Design Outsourcing in the High-Tech Industry and its Impact

on Supply Chain Strategies

by

Bindiya Vakil

Master of Business Administration

Sardar Patel University

Submitted to the Engineering Systems Division in Partial Fulfillment of the

Requirements for the Degree of

Master of Engineering in Logistics

at the
MASSACHUSETTS INS
Massachusetts Institute of Technology OF TECHNOLOGY
E

June 2005
JUL 15 2005
c 2005
Bindiya Vakil LIBRARIES
All rights reserved

The author hereby grants to MIT permission to reproduce and to

distribute publicly paper and electronic copies of this thesis document in whole or in part.

Signature of A uthor ............................................................................

Enineering Systems Division
6th May, 2005

Certified by ...................................................
Dr. Donald Rosentield
Sr. Lecturer, Sloan School of Management
Thesis pervisor

Accepted by ............................................................. -- --- ----- --- -- ......

/ 1/t Yossi Sheffi

Professor of Civil nd E ironmental Engineering
Professor of Engineering Systems
Director, MIT Center for Transportation and Logistics
BARKER
 Design Outsourcing in the High-Tech Industry and its Impact
on Supply Chain Strategies
by

Bindiya Vakil

Submitted to the Engineering Systems Division

on 6th May, 2005 in Partial Fulfillment of the
Requirements for the Degree of
Master of Engineering in Logistics

Abstract
The practice of outsourcing product design to a service provider is prevalent in fast-moving high-
tech electronics. Product design affects all subsequent activities such as procurement,
manufacturing, logistics and after-sales services. Thus, companies do design for supply chain
(DFSC), which is the process of designing products by incorporating such supply chain
considerations as supplier selection and integration, single vs. multi-sourcing, component
commonality, postponement, variety management, product reuse and recycling, planned
obsolescence, etc. Each of these strategies involves tradeoffs which become difficult to evaluate
in an outsourced environment. The levels at which companies engage with their service provider
such as fully outsourced or joint design, affect the degree of control they can exercise over the
design process. The main finding is that when products are fully outsourced, companies have
less control and therefore, a lower possibility of implementing DFSC. Certain measures are
proposed whereby companies may influence DFSC implementation even in fully outsourced
design. When some of the design is kept in-house in a joint design model, the level of control is
high. Another important finding is that design service providers actually achieve excellence in
incremental innovation and do implement DFSC strategies which have cost-saving and revenue-
generating benefits for them. The distribution of the resultant savings is primarily a function of
the company's ability to understand the service providers' cost structure and its negotiating
power in the relationship. Two case-studies are included of products where different outsourcing
levels are used and the impact of the level of outsourcing is studied for each supply chain
strategy.

Thesis Supervisor: Dr. Donald Rosenfield

Title: Sr. Lecturer, Sloan School of Management

ii
Acknowledgements
I would like to thank Dr. Chris Caplice for the opportunity to be a part of the MLOG 2005
journey, and for a life altering experience. Thank you for the challenge, the wonderful year of
intense learning and discovery and for your guidance, encouragement and mentoring.

I am thankful to Dr. Donald Rosenfield for agreeing to be my advisor. You helped me to think
things through from different perspectives and gave me direction. Thanks for reading all the
drafts and for your encouragement, guidance and candid feedback.

Many thanks to Jason Amaral of Emeraldwise, LLC., and Brian Cargille at HP for making this
thesis possible. I cannot emphasize enough how helpful you have both been. I am indebted to
you for your valuable insights, for the numerous hours you set aside and for your guidance.

Many thanks to Shaoan Chin, Randy Gomes, Damian Muzio and Glenn Ready at Cisco Systems,
Dr. Stephen Bear at HP and Jeff Erhart at Broadcom Corporation. Thanks also to Tim Russell,
Dr. Jim Hines, Professor Dr. Charlie Fine and Dr. Bill Haas at MIT.

Heights by great men reachedand kept, were not attainedby sudden flight. They,
when their companions slept, were toiling upwards in the night.
- Henry Wadsworth Longfellow

My grandfather made me memorize this quote when I barely knew what it meant. A youth from
a small village in Gujarat, you went on to be a judge in the Supreme Court of India. You taught
me that I will only be constrained by my ability to dream and toil. I know you are always
watching and are proud that I am graduating from MIT.

I want to thank my mother and my father, who taught me to think for myself and to question
everything. My mother, who is a living example of "when the going gets tough, the tough get
going", taught me to never give up. I learned from my father to be grateful for all that I have -
thank you for your wisdom. I am thankful to my sister for many treasured moments. My life
would not have been the same if not for you! Thanks to Tapu, Smita aunty, mummy and daddy
for your love.

Nothing I do is possible without Sumit, my partner in life. Thank you for the most beautiful five
years, for the million little things you do for me and for your absolute dedication to my dreams.
Thank you for letting me fly, for being there when I fall and for helping me to soar again. I
thank every little circumstance that made our paths converge and brought you in my life. I am
just very very fortunate!

Sumit, this work is for you.

iii
For Sumn
Biographical Note
I have approximately four years of experience in the high-tech electronics industry in
Supply Chain Management. I am knowledgeable about the Electronics Manufacturing Services
(EMS) providers, from the perspectives of both, being an EMS provider and being a supplier to
an EMS provider.

I worked in San Jose, CA, most recently at Broadcom Corporation

(www.broadcom.com), a leading supplier of broadband semiconductor chip solutions, as
Corporate Account Manager - EMS. I managed the strategic partnership between Broadcom and
its EMS customers. From 2000 to 2004, 1 was a Customer Supply Chain Manager at Solectron
Corporation (www.solectron.com) in Milpitas, CA. Solectron Corporation is one of the largest
EMS providers providing product design, procurement, manufacturing and distribution services
to its customers. At Solectron, I was responsible for the entire material supply chain for my
customers. This involved developing supplier relationships, negotiating pricing and supply
terms and establishing and maintaining a pipeline for my customers' products. I had ownership
of the profitability and inventory turns for my accounts and developed excess inventory
prevention, customer liability reduction, supply flexibility and lead time reduction plans. From
1999 to 2000, I worked at Prudential ICICI Mutual Funds in India. I have a Masters in Business
Administration with specialization in Finance from Sardar Patel University in India.

V
Table of Contents
Abstract .............................................................................................................................. ii

Acknow ledge m ents ......................................................................................................... iii

Dedication ......................................................................................................................... iv
Biographical Note ............................................................................................................. v
Table of Contents............................................................................................................. VI

List of Tables .................................................................................................................. viii

List of Figures ................................................................................................................ viii
1 Introduction ......................................................................................................... 9
1.1 Thesis structure ............................................................................................... 12
1.2 Research M ethodology ................................................................................. 13
2 Literature Review .................................................................................................. 15
3 Evolution of Outsourcing in the Electronics Industry.................................... 23
3.1 In-house Production by O EM s...................................................................... 23
3.2 M anufacturing O utsourcing........................................................................... 24
3.3 Procurem ent O utsourcing ............................................................................. 25
3.4 Design outsourcing......................................................................................... 26
3.5 Consolidation of Design and M anufacturing ................................................. 29
4 Engagem ent Strategies...................................................................................... 31
4.1 Strategic Considerations in Design O utsourcing .......................................... 32
4.1.1 M otivations for O utsourcing.................................................................... 32
4.1.2 Determ ining an O utsourcing Strategy .................................................... 34
4.1.3 Breadth vs. Depth of O utsourcing .......................................................... 38
4.2 Levels of Engagem ent ................................................................................... 38
4.2.1 Fully O utsourced Design ........................................................................ 39
4.2.2 Joint design ............................................................................................ 46
4.2.3 Insourced Design................................................................................... 51
4.2.4 O ne-stop shop outsourcing ................................................................... 53
4.3 Incentives............................................................................................................56
5 Design for supply chain (DFSC) ........................................................................ 59
5.1 Supplier Selection .......................................................................................... 60
5.1.1 Design O utsourcing and Supplier Selection .......................................... 62
5.2 Supplier Integration in Product Design .......................................................... 65
5.2.1 Design outsourcing and supplier integration.......................................... 68
5.3 Single/Sole Sourced versus M ulti-Sourced.................................................... 69
5.3.1 Design Outsourcing and Single versus Multi-Sourcing .......................... 71
5.4 Com ponent Com m onality .............................................................................. 73
5.4.1 Design Outsourcing and Component Commonality .............................. 75
5.5 Product m odularity .......................................................................................... 77

Vi
 5.5.1 Design outsourcing and Product Modularity ........................................... 79
5.6 P ostpone m ent .............................................................................................. . 80
5.6.1 Design outsourcing and Postponement .................................................. 83
5.7 Managing product variety................................................................................85
5.7.1 Design Outsourcing and Variety Management.......................................86
5.8 End-of-Life and Obsolete Components ........................................................ 89
5.8.1 Design Outsourcing and EOL/OBS Components.................................. 89
5.9 Economic packaging and transportation ...................................................... 91
5.9.1 Design outsourcing and Design for Logistics (DFL) .............................. 93
5.10 Product Recycling and Reuse ...................................................................... 94
5.10.1 Design Outsourcing and Design for Recycling (DFR)........................... 96
5.11 Product Obsolescence .................................................................................... 98
5.11.1 Design outsourcing and Planned Obsolescence................................... 99
5 .12 C o nc lusio n ........................................................................................................ 10 0
6 Case Studies and Observations .......................................................................... 102
6.1 Case study 1: Low-end Consumer PC............................................................103
6.1.1 Determining the Level of Engagement.....................................................104
6.1.2 DFSC Strategies for PCs in an Outsourced Design Environment..........107
6.1.3 PC Outsourcing: Summary ....................................................................... 109
6.2 Case Study 2: All-in-One Printer..................................................................... 111
6.2.1 Determining the Level of Engagement..................................................... 112
6.2.2 DFSC Strategies for AlOs in a Joint Design Environment ...................... 114
6.2.3 AIO Outsourcing: Summary...................................................................... 116
6.3 Final Observations and Conclusion................................................................. 119
6.3.1 Observations and Suggestions for DFSC................................................ 119
6.3.2 Business Observations and Suggestions ................................................ 125
6.3.3 Being a good partner ................................................................................ 127
6.3.4 Areas of Future Research ........................................................................ 128
B ib liog rap hy ................................................................................................................... 131
A Appendix A - Abbreviations ................................................................................. 136
B Appendix B - ODM Design Market Share...........................................................138
C Appendix C - OEM R&D Budgets........................................................................139

Vii
List of Tables
Table 1: Summary of DFSC strategies, Tradeoffs and Outsourcing............................. 100
Table 2: Tradeoffs in Fully Outsourced Design for a Low-end Consumer PC..............107
Table 3: Tradeoffs in Fully Outsourced Design for an All-in-One Printer......................115

List of Figures
Figure 1: Distinguishing ODMs and EMS Companies.............................................. 29
Figure 2: A Framework for the Outsourcing Decision................................................. 34
Figure 3: Strategic Value Assessment: Evaluating Key Criteria ............................... 35
Figure 4: Synthesizing Strategic and Economic Elements........................................ 36
Figure 5: Broad Guidelines for Determining a Level of Engagement......................... 37
Figure 6: Supply Chain Linkages in In-house Design................................................. 62
Figure 7: Supply Chain Linkages in Outsourced Design............................................ 63
Figure 8: Supplier Integration based on Product Characteristics............................... 66
Figure 9: Summary of Strategic & Economic Elements for Personal Computer ....... 106
Figure 10: Summary of Strategic & Economic Elements for an AIO Printer................114

viii
 Introduction

"Outsourcing" is the transfer of responsibility for a business function from an employee

group to a non-employee group, (Zhu, Hsu & Lillie 2001). Outsourcing is increasingly viewed

as a way to reduce the focus on non-value added activities by using a specialized service

provider for those functions. It allows companies to focus on their core competence. Deavers

(1997) attributes the increase in the level of outsourcing to four fundamental changes in the

competitive market environment: rapid technological change, increased risk and search for

flexibility, greater emphasis on core corporate competence and globalization. This competitive

landscape is characteristic of the global electronics industry, where outsourcing is pervasive and

expected to rise from $138 billion in 2003 to $294 billion in 2008 (Carbone 2005). This section

summarizes the evolution of outsourcing in the electronics industry and its impact on the high-

tech supply chain. It also discusses the motivation for this study and its relevance to the

industry.

"Clockspeed" (Fine 1998) is the rate of evolution of different industries. The high-tech

electronics industry is one of thefast-clockspeed industries. From insourcing of all functions in

the 1980s, vertical integration, to outsourcing almost every traditionally core activity, virtual

integration(Fine 1998; Van Hoek 1998), the high-tech electronics industry has evolved in a

relatively short span of time. The first function to be outsourced was manufacturing, followed

by procurement and logistics, then after-sales services and now many companies have begun

outsourcing design.

9
 Product design is a competence that should be considered core to the corporation. This is

because new product development provides access to a wide variety of markets, shapes customer

preferences and is a skill that is difficult to replicate (Grant 2000). However, the market for

high-tech consumer items has diminished the importance of design. The decade of the 1990s

was a period of tremendous growth for the electronics industry. Fueled by innovation, ease of

communication, low barriers to entry and globalization, many new players entered the market.

For fast-moving consumer items such as personal computers and cell phones, the market became

highly competitive as product features converged and price became the basis for competition.

Original Equipment Manufacturers (OEMs) began to look for new and better ways to leverage

economies of scale in order to reduce costs. Manufacturing was moved to countries where labor

cost was cheap. As product features converged, design lost its traditional competitive advantage.

In order to leverage the benefits of outsourcing such as economies of scale, risk pooling and low

investment, companies began to outsource product design to specialized companies called

Original Design Manufacturers (ODMs).

The current trend in electronics manufacturing is towards consolidation as manufacturing

and procurement service providers acquire design houses in order to improve their value

proposition and margins. The service providers now control design, procurement, manufacturing

and after-sales services, while the OEM brands and sells the finished products. Thus, the fates of

the two companies are intricately tied to one another. The failure of the service provider to

design and manufacture the product correctly can impact the marketing and sales of the OEM.

Similarly, a marketing snafu can result in expensive excess inventory or shortages, causing fixed

manufacturing capacity to remain idle at the service provider. Despite the inter-dependence of

the two partners, OEMs have more at stake in this equation because of their degree of

10
dependence on service providers. Due to outsourcing, they have lost visibility and control over

design, manufacturing and component supply issues and yielded some of their bargaining power.

This has reduced their importance to other upstream partners such as component suppliers, who

now target their sales efforts towards design houses. Design houses are able to negotiate price-

masking agreements and rebates that further reduce the visibility of OEMs into the true cost of

their products. One other possibility is that the service providers could make OEMs redundant,

forcing them to reconsider their business strategy. An example of this is IBM's historic decision

to sell its PC business to a Chinese manufacturer called Lenovo in February 2005.

In addition to these broad business concerns, there are many supply chain considerations

that are affected by design outsourcing. Product design impacts all the subsequent activities such

as procurement, manufacturing, distribution and repair. OEMs are contractually responsible for

demand forecasts and the resulting inventory, and in some cases distribution and after-sales

services. Hence, they need to be able to implement design for supply chain strategies like

demand aggregation, postponement, use of common parts, multiple sourcing, supply base

reduction, economic packaging, etc. in order to better manage their business. These strategies

are incorporated into the product early in the design stage and need close coordination with

different divisions in the OEM such as marketing, supply chain, warehousing and traffic, etc.

In an outsourced design scenario, these supply chain concerns may not be considered

during the product development process. Supply chain strategies which are beneficial to OEMs

may not generate interest in the design house. Since OEMs don't control design, they have to

either accept the product as is or specify requirements upfront, which is difficult. Product

reconfiguration or redesign incurs additional charges and so requires time-consuming cost-

benefit analyses and approvals. Also, since there are weaker linkages between the design and

11
marketing groups in different companies, the process of information exchange is slower. Lastly,

design and redesign requests are subject to capacity constraints at the design house.

Thus, design outsourcing has several consequences for an OEM's supply chain strategy,

both positive and negative. Two factors that determine this are the way incentives are structured

and level of outsourcing practiced by the OEM. OEMs can outsource the complete design

process or retain partial control and outsource certain chunks of the design work. This work

studies the supply chain tradeoffs inherent in a given outsourcing strategy and creates a

framework that would facilitate understanding of what strategies can be implemented. It is

relevant to OEMs which outsource design, procurement, manufacturing, etc. to service providers.

The thesis will primarily focus on companies in the high-tech electronics industry.

1. 1 Thesis structure
Chapter 2 covers existing literature about motivations for outsourcing and its risks, lays

the groundwork for the make-buy or outsourcing decision and connects product architecture with

supply chain design. It also includes literature about the different supply chain strategies.

Chapter 3 of the thesis outlines the evolution of the high-tech industry from being

vertically integrated to being virtually integrated (Fine 1998). It discusses the evolution of the

market for high-tech products and how it impacted the value-chain configuration over the course

of 25 years.

Chapter 4 outlines the strategic considerations in choosing the level of outsourcing; there

are many different ways in which companies may engage with an outsourcing service provider.

It lays out the different levels and the advantages and disadvantages of each engagement

12
strategy. It also discusses the risks inherent in one-stop-shop outsourcing, i.e. having one

company provide the full spectrum of services - design, component procurement, manufacturing,

warehousing & distribution and after-sales services. Next it brings the role of incentives in

determining the priorities of supply chain partners.

Chapter 5 talks about the different DFSC strategies; specifically outlining what each

strategy is with examples of how companies may use these strategies to improve their supply

chain performance. Lastly, it discusses the impact of outsourcing on each of these strategies.

Chapter 6 includes two case studies where the concepts discussed in chapters 4 and 5 are

applied to two products which use different engagement strategies. The choice of an

engagement strategy has a positive or negative impact on the OEM's desire to implement the

different DFSC strategies. This is discussed in the tradeoff matrix. The chapter concludes with

final observations and conclusions.

1.2 Research Methodology

In the high-tech industry, design outsourcing is most prevalent in fast-moving consumer

items such as personal computer, laptops, cell phones, printers, etc. In order to study the impact

of design outsourcing on a company's supply chain strategies, I needed to work with companies

which had outsourced design for several years and were acknowledged leaders in design for

supply chain strategies. For this reason, I decided to work with two leading companies, one in

the consumer PC business and the second in the networking equipment business. Both these

companies had much experience in the practical application of DFSC strategies.

13
 The primary methodology for this thesis is an extensive literature review, interviews with

design for supply chain personnel in high-tech companies, qualitative analysis of findings and

their application in two case studies. The literature review focused on:

" Understanding outsourcing and its benefits and pitfalls.

- Design outsourcing motivations and risks.

- Strategic considerations in choosing an outsourcing strategy.

- Design for supply chain strategies and their implementation and benefits.

The interviews were primarily focused on understanding the factors that impacted the

choice of an engagement strategy and the tradeoffs inherent in each one. Next it was important

to understand what criteria were critical for the successful implementation of a design for supply

chain effort. The literature review and the interviews facilitated the creation of a framework to

understand how the different levels of engagement impact an OEM's ability to implement supply

chain strategies. Since there is not one answer to the question of whether outsourcing is good or

bad, the case studies illuminate the dynamics of design outsourcing for specific products. By

doing two case studies focused on two different types of products in their different markets, it

was possible to create clear distinctions between the engagement strategies. In addition, the case

studies helped to clarify the application of the concepts based on an understanding of the

product, the market and the different players involved.

14
2 Literature Review

Fine (1998) explains the concept of the speed of industry evolution in terms of

"clockspeed". Clockspeed is the pace of industry evolution - fast clockspeed companies are

companies whose products are outmoded within months of their launch and whose corporate

lives are at risk almost on a daily basis, for example Dell or HP. Four lessons learned from fast

clockspeed industries can be applied to other industries: high uncertainty industries tend to

generate hedging strategies; a major proportion of economic value to customers is a result of

complex value chains comprising many different organizations, technologies, services and

capabilities; a high-opportunity strategy may not stay constant between being a component

supplier and a chain dominator; even companies that exert little control over the supply chain can

be profitable if they supply a critical component. Fine argues that all competitive advantage is

temporary; the faster the industry clockspeed, the shorter the life of competitive advantage. The

book also introduces the industry double helix, a model based on a repeating double loop that

circulates between vertically integrated industries and highly modular horizontally integrated

industries. Companies in highly vertically integrated industries face pressure to disintegrate and

focus on core competences, whereas in highly modular industries, companies feel pressure either

to control critical components or to come up with innovations that encroach on other players'

arenas.

