Managing New Product And Process Development: Text And Cases

Available through our Marketplace sellers.
Other sellers (Hardcover)
  • All (12) from $1.99   
  • New (2) from $70.56   
  • Used (10) from $1.99   
Sort by
Page 1 of 1
Showing All
Note: Marketplace items are not eligible for any coupons and promotions
Seller since 2008

Feedback rating:



New — never opened or used in original packaging.

Like New — packaging may have been opened. A "Like New" item is suitable to give as a gift.

Very Good — may have minor signs of wear on packaging but item works perfectly and has no damage.

Good — item is in good condition but packaging may have signs of shelf wear/aging or torn packaging. All specific defects should be noted in the Comments section associated with each item.

Acceptable — item is in working order but may show signs of wear such as scratches or torn packaging. All specific defects should be noted in the Comments section associated with each item.

Used — An item that has been opened and may show signs of wear. All specific defects should be noted in the Comments section associated with each item.

Refurbished — A used item that has been renewed or updated and verified to be in proper working condition. Not necessarily completed by the original manufacturer.

0029055172 New. Has the slightest of shelf wear (like you might see in a major bookstore chain). Looks like an interesting title! We provide domestic tracking upon request, ... provide personalized customer service and want you to have a great experience purchasing from us. 100% satisfaction guaranteed and thank you for your consideration. Read more Show Less

Ships from: Chicago, IL

Usually ships in 1-2 business days

  • Standard, 48 States
  • Standard (AK, HI)
Seller since 2008

Feedback rating:


Condition: New
1992 Hard cover New. No dust jacket as issued & Inventory mark on the edge. Sewn binding. Cloth over boards. 896 p. Contains: Illustrations. Audience: General/trade.

Ships from: Orlando, FL

Usually ships in 1-2 business days

  • Canadian
  • International
  • Standard, 48 States
  • Standard (AK, HI)
  • Express, 48 States
  • Express (AK, HI)
Page 1 of 1
Showing All
Sort by
Managing New Product and Process Development: Text Cases

Available on NOOK devices and apps  
  • NOOK Devices
  • Samsung Galaxy Tab 4 NOOK 7.0
  • Samsung Galaxy Tab 4 NOOK 10.1
  • NOOK HD Tablet
  • NOOK HD+ Tablet
  • NOOK eReaders
  • NOOK Color
  • NOOK Tablet
  • Tablet/Phone
  • NOOK for Windows 8 Tablet
  • NOOK for iOS
  • NOOK for Android
  • NOOK Kids for iPad
  • PC/Mac
  • NOOK for Windows 8
  • NOOK for PC
  • NOOK for Mac
  • NOOK for Web

Want a NOOK? Explore Now

NOOK Book (eBook)
$19.99 price

Product Details

  • ISBN-13: 9780029055175
  • Publisher: Free Press
  • Publication date: 1/28/1992
  • Edition description: New Edition
  • Edition number: 1
  • Pages: 896
  • Product dimensions: 6.41 (w) x 9.60 (h) x 1.87 (d)

Table of Contents

Ch. 1 Competing Through Development Capability 1
Ch. 2 The Concept of a Development Strategy 81
Ch. 3 Maps and Mapping: Functional Strategies in Pre-Project Planning 155
Ch. 4 The Aggregate Project Plan 233
Ch. 5 Structuring the Development Funnel 291
Ch. 6 A Framework for Development 363
Ch. 7 Cross-Functional Integration 457
Ch. 8 Organizing and Leading Project Teams 519
Ch. 9 Tools and Methods 595
Ch. 10 Prototype/Test Cycles 657
Ch. 11 Learning from Development Projects 731
Ch. 12 Building Development Capability 799
Notes 873
Index 881
Read More Show Less

First Chapter

Chapter 1 Competing Through Development Capability


This chapter introduces product development as a central focus of competition in the 1990s. While firms have developed new products since the Industrial Revolution, in industry after industry, the importance of doing product development well has increased dramatically in recent years. This chapter identifies the forces driving the importance of product development -- changes in competition, customer demands, and technology. An important theme in the chapter is that these forces have created a competitive imperative for speed, efficiency, and high quality in the development process.

In reading the chapter it is important to establish a basic idea of what product development involves -- both what makes it difficult to achieve, and the competitive power it creates when done well. To provide perspective on what we mean by product development, the chapter briefly summarizes the major sequence of activities involved in taking an idea from initial concept through prototype building and testing, and into commercial production. A key theme is that product development is a process involving all the major functions in a business. With the development process as background, we then use the example of the Northern Electronics Company and its problems with the A14 stereo project to illustrate the difficulties in development.

The problems on the A14 project -- missed schedules, cost overruns, and a poorly designed product -- reflect a mismatch between the way the project is organized and managed and the requirements of the development process created by the product's complexity and the rigorous and uncertain competitive environment in which Northern Electronics competed. Exhibit 1-5 summarizes the characteristics of problematic projects as well as their consequences. The exhibit also identifies key themes that characterize outstanding projects -- clarity of focus, integration across functions, a strong focus on time to market, doing things right the first time, and effective substantive leadership -- thus summarizing many of the important themes developed in the book.

Our intent in this first chapter is not only to highlight the challenge and characterize what an outstanding project might look like, but also to illustrate the competitive power created in the organizations that do development extraordinarily well. To underscore that power we close the chapter with a review of the competitive interaction between Northern and Southern Electronics in the compact stereo market. Historically these two companies mirrored one another in terms of their market approach. But in the 1980s, Southern built a new strategy around superior capability and product development. In effect, Southern embarked on a strategy to become a fast-cycle competitor. In reading through this history, it is useful to note the way in which Southern linked its product development capability with its strategies in marketing and manufacturing. In fact, the way Southern exploited its advantage in speed and efficiency over its slower Northern rival was precisely by integrating its development capabilities with its actions in marketing and manufacturing. The history also sheds light on the A14 stereo project referred to above. Here, we see what happens when a senior management team attempts to achieve substantial improvements in performance without making basic changes in processes or in capabilities. The chapter closes with a summary of the advantages that effective product development capability conferred upon Southern.

In a competitive environment that is global, intense, and dynamic, the development of new products and processes increasingly is a focal point of competition. Firms that get to market faster and more efficiently with products that are well matched to the needs and expectations of target customers create significant competitive leverage. Firms that are slow to market with products that match neither customer expectations nor the products of their rivals are destined to see their market position erode and financial performance falter. In a turbulent environment, doing product and process development well has become a requirement for being a player in the competitive game; doing development extraordinarily well has become a competitive advantage.

The New Industrial Competition: Driving Forces and Development Realities

The importance of product and process development is not limited to industries or businesses built around new scientific findings, with significant levels of R&D spending, or where new products have traditionally accounted for a major fraction of annual sales. The forces driving development are far more general. Three are particularly critical:

* Intense international competition. In business after business, the number of competitors capable of competing at a world-class level has grown at the same time that those competitors have become more aggressive. As world trade has expanded and international markets have become more accessible, the list of one's toughest competitors now includes firms that may have grown up in very different environments in North America, Europe, and Asia. The effect has been to make competition more intense, demanding, and rigorous, creating a less forgiving environment.

* Fragmented, demanding markets. Customers have grown more sophisticated and demanding. Previously unheard of levels of performance and reliability are today the expected standard. Increasing sophistication means that customers are more sensitive to nuances and differences in a product, and are attracted to products that provide solutions to their particular problems and needs. Yet they expect these solutions in easy-to-use forms.

* Diverse and rapidly changing technologies. The growing breadth and depth of technological and scientific knowledge has created new options for meeting the needs of an increasingly diverse and demanding market. The development of novel technologies and a new understanding of existing technologies increases the variety of possible solutions available to engineers and marketers in their search for new products. Furthermore, the new solutions are not only diverse, but also potentially transforming. New technologies in areas such as materials, electronics, and biology have the capacity to change fundamentally the character of a business and the nature of competition.

These forces are at work across a wide range of industries. They are central to competition in young, technically dynamic industries, but also affect mature industries where life cycles historically were relatively long, technologies mature, and demands stable. In the world auto industry, for example, the growing intensity of international competition, exploding product variety, and diversity in technology have created a turbulent environment. The number of world-scale competitors has grown from less than five in the early 1960s to more than twenty today. But perhaps more importantly, those twenty competitors come from very different environments and possess a level of capability far exceeding the standard prevailing twenty-five years ago. Much the same is true of customers. Levels of product quality once considered extraordinary are now a minimum requirement for doing business. As customers have grown more sophisticated and demanding, the variety of products has increased dramatically. In the mid 1960s, for example, the largest selling automobile in the United States was the Chevrolet Impala. The platform on which it was based sold approximately 1.5 million units per year. In 1991, the largest selling automobile in the United States was the Honda Accord, which sold about 400,000 units. Thus, in a market that is today larger than it was in 1965, the volume per model has dropped by a factor of four. Currently over 600 different automobile models are offered for sale on the U.S. market.

Similarly, technological change has had dramatic consequences. In 1970, one basic engine-drive train technology (a V8 engine, longitudinally mounted, water cooled, carbureted, hooked up to a three-speed automatic transmission with rear wheel drive) accounted for close to 80 percent of all automobile production in the United States. Indeed, there were only five engine-drive train technologies in production. By the early 1980s that number had grown to thirty-three. The growing importance of electronics, new materials, and new design concepts in engines, transmissions, suspensions, and body technologies has accelerated the pace and diversity of technological change in the 1980s. Simply keeping up with those technologies is a challenge, but an often straightforward one in comparison with having to integrate them in development efforts.

Similar forces have been at work in other traditional, mature industries. In textiles and apparel, for example, firms such as Benetton and The Limited have used information technology to create a production and distribution network which links retail outlets directly to distribution centers and back into factories and suppliers in the chain of production from fiber to finished product. The thrust of these networks is the ability to respond quickly to changing customer demands at relatively low cost. Fueled in part by availability and in part by growing demands for differentiated products, product variety has expanded significantly. In plant after plant, one finds vast increases in the number of styles produced and a sharp decline in the length of production runs. These are not changes of 10 or 20 percent; in the 1980s, it was common for apparel plants to experience a four- to fivefold increase in the number of styles produced. These increases in garment variety have pushed back into the textile plants as well. For example, the average lot size for dying at Greenwood Mills, a U.S. textile firm, declined in the 1980s from 120,000 to 11,000 yards.

Changes in markets and technologies for automobile and textile firms have accentuated the importance of speed and variety in product development. But changes in competition, customer demand, and technology have also had dramatic effects on newer, less mature industries in which product innovation has always been an important part of competition. In industries such as computer disk drives and medical equipment, already short life cycles have shrunk further and product variety has increased. In addition, competition has placed increased pressure on product reliability and product cost. In disk drives, for example, the market for Winchester-technology hard disks has expanded from a base in high-end systems for mainframe computers to include a spectrum of applications ranging from notebook personal computers to large-scale supercomputers. Even within an application segment, the number of sizes, capacities, access times, and features has increased sharply. In addition to this explosion of variety, firms in the hard disk drive industry have had to meet demands for dramatic increases in reliability (tenfold in five years) and decreases in cost (5 percent to 8 percent quarterly). These have been met in part by incremental improvements in established technologies and in part through the introduction of new design concepts, production technologies, materials, and software.

Much the same has been true in the market for new medical devices. Innovation has always been important in the creation of new medical devices, but by the 1980s success required the ability to follow an innovative product with sustained improvements in performance, application to new segments, improved reliability, and lower cost. In the case of devices for angioplasty (a procedure using a balloon on a small wire to expand clogged arteries), the initial innovation was followed by a variety of developments that offered the physician greater control of a smaller device, making access easier and creating additional applications. In concert with process changes that substantially improved or reduced variability of performance characteristics, changes in the product have opened up new applications and treatment of a more diverse set of clinical problems and patients, worldwide.

The Competitive Imperatives

Rigorous international competition, the explosion of market segments and niches, and accelerating technological change have created a set of competitive imperatives for the development of new products and processes in industries as diverse as medical instruments and automobiles, textiles, and high-end disk drives. Exhibit 1-1 identifies three of these imperatives -- speed, efficiency, and quality -- and suggests some of their implications. To succeed, firms must be responsive to changing customer demands and the moves of their competitors. This means that they must be fast. The ability to identify opportunities, mount the requisite development effort, and bring to market new products and processes quickly is critical to effective competition. But firms also must bring new products and processes to market efficiently. Because the number of new products and new process technologies has increased while model lives and life cycles have shrunk, firms must mount more development projects than has traditionally been the case utilizing substantially fewer resources per project. In the U.S. automobile market, for example, the growth of models and market segments over the last twenty-five years has meant that an auto firm must mount close to four times as many development projects simply to maintain its market share position. But smaller volumes per model and shorter design lives mean resource requirements must drop dramatically. Effective competition requires highly efficient engineering, design, and development activities.

