocks at the right time. SEJ schedules deliver- ies to each store within a ten-minute margin. If a truck is late by more than 30 minutes, the carrier has to pay a penalty equal to the gross margin ofthe products carried to the store. Employees reconfigure store shelves at least three

Fundamental to the supply chain’s operation is the close alignment between Seven-Eleven Japan’s interests and those of its partners. The incentives and disincentives are clear: Make Seven Eleven Japan successful, and share the rewards. Fail to deliver on time, and pay a penalty. That may seem harsh, but the company balances the equation by trusting its partners. For instance, when carriers deliver products to stores, no one verifies the truck’s contents. That allows carriers to save time and money, since drivers don’t have to wait after dropping off merchandise.

When Seven-Eleven Japan spots business opportuni- ties, it works with suppliers to develop products and shares revenues with them. For instance, two years ago, SEJ created an e-commerce company, 7dream.com, with six partners. The new organization allows consumers to order products online or through kiosks at SEJ stores and pick up the merchandise at any Seven-Eleven. The partners benefit from SEJ’s logistics network, which de- livers products to stores efficiently, as well as from the

The message to Seven-Eleven Japan’s partners is clear: Make the company successful, and share the rewards. Fail to deliver on time, and pay a penalty.

times daily so that storefronts cater to different consumer segments and demands at different hours.

SEJ has adapted its supply chain to its strategy over time. Some years ago, the company decided to concen- trate stores in key locations instead of building outlets all over the country. But doing so increased the possi- bility of traffic congestion every time the company re- plenished stores. The problem became more acute when SEJ decided to resupply stores three or more times a day. To minimize delays due to traffic snarls, the company adapted its distribution system. It asked its suppliers from the same region to consolidate shipments in a single truck instead of using several of them. That minimized the number of trucks going to its distribution centers, which is where SEJ cross-docks products for delivery to stores. The company has also expanded the kinds of vehicles it uses from trucks to motorcycles, boats, and even heli- copters. The effectiveness of the company’s logistics system is legendary. Less than six hours after the Kobe earthquake on January 17,1995, when relief trucks were crawling at two miles per hour on the highways, SEJ used seven helicopters and 125 motorcycles to deliver 64,000 rice balls to the city-

convenient location of stores. By encouraging partners to set up multimedia kiosks to produce games, tickets, or CDs in its shops, Seven-Eleven Japan has become a man- ufacturing outlet for partners. The company could not have aligned the interests of its partners more closely with those of its own.

• * * When 1 describe the triple-A supply chain to companies, most of them immediately assume it will require more technology and investment. Nothing could be further from the truth. Most firms already have the infrastructure in place to create triple-A supply chains. What they need is a fresh attitude and a new culture to get their supply chains to deliver triple-A performance. Companies must give up the efficiency mind-set, which is counterproduc- tive; be prepared to keep changing networks; and, instead of looking out for their interests alone, take responsibility for the entire chain. This can be challenging for compa- nies because there are no technologies that can do those things; only managers can make them happen. ^

Reprint R0410F; HBR OnPoint 8096 To order, see page 159.

112 HARVARD BUSINESS REVIEW

Harvard Business Review Notice of Use Restrictions, May 2009

Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for

the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material

in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may

not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any

other means of incorporating the content into course resources. Business licensees may not host this content on

learning management systems or use persistent linking or other means to incorporate the content into learning

management systems. Harvard Business Publishing will be pleased to grant permission to make this content

available through such means. For rates and permission, contact permissions@harvardbusiness.org.