Fine, Vardan, Pethick and El-Hout (2002) create a model to determine value-chain design

based on a project at General Motors. They provide a strategic framework for evaluating the

15
make-buy decision based, not on cost or time-to-market, but on an assessment of the market,

customer needs, speed of technological change, competitive position, supplier capability and

product architecture. This enables decision-making about whether or not a product is a suitable

candidate for design outsourcing. The framework is an integral part of this thesis and has been

covered in-depth in Section 4.1.2. It has been applied to design outsourcing decision-making for

high-tech products.

Van Hoek (1998) looks into the different elements that comprise a virtually integrated

supply chain - the application of postponement, extensive use of supplier integration and

outsourcing and integration of the value chain via effective management of information. The

important point is that the OEM achieves control of the value chain not through ownership of

different functions, but by managing the flow of information internally and externally with its

partners. He also discusses the concepts of component commonality, product modularity and a

reduction in the complexity of the manufacturing process as important enablers of postponement.

A case study of supplier integration into the manufacturing of the Smart Car reveals the extent to

which the OEM can be "squeezed out" of the physical flow of goods. The recommendation for

OEMs is that they must specify the overall product architecture and integrating the flow of

information. That is, the OEM must assume the role of a project manager in order to retain

power over its supply chain partners.

Ulrich and Ellison (2004) examine the decision of "internalizing" and "integrating"

design and production. Internalizing is including the activity within the organizational

boundaries. Integration is consolidating two or more activities into the same organizational

entity; this may or may not be the manufacturer. They discuss the interdependence of design and

manufacturing and put the design outsourcing decision in a richer framework than simply make

16
or buy. They outline the motivations and strategic considerations in outsourcing design. Based

on empirical evidence from the bicycle industry, they conclude that firms with competitive

advantage prefer to design internally, whereas firms which do not have a competitive advantage

outsource design.

Lyons, Krachenberg and Henke (1990) examine outsourcing relationships and discuss the

advantages and disadvantages for both parties in an outsourcing relationship. Different

objectives guide the motivations and actions of buyers and sellers. There are many risks

involved and contract provisions can be made to provide protection to the each company to

protect against those risks. They conclude that both parties assume risks, i.e. outsourcing is not

more beneficial to any one company. However, one point to contrast this view is that as the

number of functions outsourced increase, the vulnerability of the OEM increases in comparison

to the service provider. In addition, the service provider has the benefit of risk diversification

due to the broad portfolio of customers whereas the OEM may have one or two service

providers.

Grant (2000) dives in-depth into the benefits of design outsourcing and provides reasons

for firms' willingness to outsource design. Grant also outlines the risks of design outsourcing -

competition from the design supplier, supplier holdup, loss of knowledge and loss of corporate

and industry innovation. Using system dynamics methodology, he concludes that design

outsourcing and knowledge transfer are easier to start than to stop. He recommends

understanding knowledge intensity of the products, understanding the company's competitive

advantage, using design outsourcing to learn or even becoming a knowledge supplier to non-

competing OEMs as strategies to prevent the erosion of knowledge.

17
 Novak and Eppinger (2001) discuss the relationship between product complexity and

vertical integration based on evidence from the automotive industry. Product architecture is one

of the variables that influence product complexity. They use the property rights approach to

argue that in-house production is more attractive for highly complex products. They contend

that often, product design engineers determine product architecture and complexity independent

of purchasing agents who make sourcing decisions. They recommend that these decisions

should be jointly made in order to improve firm performance.

Cargille and Bliss (2002) discuss how supply chain analysis during product development

can result in revenue and profit enhancement. They describe the analytical tools and techniques

used by Hewlett-Packard in order to improve product development. They discuss the application

of postponement and stock-keeping unit (SKU) reduction for a CD-RW product in order to take

advantage of demand aggregation. They also develop a "rough-cut" analysis tool to evaluate the

tradeoffs in DFSC analysis.

Amaral, Billington and Tsay (2004) discuss outsourcing electronics manufacturing and

component procurement and the power plays in the partnership. They outline the means, motives

and opportunitiesframework where means are the ways in which EMS providers can profit at

OEM's expense. Motives arise because of the motivation of EMS employees and OEM

employees to generate profit for their own employer. Opportunities are all the different ways in

which EMS providers take advantage of OEM's inability to monitor their actions. Combined,

these tend to act against the OEM's best interests. They list the different risks inherent in

different levels of outsourcing and recommends strategies that can be used to counter some of

the opportunities. They also highlight the role of incentives in creating opportunities and make

18
recommendations on how certain incentives may be aligned in a procurement outsourcing

relationship.

Narayanan and Raman (2004) discuss that collaboration between partners is important for

the success of outsourcing relationships and long-term profitability of both partners. The

relationship between partners needs to be managed effectively and incentives need to be aligned

to expectations in order for the strategies to be sustainable. One reason why incentive related

issues arise is a lack of transparency in partners' actions. For example, even though Whirlpool

provides high margins to Sears for its products, it is unable to get information on effort expended

by Sears to sell its brand vs. Sears' own brand, Kenmore. Incentive related issues can also arise

when one company has information that another does not. Often, incentive schemes are badly

designed and short-sighted. One solution is to acknowledge the misalignment, identify the root

cause and redesign incentives to reinforce the behavior they desire. This may involve changing

contracts to reward partners for acting in the supply chain's best interests, gathering or sharing

previously hidden information and using personal relationships to build trust. By conducting

incentive audits at key stages, educating managers about processes and incentives and by

collaborating to solve each others' problems, companies can reduce the occurrence of incentive

misalignment.

Venkatesh and Swaminathan (2002) outline postponement concepts and provide

examples of successful postponement strategies. They introduce component commonality and

product modularity as product enablers and process standardization and resequencing as process

enablers. They also discuss the underlying factors which enable successful implementation of

postponement. Swaminathan (2003) talks about product design for postponement. As much as

19
80% of the manufacturing cost of a product is determined by product or process design. He

outlines design strategies in order to enable postponement and facilitate mass customization.

Ulrich, Randall, Fisher and Reibstein (1999) discuss management of product variety

based on a study of the strategy employed by four companies in the bicycle industry. They

discuss the strategic and tactical decisions that are involved in determining the variety delivery

system. They outline the different criteria that influence product variety strategy such as

customer preferences, perceived competitive advantage, product architecture, capabilities and

resources. Variety strategy can be capability driven or customer driven and the authors conclude

that both factors should be considered in order to be successful.

Robertson and Ulrich (1998) state that product variety management involves effectively

managing the trade-offs between the benefits of component commonality and product

differentiation. In order to effectively plan the product "platform", they propose first creating the

product plan, the differentiation plan and the communization plan. Then they propose an

iterative process to refine the three plans in order to achieve the optimal level. They strongly

recommend a strong degree of collaboration between the marketing, engineering and

manufacturing guided by top management.

Petersen, Handfield and Ragatz (2003) propose a model including the key activities that

required for successful supplier integration into new product development efforts. They reveal

that increased knowledge of a supplier can result in greater information sharing; sharing of

technology knowledge is likely to lead to increased supplier involvement and improved

outcomes. Ragatz, Handfield and Scannell (1996) discuss supplier integration into new product

development. Based on a survey of companies who employed different levels of supplier

20
integration, they discuss the most and least successful efforts based on management practices

employed.

Chuang and Yang (2004) discuss supplier integration in product design stages. They

outline the different design collaboration strategies. They contend that supplier integration is a

complex activity and propose strategies to reduce this complexity. They propose two extreme

approaches - alliance design chain strategy which involves sharing risks and resources and

virtual design chain strategy which involves using standard interfaces to enlarge the supply base.

Laseter and Ramdas (2002) discuss supplier integration in the context of product types

based on the automotive industry. Product classification is based on nature of the products, their

cost structure and nature of OEM-supplier interaction in the development process. Critical

systems are highly differentiated, high-cost, complex systems for which OEMs create detailed

specifications like brake and climate control systems. Hidden components are less

differentiating low-cost simple components defined based on physical specifications, where

suppliers are involved late in product development such as door locks and filler tubes. Invisible

subassemblies are non-differentiating, moderately costly, moderately complex systems whose

suppliers are provided information via a mix of performance specifications and detailed physical

dimensions, e.g. integral steering and bumper facia. Simple differentiatorsare high

differentiating, moderately costly, simple assemblies or components like wiring harness, ignition

control, etc. This classification has relevance to supplier integration, component commonality

and product variety management discussed in Chapter 5 of this thesis.

Christensen (1997) states that disruptive innovations are those that either create new

markets or are adopted by a company's less established customers. Disruptive innovations begin

21
in a simple application and then get better and gain wider acceptance until they change the rules

of the game and marginalize the dominant incumbents. In contrast sustaining or incremental

innovations, which are often practiced by large companies, focus on improving existing

products. Large companies are almost always better at sustaining innovation but often fail to

invest in disruptive technologies because of their size and their established processes and set

ways of thinking. They are also less flexible and react slower to emerging trends and

technologies as compared to smaller companies.

Simchi-Levi, Kaminsky and Simchi-Levi (2003) write about supply chain strategies with

case studies and examples. Chapter 7 - Procurement and Outsourcing Strategies and Chapter 9 -

Coordinated Product and Supply Chain Design have been used widely in this thesis. Chapter 7

includes the benefits and risks of outsourcing and making the outsourcing decisions. Chapter 9

covers the different supply chain strategies like Design for Logistics (DFL) including Economic

Packaging and Transportation, Standardization, Mass Customization and Supplier Integration

with examples of how different companies have implemented them.

22
3 Evolution of Outsourcing in
the Electronics Industry

The electronics industry went from conducting all operations in-house to outsourcing all

the traditionally core functions of design, procurement, manufacturing, logistics and after-sales

services. This section covers the evolution of electronics manufacturing starting with fully

insourced operations in section 3.1 to the practice of outsourcing manufacturing in section 3.2,

procurement in section 3.3 and the current practice of outsourcing design in section 3.4. Section

3.5 discusses the current consolidation taking place where the service providers are bringing all

the different outsourced functions under a single umbrella to provide one-stop-shop outsourcing

services to OEMs. The focus is on current changes and their impact on different players in the

supply chain.

3.1 In-house Productionby OEMs

In the 1960s and 1970s, electronic product manufacturing was done in-house by OEMs.

Manufacturing know-how was considered a core activity providing strong competitive advantage

since the high cost of equipment, facilities and personnel created entry barriers to newer players

and limited competition. Despite this, the current practice of close alignment of manufacturing

with design teams was absent. Due to incomplete data and rough assembly drawings, delays,

cost overruns and poor product quality were common (Schweber 2003). High quality electronics

23
products from Japanese OEMs in the 1980s changed the existing paradigm and more focus

shifted to improving quality and building product core competence.

Accordingly, many companies began focusing on creating innovations in a key portion of

the main product, a sub-assembly. Gradually, these component suppliers acquired more strategic

importance and the actual assembly of the final product lost its traditional competitive advantage.

This gave rise to the contract manufacturing industry, which later came to be known as the

electronics manufacturing (EMS) industry. Solectron Corporation (www.solectron.com), a

leading EMS provider, defines contract manufacturing as the practice of making products or

subcomponents of products to be sold under a different company's name.

3.2 Manufacturing Outsourcing

IBM pioneered the idea of outsourcing entire products in the 1980s by getting SCI to

build its personal computers (The Economist 2000). Collins and Bechler (1999) state that the

trend towards globalization, shorter product life cycles, just-in-time manufacturing, shrinking

chip manufacturing timelines and increasing pressure to accommodate customer demands are

some of the factors that have caused OEMs to re-evaluate their manufacturing strategy. As more

and more manufacturing became outsourced, EMS providers began to play an increasingly

pivotal role in the electronics industry.

In the early stages, OEMs outsourced only the actual printed circuit board assembly of

their products to EMS providers. This allowed the EMS providers to utilize manufacturing

capacity for multiple OEM customers and develop economies of scale. Additionally, advances

in capital intensive manufacturing technologies (surface mount, fine-pitch, etc.) pushed large

multinationals such as Nortel and Nokia to rely more on EMS providers who could efficiently

24
use the equipment at full capacity (Hadaya, Lefebvre and Leger 2000). EMS companies invested

in manufacturing sites and equipment worldwide and developed specialized manufacturing core

competence. The pervasiveness of manufacturing outsourcing removed the barriers to entry for

most of the small high tech startup companies. Establishing manufacturing capabilities was no

longer a deterrent to smaller players. For example, Ingram Micro, an IT product distributor,

outsourced manufacturing of both branded and non-branded build-to-order PCs to Solectron thus

competing directly with OEMs like HP and Dell using supply networks established by them

(Vande Vate, Ward 2005). This facilitated innovation and encouraged more companies to

capitalize on their core competence - new product design and development. John Chambers, the

CEO of Cisco Systems, describes Cisco's strategy of outsourcing manufacturing capabilities

worldwide as "global virtual manufacturing." Cisco uses EMS providers worldwide for its

manufacturing activities. By developing strong capabilities in information sharing and

maintaining close partnerships with EMS partners, Cisco has been able to leverage the

capabilities of its business partners to become a $22 billion company in 20 years (Simchi-Levi,

Kaminsky & Simchi-Levi 2003).

3.3 Procurement Outsourcing

While EMS companies had assembly expertise, component procurement and inventory

management were both managed by the OEM, who purchased only the required quantities of

components on a consigned basis. Consignment is the process of assembling what is provided

without owning any material (Barnes, Dai, Deng, Down, Goh, Lau & Sharafali 2000). As EMS

providers developed purchasing infrastructure to acquire components, they began offering

procurement services to OEMs. The OEM would provide the product bill of material and the

25
suppliers approved for each component. The EMS provider would purchase the components

from the approved suppliers, warehouse them, manufacture the final product and ship it to the

OEM for system assembly and testing. OEMs retained purchasing control over key components

as determined by cost, demand volume and level of customization. The EMS provider would

negotiate prices with component suppliers for low-cost generic components. Thus, the OEMs

made sourcing decisions and negotiated the price and terms of supply like lead time, delivery

terms, minimum order quantities, etc. whereas the EMS provider merely executed the

procurement functions.

With the addition of procurement responsibilities to their portfolio, EMS providers were

gradually able to develop strong relationships with component vendors. Throughout the Internet

boom of the 1990s, EMS companies won new outsourcing deals from major OEM players.

Some providers acquired greater purchasing power than their OEM customers because of

demand aggregation benefits (Hadaya, Lefebvre & Leger, 2000). By the late 1990s, the EMS

industry consolidated a lot of power within the component supply chain on account of the

volume of business. Using low prices as their value proposition, they began to offer component

procurement services to smaller OEMs. This allowed smaller OEMs to compete in the

marketplace with bigger companies and further facilitated growth. As a result, a highly

competitive marketplace with cost-conscious consumers and low product differentiation was

born.

3.4 Design outsourcing

In the late 1990s, laptops, PCs, cell phones and other fast-moving consumer electronics

became commoditized. Commoditization is the transition of a product or industry segment from

26
a niche to mass market. A characteristic of commodity products is that they were once niche

products or some of their features were niche features. In the absence of any significant

differentiation between product offerings by different players, the basis for competition are price,

delivery terms and time-to-market (DeNeufville & Pirnar 1996). As OEMs were unable to

differentiate their product offerings significantly in terms of features vis-a-vis their competitors,

price emerged as the dominating factor that determined market leadership.

The commoditization of high-tech consumer electronics led to two important

developments: the shift of high-volume electronics manufacturing to lower cost Asian economies

and the emergence of the original design manufacturers (ODMs). To reduce manufacturing cost,

the EMS industry made massive investments in China and other low-cost Asian countries and

began manufacturing high-volume, low-mix products there. As prices and product

differentiation decreased further, OEM margins declined until they were making no incremental

profits on products designed in-house. Additionally, with a shortage of engineering resources,

OEMs struggled to keep up with market demand for new product models and features (Normille

2004). According to Michael Marks, the CEO of Flextronics, in an interview with Business

Week (Engardio & Einhorn 2005), almost 80% of engineering tasks - translating prototypes into

workable designs, upgrading mature products, testing quality, writing user manuals and

qualifying parts vendors - can easily be outsourced. Thus, more and more companies began to

outsource design for commoditized products. The shortage of engineering resources combined

with increasing pressure to reduce prices led to the creation of the ODMs. ODMs tend to be

engineering-centric companies developing commodity-level products from standard building

blocks or platform designs (Coker 2004).

27
 The ODM model has cost and time-to-market advantages that are critical in a

commoditized market. ODMs enjoy economies of scale due to reuse of product designs and

engineering teams. They design products and sell the design to multiple OEMs, which splits the

fixed design cost. Additionally, they are able to allocate idle engineering resources on one

account to other customers. Thus, their costs are significantly lower as compared to the costs of

dedicated design teams at OEMs. ODMs are also able to reduce the time-to-market for new

models by using the following strategies. They build products based on existing platforms by

doing incremental innovation, i.e. changing a few features or functionality rather than by

revolutionizing technology. They also use standard or off-the-shelf items as much as possible

rather than custom or semi-custom components.

HP has outsourced design to several different ODMs for its products. Taiwan's Quanta

and Hon Hai Precision are ODMs for PCs, Lite-On for printers, Inventec for servers and MP3

players and Altek for digital cameras (Engardio & Einhorn 2005). HP's senior vice-president for

global operations, Dick Conrad, said in an interview in the same article, that it now takes 60%

less time to get a new concept to market. It allows HP to reallocate design resources to higher

value-added products such as advanced printer inks and sophisticated corporate software. The

benefits of the ODM model led to sales for the top 10 ODMs, all Taiwanese, jumping 40% to

$37 billion in 2003 (Normille 2004). The ODM segment is expected to grow to $114 billion in

2008 (Carbone 2003 & 2005).

The ODM is most successful when selling a single standard design to as many customers

as possible. Therefore, beyond a certain increase in customization, the ODM model loses its

benefits since it is based on commoditization and volume and does not lend itself easily to

customization. The lack of a global manufacturing footprint and the inability to provide regional

28
after-sales service is another shortcoming of the ODM business model (Coker 2004). The main

differences between the ODM and EMS business models are shown in the figure below based on

information by Carbone (2003).

Product Company Service Company

Robust Product Engineering Capabilities Robust Process Engineering Capabilities

More likely to own and control IP. Less likely to own IP.

Limited in Product Scope Spans across multiple product types

Assume Inventory & Marketing Liabilities Don't assume either marketing or inventory liabilities.

Concentrated in Asia Offer a Global Footprint

Design ownership usually OEM, some players moving to

Design ownership is blurred.
an ODM model

May offer competing brands. Less likely to compete with own brands.

Presence in fast-moving consumer goods. Presence in low-end storage and networking products.
Based on Carbone, J. (2003). ODMs offer design expertise, quicker time to market. Purchasing.com.

Figure 1: Distinguishing ODMs and EMS Companies

3.5 Consolidation of Design and Manufacturing

The EMS industry realized the opportunity presented by the ODMs. After the Internet

bust of 2001, the industry was characterized by slacking revenues and profitability. Typical

profit margins in the manufacturing space were 2-3%. In addition, EMS providers were unable

to differentiate their value proposition vis-a-vis their competition as all the major players in the

EMS industry developed similar manufacturing expertise and a global footprint. By adding

design outsourcing to their portfolio, EMS providers could boost their own manufacturing

29
revenues and increase profit margins. A hybrid EMS/ODM company could gain the benefits of

both models by combining their service and product capability chains and providing integrated

design, procurement, manufacturing, distribution and after-sales-services under one umbrella.

They would be able to attract business from both OEMs who were simply looking to outsource

manufacturing and customers who needed design work done on low-end or complementary

product portfolios. Thus, Flextronics acquired Microcell Group, Hon Hai acquired Ambit and

Sanmina-SCI bought Newisys (Normile 2004).

The design for a wide range of fast-moving, low-differentiated, high-volume products is

outsourced to EMS/ODMs who, based on product specifications, design the products, determine

the supply base and negotiate the pricing and supply terms with component suppliers. OEMs

gather market intelligence, create the design specifications and get the EMS/ODM to design and

manufacture the final product. Flextronics has invested more than $800 million on acquisitions

to build a 7000-engineer strong force of software, chip, telecom and mechanical designers

worldwide. In addition, Flextronics acquired Frog Design, Inc. in order to further boost its

design capabilities (Engardio & Einhorn 2005).

Thus the industry has come full circle since the early 1980s when all these functions were

kept in-house. One could say that the EMS/ODMs outsource marketing, product management

and sales to OEMs for fast-moving consumer items.

30
4 Engagement Strategies

Engagement strategies are the different ways in which OEMs choose to work with their

design partner. One constituent is the level of outsourcing which is the portion of the product

that is designed by the ODM and the degree of detail provided by the OEM in the specifications.

It also includes how many functions are outsourced and how many service providers are used.

Incentives used to reward partner behavior are also an important constituent of engagement

strategy.