Being fast and efficient is essential but not enough. The products and processes that a firm introduces must also meet demands in the market for value, reliability, and distinctive performance. Demanding customers and capable competitors mean that the ante keeps going up -- requirements of performance, reliability, ease of use, and total value increase with each product introduction. When competition is intense firms must attract and satisfy customers in a very crowded market. More and more this means offering a product that is distinctive; that not only satisfies, but also surprises and delights a customer. Moreover, attention to the total product experience and thus to total product quality is critical.

The Opportunity and the Challenge

Firms that step up to the challenge and meet these competitive imperatives enjoy a significant advantage in the market place. The development of outstanding products not only opens new markets and attracts new customers, but also leverages existing assets and builds new capability in the organization. Getting a succession of distinctive new disk drives or a string of new medical devices to market quickly and consistently requires the solution of technical problems that builds know-how. Moreover, it stimulates the creation of greater capability in problem solving, prototype construction, and testing that can be applied in future projects. All of these skills and capabilities enhance a firm's ability to compete. But there is more. Successful new products also unleash a virtuous cycle in reputation and enthusiasm within and outside the organization. Inside, successful new products energize the organization; confidence, pride, and morale grow. The best employees remain challenged and enthused. Outside, outstanding new products create broad interest in the firm and its products, enhance the firm's ability to recruit new employees, and facilitate the building of relationships with other organizations. The organization's momentum builds and reinforces itself.

While the potential opportunities to be realized in developing new products and processes are exciting, making them happen is a demanding challenge. New product or process development entails a complex set of activities that cuts across most functions in a business, as suggested by Exhibit 1-2, which lays out the phases of activity in a typical development project -- a new product. In the first two phases -- concept development and product planning -- information about market opportunities, competitive moves, technical possibilities, and production requirements must be combined to lay down the architecture of the new product. This includes its conceptual design, target market, desired level of performance, investment requirements, and financial impact. Before a new product development program is approved, firms also attempt to prove out the concept through small-scale testing, the construction of models, and, often, discussions with potential customers.

Once approved, a new product project moves into detailed engineering. The primary activity in this phase of development is the design and construction of working prototypes and the development of tools and equipment to be used in commerical production. At the heart of detailed product and process engineering is the "design-build-test" cycle. Both products and processes are laid out in concept, captured in a working model (which may exist on a computer or in physical form), and then subjected to tests that simulate product use. If the model fails to deliver the desired performance characteristics, engineers search for design changes that will close the gap and the design-build-test cycle is repeated. The conclusion of the detailed engineering phase of development is marked by an engineering "release" or "sign off" that signifies that the final design meets requirements.

At this time the firm typically moves development into a pilot manufacturing phase, during which the individual components, built and tested on production equipment, are assembled and tested as a system in the factory. During pilot production many units of the product are produced and the ability of the new or modified manufacturing process to execute at a commerical level is tested. At this stage all commercial tooling and equipment should be in place and all parts suppliers should be geared up and ready for volume production. This is the point in development at which the total system -- design, detailed engineering, tools and equipment, parts, assembly sequences, production supervisors, operators, and technicians -- comes together.

The final phase of development is ramp-up. The process has been refined and debugged, but has yet to operate at a sustained level of high-yield, volume production. In ramp-up the firm starts commerical production at a relatively low level of volume; as the organization develops confidence in its (and its suppliers') abilities to execute production consistently and marketing's abilities to sell the product, the volume increases. At the conclusion of the ramp-up phase, the production system has achieved its target levels of volume, cost, and quality. In this phase, the firm produces units for commercial sale and, hopefully, brings the volume of production up to its targeted level.

An obstacle to achieving rapid, efficient, high-quality development is the complexity and uncertainty that confronts engineers, marketers, and manufacturers. At a fundamental level the development process creates the future, and that future is often several years away. Consider, for example, the case of a new automobile. The very best companies in the world in 1990 could develop a new car in three to three and a half years. At the outset of a new car development program, therefore, designers, engineers, and marketers must conceive of a product that will attract customers three years into the future. But that product must also survive in the marketplace for at least another four to five years beyond that. Thus the challenge is to design and develop a product whose basic architecture will continue to be effective in the marketplace seven to eight years after it has been conceived.

The problems that uncertainty creates -- e.g., different views on the appropriate course of action, new circumstances that change the validity of basic assumptions, and unforeseen problems -- are compounded by the complexity of the product and the production process. A product such as a small copier, for example, may have hundreds of parts that must work together with a high degree of precision. Other products, such as the handle of Gillette's Sensor razor, appear to be fairly simple devices but, because of very demanding performance requirements, are complex in design and come out of a manufacturing process involving sophisticated equipment and a large number of operations. Moreover, products may be evaluated across a number of criteria by potential customers. Thus the market itself may be relatively complex with a variety of customers who value different product attributes in different ways. This means that the firm typically draws on a number of people with a variety of specialized skills to achieve desired, yet hard to specify, levels of cost and functionality. To work effectively, these skills and perspectives must be integrated to form an effective whole. It is not enough to have a great idea, superior conceptual design, an excellent prototype facility, or capable tooling engineers; the whole product its design system, production process, and interaction with customers -- must be created, integrated, and made operational in the development process.

But an individual development project is not an island unto itself. It interacts with other development projects and must fit with the operating organization to be effective. Projects may share critical components and use the same support groups (e.g., model shops, testing labs). Additionally, products may require compatability in design and function: models of computers use the same operating system, and different industrial control products conform to the same standards for safety. These interactions create another level of complexity in design and development. Critical links also exist with the operating organization. A new design requires the development of new tools and equipment and uses the skills and capability of operators and technicians in the manufacturing plant. Further, it must be sold by the sales group and serviced by the field organization. Of course, new products often require new skills and capabilities, but, whether relying on new or old, the success of the new product depends in part on how well it fits with the operating units and their chosen capabilities. Thus, effective development means designing and developing many elements that fit and work well as a total system.

Assessing the Promise and Reality: The A14 Stereo Project

The uncertainty and complexity that characterizes the development of new products and processes means that managing any development effort is difficult; managing major development activities effectively is very difficult. Thus, while the promise of a new development project is often bright and exciting, the reality is often quite different. The following story, based on a composite of several situations we have encountered, illustrates typical problems in product development.

In September 1989, Marta Sorensen, product manager for mid-range stereo systems at Northern Electronics Company, a large consumer electronics firm, laid out a plan for a new compact stereo system utilizing advanced technology and providing superior sound quality. Sorenson's marketing group at Northern felt that the company needed to respond quickly to the expected introduction of a new compact system by one of its toughest competitors. The plan Sorenson presented at the beginning of the concept investigation stage called for a development cycle time of one year, with volume production commencing in September 1990. (See Exhibit 1-3 for the initial schedule and subsequent changes.) This would give the factory time to fill distribution and retail channels for the all-important Christmas season in late 1990.

As the exhibit suggests, the schedule began to slip almost immediately. Because of problems in freeing up resources and scheduling meetings, and disagreements about desired product features, the concept investigation stage was not completed until November 1989, six weeks later than originally planned. At that point, no change was made to the schedule for commerical introduction or start of pilot production, but two months were added to the prototype build and test schedule. This additional time was needed as a result of the selection of a new speaker technology that the engineering group had lobbied for during the concept development stage. It was assumed that the time originally allowed for pilot production could somehow be overlapped and/or compressed.

By February 1990 new design problems had emerged. The compact size of the product created unexpected difficulties in fitting the components into a small space while maintaining sound quality. Furthermore, delays with a chip supplier and the speaker technology supplier set back the project schedule several weeks. A revised schedule, established in February 1990, called for completion of the design in April and completion of the prototype-build-test cycle by June. However, no changes were made to the schedule for pilot production or ramp-up. This meant a significant compression of the time between completion of prototype testing to commerical production; process engineering and manufacturing groups were asked to begin preparing the process for production even though the design was still incomplete.

Design engineers worked hard to solve problems with product size, and cost and completed the design in May 1990. By that time, however, new problems had emerged with the prototypes and with the production process. Part of the delay in prototyping reflected late deliveries of parts from suppliers, overambitious testing schedules, and problems in scheduling meetings for milestone reviews. But part of the delay also reflected technical problems with the introduction of surface mount technology in the printed circuit boards for the product. Moreover, process engineering had experienced difficulties with production tooling. There had been a significant number of engineering changes to accommodate changes in exterior appearance as well as performance problems with the product. As a result, the completion of prototype testing was rescheduled for August and pilot production and ramp-up were scheduled to occur in rapid fire succession thereafter.

Even the new schedule proved optimistic. As the fall months wore on and the project continued to slip, Sorenson and her marketing team realized that they would not meet the critical Christmas season deadline. Much of the latest delay had been caused by interaction between the product design and new automated assembly equipment that the manufacturing organization had installed. In order to meet product cost targets, manufacturing had chosen to move to an automated assembly system that would significantly reduce variable cost on the product. However, while design engineering was aware of the manufacturing plan, there were many subtle details of product design that conflicted with the capabilities of the automated equipment. These conflicts only surfaced late in 1990 as attempts were made to run full prototype units on the automated equipment. These problems required additional product redesign and slowed the completion of prototype testing.

Engineers eventually corrected the problems and prototype testing was completed in February 1991. While compression of the schedule had made product and process engineering operate in parallel, the completion of prototype testing did not mark the end of design changes nor the alleviation of production problems in pilot production.

Although Sorenson and the marketing group were happy to see the product make it through prototype testing, the fact that it was almost a year late had serious consequences for its potential attractiveness in the market. Sound quality and features were adequate and the cost and pricing were in line with expectations, but some of the product's aesthetics were out of synch with recent market developments. Thus, during the spring and summer of 1991 marketing pushed through a redesign of the product's exterior package to make it more attractive and contemporary. This caused some delays as engineering put through a crash program for new tooling and testing, but the redesigned exterior was put into production during the early fall. While the design of the new exterior was being developed, the manufacturing organization struggled to debug the new equipment and achieve consistent levels of quality. By September the plant had solved most of its major process problems and attention was shifted to increasing volume and filling channels for the 1991 Christmas season.

Market acceptance of the new product was satisfactory, but did not meet the projections originally laid out in 1989. Further, the engineering and manufacturing organizations soon found themselves confronted by a large number of field-identified quality problems. Exhibit 1-4 documents the engineering change history of the product from the beginning of pilot production to its post-Christmas sales period. As the exhibit suggests, there was a flurry of engineering change activity shortly after the product went into commerical production and the manufacturing organization struggled to achieve target levels of yield and volume.

Many of these engineering changes were intended to improve manufacturability. The significant peak in March 1992 reflected consumer experience with the product following the Christmas season. In February and March of 1992 design engineering launched a crash program to solve several field problems with product reliability.

The Characteristics of Effective Development

The experience of Northern Electronics with the A14 stereo system is not a pathological example. It reflects experience that is all too common in the world of product and process development. The failure of the A14 project to meet its original potential and expectations was not due to a lack of creative people, management desire, technical skills, or market understanding. The company had excellent marketing information, good relationships with its dealers and customers, recognized competence in engineering and design, and was known for its technical expertise. The A14's problems were rooted far more in the inability of the organization to bring together its insight and understanding and the expertise of its people in a coherent and effective way. In short, the A14 had problems because Northern lacked critical capabilities for integration.

Column 1 of Exhibit 1-5 summarizes typical characteristics of problematic projects like the A14, and column 2 identifies some of their implications. Problems on the A14 were rooted in the nature of the development process and its organization and the absence of a coherent and shared cross-functional plan for competing in the compact stereo market. Different functional groups (e.g., marketing and engineering) had different agendas and there was no organizational process to resolve issues before they surfaced throughout the phases of the A14 development effort. This led to delays and miscommunications throughout.

The development process itself contributed to delay and poor design. The many late engineering changes reflected in part a poorly organized and executed prototyping process. Some prototype parts came from suppliers unfamiliar with the commerical production environment at Northern and were late and poorly built. Delays getting into manufacturing were caused by a narrow focus on product performance in design choices (no design for manufacturability) and barriers to communications between engineering and manufacturing. Management treated the development of new products as the responsibility of the engineering group. Manufacturing was not of primary concern, at least not until problems with the new automated process began to surface. Without strong leadership, problems in the project went undiscovered, surfaced late, and were difficult to resolve.

In contrast to the A14 experience, column 3 in Exhibit 1-5 lays out selected themes in an outstanding development project. Objectives and accountability are clear and widely shared and stem from a concept development and product planning process that brings marketing, engineering, and manufacturing together. Moreover, early-stage development builds on clear strategies in the organization for the product line and major functions. In effect, the outstanding organization starts development projects with concept development on a firm foundation.

Once the concept has been developed and plans for the product have been laid out, execution in outstanding programs has a distinctive character. Guided by strong leadership, engineers with broad skills work in a coherent team with skilled people from marketing and manufacturing. "Integrated" describes day-to-day problem solving across departments and functional groups right down at the working level. Strong, collaborative relationships across departments are rooted in intensive communication, a shared responsibility for product performance, and an appreciation of the value to be added by each group. In this context an excellent engineering design is one that not only achieves outstanding performance but also is manufacturable and comes to market rapidly.

Indeed, time-to-market is such a critical dimension of performance in the outstanding project that all of the processes, systems, and activities in development are geared to fast action. This is particularly true for the critical design-build-test cycles that are at the heart of problem solving in development. Thus, the outstanding project has a prototyping process that creates representative components, subassemblies, and complete units of high quality. These prototypes in turn come out of a design process in which careful and simultaneous attention to the details and behavior of the product as a system catches numerous problems and identifies important opportunities early in the process. In this setup, engineers concentrate on eliminating redesigns caused by mistakes, poor communication, and lack of process understanding, and maximizing product performance and distinctiveness for its target market. "Design it right the first time" is critical because it creates products of high quality and saves valuable time.

Outstanding projects of this kind are not possible without leadership. In contrast to problematic projects where direction is lacking and responsibility diffuse, the excellent project has a project leader who gives conceptual direction and stimulates and nurtures working-level integration. Moreover, that leadership extends to linkages with critical suppliers, customers, and the market. The outstanding project leader fosters internal integration and integrates customer needs into the details of design. Effective product development is not the result of a single individual, but strong leadership makes a difference.

The Fast-Cycle Competitor

The themes that characterize outstanding development projects -- clarity of objectives, focus on time to market, integration inside and out, high-quality prototypes, and strong leadership, to name a few -- reflect capabilities that lead to rapid, efficient development of attractive products and manufacturing processes. The power of such capabilities lies in the competitive leverage they provide. A firm that develops high-quality products rapidly has several competitive options it may pursue. It may start a new product development project at the same time as the competitors, but introduce the product to the market much sooner. Alternatively, it may delay the beginning of a new development project in order to acquire better information about market developments, customer requirements, or critical technologies, introducing its product at the same time as its competitors but bringing to market a product much better suited to the needs of its customers. Furthermore, if it also has achieved speed and quality in an efficient way, it may use its resources to develop additional focused products that more closely meet the demands of specific customer niches and segments. Whatever the mix of customer targeting, speed to market, and product breadth the firm chooses to pursue, its advantages in fundamental capabilities give it a competitive edge.

For a firm like Northern -- with slipping development schedules, late design changes, and problems with field failures -- competing against a firm capable of rapid but effective product development can be a bewildering, discouraging, and ultimately unprofitable experience. Exhibit 1-6A illustrates just such an episode in Northern's history. Consider first Panel A, which graphs the price, cost, and product generation experience of Northern and its principal competitor, Southern Electronics Company, from 1978 until 1985.

Until 1985, both Northern and Southern followed standard industry cycles in new product development, pricing, and manufacturing costs. With a product development cycle of eighteen to twenty months, both firms introduced new generations of product every two years. Between major generational changes in products there were frequent model upgrades and price declines as the cost of key components and manufacturing fell with increasing volume. Thus, until the mid 1980s, both Southern and Northern had prices and costs that tracked each other closely, and both mirrored industry averages.

Improvement Efforts at Southern Electronics

In the early 1980s, changes in Southern laid the foundation for a significant change in the nature of competition in the industry. Stimulated by the efforts of Greg Jones, the new vice president of engineering, Southern embarked on a concerted effort to reduce its product development lead time. Without compromising quality, Jones and the entire organization began to develop the characteristics sketched out in column 3 of Exhibit 1-5. Stronger leadership, more effective cross-functional integration, greater attention to issues of manufacturability and design, more effective prototyping, and a revamped development process gradually led to a reduction in development lead time from eighteen to twelve months. By 1986 Southern could develop a comparable compact stereo system about six months faster than Northern.

As Panel B of Exhibit 1-6B suggests, Southern began to use its new development capability in early 1986. At that point it broke with industry tradition and introduced its next generation of stereo product about six months sooner than expected. With a more advanced system and superior performance, Southern was able to achieve a premium price in the marketplace. Although Northern followed six months later on a standard cycle, its next generation stereo was unable to command its traditional market share. As a result, Northern's volume increased more slowly than expected and its cost position began to erode slightly relative to Southern.

Southern Electronics introduced its next generation product eighteen months later in the fall of 1987. Once again the product achieved a premium price in the market. However, Southern did not fully exploit its premium pricing opportunity. Instead, it lowered prices somewhat to increase further its market share. At that point, not only was Northern behind in product features and technology, but Southern's aggressive pricing posture put even more pressure on Northern's sales volume and margins. Although Northern fought back with price discounts, increased advertising, and promotions to dealers, it was unable to stem the erosion of its historical market position. The result was an even greater disparity in the cost positions of Northern and Southern Electronics.

Northern's Competitive Reaction

In late 1988, Northern introduced its next generation stereo system, the A12. Developed under the motto "beat Southern," Northern's executives felt that the A12 would be the product to regain their former competitive position in the market. Much to their surprise, however, the rollout of the A12 in early 1989 was met by Southern's introduction of its next generation stereo system: Southern had moved to a twelve-month product introduction cycle in late 1988. At that point Northern was a full generation of technology behind Southern in its market offerings. Northern's management determined that the only course of action open was to accelerate development of the next generation system, the A13. They thus embarked on a crash development effort to bring the A13 to market in early 1990. At the same time Sorenson and her colleagues began development on the A14, which they targeted for the Christmas 1990 selling season. The A14 was to get them back into the competitive ball game on solid footing -- a "close the gap" strategy.

While Northern's strategic intent was to catch up to Southern with accelerated product development, the reality was much different. Northern brought the A13 to market in early 1990, but the development process was so hectic and the ramp-up in manufacturing so strained that the company effectively lost control of its costs. The product came to market but was much more expensive and less effective than the company had planned. Because of its many problems, scarce development resources that were to have been moved to the A14 in early 1990 were focused instead on correcting problems and cleaning up the A13's design. To make matters worse, Southern continued to follow its twelvemonth introduction cycle and actually beat Northern to the market with its next generation product. The result for Northern was a further erosion in margins and market position.

Without making fundamental changes in its development process, which management considered neither necessary nor within the charter of Sorenson and those working on the A14, Northern's attempt to push ahead with the A14 for the 1990 Christmas season was a dismal failure. The A14 product had so many problems in the field and was so expensive to manufacture that the product line became a serious financial drain on the company.

The Sources of Advantage

The key to Southern's success in the compact stereo market was its consistent ability to bring excellent products to market before its competitors. This ability was rooted in fundamental changes that Jones and others had made in its development process. These included obtaining broad-based organizational and individual buy-in to key project goals, at the onset, and empowering and encouraging development teams to modify the development process while developing the needed products. In addition it harnessed that capability to a marketing and pricing strategy that was well targeted at Northern's weaknesses. In effect, Southern changed the nature of competition in the industry; Northern was forced to play a game for which it was ill suited -- a game Northern never fully comprehended until it was years behind in capability.

Southern Electronics' ability to bring a competitive product to market more rapidly than its chief rivals created significant competitive opportunities. How Southern chose to exploit those opportunities depended on the nature of its competition and its own strategy. But the ability to move quickly in product development created at least three potential sources of advantage:

* Quality of design. Because Southern had a twelve-month development cycle, it could begin the development of a new product closer to the market introduction date than its competitors. Whereas Northern had to begin eighteen to twenty months before market introduction, Southern's designers and marketers could gather and refine an additional six months of information before setting out to design a new product. In a turbulent environment, designers face a high degree of uncertainty in the early stages of development about which set of product characteristics will be most attractive to target customers. Additional time to secure feedback on the most recently introduced generation and to learn about market developments and emerging customer preferences may mean the difference between winning and mediocre products. Although the product may use the same basic technologies, additional market information may yield a much better configuration. The product's features and aesthetics may be fresher, more up-to-date, and more closely matched to customer expectations. Thus, Southern could exploit its lead time advantage by waiting to launch its development effort until more and better market information became available. Even though its product would arrive on the market at the same time as its competitors, its product would offer the customer a superior experience.

* Product performance. A much faster development cycle gave Southern Electronics the opportunity to launch a new product program well in advance of its competitors. It could use that lead to introduce the next generation of product technology. In this case, the advantage of speed lay not in superior market or customer intelligence, but rather in the ability to exploit technological developments and bring them to market faster than its competitors. The gap in performance this created is depicted in Exhibit 1-7 for a single product generation. As illustrated in the exhibit, a six-month jump on competitors in a market accustomed to eighteen- to twenty-four-month design lives can translate into as much as three times the profit over the market life of the design. Conversely, being late to market with a new product can lead to break-even results and zero profit. This provided Southern with the leverage to control not only their own profits and returns, but also those of their chief competitor, Northern. Putting a sequence of such developments together further widens the competitive gap, as depicted in Exhibit 1-8. The slow-cycle competitor brings new technology to market every two years. The fast-cycle competitor, in contrast, achieves the same performance improvement every twelve months. While the initial advantage of the fast-cycle competitor is relatively small, the ability to move quickly to market eventually creates a significant performance gap. To the extent that customers can discern the difference in performance and to the extent that the gap offers them valuable improvements, a faster time to market creates a superior product.

* Market share and cost. A better product design and superior product performance gave Southern the opportunity to achieve premium prices in the market. However, a firm may also choose to price its product to create superior value for its customers, thereby translating advantage in design and performance into increases in market share. Where lower costs are driven by growth and increases in volume, increases in market share may translate into improved cost position for the fast-cycle operator. Thus, even if two competitors operate on the same learning curve, the fast-cycle competitor will achieve a cost advantage. However, it may also be the case that the capabilities which underlie fast development cycles create a steeper learning curve. Speed in development is rooted in the ability to solve problems quickly and to integrate insight and understanding from engineering with critical pieces of knowledge in manufacturing. This set of capabilities likewise is critical in achieving cost reductions in established products. Thus, when costs are sensitive to volume and fast-cycle capability enhances a firm's overall learning capacity, the fast-cycle competitor enjoys double leverage in improving its manufacturing costs.

How a fast-cycle competitor chooses to exploit the potential advantages in design, product performance, and manufacturing cost will depend on the competitive environment and the firm's strategy. In the case of Southern Electronics, all three dimensions of advantage were important. Initially, Southern used its six-month advantage in lead time to obtain better market information and still introduced its 1986 compact stereo about six months before its competition. In the second generation, however, Southern accelerated its model introduction and began to exploit its development capacity to achieve superior product performance. By 1990, Southern was a generation ahead of its competitors in product technology. It used its superior design and performance to achieve some price premium in the market, but it did not raise prices as much as its performance advantage warranted. The result was a superior value for customers, increases in market share, and steeper slopes on its manufacturing learning curve. Thus, Southern used its advantage in performance and cost both to expand its market share and increase its margins.

But perhaps the most powerful effect of Southern's fast-cycle capability was its ability to change the nature of competition. By improving its development productivity and shortening the time between product generations, Southern forced Northern to play a competitive game that Northern was not prepared to play. Northern would have faced competitive difficulties no matter how it responded to the Southern challenge, but it compounded its problems by failing to change fundamentally its approach to product development. By attempting accelerated development in the context of its traditional systems, Northern created internal confusion, strained its resources, and actually reduced the effectiveness of its development organization. In addition, previously enthusiastic, capable, and hard-working product managers such as Sorenson became frustrated and disappointed. Thus, at the start of the 1990s, Northern Electronics faced the challenge of undertaking a major overhaul of its development process while its margins were eroding, market position was slipping, and morale among some of its best development people was declining. Southern's fast-cycle capability had clearly put Northern and its other major competitors at a significant competitive disadvantage while generating additional enthusiasm and competence among people such as Jones and individual project contributors. Southern was continuing to build momentum as Northern and other competitors continued to lose it.

Achieving competitive advantage through effective development capability is not just a theory. Effective fast-cycle competitors have emerged in a wide range of industries. Firms such as Honda in automobiles, Applied Materials in semiconductor production equipment, ACS in angioplasty, Sony in audio products, Matsushita in VCRs, The Limited in apparel, Philips in computer monitors, Hill-Rom in hospital beds, and Quantum in disk drives have made the ability to bring outstanding products to market rapidly a central feature of their competitive strategy. Once achieved, and subsequently maintained as the organization grows, an advantage built around fast-cycle capability seems to be strong and enduring. In the first place, the advantage is based on capabilities -- human and organizational skills, processes and systems, and know-how -- that are difficult to copy. Moreover, effective, rapid development creates superior products and offers customers superior value. It therefore helps to create a market franchise and brand equity. A real product advantage rooted in difficult-to-copy capabilities and a translation of that product advantage into a fundamental market franchise that reinforces its own momentum is a powerful combination. Although product development is difficult, doing it well confers significant advantage. Furthermore, the more challenging the development requirements, the more dramatic the potential impact.

The Plan for the Book

In this book we lay out concepts for the effective organization and management of product and process development. Each chapter frames a particular problem or issue in development, provides a set of ideas for effective management, and illustrates those ideas and their application with several examples. The cases accompanying each chapter in the College version provide an opportunity to apply and develop the concepts and ideas in a practical context.