Section 4.1 presents the motivation and the strategic considerations for design

outsourcing based on research by Professor Charlie Fine in the automotive industry. His

framework for choosing an engagement strategy is applied to design outsourcing in the high-tech

industry.

Section 4.2 discusses the different levels at which companies engage with ODMs and the

advantages and disadvantages of each choice. Fully outsourced design and joint design are

covered in-depth. Insourced design is not an engagement strategy because if design is insourced,

there is no engagement with a service provider. However, a discussion about advantages and

disadvantages of insourced design is required in order to compare and contrast it with outsourced

design. Section 4.2.3 is on the advantages and disadvantages of insourced design. Section 4.2.4

discusses the pros and cons of one-stop-shop outsourcing, which is outsourcing all the functions

to a single service provider.

31
 Finally, incentives are an important aspect of the OEM's engagement strategy; they

influence partner behavior by creating convergent or divergent motivations; Section 4.3

discusses the role of incentives in managing an outsourcing partnership.

4.1 Strategic Considerations in Design Outsourcing

Outsourcing is not about reducing costs or improving time-to-market performance,

although these are most often cited as reasons for outsourcing. It is important to understand that

these are the supplemental benefits of design outsourcing. This is because product development

in the high-tech industry is all about innovation and outsourcing design is akin to outsourcing

innovation. This has long-term consequences for the existence of a firm. Thus, design

outsourcing should really be adopted with a bigger focus on strategic value of the business (Fine

2005). Thus, it is important to examine the motivations for outsourcing design and the strategic

considerations in making the outsourcing decision.

4.1.1 Motivations for Outsourcing

Simchi-Levi, Kaminski and Simchi-Levi (2003) outline the main motivations for

outsourcing as economies of scale, the possibility of risk pooling, capital expenditure sharing, the

drive to focus on core competencies and the need for increased flexibility. Ulrich and Ellison

(2004) explain that when design teams need to use specialized equipment and tools, then

outsourcing design to an ODM which specializes in that area would lead to economies of scale.

Also, risk pooling is a distinct advantage of design outsourcing. As indicated by queuing theory,

when independent sources of demand are combined, the variability in aggregate demand is less

than that of the individual sources. Since design activities occur in spurts and bursts rather than

in a constant stream, there are clear demand aggregation possibilities by outsourcing design.

32
Outsourcing design reduces the investment into product development since the OEM would not

have to create an internal team of engineers or equipment. This is especially true when the

length of the project may be small or future use of the resources may be uncertain. Design

outsourcing is also beneficial in expanding the product offering and reducing time-to-market.

Fine and Whitney (1996) and Fine (1998) summarize the key motivations for design outsourcing

into the following two categories.

Dependence for Capacity: In this case, the company has the knowledge and ability to

create the product in-house but for it chooses to outsource to make up for lack of

capacity. For example, Nokia relies on outsourcing in order to respond to demand

upswings when in-house capacity may not be sufficient (The Economist 2004). In a

survey from the American Society of Engineering Management, 83% of executives cited

lack of in-house capacity as a reason for outsourcing (Rothstein 1998).

Dependence for Knowledge: Here, the company needs the item, but lacks the skill or

capabilities to make it. Outsourcing provides these capabilities because of the range of

customers served and products designed by the ODM.

For example, Sony fully outsources design for its printers to ODMs. Sony does not have

the internal expertise in printers, but the company perceives a market need for offering printers

along with its personal computers. If Sony were to develop capability in-house and create a

product from scratch, it would be expensive and time-consuming. Instead, the company

outsources the design for printers to leverage the design knowledge of ODMs. In the absence of

design outsourcing, Sony would have to either develop the product in-house, or partner with

33
another OEM such as HP. By using an ODM, Sony can introduce a Sony branded product and

retain the profit margins (Carbone 2003).

The following framework outlined by Fine and Whitney (1996) uses the motivations for

outsourcing discussed above with the product architecture, whether integral or modular in order

to determine whether outsourcing would be risky, an option or an opportunity. A product that is

assembled from a combination of different components is a modular product, for example,

bicycles, personal computers, etc. In an integral product, the components are inter-related and

inter-dependent; rather, it is a product where the different components are designed individually

to create the final product (Swaminathan 2001; Ulrich 1995). The following figure summarizes

the outsourcing choice.

Product Dependency on Knowledge Independent for Knowledge, Independent for Knowledge

Architecture AND CapaityCapacity AND Capacity

Opportunity to reduce cost

Integral Outsourcng is Risky Outsourcng is an Opportunity through outsourcing

Modular Outsourcing is Very Risky Outsourcing is an Option Keep Product Internal

Source: Fine, C., Whitney, D. (1996). Is the Make-Buy Decision Process a Core-Competence? Working Paper,
Massachusetts Institute of Technology, Cambridge, MA

Figure 2: A Framework for the Outsourcing Decision

4.1.2 Determining an Outsourcing Strategy

While outsourcing has many benefits, it is also a risky strategy since OEMs share

competitive and technological knowledge, the transfer of which may be irreversible, create often

conflicting objectives and lose control and visibility of the product development process. These

34
risks depend on how many functions are outsourced and to how many different players. The

amount of outsourcing in turn depends on many strategic considerations outlined by Fine,

Vardan, Pethick and El-Hout (2002). Based on a dynamic value assessment framework they

provide guidelines to establish an outsourcing versus insourcing policy. Based on a strategic

value-chain assessment at GM, the model identifies the different parameters like customer

preferences, technological changes, competitive position of the company, capabilities of

suppliers and the product architecture and ties them to selecting an outsourcing strategy.

CUSTOMER High High customer

Mediumn
IMPORTANCE Low importance and fast
clockspeed indicate
high strategic value
and argues for
TECHNOLOGY Fast insourcing.
CLOCKSPEED Slow
Insourcing is indicated
COMPETITIVE Advantage in areas of strong
Parity competitive position.
POSITION Disadvantage
4
CAPABLE None
Supply-Base capability Few
must be present for S ULPPLIERS Many
successful outsourcing.
Integral POSSIBLE DECISIONS:
ARCHITECTURE * Insource
Modular * Outsource
High degree of modularity + Partner/acquire
in value-chain architecture * Partial insource
significantly eases * Partial outsource
outsourcing.
Invest
Source: Fine, Vardan, Pethick, El-Hout. (2002), Rapid-Response capability in Valu --chain Spin off
design. MIT Sloan Management Review. * Develop suppliers

Figure 3: Strategic Value Assessment: Evaluating Key Criteria

According to the model, insourcing is beneficial when customer preferences have a huge

impact, when technological change is faster, when the company has a strong competitive

position, when capabilities in the supply-base are not fully developed and when the product

35
architecture is integral rather than modular. Outsourcing would be a risky strategy in this

situation. On the other hand, when a company is in a commoditized industry, with slow

technological changes, where the supply-base has developed extensive capabilities and the

product architecture is modular, outsourcing could yield significant benefits. The different levels

between these two ends of the spectrum can then be defined. For example, in products where

customer preferences are important and technological clockspeeds are high, if supplier

capabilities are high and product architecture is modular, joint design would be a good strategy.

If customer preferences are low and technological clockspeed is slow, and if the product is

modular and supplier capability is low, then the OEM can outsource the portions of the product

with low-complexity while retaining those where it adds significant value.

STRATEGIC VALUE ADDED

High
LEVERAGE INSOURCE

Medium

OUTSOURCE HARVEST
Low

Low Medium High

(negative) (break even) (positive)

EcONOMIc VALUE ADDED

Source: Fine, Vardan, Pethick, Ei-Hout. (2002), Rapid-Response Capability in Value-chain

Design. MIT Sloan Management Review.

Figure 4: Synthesizing Strategic and Economic Elements

Another dimension in evaluating insourcing vs. outsourcing is economic value. The

guiding principles of strategic value when evaluated with economic analysis establish a

36
framework for viewing elements of the value-chain. Thus, elements with a high strategic and

economic value are better insourcing candidates, while those with low strategic and economic

value are outsourcing candidates. Elements with high strategic value and low economic value

have the potential for future leverage; elements with low strategic and high economic value have

the potential for harvesting.

In summary, the strategic and economic elements for determining the engagement

strategy can be broadly viewed as shown in the figure below.

Customer Importance High High High

Technology Clockspeed Low Medium High

Competitive Position Disadvantage Parity Advantage

Capable Suppliers High Medium Low

Product architecture Highly modular Partly Modular Integral

Economic Value Added Low Medium High

Outsourcing Model Fully Outsourced Design Joint Design In-House Design

Based on a framework developed by: Fine, Pethick, Vardan & El-Hout (2002).

Figure 5: Broad Guidelines for Determining a Level of Engagement

The above figure may be used as a broad guideline to understand the application of the

framework. Also, it is important to note that these are not the only determinants of an

engagement strategy. Many other firm or product-specific factors influence this decision, some

of which are the level of product commoditization, the motivation for outsourcing, whether

capacity driven or knowledge driven, the type of involvement desired in design and the nature of

the outsourcing partnership. Section 6.1 discusses the application of the above for a specific

product.

37
4.1.3 Breadth vs. Depth of Outsourcing
OEMs have outsourced many different activities and so in the high-tech industry, another

dimension of engagement is the number of outsourcing partners for the different functions -

manufacturing, procurement, warehousing & distribution and after-sales services. OEMs may

use an ODM for product design, an EMS provider for the actual procurement and manufacturing,

a separate third-party logistics (3PL) company for distribution and a call-center services provider

for after-sales services. This is referred to as the breadth of outsourcing. Alternatively, with

EMS providers moving to more of an EMS/ODM model, companies may use a one-stop solution

for the entire spectrum of services ranging from design to manufacturing and after-sales services.

The breadth of services provided by the outsourced partner determines the impact of the OEM's

engagement. This is because product design has an impact on all the subsequent activities.

Incentives may be aligned or conflicting depending on the product and activity and they are the

key drivers to behavior.

4.2 Levels of Engagement

The development of EMS/ODMs creates many opportunities for OEMs to reduce costs

for existing low-margin products or expand into products that complement existing product lines.

OEMs have many different options when engaging with ODMs. They may purchase an existing

design that has been developed independently by the ODM and sold to multiple OEMs. They

may choose to outsource the full design and development task to ODMs with no control over

features. Alternatively, they may provide detailed specifications to guide the ODM design.

They could outsource just a portion of the full product or do joint design with engineering

collaboration from both companies. They could do the design in-house and outsource just the

38
manufacturing to an EMS/ODM. Finally, OEMs may choose to outsource all their activities to a

single supplier, called one-stop-shop outsourcing. Thus, the EMS/ODM engagement model falls

along a continuum between two extreme approaches of vertical integration versus virtual

integration (Fine 1998). The following discussion outlines the different levels of engagement.

4.2.1 Fully Outsourced Design

Complete design outsourcing is a strategy that is usually employed by companies for

commoditized products where there is low product differentiation and where the market

competes based on price. It may also be used when OEMs depend on suppliers for knowledge,

like for example Sony outsourcing the design for its printers (Section 4.1.1). When products

have low strategic value but high economic value, OEMs may choose to outsource design.

OEMs outsourcing design to outside companies use different engagement models, which

have different cost and time-to-market implications. These are as outlined below:

Existing Design: Here the ODM has a ready-made product. The OEM buys the design

as is, and simply brands it as its own. For example, the Harvard Business School case on

Flextronics' design of a cell phone prototype called Phone 4 that the company's CEO,

Michael Marks wishes to sell to brand name companies like Motorola, Nokia, Kyocera,

etc. (Huckman & Pisano 2003).

Form Revisions: This approach differs in that the OEM makes minor form revisions to

the overall design based on the marketing strategy. For example, Dell used this strategy

for PDAs and the LCD TV. The company evaluated prototypes from different ODMs,

39
chose the one it liked, customized the look and feel of the exterior to match Dell's

product strategy (Langberg 2004).

Design to Features: The OEM may specify the characteristics of the products, i.e. the

features, performance and form factors and let the ODM figure out the process of

designing the product. Here the OEM would not try to control sourcing or influence the

design

Design to Specifications: The OEM may provide exact specifications for the end

product and let the ODM do the actual design. The specifications may be detailed to

include specific instructions about packaging, supplier selection, component selection,

compatibility requirements, etc. For example, Cisco-Linksys conducts product planning

based on its own market research and draws up detailed specifications. Product

requirements are then provided to ODMs and component manufacturers (Invest Japan,

Special Feature).

Partial Insource: A variant of the above approaches is in cases where OEMs have

intellectual property (IP) in the design for certain modules of a commoditized product.

Here, the OEM designs that module but outsources the design for the product itself. For

example, the low-end printers made by HP may be potential candidates for design

outsourcing. The printer cartridge and engine modules are highly complex devices where

HP holds the IP. Thus, HP might design the cartridge and engine modules in-house and

consider outsourcing the design for the printer itself in future. This is a different type of

engagement with an outsourced partner, where the control over the core IP-specific

modules would still remain in-house (Amaral 2005).

40
Advantages of Fully Outsourced Design
Using the outsourced design model reduces cost, investment and time-to-market. In

addition, there are many other advantages to using the outsourced design model. This section

discusses the many different advantages of outsourcing design.

Cost and Investment Reduction: OEMs' design teams are more expensive than those at

ODMs, which are typically located countries like Taiwan or China, where labor costs are low. In

addition, ODMs or EMS/ODMs are able to allocate their design resources to several different

end-customers. This reduces idle time, allows better capacity utilization and reduces the fixed

design cost component. The cost structure of ODMs has become so low that in highly

commoditized products such as low-end PCs, laptops, printers or cell phones, OEMs are forced

to outsource design to ODMs in order to compete in the market. William Wong, a senior vice-

president for marketing at Cellon, an ODM company, estimates that using a pre-designed

platform can reduce 70% of the development costs of a new model (Engardio & Einhorn 2005).

Since the typical cost of developing a new cell phone from scratch is $10 million using up to 150

engineers, designing a product that fails in the market has a huge downside. Using an ODM can

help reduce it.

Time-to-market Benefits: The ODM model offers the fastest time-to-market

proposition since these companies use reference designs to make different products. By

changing a few features on a base design to customize the look and feel of the product, the ODM

reduces development time often to just 3-4 months at a fraction of development costs (Coker

2004). In the high-tech industry where product life-cycles are often less than a year long, this is

a very important advantage.

41
 Product Range Opportunities: Using ODMs allows OEMs to increase their product

offerings and accelerate the sales of their key products. In the example of Sony (Section 4.1.1),

the company outsources design for printers so it can offer Sony brand PCs and printers to its

customers (Carbone 2003). In this case, Sony is able to divert the margin to its own bottom-line

instead of investing in a design team to make the product in-house or partnering with another

OEM to make Sony-compatible printers.

Knowledge pooling: ODMs bring the learning and knowledge that comes from

designing the same product for multiple companies. They are able to pool that knowledge and

create better products as a result. EMS/ODMs offer design for manufacturability (DFM) and

alternate component recommendations to their customers. Most top-tier EMS companies have

acquired knowledge of manufacturing and supply chain while the ODM has design expertise. In

DFM, the company's manufacturing engineers review the board design and provide

recommendations to improve automated component placement on the board and reduce fallout,

thus improving yields. For alternate component recommendations, component engineers

evaluate a product's bill-of-material (BOM) and approved vendor list (AVL) and identify parts

that are sole-sourced, end-of-life or obsolete or where prudent supplier selection practices have

not been followed and recommend replacements or alternates. Early engagement with design

teams in an EMS/ODM scenario enables products to be designed keeping in mind these

considerations. In addition, products that are nearing maturity may be rejuvenated by a fresh

design approach which can be possible by outsourcing (Grant 2000).

Benchmarking Opportunities: Companies can use design outsourcing as a strategy to

benchmark themselves vis-a'-vis their competitors. Companies may only outsource a highly

42
commoditized product and try to use that relationship to get as much knowledge about the

competition's practices as possible from the ODM.

Disadvantages of Fully Outsourced Design

While outsourced design has many advantages, it is important to understand that this

strategy has many drawbacks such as loss of control, delays, knowledge dependence and an

increase in competition.

Loss of control and visibility: The main disadvantage of the outsourced design model is

the lack of visibility into the design process which results in a loss of control. This is due to

information asymmetry which occurs when a firm outsources design (Grant 2000). OEMs have

to depend on the ODM to have efficient processes and design teams. They have no visibility into

the internal processes of the ODM and no means to gauge the accuracy of the estimates

provided. Thus, even if it were possible to achieve a lower price point, the OEM would be

unaware of it. Therefore, it is difficult to know the true cost of product design in an outsourced

scenario. In order to gauge the cost, the OEM may have to resort to audits, benchmarking or

product teardowns, which are resource-intensive, non-value added activities (Cargille 2004).

Another disadvantage is that the OEM cannot control supplier selection and component pricing.

Material cost is the biggest cost component and therefore component pricing and supplier

selection are important considerations. Suppliers provide the most competitive pricing to

customers based on the customer's strategic importance, volume of business and potential for

future revenue. Since OEMs are not involved in supplier selection, they have no leverage over

influencing pricing. Component suppliers consider ODMs to be their customers. Thus, ODMs

are able to avail of product bundling, price-masking and rebate opportunities. Further, they may

43
not pass on any ongoing cost reductions to their OEM customers. Since OEMs have lost their

leverage with suppliers, they are unable to keep track of the true material cost.

Knowledge of Market: In order for the ODMs to successfully design ready-to-brand

products that address customer needs, they need to develop market analysis skills. Since the

traditional ODM or EMS model has been dependent on OEMs for this knowledge, their products

may not address the market's needs. Since OEMs are closer to the market, they have market

intelligence about consumer preferences. However, if they decide to use an existing product

design from an ODM in order to save costs and go to market faster, they may either be unable to

control or have to compromise on features and performance.

Capacity Restrictions and Other Delays: The introduction of a new player causes a

delay in converting market intelligence into a tangible product. When insourcing, the OEM's

marketing team has direct access to the internal design team. In an outsourced design

environment, the OEM may have to kick-off a formal request for proposal (RFP), determine the

type of engagement desired, get appropriate permissions and choose the supplier with the best

offering. In addition, the project is subject to the ODM design team's capacity constraints since

the ODM prioritizes capacity depending on the size and strategic importance of each OEM.

Knowledge and Innovation Dependence: Grant (2000) states that when an OEM

outsources design, it becomes dependent on the ODM for product knowledge; this gives the

supplier additional power. In addition, outsourcing design may lead to reduced corporate and

industry innovation. This occurs when a design supplier services multiple, competing OEMs.

Since the designs are based on a common knowledge platform, there is no incremental

44
innovation. As OEMs become less familiar with the product, they become increasingly

dependent on the supplier to innovate.

The biggest risk of design outsourcing is the increase in competition. Giving up the hold

on design may give rise to a new breed of competitors. There are three ways in which this may

happen.

Competing products from ODM partners: ODMs who currently design products get a

slice of the margins. In order to retain the full margins, they may move into the OEMs' territory

by offering directly competing branded products. For example, Motorola outsourced design and

manufacturing of its mobile phones to BenQ, a Taiwanese ODM. However, BenQ started selling

its own branded cell phones in the fastest growing Asian market and acquired the second largest

market share, next only to its largest customer - Motorola. This led Motorola to discontinue its

business with BenQ (Engardio, Einhorn 2005; Vande Vate, Ward 2005). BenQ expanded its

branded operations from 25% of its business to approximately 35-40%. Acer, another

Taiwanese ODM, is the world's fifth-largest producer of personal computers. It has also

developed its own brand and is making big gains in Europe (The Economist, 2005).

Competition in the strategic alliance segment: In addition to introducing competing

products, the rise of the ODMs allowed other companies to use knowledge-based outsourcing to

introduce branded products by bypassing OEMs. For example, mobile service operators

traditionally established strategic alliances with handset makers such as Nokia, Ericsson and

Motorola to build handsets. However, the rise of the ODMs has allowed them to bypass the

established handset-makers, who have higher prices, and produce their own "operator-specific"

handsets. They apply their own brand names to the handsets and thus, differentiate themselves

45
from rival operators. An example is that of Orange, a European mobile operator. Orange sells

its own brand Smartphones that are built by HTC, a Taiwanese ODM. This allows Orange to

avoid the higher prices that the traditional handset companies charge and thus, increase its own

margins.

Competition due to low barriers to entry: In the electronics industry, the investment in

building a good design team and purchasing expensive equipment created barriers to new

entrants looking to enter the market. Since design can now be outsourced, any company wishing

to add electronics to their product portfolio can enter the market using ODMs to design products.