The first part of the book focuses on the front end of the development process. In Chapters 2 through 5 we discuss the concept of development strategy, the use of maps and mapping to chart an organization's path through the development terrain, the creation of an aggregate project plan to guide a portfolio of development efforts, and the challenge of creating an overall development process that effectively initiates and selects projects and focuses the organization's resources to bring the most attractive projects to market rapidly and efficiently. The thrust of these chapters is laying the foundation for effective development efforts. While the actual development project is a natural locus of attention and effort in organizations, individually effective development projects depend on a strong foundation in strategy, a shared understanding across functional organizations, and an overall process that effectively allocates and concentrates time, energy, attention, and resources on the most attractive opportunities.

Chapters 6 through 10 focus on the management of individual development projects. We first work through an overall framework for evaluating development efforts, including identification of the important phases of development, the measurement of performance, and the critical areas of leverage and choice for managing projects. We then examine the problems of cross-functional integration. A central theme in this part of the book is the power of integrated problem solving. Chapter 8 deals with the problem of organizing development projects. Our emphasis is on the organizational structure, the processes the organization uses to carry out development, and the impact of development leadership. We lay out four contrasting approaches to development project organization and focus particular attention on what we call heavyweight project teams.

The challenge of integration applies not only to large functional organizations like marketing, manufacturing, and engineering, but also at the working level within those organizations and across departments and work groups with different disciplines, tasks, and experiences. Chapter 9 focuses on recent developments in systematic methods and tools for product (and process) development. Concepts such as quality function deployment, design for manufacturability, computer-aided design, and computer-aided engineering represent new design and development methodologies. Much of the thrust of these methodologies is the creation of more effective integration in the development process. In Chapter 10 we examine prototyping, testing, and convergence to a final design. Much of development is a sequence of design-build-test cycles in which prototyping and testing play a central role. Effective management of prototyping is therefore a critical element of effective development capability.

In the final chapters of the book, we shift our attention from the planning and execution of specific projects to the problem of managing the improvement of the development organization and its processes. In Chapter 11 we examine the problem of learning from individual development experiences. This involves not only capturing the insight and understanding that come from current experience, but also capturing that experience in the form of changes in the development process. In addition, learning from experience involves building resources and capabilities to conduct development efforts more effectively in the future. Thus the major focus of Chapter 11 is on mastery of the building blocks for superior development capability and the associated investment in people, skills, tools, and systems.

The book concludes with a chapter on making it happen. We examine alternative improvement paths and focus on the peculiar nature of the development process and consequent issues that managers must examine in pursuing an overall improvement plan. A central theme in this final chapter -- and, indeed, throughout the entire book -- is the importance of learning by achieving consistency and balance across a wide range of development activities. There are no "three easy steps" to effective development performance. The capabilities that allow an organization to move quickly and efficiently to the market are rooted in people and their skills, organizational structure and procedures, strategies and tactics, tools and methodologies, and managerial processes. This is what makes it so difficult for organizations to improve -- and why they acquire such a strong competitive advantage when they do.

Study Questions

1. Product development has been a part of competition at least since the Industrial Revolution. Apparently, doing things faster, more efficiently, and with higher quality has always been an advantage. In what sense, therefore, has product development become more important in competition in recent years? Do you agree that increased speed, efficiency, and quality have become competitive imperatives? Are they imperative under all circumstances?
2. Consider Exhibit 1-2. What is the rationale for dividing the development process into phases? What distinguishes the first two phases from the last phases of Exhibit 1-2?
3. In the A14 stereo project, what explains the differences between the original plan of September 1989 and the third plan revision of May 1990?
4. Consider Exhibit 1-5. Pick one of the characteristics of problematic projects and compare it with its counterpart theme in the outstanding projects. What might explain the problematic characteristic? Where does it come from? What makes moving from problematic to outstanding difficult?
5. What are the key elements in Southern's "fast-cycle" strategy? What explains Northern's response to Southern's strategy? Was Northern's response predictable? Inevitable?
6. The fast-cycle strategy conferred significant advantage upon Southern. Will a fast-cycle strategy work this way under all circumstances? Why or why not?

Ampex Corporation: Product Matrix Engineering

TO: T. Burroughs
FROM: M. Hirschfeld
DATE: March 30, 1979
SUBJECT: Product Matrix Engineering -- Future Plans

I meant what I said yesterday, Ted. You and your PME Group deserve much of the credit for the success of our VPR2 program start-up. We were so involved with other matters in our discussion that I neglected to ask for your views about the future of PME. Now that you're moving into the Data Systems Division, I would appreciate your recommendations. Do you think PME should become a permanent organization, or should it be disbanded as such ad hoc organizations have been in the past? If PME does become a permanent organization, we will have to define its place in the organization and its charter very carefully.

The note from Morris Hirschfeld raised several issues that Ted Burroughs had thought about often. He was confident that the PME experiment had been a success, and he had a definite opinion about what should be done.

The Company

The Ampex Corporation manufactured magnetic tape and magnetic tape equipment, serving a worldwide market with sales of nearly $400 million in 1979 (see Exhibit 1 for financial data). Headquartered in Redwood City, California, it had manufacturing operations in several western and southern states and in a number of countries including Mexico, Brazil, Taiwan, and Japan.

Ampex placed special emphasis on its position as a technological leader in its industry, employing some of the most highly respected design engineers in its Redwood City Engineering Center, a university-like research center that was located with Ampex headquarters near Stanford University. More than half the products sold in 1979 had not existed only five years previously.

At the heart of Ampex's business was the Audio Video Systems Division (AVSD), which supplied high-performance audio and video recording and editing equipment to professional broadcasters. AVSD accounted for 35% of Ampex's sales and more than half of the company's earnings. Ampex had attained its leadership position in broadcast equipment by introducing the first professional transverse scan magnetic video recorder, the Quadruplex VR 1000, to the industry in 1956. During the next two decades, Ampex dominated the quadruplex video recording market claiming a 65% share compared to RCA's 35%. (By 1976 it was estimated that 10,000 quadruplex VTRs had been sold worldwide.) Ampex had extended its line of professional video products to include other proprietary equipment such as the slow-motion disc recorder famed for its instant replay when used to broadcast sports events such as "Wide World of Sports."

The division continued to reaffirm its commitment to scientific excellence through new product introductions every few years. Its latest technological first was an experimental digital video recorder which it demonstrated to the March 1979 convention of the Society of Motion Picture and Television Engineers (SMPTE). That product would not be commercially feasible until the mid-1980s, but Ampex saw it as the kind of product that would keep the company at the forefront of its industry.

Two Video Tape Recorder Technologies

Ampex AVSD's principal product was traditionally the Quadruplex Video Tape Recorder (VTR) for professional broadcasting. The quad machine was a large (6 feet high by 3 feet wide) piece of equipment containing expensive parts and delicate circuitry and selling for over $100,000. It recorded its video information in a crosswise (or transverse scan) pattern on two-inch magnetic tape (Figure A). The quad recorder was suitable primarily for studio use because of its size and complexity of operation. It was generally sold with accessories such as the slow-motion disc recorder ($100,000) or a computerized editor ($50,000) to form a complete magnetic tape editing system for broadcast production purposes.

Ampex also manufactured and sold a second form of magnetic video tape recorder, the helical scan VTR. Helical recorders were smaller, lighter, and cheaper (base price $5,000-$25,000) than quad. They offered features such as still frame and slow motion that quad recorders did not have, and because they recorded their information on one-inch magnetic tape (see Figure B) at slower recording speed, their tape consumption was roughly one-third of the quad's (helical: 10 square inches per second; quad: 30 square inches per second).

During the 1960s helical scan performance had been unacceptable for broadcast use. Its recording format was less forgiving to operate than that of its quad counterpart, requiring constant adjustment to reduce tape tracking errors that caused picture defects. Consequently the helical video recorder had been sold as a low-priced, high-volume product for the institutional market with margins about one-half those of quad equipment. (Helical video recorders contained fewer expensive parts and simpler electronic circuitry.) Ampex had shared the industrial recorder market with numerous Japanese recorder manufacturers including Toshiba, Matsushita, and Sony. Each manufacturer had a different tape format, but since interchangeability of tapes between machines was not crucial in the institutional market, no standard tape format was adopted.

In the early 1970s an auxiliary device called the digital time-based corrector appeared, compensating for the helical recorder's previous quality problems. With their special features, light weight, simplicity and low cost of operation, the improved helicals were ideal for television news gathering from remote locations. Major broadcasters such as Westinghouse and CBS began to demand improved-for-broadcast helical machines.

Some industry observers speculated that potential sales for helical recorders for professional uses could be 20,000 systems. Other experts had been saying for some time that at a price lower than $50,000 the distinction between industrial and professional recorders would disappear and that the total primary market for helical might reach 100,000 systems worldwide.

In 1974 Sony introduced a new model helical recorder that quickly dominated the institutional and industrial market. Ampex withdrew from that market and focused its attention on the professional recorder, launching a program to do a new helical design for broadcast use. John Roberts, product manager for the Ampex video recorder line, recalled:

It was difficult to persuade first-line engineers who had devoted their entire careers to transverse-scan recording to switch to designing helical machines. They hated to see quad sacrificed on the altar. To them, helical was inferior and would always remain so.

In 1975 the Ampex design team produced a helical video recorder designated the VPR1. The VPR1 offered "outstanding recording and playback with capability of 1/5 speed slow motion and still-frame pictures of broad-castable quality" made possible with the addition of an automatic scanning module (the Ampex AST). To avoid introducing a new tape format to a market already crowded with helical tape formats, the designers of the VPR1 retained the Ampex A format, which they had used in their latest institutional and industrial models. As Doug Grantham, systems manager, explained, "In quad the format never changed. The first quad tape made can be played on the newest machine. We wanted to push for that kind of stability in helical technology too."


At the National Association of Broadcasters (NAB) convention in March 1976 the Ampex VPR1 announcement was closely followed by Sony's revelation that it intended to penetrate the professional video recorder market for the first time with a helical machine called the BVH 1000. Sony's demonstration model equalled Ampex's in major features and price. Sony elected to introduce a different tape format from those it used in earlier helical recorders, one that was similar to Ampex A but not compatible. A third company, Germany's Bosch-Fernseh, announced its intention to market another incompatible version of the helical recorder, which it advertised as rugged, reliable, and less expensive than either of its competitors.

The reception of new recorders by potential customers was enthusiastic, but actual orders were limited because of the standards issue. Major customers wanted the ability to interchange tapes among machines without regard for make of equipment. The Fernseh product was too different for compatibility to be considered, but Sony and Ampex were enough alike to warrant standards negotiations. At the behest of the major customers, CBS and ABC, a standards working group comprising representatives from Sony, Ampex, and several other SMPTE members negotiated a compromise format in 10 months. By January 1978 the new C Format, essentially Sony's format modified for use with Ampex's automatic scanning module, was formally accepted by the SMPTE. Both companies had made concessions that would cost them some months' design rework. The helical machines that the companies had already sold could, under the terms of the agreement, be upgraded with kits to use the C Format.

VPR2 Design Program

Under pressure to produce their design speedily, the Ampex design team turned to its task of designing a second-generation helical video recorder. The Ampex team adopted three original design goals: 1) to adapt their product for use with the C Format; 2) to keep as many parts and processes compatible with the VPR1 as possible; 3) to correct known problems with the VPR1 design. Grantham discussed the effort:

We soon realized it was easier to design a new machine than to fix all the problems with the old one. After two weeks we assigned three engineers to do the upgrading kit for the VPR1 while the rest of us concentrated on producing a good C Format machine. When we were finished, the VPR2, with an entirely new scanner assembly that was three times more accurate, was a substantially better machine than the VPR1.

Of course the manufacturing people were unhappy about the new design. We had passed the design deadline, and the manufacturing people claimed the machine was impossible to manufacture. But we were far more concerned about the manufacturing aspects of the design than usual. We've not only given the factory the product design, we've shown them how to make it. One of our best engineers spent time designing tooling for the scanner. In fact, the tooling is the chief proprietary aspect of the product [see Exhibit 2 for VPR2 product information].


In the early days of the company, audio-video manufacturing had been located in the Redwood City headquarters complex. By the late 1960s taxes and union wage rates had risen sharply in northern California, and Ampex had moved a large part of its manufacturing 1,000 miles away to nonunion plants in Colorado Springs, where Ampex built a large (250,000-square-foot) two-story plant that housed both its fabrication facilities and its audio-video assembly and test operations (see Exhibit 3 for AVSD organization). Colorado had begun its operations by assembling mature high-volume audio products. By 1978 the plant housed 80-90% by unit volume of AVSD's manufacturing capacity and employed 1,800 people.

Ampex AVSD also employed 300 people in Sunnyvale, California, about 20 miles from Redwood City. Sunnyvale handled special projects and small-volume products that did not seem suited for Colorado's large-scale approach. Most of Sunnyvale's engineering staff had transferred from Redwood City.

Some members of the corporate engineering staff said the Colorado plant ought never to have been built; most expressed a low regard for its performance. They said the factory had a history of being unresponsive to their needs, and they complained that it paid less attention to quality and on-time delivery than to keeping down manufacturing costs. Terry Warner, a member of the engineering support staff, described the Colorado plant's approach to manufacturing.