For example, The Virgin Group of Australia launched the Virgin Electronics line of consumer

electronics at Target stores which included such products as CD players, MP3 and a portable

LCD TV/DVD player. These products were designed by ODMs and packaged in Virgin's

trademark deep red packaging and manufactured in China. The company expected to launch

Virgin Electronics with a staff of about 20 people focusing on what makes a difference to

consumers: how the product looks and how it works. All decisions about component selection

and manufacturing would be left to the ODMs (Langberg 2004).

Thus, while all competitive advantage is temporary (Fine 1998), fully outsourcing design

reduces the duration of temporary advantage even further, forcing companies to reevaluate their

value-proposition. The decision to fully outsource design should thus, be undertaken with

careful consideration of strategic issues rather than simply with the objective to reduce costs.

4.2.2 Joint design

A different engagement strategy used by OEMs is that ofjoint design. Here, the

engineering teams of both companies collaborate in the design process. ODMs that offer joint

46
design services are also known as Joint Design Manufacturers (JDM). Each design partner may

design specific modules and then integrate the modules to create the final product. The

intellectual property (IP) is either jointly owned by both companies, or by the OEM alone if it

uses the ODM for contract design. IP ownership in the joint design space is dependent on the

type of work done and also the negotiation power of the two partners. There are many different

ways in which IP may be split and restrictions may be placed on the ODM on rights to the

technology, knowledge sharing or the markets may be divided between the two companies.

While the IP issue is important in the joint design model, it is beyond the scope of this thesis.

In addition to IP, control over component and supplier selection is an important aspect of

the joint design model. Different JDM strategies differ by the degree of control the ODM has

over the bill of material (BOM) and the approved vendor list (AVL). A greater degree of OEM

control is appropriate for portions of the product that are core to the OEM's strategy and where

supplier relationships are important. The OEM may choose to control the entire BOM and AVL

and use the ODM for tactical repetitive design work. As ODM capabilities improve, greater

portions of the BOM and AVL may be controlled by them.

The joint design strategy is used for products where the company typically has a strategic

advantage in designing portions of its own product. This could be because customer preferences

are important, technological innovations are relatively rapid and the supplier does not have the

necessary expertise to design the entire product. Joint design is typically used for mid-range

products where margins are relatively high and market competition, though intense, is based on

product features rather than price.

47
 By jointly designing the product, the company is able to delegate the smaller, non-core

activities to the ODM while retaining control of core and strategic activities. One example is

that of Microsoft in the design for the Xbox. Microsoft designed the electronics of the Xbox and

used Flextronics to do the mechanical and system design and test engineering. In a different

case, Microsoft outsourced the mechanical and electronic portions of the design for a new tablet

PC (Costlow 2002). Thus, by outsourcing the design for non-core activities, OEMs achieve

greater efficiency in the design process and better utilization of resources.

Cisco Systems has taken joint design development to a different level. In the early

1990s, Cisco began establishing strategic partnerships with a few selected manufacturers and

opened its systems, processes and networks for them for joint equipment development. Today,

partners provide most of Cisco's component, hardware and manufacturing innovation. The

Cisco Hosting Applications Initiative (CHAI) has about 30 vendors and service providers

developing new technology for Cisco's routers and optimizing performance in hosted

applications (Quinn 2000).

Joint design must be undertaken with care and a careful strategic evaluation of the

modules being outsourced. In addition, the OEM must include restrictive knowledge transfer

clauses in the contract and acquire a financial stake in the outsourced entity in order to ensure

preferential treatment and exercise greater amount of control. This is because the transfer of

valuable knowledge may give rise to future suppliers or competitors.

Advantages of Joint Design

Joint design is a strategy that curtails many of the risks of the fully outsourced model and

has the following benefits:

48
 Control and Visibility: This strategy ensures a greater amount of control over the

design process. The OEM can specify the features, performance and form that the product

should have and provide sourcing and pricing guidelines. It enables the OEM to maintain

existing relationships with component suppliers and thus, wield greater control over the supply

chain by directly negotiating prices and supply terms for the most critical items.

OEM Objectives vs. ODM Objectives: Joint design gives greater negotiating and

decision making power to the OEM. Thus, the OEM has more opportunities to design products

to serve its own objectives rather than to facilitate the ODM's motives. Implementation of

design for supply chain strategies discussed in the next chapter of this thesis is possible in a joint

design model. The degree of incentive misalignment is lower since the decision-making

authority rests with the OEM.

Product differentiation: This model also enables OEMs to differentiate its products and

design them for its own customer base. OEMs typically use the joint design approach when they

are looking to differentiate their product offerings in the market. While the fully outsourced

model increases the homogeneity of products, joint design offers the ability to outsource non-

critical modules while still retaining control over product differentiation.

Knowledge sharing and transfer: Grant (2000) states that since product development

firms work on design projects for many different OEMs, they progress at a pace much faster than

their customers. They have their pulse on the latest design practices and trends. Thus, joint

design gives the OEM's internal design team the ability to engage with these firms and learn a

different approach. Some companies consider the bilateral knowledge transfer to be a

competitive benefit of joint design. One unintended consequence is that OEMs may realize that

49
 they do not have the competitive edge in product design that they think they do. Thus, design

outsourcing provides the means to benchmark against competitors within the industry.

Disadvantages of Joint Design

Even though joint design resolves some of the issues with the fully outsourced model, it

is a risky strategy because it may well set the stage for the product to lose differentiation, become

commoditized and cause the OEM to lose any competitive advantage.

Cost and Investment: Joint design is much more expensive than the fully outsourced

model because the OEM would have its own design team and equipment. Thus, the OEM would

incur engineering costs, both internally and NREs from the ODM.

Coordination: When the OEM development team is in the US and the ODM team is in

Asia, often, coordination becomes complicated. Different time zones have a two-sided impact -

round-the-clock development may reduce time-to-market, however, this may be offset by delays

in getting information and problems solved. In addition, the OEM would have to incur recurring

travel expenses for its engineers.

Time-to-market: Product time-to-market is longer as compared to outsourced design

projects. This is because the product may be developed from scratch and the process requires a

great degree of coordination between design teams of both companies. The joint design strategy

may affect design team morale if it is perceived that this is just the step before the product will

be fully outsourced.

Knowledge transfer: Knowledge transfer to the ODM is unavoidable in the joint design

process; in some cases, the OEM may unknowingly be transferring valuable IP. For example,

50
Motorola's design team spent a few months modifying and improving an original design from

Chi Mei, a Taiwanese ODM, for its MPx200 handset. The company now owns the specifications

for the design then belonged to Motorola so that the ODM may not transfer the improved product

to Motorola's competitors. However, the process ofjoint design educated the engineers at the

ODM (The Economist 2004).

Competition from ODM: Joint design does have the long-term risk of creating future

competitors by transferring knowledge and expertise. ODMs work with different customers and

pool the acquired knowledge which could enable them to engineer their own products. This

would reduce the competitive advantage of the OEM and commoditize the product in the market

forcing companies to eventually outsource the full design.

4.2.3 Insourced Design

At the opposite end of the outsourcing spectrum is insourced design, where all design

functions are carried out by the OEM in-house. Certain product modules may be procured from

strategic suppliers. However the product itself is designed in-house. Insourced design is most

appropriate in high-margin markets where customers value features more than cost. For

example, Apple has positioned itself in the market as a premium design company and created

pride of ownership among its customers. Apple designs the iPod in-house and prints "Designed

by Apple in California" on the back, thereby reinforcing the exclusivity of Apple's products and

the company's commitment to its customers (Burrows 2005). Its vertical integration strategy

involves controlling the design for both hardware and software products.

Other factors that impact the in-house design decision are product complexity and

supplier capabilities. Novak and Eppinger (2001) link product complexity with vertical

51
integration. Thus, as product complexity and strategic value of in-house capabilities increase,

greater the incentive to insource design. ODMs tend to specialize in developing products that are

in a mature phase of the technology life-cycle focusing on incremental rather than disruptive

technological advancements. Thus, they would have to develop the capability to develop highly

specialized complex products. The ODM model is more conducive for those fast-moving

consumer goods products where price is the important consideration. In case of highly complex

products where the volume is less, products are engineering intensive and need customization,

the ODM model would fail. For products with these characteristics, design expertise is a

valuable competitive advantage and therefore, outsourcing is not the appropriate strategy.

Advantages of Insourced Design

In-house design enables the highest degree of control and visibility. The OEM is fully

involved in supplier selection and component price negotiations. It has flexibility in designing

features, functionality, performance and product look and feel. It can tailor products to its

customers' unique requirements and react quickly to market changes or shifts.

The practice of designing products in-house can also be used as a differentiating strategy

in a market where all other companies are outsourcing. Apple's strategy of printing "Designed

by Apple in California" is an example of this type of product differentiation. Companies can

then serve a niche market which is willing to pay a premium in order to buy products designed

in-house.

Material costs and supply chain costs are the biggest components of total product costs.

Thus, in order to achieve lower price points, companies can design products to facilitate supply

chain cost savings and keep costs low, enabling them to design products in-house. Thus, instead

52
of jumping on the outsourcing bandwagon, OEMs should evaluate all possible alternatives to

reduce costs and remain competitive in the marketplace.

Disadvantages of Insourced Design

The biggest disadvantage is the high cost of design and long development time which

impacts product time-to-market. For example, in the case of Apple, the company has to pay high

salaries to its design teams based in California whereas ODMs typically have design teams in

low-cost countries such as Taiwan or China. In addition to the cost of design, in this model, the

OEM is unable to benefit from the economies of scale that are available to ODMs due to

combined volumes and shared design capacity.

Insourced design can deprive the company of the opportunity to learn from the skills

acquired by outsourced partner. Since ODMs work with many different competitors and

specialize in the design for certain products, there are some benchmarking and knowledge

gaining opportunities in outsourcing that the company may forgo by designing in-house.

4.2.4 One-stop shop outsourcing

One-stop shop outsourcing is the strategy of using a single provider such as an

EMS/ODM to provide the full spectrum of services right from product design to component

procurement, manufacturing, distribution, after-sales services and repair. Large EMS providers

such as Flextronics, Foxconn/Hon Hai, Solectron, etc. have the global manufacturing and

distribution footprint and are now becoming ODMs by acquiring design companies.

Foxconn/Hon Hai, for example, is highly vertically integrated, manufacturing components such

as PC enclosures, cable assemblies, I/O connectors and providing assembly services in China.

53
Foxconn/Hon Hai's acquisition of Ambit in 2004 signals its entry into the EMS/ODM segment

whereby it has the capability to design products, manufacture and supply components, procure

and assemble products and distribute them worldwide.

Advantages of One-Stop-Shop Outsourcing

The clear advantages of the strategy are the ability to form strategic partnerships with one

supplier, reduction in cost of coordinating and managing multiple service providers and

reduction in complexity of the supply chain. Companies which outsource all their functions to a

single service provider are able to form strategic partnerships and implement inter-corporate IT

linkages, closer executive-level management ties and gain disproportionate mindshare. While

outsourcing can result in cost savings and reduction in investments, managing the outsourced

business requires a different set of costs and investments. Contracts have to be signed, on-going

interactions between the two companies are required, periodic audits have to be done and skills

to manage an effective trusting partnership have to be developed. One-stop-shop outsourcing

reduces the complexity and cost of managing the outsourced relationship.

Value-added Manufacturing Services: EMS providers have a much larger

manufacturing footprint worldwide whereas ODMs mostly have manufacturing facilities in

Taiwan or China. Thus, EMS providers are able to offer services such as build-to-order,

configure-to-order, postponement, etc. which allow OEMs to achieve better supply chain

coordination.

Negotiating Power: By outsourcing everything to a single EMS/ODM service provider,

an OEM can increase its strategic importance to that supplier. Splitting up the business can

result in reduced mindshare from both service providers. The higher value of business would

54
enable the OEM to negotiate better pricing and other terms as the service provider may be

motivated to provide the OEM preferential treatment.

Competition: One important advantage of the one-stop-shop outsourcing strategy by

using EMS/ODMs is that these companies are less likely to become direct competitors as this is

counter to their business model. EMS/ODMs view themselves as non-brand intensive

companies with a greater focus on providing outsourcing solutions to OEMs, although this may

well change in the future. Also, the portfolio of customers using EMS services is diverse. An

ODM however, views itself as an engineering company and hence is more likely to begin

branding and competing with OEM customers.

Disadvantages of One-Stop-Shop Outsourcing

Despite its advantages, this strategy is risky since it allows the negotiating power within

the supply chain to consolidate with one entity - the service provider. Outsourcing involves

giving up control to an external entity which has its own priorities and motivations. It creates the

dependence on an outside company for the full gamut of services, which increases the many

opportunities that the service provider has to profit at the OEM's expense. Grant (2000) says

that if a supplier perceives that it has greater bargaining power than the OEM, then it could use

its power to negotiate additional concessions.

Switching Costs: While it is possible to move business from one supplier to another in

the event that the EMS/ODM becomes monopolistic, this incurs substantial costs. First the OEM

would have to create an RFP, a request for proposal, receive the proposals, evaluate them,

qualify the supplier and negotiate a contract. Following that, the company would have to

transfer all capital equipment, inventory, component supply agreements and undertake an

55
extensive training process. All of the above create barriers to discontinuing business with the

existing provider. This further augments the service provider's monopolistic power.

Lack of Redundancy: The lack of redundancy and the total reliance on a single service

provider creates supply chain vulnerabilities that may create problems. For example, capacity

constraints at the EMS/ODM in a rising market may jeopardize the revenue stream of the ODM.

Also, the practice weakens the OEM's ability to negotiate cost reductions and demand

performance improvement.

4.3 Incentives
Incentives are a very important factor in determining partner behavior in an outsourcing

relationship. Amaral, Billington and Tsay (2004) argue that the agenda for the OEM and the

service provider may not be completely conflicting, but it is naive to presume that they will be

perfectly aligned. They contend that outsourcing creates this divergence, as a group of

employees with a common stake in their employer's success is replaced by an external entity that

has to satisfy its own investors in the short and long run. In an insourced environment, the OEM

makes all decisions with the objective of furthering its own interest. However, in an outsourcing

situation, the objectives of the OEM may conflict with those of the service provider. Since the

service provider's employees have a vested interest in ensuring success for their own employer,

the interests of the OEM may be compromised. When OEMs outsource all the functions of

design, procurement, manufacturing and after-sales services, they create more opportunities to

widen the incentive gap.

In a design outsourcing relationship, the ODMs have an incentive to design a generic

product with the lowest product differentiation and try to sell it to as many customers as possible

56
in order to achieve the lowest possible fixed cost. From an OEM standpoint, this further reduces

any product differentiation and thus, eliminates their importance in the marketplace.

Thus, it is important to align incentives between partners to foster cooperation and to

encourage desired behavior. Narayanan and Raman (2004) argue that by changing how, rather

than how much, they pay partners, OEMs can improve supply chain performance and profits for

all partners. For example, in 2001, Cisco scrapped around $2.5 billion in excess inventory,

making it one of the biggest inventory write-offs in US history. Cisco rewarded its EMS

providers for on-time delivery and higher cost reductions. In order to achieve a high service

level, these companies carried large portions of safety stock and to get components at lower

prices, they made large volume purchases. When demand slowed in 2000, the EMS providers

had high levels of inventory earmarked for Cisco leading to the massive inventory write-off.

Had Cisco tied incentives to inventory turns as well as to on-time delivery, it would have created

tradeoffs which would have ensured that the service provider acted in the best interest of both

companies.

In addition to creating incentives or penalties that impact performance, incentives that

encourage risk sharing create stronger alignment of objectives. Linder (2004) gives the example

of a heavy equipment manufacturer which had strategic partnerships with some companies that

make critical subassemblies for its products. Instead of purchasing the subassemblies at a fixed

price, the OEM shared an agreed-upon percentage of its revenue from the sale of its equipment.

Thus the fate of both companies was linked to the sale of the equipment (Linder 2004). In

addition to revenue sharing contracts, risk sharing contracts are also used. Linder states that in a

risk sharing contract, both companies invest resources and share the benefits of a successful

57
outcome. By having a financial stake on the same outcome, the interests of the two companies

converge and this makes them more likely to act in the best interest of the partnership.

The next chapter outlines the different ways in which an incentive gap exists in

implementing supply chain strategies at the design stage in an outsourcing relationship. OEMs

outsourcing design should be aware of their partner's priorities and modify incentives in order to

promote behavior that is best for the supply chain.

58
5 Design for supply chain
(DFSC)

Product design impacts all future activities such as component procurement,

manufacturing effectiveness, economical distribution and reduced after-sales services. Thus, it is

important to design products such that supply chain considerations are factored in at the design

stage. It is not sufficient to simply build products based on market considerations. In order to

maximize shareholder value, it is important to design products such that they can be delivered to

the end customer in the most efficient way possible without compromising quality.

The process of designing products keeping in mind supply chain considerations is known

as design for supply chain (DFSC). There are many DFSC strategies that companies use and this

section discusses the benefits of each one with examples of companies implementing them.

Traditionally, OEMs have been responsible for product design and many companies in the fast-

moving consumer electronics industry have developed the capability incorporating supply chain

thinking into product design phase. With design outsourcing, they lose control of the design

process. Also, outsourcing causes incentive misalignment when companies are not careful about

the way they reward their service providers. The two companies may have conflicting

motivations with respect to implementing DFSC strategies. Thus, the hypothesis for this thesis is

that design outsourcing adversely affects companies' ability to envision, evaluate and implement

DFSC strategies.

59
 As discussed in Chapter 4, companies use different levels of outsourcing. The OEM's

engagement strategy impacts the amount of control they can exercise over the design process and

thus, their ability to evaluate the benefits of DFSC strategies and implement them. This section

outlines the different DFSC strategies, the tradeoffs inherent in evaluating their impact and how

the different levels of outsourcing impact their implementation. The following strategies are

covered.

Section Topic
5.1 Supplier Selection
5.2 Supplier Integration in Product Design
5.3 Single/Sole Sourced vs. Multi-Sourced
5.4 Component Commonality
5.5 Product modularity
5.6 Postponement
5.7 Managing product variety
5.8 End-of-Life and Obsolete Components
5.9 Economic packaging and transportation
5.10 Product Recycling and Reuse
5.11 Product Obsolescence

5.1 Supplier Selection

Component suppliers play a very important role in the high-tech supply chain. They act

more like joint design partners than suppliers because most high-tech products are highly

modular with a lot of product innovation being carried out by component suppliers. For

example, Intel is a component supplier however the introduction of a new Intel processor can

make PCs with the old ones obsolete. In addition, the high-tech industry experiences periods of

high and low demand. While the boom times of the late 1990s were characterized by low supply

capacities and therefore, allocation markets, the period between 2001 and 2003 was

characterized by excess capacity and inventory. Strong relationships with certain key suppliers

60
are a competitive advantage since OEMs with strategic partnerships with suppliers are assured

supply even in allocation markets. Similarly, suppliers are more than willing to accept returns or

cancel open orders for certain OEMs during periods of excess demand. These types of problems

in procurement and manufacturing can be avoided by using prudent principles when choosing

suppliers especially for the critical components.

Material cost is usually the primary driver in supplier selection. However, considerations

such as a supplier's strategic importance, past delivery performance, willingness to offer material

cost reductions, lead time, minimum order sizes, willingness to participate in lead time reduction

programs, IT integration capabilities and other delivery terms should also be incorporated into

the decision-making process.

Incentives play a big part in determining the choices that design teams make when

choosing suppliers. For example, if low material cost is rewarded, then design teams choose

suppliers who provide the lowest cost. However, the low cost may be accompanied by long lead

times, a low on-time delivery record, lack of responsiveness to issues and unwillingness to

participate in strategic lead time reduction or information sharing activities. If the contract

stipulates stiff penalties for delivery reliability instead, then suppliers who have a strong delivery

performance and a history of collaborating with the ODM on supply issues will be selected.

Similarly, design engineers working autonomously may design-in suppliers with

technological superiority but high material cost and long lead times. For high value items, it is

imperative that suppliers be strategic partners who assume a portion of the risks. Supplier

selection is one of the most important DFSC decisions with the most impact to future activities

and performance.

61
5.1.1 Design Outsourcing and Supplier Selection
EMS and some ODMs are still viewed by most suppliers as companies that execute the

OEMs' design. The supply chain for electronics manufacturing in an in-house design situation

looks as below where the OEM controls the primary sourcing decision and the EMS or ODM

controls the execution. Thus, component suppliers follow the traditional model of pitching and

selling products to OEMs directly rather than to EMS providers. According to Jeff Erhart, the

Director of Worldwide Channel Sales at Broadcom Corporation, most leading semiconductor

chip suppliers consider OEMs to be their strategic customer even when design is outsourced to

ODMs. They have developed historically strong relationships with OEMs and therefore, pitch

their products directly to them, working directly with them on pricing and supply issues. He says

this perception will have to change and component suppliers will have to create sales teams

targeted at developing EMS/ODM relationships. Thus, in the short-term, OEMs may be able to

retain their leverage with their component suppliers until the ODMs form their own partnerships.