The philosophy at Colorado has always been to deskill the operator; make the machines do all the thinking. When it comes to assembling new products, that's a problem because the assemblers lack the skills and the necessary judgment to assemble tight tolerance subassemblies, for instance. It's no wonder manufacturing engineering insists that tooling be foolproof.

Ken Leonard, plant manager at Colorado Springs, responded to the criticisms from Corporate Engineering:

Manufacturing has to worry about reliable details and unrealistic schedules. We are responsible for keeping materials costs down and getting volume production up on time. We are always squeezed between an optimistic product delivery date set by marketing and a pessimistic part delivery date set by a supplier with a long lead time. Designers are impractical, undisciplined people who can never make up their minds. Their prototypes are put together by engineers with sandpaper fingers who know exactly what effect they are trying to achieve. These designs defy assembly by a bench worker working for $5 an hour from a vague set of blueprints and a list of approximate specifications.

A major burden on AVSD manufacturing at Ampex were the engineering change notices. (ECNs were official notifications from the design team to the factory engineering staff and purchasing people that a part substitution had been made.) The Colorado plant typically had 31,000 separate parts on hand of which about half were purchased. Typically 15% of this number changed over in one year. To cope with the task of inventory control the factory installed a computerized materials requirements planning (MRP) system in the early 1970s, but recording and updating the data base placed such demands on the materials people that the system had not been reliable.

In 1977 Gary Herold, who joined Ampex from the computer industry, was appointed corporate director of materials and given the task of gaining control of the materials situation in Colorado. Herold said:

When I arrived in October 1977, we had a horrendous inventory problem. We had 300 people in materials handling, and they were costing us 20% of direct materials. We were coping with 80 ECNs per week of which 60 were marked mandatory. It took a week to put a change through.

I came in and set up a new system. First we separated new product parts from the rest and controlled them off line. We made a rule that we would only enter a part in the data base when the part had already been tested out and the design released, and we wouldn't accept an ECN until there was a high degree of certainty it would be used. We are down to 20 ECNs a week. I think we've influenced the engineers to become much more accurate.

Product Transfer

New product transfer from design to manufacturing had long been a problem area for Ampex. When audio-video manufacturing had been located in Redwood City, the transfer had been accomplished by putting design engineers in the plant to do the early training of assemblers when necessary.

Later, when manufacturing had moved away from headquarters, several different approaches to transferring the product had been tried. First, a limited production line had been set up in Redwood City, where the design was debugged and then transferred to the Colorado plant. When that had resulted in friction between engineering and manufacturing, Sunnyvale had been used as a start-up facility from which the product was transferred to Colorado Springs only when the design was considered firm. Redwood City's engineers could easily visit Sunnyvale.

The VPR1 project had gone through start-up in Sunnyvale. The experience had been costly in both time and money. Sunnyvale procedures and processes were typically quite different from those employed in Colorado, and the difference was accentuated because the VPR1 was a higher-volume product (at 75 units a month) than either plant was accustomed to turning out in video products.

The two-phase transfer had resulted in two complete workups for the product. Not only had VPR1 been totally redocumented for Colorado but certain tools and test equipment had been purchased twice. Colorado blamed the loss of learning on the early units for keeping the factory from bringing down costs and achieving target production volume on schedule. The transfer of information between the two plants' materials control systems had also caused substantial losses on materials. It was rumored that the project had dropped $37,000 in two weeks because of typographical errors alone. Overall, inaccurate part numbers and unrecorded or belatedly recorded changes pushed the materials wastage figure to $150,000 over budget by $100,000. Eighteen months had elapsed before saleable units were being produced in Colorado Springs, and customers had waited for their VPR1 units as much as a year after ordering them.

Gary Herold commented on the types of problems that had occurred in the VPR1 program:

Engineering worried only about its design. They let manufacturing worry about how to make it work: if manufacturing couldn't make the design work then manufacturing was incompetent. Here at Ampex, engineers have always missed design schedules, and they've always been allowed to run over because to enforce a design date would be to stifle creativity. Meanwhile, the factory has always been held to its original product introduction date. This has left the factory with the largest amount of work to do in the shortest amount of time.

As a result of the experience on the VPR1, it was decided to put the VPR2 directly into Colorado Springs from Redwood City. The new Colorado Springs plant manager, Ken Leonard, set up a New Product Manufacturing Team especially to handle the interface with Redwood City engineering. The new product group contained two project engineers and one purchasing manager: the first time these functions had answered to a common boss in Colorado (see Exhibit 3). Previously, new products had been handled by the regular staff.

Everyone knew that the VPR2 was going to be a very important product for Ampex and much different from the VPR1 Original marketing projections had been 600 units the first year, but by January 1978 1,200 units were on backorder and revised forecasts kept appearing from marketing as further orders came in from new types of customers. To meet new projections Colorado would have to produce 260 VPR2 units per month. At the new rate the VPR2 would constitute 70% of the Colorado plant's total unit volume when it was being produced at full volume.

Ampex manufacturing personnel saw the VPR2 as a formidable manufacturing task in other ways as well. Attention to quality would be paramount, since Sony was known for its high-quality manufacturing. Cost control would be even more significant than it had customarily been because helical recorders were lower-margin products than quads had been. Scrap was budgeted not to exceed $500,000.

The schedule established in July 1977 allowed four months to complete initial design and nine more months before the first 20 sale units were to be delivered early in September. Full-volume production was to be reached in January 1979, by which time it was hoped that labor content might be approaching the steady state goal of 21 hours per scanner. Prompt delivery would be crucial for broadcasting customers, especially those who had ordered hundreds of units for use in the winter and summer Olympics of 1980. According to marketing, inability to promise early delivery might be the determining factor in European sales, where Bosch would be a strong competitor.

Product Matrix Engineering

By October 1977 the VPR2 design team was still not willing to give out much information about its design. Advance indications of product specifications alarmed the manufacturing staff. The tolerances of the new scanner mechanism were said to be 10 times more exact than those of the VPR1. Eighty to 95% of the VPR2 parts were to be new. Essentially only the motor and the transport mechanism remained compatible. In all likelihood these differences would mean major changes in the way the product had to be manufactured. Higher skilled assemblers, an environment controlled for temperature and humidity, and a complex testing procedure might all be necessary.

The Colorado plant's New Product Manufacturing Team members were anxious to gain information to give to their prospective vendors, especially since components were likely to be higher precision than usual. However, the designers would commit to very little, and they resisted taking time to put their specifications on paper until they were ready to call the design firm. Lacking definite information, the manufacturing people could do little in advance without risking substantial materials wastage.

In October a decision was made to establish a liaison effort between new product engineering and manufacturing on the VPR2 project. The frequent arguments between engineering and manufacturing were becoming heated. Although it was doubtful that the design team would meet its scheduled deadlines, Richard Rothmann, Ampex's executive vice president, had warned the Colorado plant that no slippage would be tolerated in the September 1978 product shipment date.

Ted Burroughs, a former Ampex employee who had left to manage a smaller company several years before, was hired to coordinate the liaison function as an engineering support group. Named assistant to the chief engineer AVSD, Burroughs decided to adopt a matrix form of project management, which he called product matrix engineering.

Burroughs explained the philosophy of his group as follows:

PME was envisioned as a natural buffer between Manufacturing and Engineering. Previous attempts to establish a high-level liaison had been unsuccessful. When I was hired there was more discussion as to whether the head of PME should be at general management level; but the key idea was to start at the grass roots level this time, to resolve conflicts before they reached higher levels. It was important that neither side felt that the PME head was in the opposing camp. We represented the factory to Engineering and Engineering to the factory.

Burroughs's criterion in selecting people to be on the PME team was what he called the "green beret principle," requiring each to have experience in more than one function. Bob Ferrante, for example, Burroughs's next-in-command, had been with Ampex for 17 years. He had worked first on the limited production line in Redwood City and later at Sunnyvale, always with skilled people on small-volume or custom products. Later, as a junior engineer, he had become involved in interfacing with the Sunnyvale factory on several projects.

At first the role of PME was loosely defined. In general the group concerned itself with aspects of the product start-up that mattered most to the factory: documentation, cost structure, and materials control. Its responsibilities evolved partly in response to the VPR2 design. As the design became more complicated and the start-up and production tasks it required began to exceed the previous experience of the manufacturing group, PME assumed more responsibilities.

PME devoted several months to improving the amount and quality of information that passed between engineering and manufacturing, especially discovering the advance information the factory needed about critical parts before the design was released. (Critical parts were those that had long lead times or were exceptionally difficult to procure either because there was a limited number of qualified suppliers or because a part was unusually exacting in its specifications.) PME members also notified the New Product Manufacturing Team when meetings were held on program decisions that would affect them. Gradually they gained the confidence of the Colorado New Product Team, but Engineering also had to be convinced. Terry Warner recalled that it was six months before the design team also trusted him.

To assure continuity and accuracy of information, the PME team developed new procedures and documentation devices for Ampex. A drawings tree (the Xmas tree) identified by number the drawings that were needed for each subassembly. A material status chart was a bill of materials color-coded to indicate PME's confidence level concerning the likelihood of a part being used in the final design. Such documents formed the basis for frequent project reviews and served as control feedback documents. Warner explained:

As soon as we could get any information at all about a part of the VPR2 design, we would publish an initial compilation which would stimulate responses from the design engineers. Then we would update and republish our charts and documents immediately and repeat the same procedure. Each new piece of information was factored into the schedule to come up with some idea of a critical path. And each time we were reasonably confident about a new part number, we would pass it along to Colorado.

Procedures were also established for document control of engineering changes. Every ECN had to go through the same people both in Redwood City and in Colorado in the same order. Setting up documents and procedures alone took several months of full-time work.

By March 1978 most members of the PME had been accepted as individuals, and the group began taking on tasks that were more than simply interface responsibilities. Burroughs insisted that the group farm out everything it possibly could to the factory, but a number of tasks had arisen, such as precision part purchasing or fixture testing, that the factory did not have the resources or the time to handle. Since an expanded role for PME would require top management approval, Burroughs proposed a more formalized PME structure that would also act as project interface for other new products to be transferred between Redwood City and satellite factories. The VPR2 program would continue to receive highest priority until it was completed.

Burroughs requested and received a budget of $500,000 for equipment and permission to hire assemblers. This allowed him to set up a small scanner assembly laboratory and to take on a supplementary role in purchasing and prototype building.

The new PME organization consisted of 17 people (see Exhibit 4). Initially it was agreed that PME should order parts for, build, and assist in testing 14 engineering models of the VPR2 system.

As soon as the initial design was completed, the first prototypes were built. The PME assembly group proved to have real advantages as prototype builders and fixture testers. They could give engineers instant feedback on the product design and fixtures, from people who were more representative of bench work labor available in Colorado than the engineers would be. The nine PME assemblers were medium-skilled technicians paid $8.50 per hour. They could read blueprints, and they each learned to do a variety of jobs. All of them learned to put together major parts of an entire scanner.

As time became short, Burroughs proposed to do 100 scanner modules in Redwood City, while Colorado geared up for the system assembly first. PME would calibrate scanner assembly fixtures, assemble and test the engineering models, and then assemble and test the first saleable units. Then personnel from the Redwood City scanner laboratory would transfer to Colorado Springs to help assemble the first scanners there. Meanwhile, PME would assemble and ship the 20 saleable units to be delivered in September.

Disagreement arose over the number of scanner units that Redwood City should build. The engineers were pleased with the PME assembly's early performance and argued that at least 200 saleable scanner modules should be produced by PME. Colorado argued that it needed to have control over those important parts of the VPR2 systems. Prime costs, both materials and labor, would be twice as high in Redwood City. Labor rates in general were higher in Redwood City than in Colorado, and parts were much more expensive. While Colorado purchased in larger lots and could promise longstanding relationships with reliable suppliers, Redwood City bought only small numbers and was known among vendors for its willingness to pay a high premium for immediate delivery. It was also known to have no entry check on parts so that material delivered there often had to be reworked in the model shop. The adapter casting, one critical part of the scanner, cost $90 when purchased by PME in a lot of 150, $25 when purchased in an 800-unit lot by Colorado from the same vendor. In the end, the number of 70 saleable scanners was accepted as a compromise figure for PME to assemble.


The VPR2 program met each of its stated goals. Twenty completed systems were shipped on schedule in September 1978. The Colorado plant was turning out completed systems at the rate of 100 per month by January 1, 1979, though it had yet to assemble saleable scanners. Seventy scanners had not been enough to debug the new tooling, and Burroughs had transferred them later than planned. The material scrappage cost charged on the VPR2 product start-up was $10,000, a fraction of the amount budgeted.

Manufacturing engineers at the Colorado plant noted, however, that there had been additional costs associated with PME's handling of the VPR2 start-up that had not been captured by conventional measures. Since printed circuit board purchasing had been handled by the PME Group, for instance, rework in the model shop of perhaps $50,000 had been absorbed into engineering overhead rather than charged to manufacturing (see Exhibit 5 for unit cost information about the first 150 scanners assembled). The 11 complete systems assembled at Redwood City had prime costs 50% higher than expected steady state system costs were likely to be, while the first 10 assembled in Colorado were only 25% higher than steady state.

Customarily the Manufacturing Engineering Group in Colorado handled tooling for new product assemblies in their own plant. It had been a blow to their morale to be passed over on VPR2. Redwood City's decision to design the fixture for the VPR2 scanner was interpreted as a sign of low esteem for their capabilities at headquarters. They complained that the fixtures designed in Redwood City were $100,000 more expensive than they might have been, and that the tooling required too much judgment from assemblers. Peter Davis, an engineer at the Colorado plant, explained the consequences:

So far, we've been trying to put together the scanner here at the plant using selected employees with their old job grades and giving them merit pay for scanner work. But we may have to hire a more skilled grade of employee to get the consistent quality output we have to have. An extra $1 per hour will increase the prime cost significantly, but getting good output the first time will put less strain on our testing capacity and improve our volume performance considerably.

Because of the evident success of the VPR2 program, many key PME members were promoted out of the group. Burroughs was named operations manager in the Data Systems Division. Ferrante was given the title of engineering section manager when he succeeded Burroughs as PME head. Jerome Topping, head of Manufacturing Engineering in Colorado, praised the PME Group but expressed concern about the effect the promotions might have:

PME worked well as our eyes in Redwood City. But I am worried about what's happening now. The key to this type of group is continuity. It took us a long time to learn to trust the PME people as individuals to represent our interests fairly when we weren't there. Of course, the matrix structure and the new control procedures were partially responsible for PME's success, but I'd have to say that it was probably 70% people and 30% procedures. PME needs very strong interface types with a high level of authority, people who can insist on getting a part order released, for instance, where it's appropriate.

Even though no new products as critical as VPR2 were expected during the next year, there was clearly more than enough work to keep a PME group occupied. Already under Ferrante, PME was being deluged with requests to take on new types of work that the group had not originally been set up to do. Because of the small assembly group's skill and flexibility, several product managers wanted PME to build their saleables for custom products. PME had also earned a reputation as a good training center for lead assemblers through its training of the scanner assembly team from Colorado. Ampex's plant in Juarez, Mexico, wanted to send a group of leadmen to be trained by PME to assemble a new solid-state computerized editing system. Taking on either the training work or the assembly of saleables would soon require a larger PME staff, but Ferrante thought he could easily find 20 more people (mostly college students) to work as PME assemblers during the summer. The old PME quarters, located between Design Engineering and the model shop, would be too cramped to accommodate any extra people, but Ferrante knew of some unused storeroom space on a floor two stories below, where PME could set up expanded quarters.

PME was having some problems with other new product transfers it was coordinating. The VPR20, portable version of the VPR2, was supposed to be transferred to Sunnyvale, but had slipped far behind schedule at the prototypes stage. The design engineer in charge of the VPR20 had taken the project out of PME to build the demonstrator models in Engineering on a six-week crash effort before the NAB convention in March 1979. PME members blamed the delay on a single-source part that had not been delivered on schedule, while the designer blamed PME for paying so much attention to building saleable VPR2 scanners that they missed their deadlines on his lower-priority project.

PME Future

Inside Ampex, ideas differed as to the role PME should play in the aftermath of VPR2. Certain design engineers argued that their new products should now go directly to Colorado. They cited recent quick response time from the plant as evidence that Colorado's New Product Manufacturing Team was becoming effective. Warner wondered whether institutionalizing PME would make it a less effective form of organization to use in a crisis -- "a peacetime army."

Doug Grantham, as new engineering manager, wanted to keep PME as a short-run production facility under the control of Engineering. It could continue to produce prototypes and demonstration models, and it could do saleables when its services weren't otherwise required. "Of course what I'm really saying is I trust Bob Ferrante, not PME. PME without Bob Ferrante, I'm less sure of. Who knows whether the person who succeeds him will be as capable?" Herold commented:

Ampex can do better than PME. What really needs to happen is for the factory to be involved directly in the design process. Now that the New Product Manufacturing Team has some experience, they should be able to manage the transition alone. We might consider a small liaison group at Redwood City, but they should be factory generated and under factory control.

Of course, it suits Engineering to keep manufacturing capabilities limited. Your average high-technology company represses innovation in the factory because it inhibits innovation in design. But we're no longer in a controlled marketplace willing to wait for its products.

Past and present members of PME also disagree as to if and how their group should continue. Ferrante wanted an ongoing organization with a flexible role, "You can't have fixed and hard rules about what should be done for every project ahead of time. On some we just check the documents; on some we can proof the design; for some we handle purchasing, and for some we might go so far as to build saleable units."

Burroughs' Response

Burroughs began writing his answer to Hirschfeld:

TO: M. Hirschfeld
FROM: T. Burroughs
DATE: March 31, 1979

As I see it, Morris, you have three options. You can keep PME as it is, a grass roots organization reporting to Engineering; you can put an end to it and leave its most important functions to be picked up by the New Product Manufacturing Team, or you can give PME the status I think Ampex's business now demands and appoint the head of PME as a general manager reporting directly to you.

Such a person could be involved in the earliest project planning session and could take part in new process decisions that would affect many new products. The decision whether to adopt automatic component insertion equipment for electronics subassembly is an example. This person would have the power to keep good integrators in the group and develop them, not have them promoted away when they showed capability. Finally this manager could insist that it be standard procedure for all new products to go through PME before entering the factory.

Braun AG: The KF 40 Coffee Machine (Abridged)

"If we're going to do it, we've got to quit stalling," exclaimed Albrecht Jestädt, head of development for a new coffeemaker at Braun AG. "I've said all I can about polypropylene, and I'm convinced we can go with it," he added, taking another sip of his beer.

At the end of the day in January 1983, Jestädt and his colleagues were discussing Braun's newest design: an elegant, cylindrical coffeemaker, called the "KF 40," destined for the mid- and upper end of the mass market. To meet management's cost targets, however, they would have to use polypropylene, a much less expensive plastic than Braun's traditional material, and whether so doing would jeopardize Braun's reputation for quality was a matter of intense debate throughout the company. Unlike the very expensive polycarbonate, Braun's traditional material, polypropylene could not be molded into large, complicated parts (like the KF 40's "tank") without suffering so-called '"sink" marks on surfaces that were supposed to be flawlessly even. So the designers had devised a solution that involved a major departure from the smooth, winter-white surfaces characteristic of all Braun household products. (See Exhibit 1 for a prototype.)

"The decision is obvious," claimed Gilbert Greaves, business director for household products. "We need this product now, and we have to stop being quite so picky."

"I think we should be picky," said Hartwig Kahlcke, the industrial designer on the project. "But we feel that the rippled design for the tank actually enhances the surface appearance, without compromise."

"Maybe," said Hartmut Stroth, recently appointed director of corporate communications. "But it's no trivial matter. It's true that if we lose a year, we might not get in the market at all. Yet nothing is worth losing our reputation for superior quality. Not even the mass market." Stroth, who had served for over a decade in various communications positions at Braun, was very sensitive to the importance of Braun's "visual equity" and the need for maintaining it: "Not only do we have to think hard about how this corrugated surface design would fit into the Braun 'look,' but also about what that look represents. The idea of using design to mask sink marks bothers me in principle, and it may not work in practice, especially if the stuff doesn't hold up. I'm anything but risk-averse in this business, but I need to be convinced."

"Then let's go ahead with the trial tooling," Kahlcke responded. "The chairman has already okayed it; maybe that will convince you." Not waiting for Stroth's response, Kahlke inquired about the chairman's views to date. "I know he liked the design, and I know he wants the product. What does he think about the material at this point?"

"You tell me," answered Lorne Waxlax, chairman of the board and Braun's CEO. He would have to make the ultimate decision and had just dropped in, as was his custom, to get the latest thinking on the KF 40.

Company Background: Braun by Design

Braun AG began as a family-owned radio and small appliance business founded in 1921 by Max Braun. In the 1940s, Braun developed a novelty, the electric razor, which he introduced in 1950. After Braun's death in 1951, his sons, Artur and Erwin, took over, and three years later they asked their friend, Fritz Eichler, an artist then working in the theater, to help them find a new approach to their struggling business. In 1955, looking for an architect to help build a new office building, the company hired Dieter Rams, just two years out of architecture school.

Rams became Eichler's protégé, and together they built a small, intense design department at the company's headquarters in Kronberg, Germany. Convinced they could change the taste of their fellow citizens, Eichler, Rams, and colleagues set out to design and build a new kind of product. (See Exhibit 2 for Rams' ten commandments of good design. )

Eichler and Rams believed that their design philosophy should permeate the company, providing a recognizable identity not only in its products, but in every aspect of its relations with customers. (See Exhibit 3, "The Principles of Braun's Corporate Identity.") In Rams' view, achieving that identity required top management support of good design, and team work -- constant interaction among disciplines. But designers also needed certain responsibilities and authority; otherwise, they would arrive, at most, at "superficial product cosmetics."

According to Rams, designers needed four things. First, they had to be responsible for configuring all elements of the product that would influence its final appearance. Second, designers needed the authority to determine the dimensions of a product -- e.g., the positioning and ergonomic design of its operating functions; third, they must be the ones to decide on surface structures, colors, product labeling and imprinting; fourth, they needed to cooperate with the engineers on construction problems -- e.g., manufacturability -- whenever the form of a product directly depended on the construction.

Although Rams was not without critics, his work was an effective counterweight to the popular assumption that designers merely dreamed up the external form of a product. Moreover, he adamantly stood by his own definition of "functional": that the purpose of good design is to fulfill the primary function of a product, including its need to be appealing to the user so it would be a welcome object in his or her environment.

By the mid-1970s Braun had built a thriving business, primarily in small home appliances (e.g., shavers, coffeemakers, and mixers), with additional sales in consumer electronics (e.g., cameras and hi-fi equipment). Further, the Braun design group was succeeding in its mission. One of their first products, a heavy-duty kitchen mixer (1957), was still in production and selling well. Most famous was their shaver, familiar to men all over the world. Many Braun products had won design awards; 36 of them, including Braun's first coffeemaker, had found a permanent place in New York's Museum of Modern Art.

The company's mission was carried out not only in its products, but in its people. The company's principles had permeated its corporate consciousness and were second nature even at lower levels in the organization. Almost any employee could tell a visitor that Braun's values were embodied in its products, which had to have three characteristics: (a) first-class design; (b) superior quality; and (c) functions or features ahead of the competition. "We'll never bring out just a me-too product" echoed in every department.

The Gillette Connection

In 1967, the Braun brothers sold the company to an American consumer products giant, Gillette, well known for its mass-produced, mass-marketed products like razors, blades, and toiletries that had been marketed in Europe since the turn of the twentieth century. For the first several years, Gillette left Braun's product strategy intact while infusing some of its management expertise into the organization. In fact, very few people knew that this German company par excellence had an American owner. But Braun soon began to expand its operations in other countries and extend its target markets beyond the opinion leaders it had originally cultivated.

For example, in 1971, Lorne Waxlax, a Gillette manager since 1958, took charge of Braun's Spanish plant. He largely refocused the operation, emphasizing product development, sophisticated manufacturing, market research, and television advertising. The plant manufactured Braun's first successful mass-produced kitchen appliance, the hand blender, and served as the training ground for Braun's mass-produced appliance motors. By the early 1980s, Braun's sales exceeded $400 million (see Exhibit 4).

Braun's Organization and Operations

Braun AG in 1983 was organized into three main functions: business management, technical operations, and group sales. (See organization chart in Exhibit 5.) Business management, a coordinating group established in 1976, was essentially strategic marketing. Until Gillette came along, Braun had assumed that if one made a good product, it would sell. And it generally did. But in 1975, marketing became more important, as the domestic and international marketing people came together under a single group. A director for each product group reported to the head of business management, as did the director of communications, which included packaging.

Braun invested heavily in technology, and all key technically related disciplines were based at company headquarters under Dr. Thomas H. Thomsen, recently appointed director of technical operations. Previously, he had been head of engineering for Gillette in both London and Boston. Technical operations comprised four functional groups: R&D, Manufacturing, Quality Assurance, and Industrial Design.

The research and development department employed 220 people and included scientists and engineers working on advanced technology, as well as those involved in product development. R&D was headed by Dr. Peter Hexner, an American ex-army colonel who had directed Gillette's advanced technology department for 10 years. He commented on the challenge of balancing technology with the demands of design: "We have the classic conflicts. Design wants an elegant shaver a centimeter thick, and I have to knock reality into their heads: 'You can't fit a motor into a case a centimeter thick.'"