Component
OEM Supplier

EMS or ODM

Figure 6: Supply Chain Linkages in In-house Design

62
 Fully Outsourced Model: Here the OEM depends on the ODM to create the entire

product design with minimal involvement; it has neither control nor visibility into the suppliers

chosen for the different components. The supply chain view in a fully-outsourced design is as

below where the OEM is dependent on the ODM for developing and maintaining strategic

linkages with component suppliers.

Component
Supplier

Figure 7: Supply Chain Linkages in Outsourced Design

The OEM may have developed strong relationships with certain suppliers due to the long

history of business and integration projects with them. They may prefer to work with these

suppliers which would enable them to negotiate ongoing cost reductions and during allocation

markets, to negotiate preferential component allocations. However, ODMs may select suppliers

who have better linkages with them in order to retain control over pricing and supply terms.

Thus, the OEMs' preferred suppliers may not be selected. Since OEMs are usually held

accountable for their forecasts, any inventory purchased based on the forecast is their liability. If

they have no former history of business with a supplier, then they cannot influence the supplier

to provide cost reductions or participate in inventory risk-sharing agreements.

63
 The above problems highlight how outsourcing may result in non OEM-preferred

suppliers being selected. However, outsourcing provides many opportunities to leverage

economies of scale due to demand aggregation. Because of this, ODMs have also developed

strong supplier linkages and they may be depended upon for having the necessary supplier

leverage to get cost reductions, preferential allocations and favorable supply terms. If OEMs

work with large established ODMs who have high volumes and greater strategic importance to

suppliers, then they can be assured of higher cost reductions and preferential supply terms. The

distribution of these benefits is a function of the OEM's negotiation power and strategic

importance to the ODM. OEMs can influence the supplier selection process by providing

preferred vendor lists and negotiating supply agreements that make the ODM responsible for raw

material and work-in-process inventory (WIP).

Joint Design: In a joint design effort, the OEM is able to exert a greater degree of

control over the supplier selection process. The ongoing collaboration between the two

companies in the design process facilitates the exchange of information and set priorities in

supplier selection. It also enables the two companies to choose suppliers where each has

leverage and thus, maximize the number of strategic suppliers.

In summary, while design outsourcing does make the task of supplier management more

difficult from the point of view of the OEM, the OEM can influence the process through the

choice of the ODM partner, by the terms of the partnership contract and by the targets and

deliverables they establish.

64
5.2 Supplier Integration in Product Design
Supplier integration in product design is a critical component of strategic supply chain

management. Ragatz, Handfield and Scannell (1996) found in a survey that more and more

companies are realizing financial benefits from involving suppliers in the design process and

plan to integrate suppliers earlier and in more depth. The companies responding to their survey

noted that some of the benefits of supplier integration were the generation of new and better

ideas, development and deployment of new technologies, reduced cycle times, improved product

quality and a reduction in material purchasing costs, leading to an overall competitive advantage.

Supplier integration during design can also bring up any potential end-of-life (EOL) parts or

components that may become obsolete (discussed in Section 5.8).

The authors report that companies involve suppliers in many different ways. Some

companies hold deliberations with suppliers in an informal setting when conducting product

design, while others collaborate closely with joint design teams. A higher level of involvement

occurs when companies outsource a subset or module of the product to a supplier for

independent design with joint integration into the final product. This kind of involvement should

be practiced where there is a close strategic partnership and trust between the two partners. The

decision of the level of integration should be a function of key internal core competencies,

current and future new product developments and external development and manufacturing

needs. Laseter and Ramdas (2002) provide a framework for deploying supplier integration based

on a classification of the different subassemblies and components as critical systems, hidden

components, simple differentiators or invisible

65
subassemblies.

Hidden Simple Invisible

Critical Systems
Components Differentiators Subassemblies
Highly Less Highly Invisible,
Differentiating, Differentiating, differentiating, Moderately Costly,
Description High Cost, Highly Low Cost, Simple Moderately Costly, Moderately
Complex Systems Components Simple Assemblies Complex Systems

Supplier
Early and Few & Simple Few & Simple Multiple but Simple
Involvement Continuous Interfaces Interfaces Interfaces
Timing

Few, Possibly
Product Line Single Version for Many Versions for Few and Shared
Unique to the
Complexity the Product each Product Across Products
Product

Designed by Expert
Design Catalog or Supplier Concept Inside, Black Box to
(in-house or
Philosophy Design Supplier Details Expert Suppliers
external)

Modified from Laseter & Ramdas (2002). Product Types and Supplier Roles in Product Development: An
Exploratory Analysis. IEEE Transactionson EngineeringManagement. 49:2. Pg. 116.

Figure 8: Supplier Integration based on Product Characteristics

The process of supplier integration should be a carefully orchestrated systematic process

beginning with a deep understanding of the suppliers' capabilities. Also, it is a long-term

process involving strategic information sharing of emerging technologies and their applicability

to future products. Finally, supplier integration is an effort spearheaded by management in both

companies to ensure that the supplier's insights are considered in overall decision-making

(Petersen, Handfield & Ragatz 2003). One example of this is that of Broadcom Corporation, a

leading broadband semiconductor company which supplies a range of semi-custom to custom

integrated circuit (IC) chips to its customers. Broadcom chipsets are usually sole-sourced and

expensive components in the product. For these chipsets, supplier capabilities, past performance

66
record and product strengths are the most important considerations during the design phase in

addition to price and supply terms. Broadcom's marketing and product development personnel

remain in close contact with the design team at the OEM throughout the design process. They

provide information about the company's product roadmap and evaluate the best-fit products for

the OEM's requirements. They ensure the chipsets will be compatible with past and future

generations of the product. By facilitating engineering collaboration, they develop new custom

products specific to OEM requirements. They provide ongoing support through product

development and negotiate pricing and supply terms with their customers (Erhart 2004).

While supplier integration is a sound strategy, it does have a few risks. One is that by

integrating suppliers closely in product design and manufacturing, the OEM's role may get

increasingly devalued as suppliers take over the design of more and more subassembly modules.

Most ODMs were component suppliers (Carbone 2003). Many ODMs such as Foxconn/Hon

Hai, Inventec, Compal, Delta Electronics, Industrial and Gigabyte were suppliers of

motherboards and other computer components. They gradually moved into computer systems

and then into PCs, desktops, servers and telecommunication products. Van Hoek (1998)

proposes that the OEM must maintain a lead over suppliers by designing the product architecture

and facilitating the information flows. Another risk is that close supplier integration results in

total reliance on a single supplier which leads to supply chain vulnerability. Thus, supplier

integration should be practiced for high value subassemblies or modules and incentives should

be aligned in order to ensure that partners' actions are beneficial to the entire chain.

67
5.2.1 Design outsourcing and supplier integration
Design outsourcing changes the players in the supplier integration process. While

previously, the OEM's design team had close ties with strategic component suppliers, now it is

the ODM which is more aware of new components, innovations and developments. This creates

information asymmetry because the OEM is close to the market but is unaware of the supplier's

product capabilities while the ODM is aware of supplier capabilities but unaware of market

requirements. Thus, new products may be over or under engineered relative to market

requirements.

Innovation: Innovation in the high-tech industry is often facilitated by or an outcome of

supplier integration. With ODMs controlling the design process, innovation also gets

outsourced. This has more adverse consequences in the long-term when OEMs are more

dependent on the ODMs for knowledge. One argument for the design outsourcing trend is that

outsourcing the more mundane design functions allows OEMs' design teams to focus on the

more value-added complex development work. However, Engardio and Einhorn (2005) noted in

their Business Week article that in practice, OEMs' R&D budgets have been shrinking. For

example, despite the recovery in the high-tech industry since 2004, many OEMs' corporate R&D

budgets as a percentage of sales have shrunk. HP's R&D spending used to range around 6% of

sales but is at 4.4%. Cisco Systems' budget shrank from the previous mean of 17% to a current

level of 14.5%. Other OEMs such as Motorola, Lucent Technologies and Ericsson have also

reduced R&D spending (see Appendix C). In the fully outsourced design model, there are fewer

opportunities for supplier integration with OEMs, which are cut out of the design process.

Fully Outsourced Model: Here the onus of supplier integration is on the ODM. Even if

the OEM provides guidelines on supplier selection, it cannot control integration because it

68
requires commitment at every level within the two companies. ODMs focus on incremental

innovation which does not call for a deeper level of integration.

Joint Design: Joint design provides many opportunities for OEM control over supplier

integration. Since the OEM dictates product architecture and outsources only non value-added

portions of the design, it can retain its supplier linkages. For many products, forward and

backward compatibility of products is essential. Supplier integration facilitates knowledge

sharing of product evolution and roadmaps with suppliers and the ODM. This pooling of

knowledge is an important advantage of the joint product development model.

In summary, supplier integration is a very important activity which determines and

facilitates innovation. When companies outsource design for core products, they risk being

pushed out of the innovation game in the long-run due to the gradual erosion of capabilities and

supplier linkages. To some extent, joint design helps to retain control over this process;

however, in the long-run, even joint design is a risky strategy.

5.3 Single/Sole Sourced versus Multi-Sourced

Components that are sourced from a single supplier make the company dependent on the

performance of that one supplier. The fewer suppliers that exist for any outsourced element, the

more the leverage those suppliers have over the OEM. (Fine, Vardan, Pethick & El-Hout 2002).

For example, after IBM outsourced the processors to Intel, it was completely dependent on Intel

and had no control over the prices charged by Intel or the sourcing conditions such as lead times

or minimum order quantities (MOQ) imposed. Also, an industry-wide single sourcing strategy

may create monopolistic power and reduce the supplier's motivation to innovate.

69
 Another risk of sole sourced parts is the lack of redundancy leaving the supply chain

vulnerable to any disruptions. A case in point is of Nokia and Ericsson from March 2000. A fire

destroyed a Philips plant in Albuquerque that supplied radio-frequency (RF) chips for mobile

phones to Nokia and Ericsson. Nokia did not depend only on Philips for the chips and so it was

able to quickly shift its orders to other suppliers. Ericsson, on the other hand, had a single-source

strategy in order to cut supply base costs, and so it was forced to cut back its own production and

cede market share to competition (Vande Vate & Ward 2005). In another example, in February

1997, an Aisin Seiki brake manufacturing plant supplying to Toyota was destroyed by fire.

Since Toyota practiced 100% just-in-time supply which required single sourcing, it only had the

bare minimum inventory. The cost to Toyota was USD 195 million (Mortimer 2001).

Because of these two risks, a cautious strategy is to design-in multiple suppliers for

strategic components with the possibility of using their products interchangeably in order to

hedge against supply disruptions. In addition to building a robust and flexible supply chain, this

strategy helps OEMs to retain negotiating power. Suppliers can no longer dictate pricing or

delivery terms as the company can easily switch to a different supplier who will give a better

deal in order to get the business.

While this strategy can be used for every kind of component, it is especially important

evaluate the tradeoffs of multi-sourcing expensive and strategic subassemblies or components.

For criticalor invisible systems, when the company is dependent on suppliers for knowledge-

based products where suppliers own IP or differ by standards they use, multi-sourcing can be a

challenge. For example, HP outsourced the engine for its network printer to a supplier in Japan,

who owned the patent for the specialized product. The supplier desired a lead time of 14-weeks

for the engines. HP's network printers had a total lifecycle of 18 months. The task of

70
forecasting demand in order to source the engines with a 14-week lead time thus became a major

challenge for HP (Feitzinger & Lee 1997). Developing an alternate supplier would require

investing valuable time and design resources. It could also have an adverse effect on the

relationship with the incumbent supplier.

Another situation where multi-sourcing can be a challenge is when companies employ

just-in-time (JIT) strategies and integrate suppliers into the design process (discussed in Section

5.2). In such cases, the establishment of a single strategic vendor partnership is essential in order

to build trust, cooperation, integration and information sharing. However, these partnerships

should be entered into with suppliers of key design-intensive products which are critical to the

performance of the end product. These partnerships can be formalized with risk-sharing

incentive strategies in order to create commitment from the supplier to the OEM. Thus, there are

tradeoffs to be considered in choosing the best strategy.

5.3.1 Design Outsourcing and Single versus Multi-Sourcing

ODMs and EMS providers have the capability to identify alternate suppliers for most

components because they have accumulated component selection information about their

customers. For example, large EMS companies provide alternate parts recommendations to their

customers. Their component engineers work with OEMs to identify single/sole sourced

components and recommend alternate parts/ suppliers. Thus, they are capable of designing

products with multiple sources in order to build supply base flexibility.

Motivation vs. Incentive: Since ODMs mostly design commoditized products, the

pressure to reduce cost is high. Designing multiple sources gives them the power to foster

competition between suppliers and drive prices down. Therefore, they have the motivation to

71
multi-source components. They are able to do this effectively for components which are simple,

pin-to-pin compatible, form-fit-function replacements. However, some semi-custom or custom

may require board redesign in order to be qualified as alternate parts. The approval process is

time-consuming since it involves requesting samples and drawings from the supplier and

conducting tests to ensure that they are compatible with the existing component. Thus, the

process of qualifying alternate parts is time and resource intensive and impacts the ODM's

incentive to implement. Redesign may increase the cost of the product diminishing the

importance of multi-sourcing. One important point to note is that multi-sourcing is a more viable

option in the outsourced model because the ODM's cost structure is much lower than OEM's

cost structure.

Tradeoffs: For any custom components where suppliers own intellectual property (IP) or

are strategic partners in design integration, it is necessary to evaluate the benefits of developing

alternative sources. Multi-sourcing can harmfully affect existing supplier integration

partnerships. This requires careful management of existing relationships and strict

confidentiality guidelines to be implemented and enforced. As described above, an increase in

cost of redesign may adversely impact the cost savings that may accrue from multi-sourcing.

Thus, long-term redundancy, resilience and supply assurance concerns may take a back seat to

the more immediate cost pressures.

Fully Outsourced Design: Here, OEMs do not have the ability to influence sourcing

during the design phase unless they provide guidelines in the design specification. However,

alternate parts can be added to the approved-supplier list (AVL) at any point in the product life

cycle. Product reviews can also create the checks and balances necessary to ensure that sole-

sourced components get the requisite attention. OEMs should have product review meetings

72
where all critical sole-sourced parts are marked as action-items with a set timeline for approving

an alternate vendor.

Joint Design: Since the ODMs have the capability to identify and implement alternate

parts, and because their cost structure is lower, this task should be outsourced to them in the joint

design model. This would allow the OEM's design resources to be freed up to focus on more

value-added functions.

In summary, ODMs have a stronger likelihood of implementing multi-sourcing because

of their lower cost structure, greater cost pressures and higher frequency of product redesign.

However, it should be noted multi-sourcing has inherent tradeoffs that need to be evaluated in

order to effectively balance supplier integration with supply assurance. For the strategic sole-

sourced components, a risk-sharing agreement may be entered into where each partner

(component supplier, ODM and OEM) would share the losses from any supply disruptions.

5.4 Component Commonality

Lee (1993) states that the success of postponement (section 5.6) depends on the viability

of storing semi-finished goods, which depends in turn on component commonality and product

modularity (section 5.5). Component commonality is when two different components used in

different but related products are combined to create a common component that can be used in

either product. A slight variation of component commonality is using common parts in many

different products - also known as procurement standardization by Swaminathan (2001).

Common parts make it easier to create semi-finished goods that can then be customized into

different end products. The purpose of component commonality is to reduce the number of

73
SKUs, thereby reducing inventory. It also enables demand aggregation, risk pooling and

facilitates better forecasting.

When designing products, component commonality is a very important consideration

because of its many benefits. Having common parts facilitates demand aggregation and risk

pooling and prevents inventory buildups. The strategy is used frequently in the PC industry

since most computers are fairly homogenous. Hidden components or invisible subassemblies are

the primary candidates for commonality projects. Thus, companies use the same component

from the same supplier on each version of their PC. In the high-tech supply chain, most

components come in reels with minimum pack quantities (MPQ). Also, many suppliers dictate

MOQ requirements in order to achieve lower price points. If the demand for multiple products is

combined, then inventory turns can improve. Also, risk-pooling can be done by offsetting the

risk of shortages in one product portfolio with the occurrence of excess in another. Thus,

common parts simplify the procurement process and enable companies to lower their costs.

Feitzinger and Lee (1997) discuss the network printer design at Hewlett-Packard. The

assembly for the network printer was outsourced to a Japanese partner. The components

included an engine, the printed circuit board, a power supply and a fuser unit. The final product

was sold to both Europe and North America. Printers sold in Europe would have a 220-volt

power supply whereas the ones sold in North America would have 110-volt power supply. On

account of the long lead times for the product and the shipping times from Japan, it was

increasingly difficult to predict the demand in the two continents. By investing resources in

R&D, HP came up with a universal power supply. This would allow them to pool the risk of

demand shortfalls or overruns in each region. By using the common power supply, HP could

delay the decision making process of the quantities to ship to each region until demand was

74
known. Thus, even though the universal power supply increased the cost of the product, HP

saved almost 5% in total annual costs of manufacturing, stocking and distribution worldwide.

5.4.1 Design Outsourcing and Component Commonality

ODMs have an incentive to implement component commonality in cases where it would

simplify future product design, reduce design time and cost. The possibility of demand

aggregation increases their ability to request greater cost reductions from the component

supplier. Since they design and redesign products frequently based on the requirements of their

customers, implementing commonality would simplify their redesign process.

Fully Outsourced Model: Component commonality is a favored strategy of ODMs

because it helps them to cut costs and design time, their two most important incentives. If the

OEM requires additional commonality beyond what the ODM typically provides, they may have

to establish those requirements in the product specification. However, this is difficult to do,

especially in the long-run when they are dependent on the ODM for product knowledge. Where

the OEM uses several ODMs to manufacture similar products, it may be difficult to implement

commonality across the product due to conflicting objectives.

Multiple ODMs and Component Commonality: When an OEM uses many different

ODMs for its products, different suppliers may be designated to supply similar components. For

example, some leading OEMs outsource the design for some similar products to multiple ODMs.

Using different ODMs for similar products allows them to benchmark ODMs, encourage

competition and develop multiple suppliers. One outcome of this strategy is component

proliferation; since each ODM has strategic relationships with different suppliers for the same

component, they may be unwilling to use suppliers that another ODM designed. Thus, the OEM

75
ends up dividing its total volume among the ODM-chosen suppliers. Since OEMs are

responsible for demand forecast and inventory, the increased number of components results in an

increased inventory liability. The OEM can get the ODM to achieve some commonality by

requiring multi-sourcing of any sole-sourced components and then recommending alternate

parts. OEMs should provide component sourcing guidelines in their design specifications;

otherwise, the risk of component proliferation exists.

Joint Design: Here, the OEM controls the design process and can require the ODM to

work on a component commonality project. The OEM has a greater degree of control and

knowledge about the product and the design process. The ownership of the non-recurring

expenses (NREs), i.e. the fixed cost of the redesign, depends on the negotiating power of the two

partners and the importance of the OEM to the ODM's future growth. Also, the ODM may be

willing to absorb the NREs if it perceives significant future cost savings due to the redesign.

Incentives: If material cost increases due to component commonality, this may offset

some of the benefits of this strategy, especially if the OEM rewards the ODM based on material

cost reductions. For example, in HP's implementation of the universal power supply, the

product cost increased by $30 per unit. For highly commoditized products, the pressure on cost

reductions is high; thus, the possibility of an increase in product cost reduces the motivation to

redesign in order to implement commonality. Also, if the benefits of a redesign effort

disproportionately favor the OEM, then the ODM might negotiate in order to higher NREs to

share some of the benefits of commonality. The OEM may either demonstrate the benefits of

implementing commonality to the ODM on future designs and revenue generation or divide the

pie.

76
 Tradeoff - Commonality vs. Differentiation: Component commonality's economic

benefits are offset by a possible loss of sales which can occur from the reduced product

differentiation. If it is to be implemented in criticalsystems or simple differentiatorswhich

visible to the customer because of form or performance changes, then it has to be done in

conjunction with marketing input since the reduced differentiation may negatively impact sales.

This is relatively easy when the design and marketing teams are in-house. However, in an

outsourced relationship, where each group resides in a separate company, the OEM's ability to

get accurate information in order to evaluate the costs and benefits is limited since it is dependent

on the ODM for financial data.

5.5 Product modularity

Swaminathan (2001) defines a modular product as one that can be made by appropriately

combining the different components or subassemblies (modules) that are used in the product and

where customers are interested in alternative options for the different modules. Thus, product

modularity is a function of product architecture. The purpose of product modularity is to

develop products within a short lead time, enable product variety and facilitate postponement.