Process development and manufacturing engineering were part of the manufacturing organization, which managed Braun's component and assembly plants. Over three-quarters of Braun's manufacturing activity took place in its two large state-of-the-art plants in Walldüurn and Marktheidenfeld (smaller operations were located in Spain, Ireland, Mexico, Argentina, and Brazil). Because of German labor's high cost, the manufacturing organization was continually pressured to produce efficiently, particularly in a plant like Marktheidenfeld that produced a wide variety of low volume products including kitchen appliances. (For example, the plant produced around 700 units of the KF-35 coffeemaker per day.) While efforts to improve operations included automation, especially assembly, more challenging was designing a product so it required minimal assembly. "Anyone can make a cheap product with many parts and hire cheap labor offshore to screw them on. But not everyone can reduce as many parts as possible to one," said Bernard Wild, plant director at Marktheidenfeld. "We can prove that advanced industrial nations don't have to forfeit manufacturing just because their labor costs are high. Our resources are in our brains and imaginations-our know-how."

Quality assurance was responsible for analysis of competitive offerings and rigorous testing during the product development process. Because Braun insisted that all its products be better than those of all the competition, the quality group relentlessly pursued the smallest detail with very high standards. As Werner Utsch, a quality engineer, commented, "We take them apart down to the last screw."

The fourth group in technical operations, industrial design, had an impact on the company far beyond its 16-member size. Indeed, Dieter Rams, head of the department, felt that small size was an important ingredient in its success. The department employed seven designers, most of whom had won the Braun Prize, a design award the firm had offered to design students since 1968. By 1983 Rams' international stature often resulted in his being equated with Braun design almost exclusively. Yet he found this star status awkward: "I constantly have to stress that I don't do everything; I'm simply the motor that drives the department. I try to give other people the credit they deserve."

Until recently, Rams had reported only to the chairman. But because of time limitations, Waxlax assigned industrial design to technical operations, where most problems could be solved. The direct line to the chairman remained, but was used only for the most important issues and impasses. Rams noted, "I've had a good understanding with every chairman I've worked with. But often designers aren't so lucky. We often educate business management people to the point where they begin to understand design and are supportive to us, but then they leave."

Product Development: The Triangle of Power -- Design, Technology, Marketing

Product development had been relatively informal until 1980, when three people, representing R&D, business management, and manufacturing engineering, came together to develop procedures to make the process more operational and efficient. The result was a product development manual, introduced in 1981, that covered the responsibilities of key persons in a team (called a MTS team, for marketing-technology-strategy), definitions of elements in project development (e.g., different kinds of models), product specification guide, stages ("categories") and signoff points in the process. (See Exhibit 6 for the project manual's table of contents.)

The "product program manager" (PPM) was responsible for maintaining these procedures; he or she chaired team meetings and represented the team vis-&$224;-vis management, reporting directly to the head of technical operations or business management. The team itself had no formal leader. Various people took over as the stage in the product development process dictated, and stronger personalities could be influential; Jestädt, for example, first as product program manager, then as R&D manager for coffee, had quickly emerged as the de facto leader of the coffee machine project. (See Exhibit 7 for the PPM's and team's formal responsibilities.) In addition to the core team, people from other groups and disciplines -- sometimes as many as 40 -- became involved as the project proceeded.

The team's monthly reports to the chairman had a standard format, divided into four sections: Description, Status, Further Steps, and Problems (or Risks). Although these monthly reviews were considered effective in motivating people to move toward the project goal, Waxlax did not like to use them as a threat: "The trick is to know whether the deadline is truly viable or not. It's easy for marketing to insist on a deadline -- they don't have to do the work. I believe the engineers know better than I how fast the team can go, and for that reason I don't want to force it unduly."

Waxlax saw the meetings as an efficient way to keep up to date on all that was going on and to keep on top of problems and conflicts as they arose. He didn't believe in minimizing conflict, but saw it as positive for the company: "It's often the guy who is against something who forces it to become better." lie also viewed the monthly meetings as "a chance for me to encourage people," he added.

The point at which a project became formal and began to adhere to the Projektablauf [Project Procedures] varied. If, for example, a project had proceeded informally rather far in its development before entering the formal product development process, it might simply be formalized and have product specifications delineated. In its early stages, a project like the KF 40 might have provided monthly reports to the chairman for some time before becoming a formal project. (See Exhibit 8 for the Braun product development process line. )

The industrial design department played a central role in development, particularly at the front end of the process. Because most key disciplines at Braun were located in the same building, much communication about development took place informally, and no one really kept track of where ideas came from and when they first got together with a colleague from another department. That design, because of its reputation, often received disproportionate credit for a product occasionally irked some engineers and scientists, whose contributions were less visible. Well aware of this problem, the company's chairman accepted the responsibility of keeping the rivalry healthy.

Industrial design's relationship to other departments varied. Within the "triangle of power" (design-technology-marketing), design felt most akin to technology. The designers kept up with new developments in such fields as materials science, for example. "We understand technology, so when we have an idea, it is not unrealistic technically. We don't come up with totally impossible ideas," explained Rams. Likewise, with manufacturing, the group knew what it meant to design a product for manufacturability; if they were having difficulties, all they had to do was to go down the hall and across the parking lot to the engineering building. Such interaction among all disciplines was daily fare at Braun.

Marketing was something else, however. Business management often had conflicts with design because, said the marketers, the latter insisted on certain principles that were not always viable in the marketplace. "The problem with designers," Greaves, director of household products, sighed, "is that they think they design for eternity. Rams will hand me a 1965 design and expect me to go for it today." Sometimes the conflicts were trivial. For example, "one time we argued with Kahlcke (an industrial designer) over the baseplate of the mixer. Because of his obsession with details, he wanted it changed. I told him that was ridiculous, since no one would ever see it," recounted Greaves. "The cord storage was not in the base, so there was no reason whatsoever to turn the mixer over. But Kahlcke got his way!"

People in industrial design had a different perspective. Noted Rams: "I don't mind if technology has greater influence than design; we understand each other and can work things out. But when marketing gets power, it can be bad." For example, sometimes marketing got its way with regard to color. "Why should we pay attention to color fads? Just because red cars are popular one year, why should we have a red hair dryer? It is not integral to the design."

A New Strategy

Lorne Waxlax became chairman of Braun in 1980. After his five-year stint in Spain, he had managed Braun's non-Central European export business for three years, and then headed business management for two. Waxlax had long wanted to encourage Braun to get rid of cameras and hi-fi and to focus more effectively on its core technologies in the personal care and appliance businesses. As chairman, Waxlax could proceed with this strategy.

While narrowing the product line, Waxlax also saw opportunity in six segments of the consumer appliance business: coffeemakers, irons, toasters, hairdryers, shavers, and food preparation products. They were big, and they were constant; the market for coffeemakers in Europe alone was over 9 million units per year. Even a small percentage share would make a good business, but Waxlax was "always going for a big share." Both business management and Braun's designers eagerly embraced this new strategy. Rams made very clear his philosophy: good design should be for everyone.

By 1983 Braun was well established in several product families. Electric shavers were their biggest and most widely marketed product line, accounting for half the company's revenues. In many countries Braun held first place in market share; wherever they were present, they were among the top three. The household division, whose image was represented by its classic kitchen machine, produced coffeemakers, mixers, juice presses, food processors, food choppers, and irons as well. In the personal care area, hairdryers and curling irons were the most successful, having achieved market leadership in Europe. Braun's exports were continuing to grow: in 1982 exports accounted for 75% of its turnover.

Waxlax and his top managers wished to focus on Braun's core products and expertise, its reputation for excellence in design, and the opportunities within grasp at the upper end of the mass market. Balancing these three dimensions of the company -- technology, design, and business management -- while maintaining the integrity of its corporate mission was management's key challenge.

Coffeemakers: The KF 40 Project

One August day in 1981, when half of Germany was on holiday and the other half getting ready to leave, Waxlax wrote a memo to Gilbert Greaves, asking him to check into the possibilities for Braun in coffeemakers, a key element in the new strategy. Braun had entered the coffeemaker business in 1972 with the KF 20, a novel cylindrical design that won many design awards and was enthroned in the Museum of Modern Art in New York. (See Exhibit 9.) Available in strong red and yellow, it had entered the consciousness of upper-income coffee drinkers in both Europe and the U.S. It was, however, a very expensive machine, and expensive to produce, retailing at about DM 120.

A few years later, the company introduced the KF 35, a sleeker version of the then popular "L-shape" epitomized by Mr. Coffee. It cost about 40% less than the KF 20 to produce, retailing at about DM 90. The design department was not fully satisfied with it: "I always thought it looked like a chemical lab sitting on the table," declared Rams, disdainfully. (See Exhibit 10.) The unit enjoyed only average sales, about 150,000 units annually.

Braun's major competitors in the middle-to-high-end coffee machine segment were two German companies, Krups and Rowenta. Braun's market research defined this segment in terms of price points: DM 70 retail or above. In Germany and France, two of the biggest markets for coffeemakers, half the units were sold in that range. The market researchers were confident that a new Braun coffeemaker family offered across the entire spectrum from DM 72 to DM 136, would be competitive in Europe, where the greater part of the market (about 70% of 9 million units annually) was for replacements. An open question was how soon Krups and Rowenta would copy Braun's design, as was their custom.

The U.S. market was considerably less certain yet crucial if Braun was to attain a volume permitting a tolerable return on investment. "You don't go for the small appliance business because of the margins, You have to have high volumes," remarked Waxlax. Americans had been introduced to filter coffee systems through Mr. Coffee, a low-end product. Would they be willing to pay for Braun quality? With the currency exchange rate at DM 2.40 to $1.00, it was reasonable to expect imports to the U.S. to grow. The market for filter coffee machines was already running at about 11 million units, and penetration was still low. Braun's distribution system in the U.S. was practically nonexistent, however, and Waxlax wondered if it could be sufficiently developed in time.

Upon receiving Waxlax's coffee machine memo in August 1981, Greaves set to work with his people and in two months came up with a rough description of a product Braun could sell:

It should have a shortened filter, slimmer jug [than the competition]; it should come in different colors, the water tank should be opaque, the tubes should be completely covered up, the filter should be tight and compact, the thermal jug should be more elegant, lighter, handier, taller, slimmer, and presentable on the coffee table.

That idea, which marketing articulated in October, 1981, evolved into a "product profile" that Greaves circulated to key people on December 2, 1981. In this memo Greaves discussed, as Waxlax had asked, the issue of the cost/volume relationship, and presented the direct costs and price points in connection with assumed volumes. He also analyzed the market segments and defined those segments where Braun could realistically compete. He determined that (a) a range of models -- at least two -- would be necessary; (b) this range was defined so that it could be constructed on a "building block system" to "minimize tooling investments"; and (c) the range would enable Braun to compete in medium to high price segments, at retail prices about DM 70. This would mean that Braun had to compete with key players -- Krupers, Rowenta, Siemens, and AEG -- on feature, not price. To be profitable, it would have to cost fully one-third less in direct costs than the KF 35, or 60% less than the KF 20. (See Exhibit 11 for summary of Greaves' memo.)

When the document was sent to R&D for feasibility analyses, R&D's first reaction to the target costs was "Nonsense! You can't make a coffeemaker for DM 23 in this company!" Nor did engineering take the idea well: "To be honest," confessed Hans-Jürgen Dittombée, manager of industrial engineering, "we thought the cost targets were impossible. We are responsible for technical planning and didn't see how we could get there."

Working with Greaves on the project was a young, energetic product program manager, Albrecht Jestädt, a mechanical engineer with experience in production and engineering. Upon hearing R&D's and Engineering's reactions, Jestädt refused to take "no" for an answer, and set about looking for alternatives. If Braun can't manufacture it, at least we can sell on OEM product that we design, he reasoned. Over the next year, he explored options in and outside of Germany and managed to find a manufacturer in Switzerland who could meet the cost requirements.

The KF 40: Problem Solving in Development

In the meanwhile Jestädt and the designer for household products, Hartwig Kahlcke, teamed up and began to develop the product. Kahlcke, a quiet contrast to the ebullient Jestädt, had come to the design department 10 years earlier, and worked on the KF 35. Kahlcke had also dealt extensively with the Spanish group because they did so many household products. "We share a vision," Jestädt declared, "and we're both willing to do what we have to to realize it." That vision had as its starting point the KF 20 and its still novel cylindrical design. How could they use the cylinder within the cost parameters? They had to use less material and only one heating element to start with. "Our first design was really terrific -- the water tank completely surrounded the filter. But then we realized that we had to think modularly, so manufacturing costs would be minimized, and so we had to drop it," Jestädt recalled.

Jestädt and Kahlcke knew that the cylindrical form not only was appealing, but it used less material than the "chemical lab," the KF 35. Going back and forth they came up with five or six blue foam models before settling on what they believed was the optimal configuration: a cylinder within a cylinder, operating on the same principle as the KF 35 but much more compact. The main novelty: It would be operated from the front, and thus it would take less space and look even slimmer on the kitchen counter. (See Exhibit 12 for initial concept.)

At the same time, other disciplines continued working on the project. R&D, after exclaiming "impossible" at the very idea, took up the challenge and looked at how to get the cost out of the heating element and many other dimensions of the machine. The cost target presupposed a single heating element for both heating the water and keeping the coffee hot, rather than the two needed for the KF 20. Within those parameters, they finally decided that they could go for aluminum rather than copper in the heating element, which would be cheaper, but it would mean different dimensions for the various parts because of differences in conductivity. Keeping the coffee temperature at 82°C was considered an absolute must by the designers and marketing alike, even though they knew it was essentially an insoluble problem because of the length of time that coffee might be held.