Designing modular products requires intense and continuous collaboration between customers,

design engineers, manufacturing engineers, procurement and marketing. Bicycles are an

example of a product with a modular architecture (Venkatesh & Swaminathan 2002).

Manufacturers offer a wide variety of options of hubs, brakes, crank sets, seat posts, etc.

assembled on a frame.

One example of product modularity is the case of Sears/Whirlpool (Lee & Tang 1997)

where the company split the color-differentiating module in dishwashers into two - one frame

77
module common to all dishwashers and the other color sheet module which is inserted into the

frame. While the frame module remains the same, the color preference is made by the customer.

This enabled Whirlpool to hold generic product inventory at the warehouse and customize every

dishwasher to the customer's color preference.

Product modularity also offers the benefits of demand aggregation, risk pooling and

inventory control. It enables parallel manufacturing. Where before, the product would go

through a sequential manufacturing process, now, each module can be simultaneously

manufactured. This reduces the lead time and work-in-process inventory. In addition, it is a

valuable strategy to enable postponement and also mass customization. Mass customization is

the delivery of a wide variety of customized goods or services quickly and efficiently at low cost

(Simchi-Levi, Kaminsky & Simchi-Levi 2003). In order to implement mass customization,

products must be modular and able to be linked or assembled quickly with little additional cost

or overhead. Dell is a good case in point. The company has completely changed the customer's

PC buying experience by using product modularity and strategic supplier partnerships to

customize every PC sold to customer specifications. Dell begins to build the PC after receiving

customer orders through its website. Customers can choose all the different modular

components to create the PC that suits their requirements.

Product modularity enables renewal or upgrades of the product during its lifetime,

effectively extending it (Van Hoek 1998). This is because a modular architecture makes it easy

to add new features by replacing or upgrading modules without replacing the product itself. For

example, in a PC it was possible to replace a floppy drive with a CD drive, replace that with a

CD-RW drive and then replace that with a DVD/CD-RW drive as these products were

78
introduced in the market. Thus, at every stage of innovation in a module of the product, the

product itself does not become obsolete.

Product modularity creates many collaboration opportunities within the design chain.

Depending on product complexity, companies may choose to share design resources to develop a

product or simply outsource the design to suppliers who may then develop the product based on

their expertise and technology. Intense competition and the need to achieve excellence in a

specific technology have created much modularity in high-tech products. Companies are able to

successfully outsource manufacturing and design because the different modules have greater

strategic importance than the assembled product itself - the "Intel Inside" phenomenon.

5.5.1 Design outsourcing and Product Modularity

Product modularity is not a necessary condition for design outsourcing. However, most

products in the high-tech industry are modular and this makes outsourcing relatively easy.

Products with an integral architecture may have to be made modular in order to do postponement

and reduce lead times. Changing product architecture requires rethinking the product structure

and redesigning it. Since ODMs are engineering companies, they may have the sophistication

needed to develop products from scratch. Since modular products are easier to customize, they

may be able to sell them to more OEMs without significant redesigns. Therefore, if developing

the new product creates greater opportunities to increase their market share, ODMs are likely to

invest in product modularity.

Fully Outsourced Design: Outsourcing the design of products with an integral

architecture takes away control of the entire product, and with it, opportunities to create modules

and retain control over the design for those modules which have significant IP. Fine and

79
Whitney (1996) emphasize the risk inherent in outsourcing modular products where the OEM is

dependent on the service provider for knowledge and capacity. When an ODM designs an

integral product from scratch, it may have an incentive to make it modular in order to sell the

same design to more customers by customizing the product during assembly.

Joint Design: A modular architecture facilitates joint design. Here, the company can

outsource those modules of the design that are low-tech or non-core and retains those that are

complex or core. Thus, modules that are criticalsystems are designed in-house. If these are

sourced from suppliers, it is important to control the supplier integration process with close long-

term partnerships. Also, modules where future performance improvements can be achieved by

innovation should be retained in-house. A modular architecture facilitates the development of

subsystems with relative independence for both companies with maximum interaction occurring

during the initial design and final integration phases.

In summary, ODMs have the incentive to create change product architecture and creates

more modular products. However, OEMs should evaluate the product, engage in a joint design

model and retain control of the complex and core portions of the product in-house.

5.6 Postponement
The strategy of delaying the point at which the final customization of a product takes

place is known as postponement (Swaminathan & Lee 2003). This strategy enables companies

to better manage demand uncertainties for each individual product configuration. Postponement

allows demand aggregation for the many different versions of a single product. Since the

combined demand is easier to forecast than that of each individual configuration, greater

accuracy in planning procurement and manufacturing requirements can be achieved. Products

80
are inventoried in their standard form until actual demand for a specific configuration is known;

this builds flexibility into the system and allows risk pooling. Risk pooling is mitigating the risk

of excess demand for one variety with the risk of running short for another variety of product.

Postponement also eliminates the need to buffer inventory for many different products and thus,

reduces inventory carrying costs.

Dapiran (1992) describes the example of postponement at Benetton. The production

process involved dyeing the yam into different colors and then knitting the sweaters. Since the

knitting process was slow, it was necessary to predict demand of each color well before actual

trends materialized and to hold high levels of finished goods inventory to meet desired service

levels. Benetton resequenced the production process to do the knitting process first and then

hold inventory in undyed garments. Once the seasonal trends were realized, the sweaters would

go through the dyeing process. The company had to invest in dyeing technology in order to

ensure that the quality of its garments would be maintained.

Brown, Lee and Petrakian (2000) discuss postponement at Xilinx, a manufacturer of ICs

called logic devices. Xilinx is several stages upstream from end-customers and hence, demand

for its products is highly unpredictable. Simultaneously, there is a wide variety of ICs

differentiated by functionalities such as speed, number of logic gates, package type, pin count

and grade. Using proprietary design techniques, Xilinx enabled customers to do the final

configuration of the ICs using software, thereby eliminating the need to predict demand for

individual chips. The ICs can be programmed in many different combinations with this strategy.

Since customers do the final software configuration, this is a more extreme case of

postponement.

81
 Zinn and Bowersox (1998) describe the different types of postponement strategies that

are used by companies. These include labeling postponement, packaging postponement,

assembly postponement, manufacturing postponement and time postponement. Labeling

postponement is applying labels on a standard product differently based upon the actual demand,

for example, applying labels in different languages or for different brands right before shipping.

Packaging postponement is delaying packaging until final demand is known. For example,

products are localized with manuals in different languages, power supplies and accessories

before shipping to final customer.

When final assembly or some manufacturing activities are delayed until demand is

certain, the strategy is assembly or manufacturing postponement. The final processes may be

carried out either at the assembly plant, a warehouse or in some cases, by the customers

themselves. Early in the manufacturing process, the firm builds to forecast. The point at which

the firm begins to build to actual customer demand is known as the push-pull boundary. Brown,

Lee and Petrakian (2000) discuss process postponement, where the firm delays product

differentiation by moving the push-pull boundary toward the final customer. They describe the

strategy used by Xilinx which involves two stages. First, unfinished products called dies are

manufactured based on demand forecast and held in inventory. Based on actual orders from

customers, the dies are pulled from inventory and customized into finished ICs.

Manufacturing or assembly postponement may involve R&D expenditure in product and

process redesign, depending on the type of product and degree of postponement. For example,

when the product life cycle is short, product redesign can be incorporated fairly easily since the

changes can be implemented in the new generation product (Brown, Lee & Petrakian 2000). In

order to implement process postponement, Xilinx had to invest in inventory modeling expertise,

82
supply-mix prediction and cycle-time reduction in its wafer-fabrication facilities. Process

redesign involves resequencing production stages in order to move the customization stages

towards the end of the production process.

5.6.1 Design outsourcing and Postponement

According some industry executives, ODMs constantly revise and redesign their products

in order to stay ahead of their competitors. Thus, it would be easy for them to incorporate design

changes that would make products easier to postpone.

Capabilities: ODMs may be better at postponement than their EMS counterparts

because they control product design. Also, since they are in highly commoditized product

markets where demand is difficult to predict, most of their customers ask for postponement.

However, not all ODMs have the capability to do postponement which requires a strong global

manufacturing footprint. Many small ODMs are fairly localized within Asia and may not be able

to offer postponement.

Benefits: Postponement benefits are typically more favorable to OEMs. This is because

postponement primarily resolves the inability to forecast individual SKUs accurately. Demand

forecasting and finished product inventory is typically the responsibility of OEMs and hence

they have a vested interest in implementing postponement. Whether fully outsourced or joint

design, postponement meets with resistance from the ODMs.

Motivation: ODMs are reluctant to redesign products simply to facilitate postponement

since redesign increases cost and reduces margin. From a manufacturing perspective, making

the final product in Asia is the most cost-effective strategy because labor is cheap. In addition,

83
customizing the product closer to the end market increases the complexity of the supply chain

since ODMs would have to develop customization and final assembly infrastructure in the

different markets. The benefits of postponement disproportionately favor OEMs, whereas the

cost of implementing postponement becomes the responsibility of the ODM. Thus, ODMs have

very little motivation to do postponement. OEMs may be able to sell the idea of postponement

for a highly modular product with little customization where the ODM may be able to customize

the product for their many other customers.

Fully Outsourced Design: ODMs may have some benefits from designing products for

postponement in fully outsourced design. This is because they may be able to sell the products

to many customers with the least amount of inventory if they design the products such that any

customization only happens in the final stages of the process. However, if this is not a benefit,

then ODMs may not have the incentive to invest design resources into facilitating postponement.

Joint Design: In the joint development model, the OEM may be able to exercise control

over the products' ability to be postponed. This is because strategic product decisions are

controlled by the OEM. The OEM also has product knowledge and capabilities internally. In

case the redesign threatens any core IP, the OEM should do any postponement design internally.

Incentives: ODMs and EMS providers typically charge high rates for implementing

postponement. However, some offer postponement services at highly competitive rates. OEMs'

ability to negotiate the cost of doing postponement depends on their strategic importance to the

ODM. Larger companies with high volumes may be able to get postponement as a value-added

free service, whereas, others may have to share some of the gains of postponement.

84
 If ODMs are paid based on shipments they have an incentive to manufacture and ship

final products quickly in order to get paid faster. OEMs who have strategic partnerships with

their ODM partners may want to consider revenue-sharing contracts in order to align incentives.

In a revenue-sharing contract, the ODM would be paid a percentage of revenue generated based

on the final sale to the customer. This might help to align the incentives whereby both the

companies would be compensated based on the final sale. However, revenue-sharing contracts

have tradeoffs that must be considered before implementation in an ODM environment.

To summarize, postponement is one area of major incentive misalignment where the

costs are borne by the ODM and the benefits accrue to the OEM. Revenue-sharing contracts,

postponement "fees", cost-savings sharing are some ways OEMs may compensate the ODM.

Postponement for certain products may be beneficial to the ODM if they can build the

homogeneous product and customize it locally for their many customers.

5.7 Managing product variety

Robertson and Ulrich (1998) state that good product development involves developing a

family or "platform" of products, targeted at different customers, in a flexible production

process. The product family shares components, knowledge and production processes. Since

customer preferences are unpredictable, increasing product variety is an important marketing

strategy which allows companies to satisfy many different customers' needs. However, as the

number of products increase, the inventory needed in the supply chain increases and forecasting

the demand for each individual SKU becomes increasingly difficult. Thus, effectively managing

product variety involves evaluating the tradeoff between the benefits of implementing

component commonality, which is cost effective, and the benefits of differentiating products to

85
match customer preferences. They contend that the nature of this tradeoff can be a function of

product architecture with an example of a product architecture redesign involving the instrument

panel of a coupe and sedan. Here the engineering team was able to incorporate commonality in

many different subassemblies. One redesign involved the heating, ventilation and air-

conditioning (HVAC) system. Designing the duct system using a modular architecture permitted

reuse of many HVAC components. Also, with careful packaging, the team was able to reuse the

support structure for the entire instrument panel resulting in savings of $10.4 million in

development and tooling costs. By implementing commonality into the components that are

invisible to the customer, the company was able to ensure that the product was not homogenized.

Ulrich, Randall, Fisher and Reibstein (1999) argue that a variety strategy can be

evaluated based on the whether the product offers perceived value to the customer based on its

distinctiveness, whether the chosen product architecture minimizes cost for a given dimension of

variety and whether the firm possesses design and operations capabilities and resources to

support the dimensions of variety. Reducing the number of end-products in order to achieve

economies of scale using component commonality has its risks. The reduced product

differentiation can result in reduced sales because customers may find products better suited to

their specific needs with other competitors.

5.7.1 Design Outsourcing and Variety Management

Clearly, managing product variety effectively involves close coordination between

marketing and engineering teams. In an outsourced environment, this is very difficult to achieve.

The OEM's objective is to increase variety in order to target a wider market while an ODM's

concern is to incorporate features that other customers also want in order to allocate its design

86
cost among a wider customer base. Thus, if the NREs for additional redesign are high, they

might make the variety increase unviable. However, it is important to note that an ODM's cost

structure is lower than the OEM; therefore, the NREs in outsourced design are lower. Thus, in

an outsourced environment, more redesign can be done relative to in-house design.

Existing Design: When OEMs choose an existing ODM design and brand it, the

limitations on product variety are the greatest. This is true when products have an integral

architecture where increasing variety involves significant redesign. In products such as PCs, the

architecture is so modular that variety can be incorporated at the assembly stage. In this case,

outsourcing design does not have a negative impact on the OEM's ability to manage variety.

Thus, the ability to manage product variety increases with the increase in product modularity.

Design to Specifications: Here, incorporating both product variety and commonality

criteria into the design spec is challenging because the tradeoffs are relatively unknown that early

in the product development stage. This problem is more acute in the long-term as the OEM loses

internal product design expertise due to a long standing design outsourcing strategy. In this case,

the company does not have the internal expertise to understand product architecture and evaluate

commonality and variety considerations.

Non-Recurring Expenses (NREs): Any modifications to the base design to increase

product variety may incur non-recurring expenses (NREs). However, the ODM may be willing

to negotiate NREs depending again, on the size, importance and power of the OEM. The OEM

may be able to get the ODM to absorb or share NREs because the redesign might be beneficial to

its other customers. Ownership of IP for product redesign also impacts this decision. If the

ODM owns the IP, they can sell the product to other customers. If the OEM owns the IP, then

87
they might have to bear a higher proportion of the NREs. These expenses would also be incurred

in doing any redesign in-house; however, NREs in outsourced design are lower.

Joint Design: Here, the risk of transfer of knowledge or competitive differentiation

information exists. Since the ODM works closely with the OEM's design team, a lot of

knowledge is exchanged through formal and informal channels. Thus, it is very important to

negotiate and enforce strict confidentiality terms from the non-disclosure agreement (NDA).

Tradeoffs: Effectively managing product variety requires the ability to understand and

evaluate the many different trade-offs involved. This requires, in addition to strong analytical

capabilities, access to information spanning the two companies. This depends to a great extent

on the strategic partnership, level of trust and integration between the companies. OEMs have

experience with inter-departmental collaboration, information sharing and analysis. Marketing

collaboration and information sharing with external design teams to a similar extent is difficult.

Despite all the concerns about the ODM's unwillingness to customize their products and

processes for a specific OEM, when strategic partners make a request, ODMs do bend over

backwards in order to satisfy them. Burrows (2005) gives the example of an ODM

manufacturing Apple's notebook PCs. Due to Steve Jobs' requirement that no screws be visible

on the product, the ODM had to invent a tooling process in order to do the job. In such a case,

the concern is that unless Apple acquires the rights to the technology, it could be made available

to other customers reducing Apple's differentiation in the market.

88
5.8 End-of-Life and Obsolete Components
When designing new products in an environment where component commonality or

product modularity are used, it is important to design in new components for any components

that have either been discontinued by the manufacturer or become obsolete (OBS). In the high-

tech industry, this is an especially common occurrence since product and component life-cycles

are very short. OEMs that design new products on old platforms often find that newly designed

products have obsolete components or EOL parts. For these parts, usually manufacturers

announce last-time-buy (LTB) dates. This indicates the last date when customers may place

orders. Thus, OEMs have to forecast the demand for the entire life of their product and place an

order to cover that demand. Mistakes in forecasting this can be expensive. Also, the inventory

buildup due to EOL purchases can increase inventory carrying costs and incur related expenses.

While placing LTBs can be a costly proposition, on the other hand, redesigning products to

replace EOL parts can be expensive and time-consuming. Thus, a thorough evaluation of

components at design stage is required to eliminate any potential EOL or obsolete parts.

Processes should be outlined to identify these parts and either, find replacement parts or conduct

product redesign based on a thorough analysis of the tradeoffs involved.

5.8.1 Design Outsourcing and EOLIOBS Components

The problem of EOL/OBS parts exists in any design organization in the high-tech

industry, whether insourced or outsourced, because of the short product life-cycle for most of the

components. However, they can be a problem when outsourcing design because the ODM uses

reference designs to build the product. This vanilla design may or may not be revised at each

upgrade or redesign. Thus, there is a risk that new models of the product are unable to handle

89
the additional demands placed on them by the software and other components. Also, smaller

ODMs may not be as effective as their larger counterparts at developing the type of strong

supplier linkages that are required in order to proactively capture and manage EOL and OBS

parts.

Motivation: ODMs are unwilling to carry much inventory and therefore, making end-of-

life purchases and holding the inventory over the life of the product may be an unattractive

option for them. ODMs' cost of redesign is lower and frequency is higher than OEMs. Also, the

presence of such components in a product may affect their chances of selling their product to

new customers. All these reasons make them more likely to replace EOL or OBS parts with new

versions.

Infrastructure: PCNAlert.com of Pasadena, CA is a company that sends alerts about

component obsolescence, composition changes or EOL notifications to its subscribers. EMS,

ODMs and many OEMs can subscribe to receive these alerts. However, often these are difficult

to keep track of for several reasons. There are many different types of components that are used

in high-tech products and the company needs to store these alerts in a database with a lookup

function that can be used to reference the component change history. The company also

evaluates its customers' AVL for free and provides guidance about any obsolete components.

Some electronics component distributors such as Arrow Electronics and large EMS providers

also provide AVL reviews as a value-added service. For OEMs unwilling or unable to create an

inhouse infrastructure for these components, using the evaluation service may be a better option.

Risk Mitigation: Contracts that are designed to hold the ODM responsible for any parts

becoming obsolete or EOL within a designated fixed period after completion of design can help

90
protect the OEM from the costs resulting from such errors. This would mitigate some of the risk

associated with these parts for the OEM. In an outsourced design environment, where the ODM

constantly revises the product, an EOL or OBS part can be replaced at the next redesign. Often,

OEMs designing products in-house are slower to redesign existing products than are ODMs for

EOL or OBS parts.

In summary, outsourced design does not have a significant adverse impact, on EOL and

obsolete parts. The larger ODMs and EMS providers may have a better infrastructure,

capabilities and resources to capture and replace potential EOL or OBS parts. Also, the

frequency of redesign in an outsourced environment creates more opportunities to replace such

parts.

5.9 Economic packaging and transportation

The key cost drivers in a supply chain are usually transportation and inventory costs.

When high service levels are desired, high inventory levels must be maintained. Lee (1993)

identified economic packaging and transportation as a key design for logistics component. This

involves designing products so that they can be packed and stored in the most efficient manner

possible. Products that are light-weight will fill up the available space on a truck faster, causing

a truck to cube out. Heavy but compact items will quickly achieve the maximum weight of a

truck even when there is space available for more, causing a truck to weigh out. Thus, if

delivery trucks cube out rather than weigh out, then designing products to be packed more

compactly reduces the transportation cost (Simchi-Levi, Kaminsky & Simchi-Levi 2003). The

strategy of designing products keeping in mind such logistical issues as transportation and

warehousing is known as design for logistics (DFL).

91
 For example, Swedish furniture retailer Ikea has over 10,000 different products. Ikea

designed its products' packaging to be as compact and efficient as possible in easy-to-assemble-

at-home kits. Since these kits are easy and cheap to transport, products can be manufactured

efficiently in a small number of factories and shipped relatively inexpensively to stores

worldwide because of economies of scale (The Economist, 1994).

In addition to the above, efficient packaging enables products to be stacked easily in a

relatively small storage area. This in turn reduces inventory handling costs, space per product

and rent for the space (Simchi-Levi, Kaminsky & Simchi-Levi 2003). Keeping this in mind,

companies either design packaging efficiently for efficient storage, or in some cases, the

products themselves are designed for efficient storage. For example, large light-weight plastic

items such as storage bins, containers or patio furniture are designed so that they can be stacked

in order to reduce the storage space.

Another alternative is to ship goods in bulk and only complete final packaging at the

warehouse or at the retailer. This also makes transportation cost effective since bulk goods are

cheaper to transport. In India, packaging for certain goods such as rice, flour, etc. is delayed

until the products are sold to customers in retail stores. Customers decide the quantity they

would like to purchase by weight, upon which the retailer packages the goods and sells them.

Packaging design can also be done to facilitate other transportation strategies such as

cross-docking. Cross-docking involves moving goods from one truck to another set of trucks

going to different locations often without storing them in a warehouse. However, sometimes, it

may be necessary to break pallets and mix goods to create new pallets with different products. In

92
such cases, packaging design can enable quick and easy unpacking and repacking of pallets in

order to facilitate cross-docking.

5.9.1 Design outsourcing and Design for Logistics (DFL)

When design is outsourced to an ODM partner, DFL may not be a priority for them. This

is especially true when the ODM is not responsible for product distribution. Typically in an

outsourcing relationship, ODMs are judged based on packaging cost and damage to cargo.

These can both be met by using cheap packaging, load fewer boxes on a pallet and wrap the

product with a cheap protective covering in order to reduce damage. However, this increases the

number of pallets that the OEM needs to transport and thus, the transportation cost. The number

of boxes on a pallet can be used as a metric for high-volume products that are transported to a

single destination. However, where volumes are split up and transported separately to different

destinations, this is not a valid metric. Thus, packaging guidelines need to be provided in the

design specification. Since packaging can mostly be designed even later in the design stage,

products with fully outsourced design can have economic packaging based on the OEM's

experience.

Geographic issues in Packaging: Another challenge in packaging design in a global

supply chain is that the ODM may design packaging keeping in mind convenience in one

geographic area but this may not be the best solution for another. For example, at a leading high-

tech consumer products OEM, an ODM designed packaging such that the boxes could be quickly

hand-loaded on the pallet in China. However, once the products reached the US, the pallets were

not compatible with the automatic unloading machines used. Hand-loading of pallets was cost-

93
effective in China, but prohibitively expensive in the US. Since unloading had to be done

manually, this increased the cost for the OEM (Amaral & Cargille 2005).

Logistics Cost Ownership: Leading OEMs take full responsibility for freight costs

because of the perception that they can negotiate better prices because of the volumes they

transport. In order to make DFL a priority for ODMs, they should be given responsibility to

manage transportation. This would have two important benefits: the OEM would find out if its

rates are truly as competitive as perceived, and the ODM would make DFL a top priority because

transportation costs are generally high. One leading industry expert proposed that the OEM then

require that the ODM use its preferred freight provider and negotiate rebate agreements with the

freight provider. Here, the ODM would pay a standard quoted price to the transportation

company. Periodically, the transportation company would refund a portion of the money to the

OEM. The obvious risk with this strategy is that it takes the OEM closer to the one-stop-shop

outsourcing model where there is greater dependence on a single service provider and a lower

ability to exercise control because of more outsourcing.

In summary, DFL may not be a priority for ODMs. However, they may have some

knowledge of industry practices because of their work with their many customers. By designing

appropriate metrics and making them responsible for logistics, this can be made a priority for

them.

5.10 Product Recycling and Reuse

The auto and computer hardware industries are global leaders in consumption of basic

materials that become obsolete in less than a decade. While the auto industry has a robust

system for pushing products through recycling, the computer industry only recycles certain

94
components such as chips, power units, etc., while the others usually end up in landfills (Taylor

2002). Thus, recycling and reuse principles are an important consideration during product

design. Dillon (1994) calls the strategy of designing products to facilitate recycling and reuse

design for recycling (DFR). Component reuse and recycling and product remanufacturing can

have a positive and/or a negative impact on cost, product reliability and the environment. Thus,

the design decision to incorporate product or component reuse must consider tradeoffs among

cost, reliability and environmental impact (Mangun & Thurston 2002).

For example, HP has incorporated recycled material into certain new products and found

ways to increase the economic lifetime of materials. NewEnvironmentalism.org's company

spotlight on HP reported certain eco-minded design features of HP's products. HP uses light-

emitting diodes instead of a mercury lamp for scanners. In addition, HP also uses an innovative

foam chassis to reduce the number of components needed for certain products and to make them

easier to dismantle for recycling. HP eliminated nickel-cadmium batteries from all current PC

models since it is difficult to dispose of hazardous waste. In order to reduce the use of expanded

polystyrene in packaging, HP uses many different packaging solutions. For example, suspension

film and molded paper pulp trays are made with 100% consumer recycled content. In Asia, HP

has replaced traditional packaging with wheat-based foam pellets. In addition to the above, HP's

LaserJet toner cartridge and consumables recycling program provides consumers the means to

return empty cartridges to HP in order to dismantle and recycle components.

Another example is that of Cisco Systems' use of modules in order to facilitate system

upgrades without having to replace the existing system chassis and back-planes. Cisco's DFR

program evaluates cost-benefit tradeoffs, material diversity, ease of dismantling and end-of-life

95
and disposal concerns during the product design phase (Cisco Systems, Inc., Product

Stewardship Program).

In addition to consumer preferences for environmentally-friendly products, legislation in

certain European countries has made electronics recycling mandatory. In countries like the

Netherlands, Germany, Austria, Italy, Switzerland and France, legislation based on the "producer

responsibility" principle has been enacted in order to hold product design companies accountable

for facilitating recycling (Dillon 1994). However, as pointed out by Dillon in the same paper,

cost-effective recycling involves reconsidering the product design process. Product composition,

material variety, purity of recyclates and hazardous constituents of electronics products have to

be evaluated. Product design changes have to be made in order to reduce dismantling time and

effort, and to improve the reusability of components. Material types will have to be

standardized, parts have to be made easily detachable and the number of modules which require

dismantling will have to be reduced.

5.10.1 Design Outsourcing and Design for Recycling (DFR)

There are many economic benefits of reusing components since reuse increases the

economic life of components. In addition, in niche markets, consumers buy products from

OEMs who have a reputation of designing "green" products. ODMs may not have the same

knowledge or ability to incorporate such considerations into product design.

Capability: It is clear that many OEMs have the capability to design environmentally

friendly products. The reverse logistics capability chain in the electronics industry is unique to

each OEM. There is no consolidated industry-wide infrastructure to accumulate used equipment,

dismantle it and resell components. Having multiple OEMs with their own recycling facility

96
creates supply issues since one OEM may not want to supply components for a competitor's

products.

Fully Outsourced Model: When ODMs design a generic product to sell to multiple

ODMs, they may have no knowledge about which specific OEM will buy their products. Thus,

they may not have a strong incentive to design-in recycled parts. In addition to this information

gap, ODMs do not have the same pressures from their customers that OEMs do. For example,

HP's customers expect HP to design environmentally-friendly products. However, all the

ODMs' customers do not have the same expectations.

Joint Design: This model offers an opportunity to collaborate with the ODM in

evaluating the tradeoffs involved in designing products that can be recycled and reused. By

sharing knowledge and capabilities, products can be designed using DFR considerations. Since

industry standards are evolving partly on account of government regulations, joint design offers

opportunities to acquire new capabilities if these are absent from OEMs' design teams. By using

an ODM that has capabilities in the DFR area, an OEM can reduce the learning curve relative to

developing the capabilities in-house.

It is important to note that not all OEMs are as focused on product reuse and recycling as

some of the bigger companies in the consumer electronics segment. Thus, often, ODMs may

have more capabilities than their OEM customers to design recyclable products since they work

with a range of companies.

97
5.11 Product Obsolescence
The high-tech industry faces two different types of obsolescence; one is that which can be

attributed to new innovations and the other, which is planned. Computers become obsolete due

to the demands placed on them by newer software programs with their constant upgrades (Taylor

2002). Computers also become obsolete because of innovations in components that go into

them. For example, when Intel introduces a new processor, prices for computers with the old

processor drop. OEMs must quickly introduce new PCs with the latest processor in order to sell

to the early adopters. Due to the many different components and software that go into high-tech

equipment, the rate of product obsolescence is high and product development engineers and

product managers should plan to manage it during the design phase.

For example, an executive at a leading high-tech networking equipment OEM said they

always used the latest and state-of-the-art components in their products to ensure that the

products worked longer and remained usable through many software and hardware upgrades. He

discussed his experience with some customers who have been using the equipment for a decade.

Since it still performs to expectations, they don't need new equipment. The company always

replaces any underperforming components with the latest in order to ensure that the product lasts

longer. This is one reason why their products also cost more than other offerings in the market.

"Planned" obsolescence is originally attributed to the automotive industry. It is the

practice of designing products such that they would breakdown or experience mechanical

failures within a fixed number of years in order to ensure future sales. The automotive industry

faced criticism from the market in the 1970s because it was perceived that the industry practiced

planned obsolescence. In the high-tech industry, smaller startups are dependent on planned

98
obsolescence to compete with behemoths like Cisco. Many startup companies offer products

with better features and lower prices than their larger competitors. According to an industry

source, one way they can do this is by using inexpensive underperforming components that will

become obsolete in a few years. Once companies purchase the equipment from one provider,

they give preference to the incumbent in awarding future business. Thus, the strategy of planned

obsolescence has important pay-offs. Some companies consider planned obsolescence a

necessary strategy to compete and get long-term revenues.

5.11.1 Design outsourcing and Planned Obsolescence

In an outsourced design environment, product obsolescence has a different set of

consequences. This is because product redesign offers additional revenue opportunities to the

ODM. ODMs constantly redesign and upgrade their products, either due to requests from

customers or in order to incorporate new innovations. These constant redesigns make the older

versions obsolete. When OEMs outsource design, the ODM's frequent redesigns incorporating

new and better features would prompt the OEM to award new business. Thus, obsolescence has

revenue-generation consequences for ODMs.

Fully Outsourced Design: ODM products may become obsolete if they use a generic

design and customize it for different OEMs. The components that get designed into the generic

design may have subsequent, better performing versions available in the market which would

support the most current software and applications. If this reference design is used, the products

may become obsolete faster leading to future redesign revenues.

Joint Design: With joint design, OEMs do have a higher degree of control over this

process since they have knowledge of the overall product and the different modules that get

99
designed by the ODM. Here the OEMs may have greater opportunities to interact with

component suppliers and get information about forward and backward compatibility of current

and upcoming products. Joint design also opens up opportunities to integrate key suppliers into

the design process.

By specifying product features, performance requirements and future product evolution

and capabilities planned obsolescence can be managed. In summary, planned obsolescence is a

risk when OEMs buy an existing product design from the ODM without providing detailed

specifications. However, any possible incentive misalignment can be managed and controlled by

issuing clear product design specifications.

5.12Conclusion
The DFSC dynamics in a design outsourcing environment play out differently for the

various levels of engagement. Depending on the level of engagement, type of incentives and

allocation of benefits, the ODM may or may not be motivated to implement DFSC. The table

below outlines each strategy and the corresponding tradeoffs and considerations in outsourcing.

Table 1: Summary of DFSC strategies, Tradeoffs and Outsourcing

Supply Chain .

100
 Easier to manage in an ODM environment
because of frequent redesigns.

101
6 Case Studies and
Observations

In order to validate or invalidate the observations noted in the section on design for

supply chain (DFSC) strategies, it is essential to review the outsourcing process in context of a

specific product. Thus, this section includes a case study about how the outsourcing choice for

two different products impacts the ODM's opportunities for DFSC.

Section 6.1.1 outlines the process of selecting an outsourcing engagement model. For

each DFSC strategy, there is a summary of the impact of outsourcing, whether positive or

negative; section 6.1.2 is a framework for understanding the positive and negative effects of

outsourcing on different DFSC strategies. It suggests some measures that an OEM might take in

order to alleviate the negative impact of outsourcing. Lastly, section 6.1.3 summarizes the

discussion, outlines the observations and conclusions.

In order to do a well-rounded analysis of the different levels of outsourcing, it was

essential to take two different products serving two different markets. Accordingly, the two

products chosen here are a low-end consumer PC (Section 6.1) and an all-in-one printer (Section

6.2).

102
6.1 Case study 1: Low-end Consumer PC
The major players in the personal computer market are Dell, HP, Gateway, IBM, Sony,

etc. The PC market is characterized by intense competition. Cost and time-to-market are

important considerations. Time-to-market is typically associated with incremental innovation or

development in the components that are used in the PC. For example, when Intel develops a new

and faster processor, the PC manufacturers have to quickly revisit their existing design, develop

a PC with the newest processor and release it in the market. In addition to the processor, the PC

has many other components which follow different product life cycles and innovation

timeframes. The past five years have seen many technological advances in video graphics,

memory usage, media such as DVD/CD-RW drives, displays, etc. Each new innovation is not

always backward compatible with older hardware and software. Thus, the fear of product

obsolescence is an important factor in the PC adoption rate. In addition to cost and time-to-

market, PC consumers value product reliability, forward and backward compatibility, form

features and ease of use.

On the one hand, the modular architecture of the PC facilitates quick response to

innovations. Thus, new innovations can be incorporated into new designs quickly. Changes to

the assembly process are less time and cost intensive. On the other hand, the low-end PC is

consumed in households where people value reliability of their machines. Consumers want PCs

that have the latest products so that their machine lasts for a long time. Thus, the different

timelines and frequencies of each incremental innovation make frequent product development

expensive. This is because OEMs have to integrate their component suppliers in conducting

detailed life-cycle planning to ensure that current models will be compatible with future product

103
innovations. Product modularity in the PC has decreased the differentiation between product

offerings between different players. All the products have the same features and characteristics.

The industry has tried cost saving measures such as manufacturing products in low labor-

cost countries such as China and using many supply chain innovations to cut costs. As new

emerging countries such as China, India, Russia and Brazil are added to the PC market, the

downward pressure on costs increases. OEM margins on personal computers are already low,

and the wide-scale adoption of PCs in emerging economies demands even further reductions.

The low price points achieved by manufacturing outsourcing and offshoring have been replicated

by all the players in the PC industry. The next major cost component is design. As the product

is highly modular and standardized, there is less product differentiation between a PC offered by

Dell versus a PC offered by HP. Thus, in-house design does not carry an incremental advantage.

Design outsourcing offers many economies of scale that can further reduce the cost of the PC

and facilitate wider adoption.

6.1.1 Determining the Level of Engagement

In order to determine the level of design outsourcing the strategic value assessment

framework (Fine, Vardan, Pethick & El-Hout 2002) outlined in section 4.1.2 is applied. The key

elements of this framework are customer preferences, technology clockspeed, competitive

position, capable suppliers and product architecture. Customerpreferences about product

features are very high in each type of customer segment. The early adopters prefer the latest

technology while the late adopters prefer long-lasting features and product reliability (Brown,

Venkatesh 2003). Technology clockspeed in the PC industry itself is slow. This is because while

incremental innovations do happen for the different components, the PC itself has remained the

104
same over the past two decades. The competitive position of each OEM in the PC industry is

weak because design does not carry significant competitive advantage. Component suppliers

have developed strong technological capabilities. The fewer suppliers that exist for any

outsourced element, the more the leverage those suppliers have over the OEM (Fine 2002). In

the PC industry, all components have multiple suppliers. Even the historical power enjoyed by

Intel has been somewhat diluted with competition from AMD. Simultaneously, ODMs have

developed strong capabilities in PC design. Productarchitectureof the PC is highly modular

and the role of the PC design team is to integrate all the different modules into the most optimal

performing product.

High consumer preferences are offset by the high level of market competitiveness. This

is because even though consumers have specific preferences in terms of features and

performance, the differentiation between competitors is very low. Each company can put

together an offering with the same configuration. In addition, the slow technological clockspeed,

high supplier capability and modular product architecture indicate that the risk of design

outsourcing is low. The high level of price sensitivity and competition in the PC industry has

resulted in low margins for the OEMs. While either one of the fully outsourced design or joint

design models may be used, joint design for a PC may be too expensive because of the higher

cost structure of OEMs. Since margins for PCs are slim, joint design may not be a viable

strategy. Thus the economic and strategic value elements indicate that fully outsourced design

for the PC would be an appropriate strategy.

105
 Customer Importance High
Technology Clockspeed Low
Competitive Position Disadvantage
Capable Suppliers High
Product architecture Highly modular

Economic Value Added Low

Outsourcing Model Fully Outsourced Design

Based on a framework developed by: Fine, Pethick, Vardan & El-Hout (2002).

Figure 9: Summary of Strategic & Economic Elements for Personal Computer

Fully outsourced design has a hierarchy of engagement strategies that a company might

use. Existing design, form revisions, design to specifications and partial insource are different

ways that an OEM may engage in the PC design process. These are defined in section 4.2.1.

Due to the level of supplier involvement and coordination required for a PC, complete

dependence on an ODM for PC design is not a recommended strategy.

As observed in the strategic and economic assessment of the PC, customer preferences

are important even though the rest of the elements favor outsourcing. This implies that the OEM

needs some degree of control over the design process in order to ensure that marketing

intelligence about the product gets considered in design. Of the different engagement models

outlined in section 4.2.1, the two best suited to the PC are design to features and design to

specifications. These models provide some degree of control over selection of features and

performance characteristics of the product. A company looking simply for the best product in

terms of features might decide to go with the design for features model. However, when the

objective is to design a product with market preferred features and also incorporate supply chain

106
 considerations, the design to specifications model offers the best solution. Here, OEMs have the

ability to influence the design process while simultaneously getting the benefit of economies of

scale and faster time-to-market. Providing detailed design specifications ensures that DFSC

considerations are incorporated into the product.

6.1.2 DFSC Strategies for PCs in an Outsourced Design Environment

Having determined the outsourcing engagement strategy, it is important to consider the

DFSC strategies listed in Chapter 5 for the PC supply chain. Design outsourcing can have

positive and negative impacts on the OEMs' ability to implement these strategies. The following

table highlights the positive and negative impacts of design outsourcing on various DFSC

strategies.

Table 2: Tradeoffs in Fully Outsourced Design for a Low-end Consumer PC

I I Pros Cons Recommendations

- ODMs control sourcing.
They may request proprietary - Provide preferred vendor
- PC component pricing and supply terms. list for key components.
suppliers have a
- If the ODM only designs, - Conduct pricing audits,
history of business
material cost rather than benchmarking with other
with major OEMs &
DFSC is a priority. ODMs; keep pressure on
ODMs.
- Using multiple ODMs cost reductions.
divides up volumes.

- ODM may not share - Have product evolution

- Conduct three-way integration benefits or and roadmap discussions

integration with key knowledge. with key suppliers.
suppliers. - Allowing ODM to lead - When using multiple
supplier integration can result ODMs, design spec should
in planned obsolescence. list strategic vendors.

107
 Pros I Cons Recommendations
- If the ODM only designs,
- EMS & ODMs have there is no incentive to multi-
multi-sourcing ability - Provide multi-sourcing
source.
and motivation guidelines based on past
because it enables cost - Outsourcing makes experience with suppliers.
reductions. evaluation of tradeoffs
difficult.

- Control sourcing of key

- EMS and ODMs - ODM may oppose if
items via spec.
have the ability and material cost increases.
motivation to design - Make ODM responsible
- High NREs may reduce
products with common for inventory or reward
cost savings.
parts since they are inventory turns in contract.
material cost - Using multiple ODMs
- Demonstrate cost
conscious. builds up inventory.
savings on future products.

- Postponement for a - All ODMs do not have a

PC does not need global footprint. Mostly - With strategic partners,
redesign. manufacture in Asia. align incentives by
- ODM can customize - Unwilling because costs revenue sharing contract or
and complexity for ODM share cost savings.
base design and sell to
different OEMs. while benefits go to OEM.

- Redesign may incur NREs.

- Modular product - Outline differentiation in
design is easy to - Exchange of information to design specifications.
collapse or aggregate. evaluate tradeoffs is difficult.
- Provide an incentive to
- Benchmarking - ODM will generally want ODM to generate new
opportunity. to design features that other ideas.
customers also want.

- Make ODM responsible

- 0DM has incentive Typical contract does not
to do DFL if for freight. Negotiate
include packaging rebates with carrier.
responsible for specifications.
distribution. - Provide DFL
- May not be willing to requirements in spec.
-OEM caninfluence invest in or have expertise in
if packaging is custom - Provide cost targets and
DFL.
designed. use appropriate metrics.

108
 Pros Cons Recommendations
- Some ODMs may not have
DFR expertise. - Provide guidelines and
- ODM can be mandate DFR.
motivated to do DFR - No centralized facility to
since reuse would save source reused components. - Transfer DFR
material cost. knowledge to ODM to
- Unique DFR request may build their capabilities.
involve NREs.