R&D also responded to new design concepts. Kahlcke and Rams, for example, wanted to glue the handle on the pot, and asked Engineering to explore adhesives. The design reasons were both aesthetic and functional: The conventional means of attaching the handle to the coffee jug was the metal band, which both interrupted the line of the jug and collected dirt. In the course of working on the adhesives, it had become clear that manufacturing engineering would have to design an automated gluing process, in order to keep the costs down. The good news by spring 1983 was that the design and manufacturing process was expected to cost less than the conventional metal-band method. R&D still had not found the ideal adhesive, however, one that would hold for years under heat, impact, and moisture.

R&D had other challenges. The marketing people had found that an anti-drip device would be very attractive for customers, but Braun wanted to go at least one step beyond the competition. The idea was to prevent drips either from the filter (when one pulls out the coffeepot) or from the water tube (when one swings the filter out). It was supposed to be a relatively easy assignment but, as Gunter Oppermann, head of R&D for household products, pointed out, "Simple is most difficult," and that was what the project was about. The dripstop was a case in point. "It has to be dual-action (stopping the flow when either the pot or the filter was pulled out), and we have to go around some outside patents. We thought about toilet flushers as a model and started from there. We didn't want the device to stick, and yet it must be sturdy."

Quality assurance was working on several aspects of the new machine, including its end product. "We found that we didn't really know anything about coffee," quality engineer Werner Utsch confessed, "so we had to analyze and test some more, and that has led us to work with the coffee producers." The tests revealed valuable information: "We have found that our competition doesn't know much about coffee either." The next step would be blind taste tests, for which they needed a functional model.

The market researchers continued gathering data as well. In October 1982 they tested the thermal jug concept and determined that it would be an essential selling feature. The next month they tested filter systems; the swivel filter won hands down. At this time, contrary to results a year earlier, the market wanted a detachable, transparent water tank. It was "significantly preferred over a nontransparent one" and "should be included...if the price is not prohibitive to the customer." (See Exhibit 13 for market test results.) Jestädt and Kahlcke had, however, already developed a modular design that could not accommodate a transparent tank.

Operating from the principle that "no parts = no assembly costs," Jestädt and Kahlcke were striving to collapse the number of parts into as few as possible. This was where Rams' motto, "To design means to think [Designarbeit ist Denkarbeit]," converged with Bernard Wild's view of Braun's know-how. Working with machine tool experts, headed by Friedhelm Bau, Jestädt and Kahlcke had designed a configuration that incorporated many large and small parts that in the past would have been screwed together. The water tank was now part of the appliance housing, and the whole large piece, known simply as the "tank," was now central to their product concept, for it accounted for a good chunk of the savings in assembly costs. (See Exhibit 14.) It was, however, the largest, most complicated part ever attempted in polypropylene injection molding at Braun, and as such would be risky. (Exhibit 15 explains injection molding.)

Manufacturing and toolmaking engineers were involved from the beginning of the project. Bernard Wild had prepared an analysis of plant requirements in order to achieve the projected volumes for the new coffeemaker. As soon as polypropylene was proposed, Bau's toolmaking department started working with plastic suppliers and toolmakers in Berlin, who had experience with designing large tools for polypropylene. Bau was convinced that the large tank could be molded on the three 330-ton molding machines (presses) available at Marktheidenfelt. One machine could make 1,500 moldings (tank units) per day. With estimated volumes at 500,000 the first year, ramping up to 2,000,000 units the fourth year, the plant needed to be prepared to manufacture 10,000 units per day, given the 220 days per year that the plant operated. They were assuming a one-minute cycle time for the "tank" part, but could not be certain of it. They could start with three molds (or "tools"), one each for the 10- and 12-cup units and one for the thermal carafe, but they preferred to have the flexibility offered by five molds -- two each for the 10- and 12-cup units.

If the product took off as expected, they would need four more molding machines and at least as many molds. Each machine cost about DM 500,000. The estimated cost for each tank mold was DM 250,000 and the lead time for tooling was around nine months for the large molds. Because the molds were not interchangeable for various types of plastics, the choice of plastic was crucial to engineering's planning.

Polypropylene: A Question of Braun-ness

Braun had pioneered in the use of plastics as early as the 1950s, when it rejected fake wood and overstuffed designs for its products. Its designers, engineers, and toolmakers were experienced in making both clear and opaque parts from several different kinds of plastic. For the outer housing of its appliances, the company had traditionally used polycarbonate, a dense, stable material that could be fashioned into precision parts with smooth surfaces. Polycarbonate was, however, too expensive for the new coffee machine's requirements.

For that reason, Jestädt had begun working with ABS, which sold for about half of polycarbonate's going rate (see Exhibit 16). Even that, as it turned out, would probably be too expensive. The alternative, polypropylene, was the material of choice for low-end producers, but had never before been considered by Braun, except for interior parts that could benefit from its lower density and other features. The amount of polypropylene needed for each KF 40 unit was estimated at 700-950 grams. The problem with polypropylene for use in injection molding was its instability during the cooling process. Having a lower specific weight than the denser plastics, it tended to shrink unevenly and fall off, or "sink," at edges and meeting points. The resultant "sink marks" marred the surface and looked "cheap." Nor did polypropylene become as rigid as the more expensive materials, thus posing additional design challenges. Large parts were therefore especially vulnerable to a flimsy feel and had to be designed with the need to control that problem. It might mean thicker walls or a shape in the mold that would buttress the form from within.

When polypropylene was first suggested, many colleagues familiar with its problems immediately objected: It will not be a Braun product if we use this cheap stuff, they warned. Despite such adamant objections, Jestädt and Kahlcke began working with chemical suppliers and toolmakers to explore ways of improving the quality of polypropylene parts. In fall 1982 they achieved a breakthrough: Why not let necessity be the mother of design in this case? If we can't get a perfectly smooth surface, let's minimize the effect of the sink marks by treating the surface in some way. This inspiration led to the idea for a corrugated surface that would both mask flaws and actually enhance the design as well.

NO! said the purists, for whom Braun design was synonymous with absolutely smooth, winter-white surfaces. "It's a compromise," said Utsch, "and I don't like compromises." Utsch, head of quality assurance for the project, kept pointing to polypropylene's tendency to scratch: "It's just too soft. Even a fingernail can scratch it. And if you wipe it off with the same sponge you wiped the counter off with, you can scratch it with food particles or coffee grounds." Even Rams was skeptical at first, but eventually came to support the solution. "It is the obvious way to go, given the project requirements."

Polypropylene did have some advantages other than its price, Oppermann pointed out: "It doesn't absorb water, so it won't stain easily. And, as far as we can tell, it won't get brittle as fast as polycarbonate, so it won't chip easily."

Jestädt, ever confident, explored further. Wiling to take risks, R&D director Hexner supported R&D's involvement in trying to make polypropylene work. Like everyone else, he knew that if it didn't work, it would be extremely costly. "They are talking about a huge and very complicated tool for the tank. If it doesn't work, we'll have to throw it away and be another year behind." But Hexner didn't see any choice: "We've been given the job of making this thing at a ridiculous cost. My people say that it's possible only with polypropylene, and I agree." To Hexner the "purists" were entirely unrealistic. "If a Braun product has to have a smooth surface, then you have two choices: Go with flaws, or forget it. And that is ridiculous!"

Hexner's boss, Dr. Thomsen, did not think it ridiculous to consider further choices. Nor did Waxlax. "We could make a business with, say, ABS. But it would be a different business," Thomsen contended. Waxlax was worried about the U.S. market implications: "We'd either have to drop the U.S. market, and that means low volumes, or restrict it to the higher-priced department store segment."

Jestädt and Kahlcke, meanwhile, were not insensitive to the design concerns. The ridges of the corrugated surface would have to be absolutely smooth, with no peaks or valleys, so that they would not catch any dirt. That job was turned over to the toolmakers. By the end of 1982 Bau's department was confident that the job could be done using the 330-ton molding machines at Marktheidenfeld. An outside consultant had suggested that the molding machines should be larger (500-ton) for a part the size of the tank, but that would mean an additional investment of DM 2 million for two new machines and upgrades of the old machines. Because the larger machines were much slower than the smaller ones, it would take five of them to produce the same number of units per day as the three 330-ton machines could produce.

In December, 1982, Jestädt had presented his plan for a Swiss company to manufacture the new design. At the same meeting, someone brought in a cheap DM 29 coffeemaker from a supermarket and challenged those present, "If these guys can sell a coffeemaker for DM 29, you can surely make one for DM 23." As the discussion proceeded, the group realized that the new design was so special that it would be dangerous to let it out to a subcontractor; they would have to keep it inside in order to assure a competitive lead.

At that point it was proposed to take three months and build a trial tool to test the material; Waxlax approved DM 140,000 for the test and the tool, if the team chose to take that step. The proposal was, according to a project report for December 12, simply to "clarify if polypropylene is suitable for the appliance housing material." That was the point, according to Dittombée of industrial engineering. "I am confident that we can master polypropylene technically," he said, "but the discussion is about whether Braun can -- or should -- use it." For the purists, such a trial was far better than ordering the production dies and finding out polypropylene wouldn't work in this design and product.

A Material Decision

Over the next four months the coffeemaker project became more intense. At the report to management at the end of January 1983, drawings for the functional model were presented, and a schedule established (see Exhibit 17). The new 10-cup coffeemaker now had a name: the KF 40. A second model, the KF 45, would have a 3-4 cup switch, costing one DM more. According to this schedule, the functional model would be ready by the end of March, with final tool drawings complete on May 16. The formal go/no-go decision would be made on May 17, followed by a "category I" signoff, which released the drawings so that tools could be ordered. Production ramp-up was estimated to begin in April of 1984, to reach 3,000 units per day within three months.

All this assumed that the KF 40 could be made with polypropylene and that all the other problems, such as the drip-stop, could be solved in time. By producing this schedule, business management had already cast its vote of confidence. Waxlax knew that Greaves tended to be conservative in his forecasts, and therefore one didn't have to worry about unrealistic figures in his analyses. Neither engineering nor design wanted to be pushed, however, and that Waxlax respected. The decision was a strategic one: a big risk -- but one with a big payoff if they succeeded. The risk was not so much financial, though a million DM in molds and two years in development costs would not be insignificant. Should they go ahead without trial tooling, take three months for the trial test, rethink their positioning with a more expensive plastic, or walk away from the project? What risks were they willing to take and how far should they go before modifying the business strategy? Waxlax intended to take his time in listening to all points of view.

Copyright © 1993 by Kim B. Clark and Steven C. Wheelwright

Read More Show Less

Customer Reviews

Be the first to write a review
( 0 )
Rating Distribution

5 Star


4 Star


3 Star


2 Star


1 Star


Your Rating:

Your Name: Create a Pen Name or

Barnes & Review Rules

Our reader reviews allow you to share your comments on titles you liked, or didn't, with others. By submitting an online review, you are representing to Barnes & that all information contained in your review is original and accurate in all respects, and that the submission of such content by you and the posting of such content by Barnes & does not and will not violate the rights of any third party. Please follow the rules below to help ensure that your review can be posted.

Reviews by Our Customers Under the Age of 13

We highly value and respect everyone's opinion concerning the titles we offer. However, we cannot allow persons under the age of 13 to have accounts at or to post customer reviews. Please see our Terms of Use for more details.

What to exclude from your review:

Please do not write about reviews, commentary, or information posted on the product page. If you see any errors in the information on the product page, please send us an email.

Reviews should not contain any of the following:

  • - HTML tags, profanity, obscenities, vulgarities, or comments that defame anyone
  • - Time-sensitive information such as tour dates, signings, lectures, etc.
  • - Single-word reviews. Other people will read your review to discover why you liked or didn't like the title. Be descriptive.
  • - Comments focusing on the author or that may ruin the ending for others
  • - Phone numbers, addresses, URLs
  • - Pricing and availability information or alternative ordering information
  • - Advertisements or commercial solicitation


  • - By submitting a review, you grant to Barnes & and its sublicensees the royalty-free, perpetual, irrevocable right and license to use the review in accordance with the Barnes & Terms of Use.
  • - Barnes & reserves the right not to post any review -- particularly those that do not follow the terms and conditions of these Rules. Barnes & also reserves the right to remove any review at any time without notice.
  • - See Terms of Use for other conditions and disclaimers.
Search for Products You'd Like to Recommend

Recommend other products that relate to your review. Just search for them below and share!

Create a Pen Name

Your Pen Name is your unique identity on It will appear on the reviews you write and other website activities. Your Pen Name cannot be edited, changed or deleted once submitted.

Your Pen Name can be any combination of alphanumeric characters (plus - and _), and must be at least two characters long.

Continue Anonymously

    If you find inappropriate content, please report it to Barnes & Noble
    Why is this product inappropriate?
    Comments (optional)