- Faster the - Engage the ODM in

- Higher the rate of product evolution
obsolescence, the
obsolescence, greater the discussions.
smaller the timeframe
future revenue potential for
to resell existing - Provide compatibility
ODMs.
designs. guidelines in design spec.

- ODMs capable & - Base designs may not be - Hold ODMs responsible
better for finding modified Thus, parts used on for parts becoming
replacements because the original product may EOL/OBS within a
higher rate of redesign. become EOL/OBS. designated time horizon.

6.1.3 PC Outsourcing: Summary

The PC is a highly commoditized product that is a good outsourcing candidate.

Depending on their objectives, OEMs have two good engagement models - design to features

and design to specifications. If the objective is to design a product that has the required features,

etc. then design to features is a suitable strategy. If the objective is to optimize the supply chain,

then design to specifications provides the requisite influence into the design parameters.

OEMs can influence most of the supply chain considerations by providing guidelines in

the design spec. However, since two companies are involved, there are inherent information and

incentive misalignments that create artificial barriers to optimizing the chain as a whole. This

109
also makes the task of evaluating supply chain tradeoffs and allocating the costs and benefits of

the different strategies quite challenging.

For some strategies, such as supplier selection, single versus multi-sourcing,

commonality, packaging and DFR, guidelines can be provided in the design specifications.

However, OEMs must guard against the risk of over-specifying. Coker (2004) notes that the

ODM model loses some of its benefits as products are customized. In addition, certain DFSC

strategies provide a competitive edge to OEMs in which case sharing the knowledge with the

ODM is risky since it will most certainly reach their competitors.

Other strategies such as supplier integration, mass customization or variety reduction,

postponement, planned obsolescence and EOL/OBS occurrence depend on the degree of

interaction between functional groups within the OEM and the ODM. For instance, mass

customization and variety reduction requires collaboration between OEM marketing and ODM

design teams. EOL/OBS, planned obsolescence and supplier integration require three-way

communication and linkages between the OEM, ODM and component suppliers. The efficacy of

such interactions depends on the level of commitment of both partners and the importance of the

relationship to each. The importance of relationship management in the context of design

outsourcing is discussed in section 6.2.1.

As OEMs depend on design outsourcing, they experience gradual capability erosion

because of which, eventually, the ODM will simply make better products than the OEM can

envision or specify. The best strategy would then be to simply outsource all these considerations

to the ODM with minimal micromanagement. In this case, the OEM should align incentives to

ensure equitable distribution of the benefits of DFSC implementation.

110
 One major risk of design outsourcing as seen in section 4.2.1 is competition from

outsourcing partners. In the PC market, there is still about 50% market penetration (Brown,

Venkatesh 2003) within the US itself with major emerging economies such as China, India,

Brazil, Russia and many African countries still vastly untapped. Thus, there is tremendous

business potential still to be exploited. The pressure to reduce the cost of a PC will constantly

increase in order to make them attractive to consumers in emerging economies. Thus, the risk of

ODMs competing directly in these markets is extremely high since this would reduce an

intermediary in the PC supply chain. Thus it is imperative to choose ODM partners whose

business model is to not compete directly with their OEM customers.

The PC is a highly modular product with most of the components being fairly

standardized. An important observation about the PC is that almost all the innovation is

incremental i.e. new innovation improves existing features without redefining the product itself.

Thus, there is potential for disruptive innovation in this segment. Christensen (1997) discusses

sustaining or incremental innovation in the context of large existing corporations. Since most

large companies are extremely good at developing better versions of existing products, new

players can rarely defeat incumbents in this arena. Established companies often fail to see

disruptive technologies taking root in their smaller customers and miss the boat when these

innovations become widely accepted. Outsourcing the design of PCs to ODMs puts the onus of

incremental innovation on them.

6.2 Case Study 2: All-in-One Printer

The all-in-one (AIO) printer combines three to four different products - a printer

(document, photo or both), scanner and copier, and on some machines, fax - into one. The

111
advantages to the consumer are savings in terms of cost, space and training time. HP, Cannon,

Xerox, Epson, Dell, Lexmark, etc. are the major players in this market. The product is targeted

at home consumers and small & medium enterprises (SMEs).

The market is driven by product characteristics such as size, functions, ease of use,

software capabilities, standalone versus networking capabilities, print quality, cartridge price,

etc. While price is an important part of the consumer's purchasing decision, features and

functionality are more important. Devices can be high or low-end depending on the number of

features included. Each company differentiates its product based on features and performance in

the above mentioned areas. For example, some AIOs use a single cartridge for all types of

printing while others offer four different cartridges for different types of printing. The high-end

AIOs are capable of being standalone units while the low-end ones need to be connected to an

operational PC in order to be operational. Some units can be used as network printers while

others cannot. Ease of use and need for training are important as most of the users are novices.

While product features are extremely important, many of the capabilities have begun

converging. Most of the offerings with similar features compete on price. As AIOs gain wider

acceptance, price will become the more important determinant of market dominance. In order to

make these devices more mainstream, it is important to decrease the price further and make them

more affordable for regular home users. This analysis is for the low-end AIOs targeted at homes

and SMEs.

6.2.1 Determining the Level of Engagement

Once again the strategic value assessment framework (Fine, Vardan, Pethick & El-Hout

2002) outlined in section 4.1.2 is applied in order to determine the level of engagement. The

112
product is assessed for the key elements, customer importance, technology clockspeed,

competitive position, capable suppliers and product architecture. Customerpreferences for

product features are very high. Mainstream customers are not product experts and thus, ease of

use and quality are important considerations for them. The companies designing these products

compete on features rather than price. Technology clockspeed for the product is faster than in the

PC industry however, it is gradually slowing. The products have seen much functionality being

added since 2000. However, most of the offerings have only seen incremental improvements in

terms of quality and performance since then. The competitive position is different for different

companies. HP is the leader in the printer market, with Xerox and Cannon being other major

players. The Japanese printer company Lexmark is also a competitor and now has a partnership

with Dell, which has begun its foray into the printer market. Component suppliers are

developing technological capabilities. Most of the design capabilities are with OEMs. While

ODMs have developed capabilities for the different modules individually, they don't have the

product integration ability yet.. Productarchitectureis partly modular since the AlOs involve

integrating different modules such as printing, copying, faxing, etc. One of the important

requirements for the AIO is size. Therefore, some modules may have to be modified in order to

achieve the best fit into the enclosure.

While consumer preferences are high, in order to penetrate the low-end market further,

OEMs will have to achieve lower price-points. Thus, competition in this market segment will be

based on price which will require some degree of design outsourcing. However, since supplier

capabilities are not well developed and technological advancements are happening at a faster rate

relative to the PC, design cannot be fully outsourced. Also, the products are economically

113
valuable since an AIO offers higher margins than low-end printers. All the above indicate that

joint design would be a good strategy for the AIOs. This can be summarized below.

Customer Importance High

Technology Clockspeed Medium
Competitive Position Parity
Capable Suppliers Medium
Product architecture Partly Modular

Economic Value Added Medium

Outsourcing Model Joint Design

Based on a framework developed by: Fine, Pethick, Vardan & El-Hout (2002).

Figure 10: Summary of Strategic & Economic Elements for an AIO Printer

With joint design, OEMs can outsource design for the standardized modules and keep the

task of integrating them in-house. This would allow them full control over introducing new

features and improving their offering without significantly increasing the size of the design team,

and thus the cost. The economic and strategic value elements indicate that outsourcing design

for certain modules of the AIO would be a low-risk strategy while integration of the final unit in-

house would be more prudent. Since ODMs have the internal capabilities to design a printer

module, a fax module, a copier module, etc., the OEM does not need to use its own resources.

6.2.2 DFSC Strategies for AlOs in a Joint Design Environment

Having determined the outsourcing engagement strategy, it is important to consider the

impact of outsourcing on the company's ability to implement the different DFSC strategies listed

in section 5 for the AIO supply chain. The following table highlights the positive and negative

impacts of joint design on various DFSC strategies.

114
 Table 3: Tradeoffs in Fully Outsourced Design for an All-in-One Printer

Pros Cons Recommendations

- Ability to choose - Conduct price reviews
preferred suppliers. - Joint design dilutes the with strategic suppliers to
- More control over strategic importance of the ensure that the pricing
OEM to the supplier. provided by the ODM is
proprietary pricing and
supply terms. accurate.

- Knowledge transfer,
- Ability to ensure possibility of creating a - Outsource standardized
commitment to supplier future competitor by modules. Integrate key
integration. helping ODM build suppliers in-house.
integration capabilities.
I__ ____ -I
- Use ODM to qualify
- Both ODM and EMS - ODM will recommend multiple vendors.
providers have the ability alternate vendors who are
- Implement
to identify alternative their strategic partners.
components. recommendations to
leverage ODM relationships

- Redesign should be - ODM might want to

negotiate a non-exclusive - For strategic modules,
outsourced if no IP threat
to get low NREs. contract for its own work. control redesign even if
outsourcing is more
- ODM may offer deals - Knowledge transfer economical.
to acquire knowledge. through informal means.

- Modules may have to be

- ODMs already have
modified to enable - Make software control the
the skill to build the
interoperability. hardware forcing the ODM
different modules.
- ODM can acquire to depend on the OEM.
integration skills.

- Postponement benefits
- With strategic partners,
the OEM more than the
- Not too much variety do profit or revenue sharing
ODM.
as product is partly to align incentives.
integral - Product may need
- Share inventory reduction
redesign in order to enable
benefits.
postponement.

115
 Pros Cons Recommendations
- Not much variety in
- Control product
the same market - ODM's role is mostly
characteristics in-house.
segment. execution of OEM's
- Use software to improve
- Greater collaboration design requirements.
performance.
with marketing possible.

- Make ODM evaluate the BOM and AVL for EOL/OBS parts and find alternates.

- May not have expertise - Offer rewards for

- Cheaper to make ODM
in designing economic improving existing
do DFL to specifications.
packaging. packaging design.

- Possible to outsource - Provide specific DFR

- All ODMs may not have
portions of the module to guidelines and incentives in
DFR capabilities.
design for recycling. design spec.

- ODM may improve

- Easier to evaluate - Engage the design team
existing product by
ODM's design for in product evolution
learning from OEM and
obsolescence. discussions with marketing.
other customers.

6.2.3 AIO Outsourcing: Summary

The AIO is a new product that is being targeted at the home and small business markets.

The pressure to cut costs is high and the different product offerings in the market have begun

converging in terms of features and performance. The strategic and economic elements indicate

that the product is not yet commoditized but that there may be some gains to outsourcing certain

standardized portions of the design. A joint design model might have certain important benefits.

OEMs may choose to engage initially by retaining full control over the product

architecture and design, the BOM and AVL. Non-complex standardized modules may be

outsourced. Also, Table 3 above indicates that letting the ODM do some of the commonality,

116
DFR, DFL, multi-sourcing and EOL/OBS work may leave the in-house design team to focus on

the more important activities of the product such as architecture, features, performance,

integration and software.

In the joint design model, OEMs control all of the supply chain considerations. To some

extent, there may be incentive misalignment; however the OEM has most of the power in the

supply chain because it controls all the important tasks such as product architecture, supplier

selection, pricing, supplier integration, etc. The task of evaluating supply chain tradeoffs is

easier. Since the ODM is trying to acquire knowledge of the product and build design expertise,

it may be willing to provide certain value-added services such as alternate part recommendations,

evaluating the BOM for end-of-life and obsolete parts and getting alternate parts qualified.

These are activities that can be outsourced to the ODM in a joint design model.

In addition, certain DFSC strategies provide a competitive edge to OEMs in which case

sharing knowledge with the ODM is risky since it will most certainly reach their competitors.

Component commonality and variety management are two such areas. Increasing component

commonality may involve redesigning portions of the product design that the OEM may have

expertise in. In this case, the redesign must be kept in-house even if outsourcing it may save

costs. Also, as much as possible, software should be used to integrate the different modules of

the product to inter-operate and perform. Software can be licensed to the ODM even when the

product is commoditized and the ODM begins building better products than the OEM.

In the current scenario, AlOs are not offered in a wide variety. There are specific models

performing set functions for customers to choose from. Despite that, there should be close

117
collaboration between the marketing, design and manufacturing teams to make the product more

modular, increase its ability to be postponed and optimize component commonality.

One important caveat is that often joint design is the precursor to fully outsourced design.

Thus, a complete product review must be undertaken to determine the strategic segments of the

product. For example, the color cartridge requires engineering expertise to achieve the best

results. The mechanism used to move the cartridge to and fro must be precise and exact.

Software may be used to achieve interoperability of the different modules and to make the

product function. Thus, when joint design is done, strict IP infringement guidelines should be set

around these modules and software to ensure that even when the ODM develops its own product

based on knowledge acquired working with the OEM, it will need the OEM to supply the

cartridge, the engine and the software.

The main reason to do joint design is to optimize the use of engineering resources by

outsourcing the standardized and non-core design work and retaining complex and strategic work

in-house. However, this strategy may affect morale within the engineering team as design

engineers may perceive that they are training their replacements. This might motivate them to

attempt to keep as much of the work in-house as possible. In extreme cases, engineering teams

leave and find better opportunities. Thus, the objectives of joint design should be clearly

outlined and effectively communicated. A long-term design outsourcing strategy should be

formulated and articulated. A clear vision for the role of the existing design team for the future

should be shared in order to ensure that the objectives of joint design may be achieved.

118
6.3 Final Observations and Conclusion
As with any other outsourcing, design outsourcing has its benefits and risks. The

introduction of an external entity creates incentive and motivation issues in many projects that

would be simple to implement in-house. The thesis outlined these in detail in Chapters 4 and 5.

The case studies provided clarification by analyzing the results for two products. Section 6.3.1

below summarizes the observations and suggestions for DFSC strategy implementation. Section

6.3.2 then suggests some measures that OEMs may take in order to protect against some of the

risks of design outsourcing.

6.3.1 Observations and Suggestions for DFSC

The hypothesis for this thesis was that design outsourcing adversely affects companies'

ability to envision, evaluate and implement DFSC strategies. However, the main observation

was that this is not true for all DFSC strategies. Design outsourcing is beneficial for some DFSC

strategies and has an adverse impact on others. The level of outsourcing chosen, the distribution

of benefits and the ability to evaluate tradeoffs determine the OEM's ability to implement DFSC.

For some strategies, the economies of scale in outsourcing make many unviable redesign

projects viable and actually facilitate supply chain thinking. This is true when the service

provider has multiple customers for the same product. The service provider actually becomes

better than the OEM at improving the product and incorporating certain supply chain

considerations because of the following reasons: first, the ODM's redesign frequency is much

higher than that of an OEM; second, multiple OEM customers request DFSC measures; third,

many of the DFSC strategies are beneficial to the ODM.

119
6.3.1.1.1 Not adversely impacted by Outsourcing:
ODMs will be willing to implement those strategies that provide benefits to them. These

include multi-sourcing, component commonality, product modularity and reduction of EOL/OBS

parts. Multi-sourcing and component commonality have cost-saving advantages while product

modularity have revenue-generating possibilities for the ODM. EOL/OBS components occur in

any design effort, whether in-house or outsourced. Thus, the OEM would not have to provide

additional incentives to implement these. In outsourced design, the ODM has the capability to

identify and replace such parts; also, the ODM is unwilling to do any last time buys. Frequent

redesigns by the ODM create opportunities to identify and implement replacements. ODM's

redesign cost is lower than OEM.

6.3.1.1.2 Adversely impacted by Outsourcing

Strategies that are adversely impacted by fully outsourced design are supplier integration,

product variety management, product obsolescence and postponement. Joint design does risk

loss of control in these important areas however the impact is less severe because the OEM still

controls the overall product architecture. Supplier integration leads to innovation and new

product development. Outsourcing severs these links. Product variety management is relatively

easy for a modular product, however, for an integral product, the ODM will only want to

implement those differentiating features which other customers also desire, thus defeating the

purpose. The NREs charged by the ODM to increase variety may make the project unviable. A

faster rate of product obsolescence generates revenue for the ODM; thus they may have the

motivation to plan some level of obsolescence.

120
6.3.1.1.3 Incentive Misalignment
Postponement is a classic case of incentive misalignment due to outsourcing. The costs

of designing products for postponement and then implementing it lie with the ODM whereas the

benefits due to finished product reduction and mass customization favor the OEM. Thus, the

OEM may have to provide incentives in the form of postponement fees, revenue-sharing or

sharing of cost savings. The type of incentive depends on the nature of the partnership; revenue-

sharing necessitates some degree of information sharing which requires a strategic partnership.

For highly commoditized products, an OEM may outsource to multiple ODMs in which case

postponement fees may be more appropriate.

6.3.1.1.4 May/May not be Adversely Impacted

Design for logistics (DFL) and design for recycling (DFR) may or may not be casualties

in outsourcing. DFL may be a cost-saving opportunity for an ODM who manages logistics, but

not when the OEM is responsible for logistics expenses. When DFL and DFR are requested by

many customers, the ODM may have the skills and motivation to incorporate these

considerations in all its designs. The OEM may have to incur redesign cost however, this cost is

lower than in in-house design.

Supplier selection is not significantly impacted by outsourcing. The risk of not getting

OEM-preferred vendors is offset somewhat by the ODM's own negotiating power with its

strategic suppliers. The ODM may negotiate proprietary pricing, supply terms, etc. however, the

OEM may already be at lower price points because of the economies of scale achieved through

outsourcing.

121
6.3.1.2 Observation 2:
The level of outsourcing, tradeoffs, incentives and negotiation power impact the

implementation of DFSC strategies.

6.3.1.2.1 Fully Outsourced Design & DFSC

The impact of the levels of design outsourcing on the various DFSC strategies depends

on the reason for outsourcing, type of product and market, ability to exchange information and

evaluate tradeoffs, negotiating power and strategic importance and the ability to trust and partner

with the service provider. The OEM may choose fully outsourced design for products that

complement its core product portfolio and where it is dependent for knowledge. Since ODMs

already incorporate certain DFSC considerations in highly commoditized products, any

additional requirements may be outlined in the design specification. To that extent, design to

specifications facilitates the communication of the OEM's priorities and helps to set expectations

about DFSC requirements. In the fully outsourced model, design to specifications is the better

model. When dependent on knowledge however, the OEM may not have the ability to provide

detailed DFSC requirements. Also, a long-term design outsourcing strategy results in erosion of

OEM capabilities and expertise. With this, the ability to provide detailed product specifications

also decreases.

6.3.1.2.2 JointDesign & DFSC

Joint design is a better strategy for an OEM that is independentfor knowledge and

looking to implement DFSC; it gives the OEM the ability to control the design process, set

priorities and protect its own interests. In joint design, implementation of component

commonality, multi-sourcing, DFL, DFR and EOUOBS should be outsourced to the ODM to the

122
extent that it doesn't threaten IP. The caveat is that joint design does threaten knowledge

transfer through formal and informal channels. Also, due to increasing cost pressures on highly

commoditized products, joint design may not be a viable proposition unless the OEM is able to

differentiate its offering in the market. For example, Apple Computers uses joint design for its

products. In order to protect its competitive edge, Apple outsources only the most mundane and

standardized of its design activities and is intensely protective of its product development

process. It differentiates its products by catering to a niche market, emphasizing the security and

reliability of its software and by positioning itself as a higher-end product with aesthetic appeal.

(Burrows 2005).

6.3.1.2.3 Tradeoffs in DFSC

Many large OEMs in the high-tech consumer goods segment are recognized as leaders in

supply chain innovation. Most supply chain decisions involve computing and evaluating the

inherent tradeoffs. Implementing DFSC in an outsourced environment makes the evaluation of

tradeoffs difficult because of the introduction of an external entity with its own motivations,

deliverables and priorities. In addition, the OEM is dependent on such an entity for information

critical to the analysis. Depending on the type of relationship between the two companies, the

ODM may or may not be willing to share its true costs with the OEM. OEMs may use audits,

benchmarking or product teardowns in order to evaluate the cost. But these are resource-

intensive non-value added activities. Another alternative is to create cost models to arrive at the

cost of design.

123
6.3.1.2.4 Distributionof benefits of DFSC
The distribution of benefits DFSC strategies creates incentive misalignments because

their costs and benefits may be disproportionately allocated to each player; this impacts the

ODM's motivation to implement these strategies, especially when the benefits disproportionately

favor the OEM. Thus, time-consuming negotiations have to take place in order to equitably

distribute the benefits of DFSC strategies. This may be in the form of cost, profit or revenue-

sharing contracts or NREs. If the ODM wants to recover the full cost of redesign, then the OEM

may ask to own any IP generated so that the ODM does not use the same technology to improve

competitor products. If a portion of the costs are charged, then some kind of restrictions may be

negotiated in order to get exclusive rights to the technology for a set period of time.

It is important to remember that most DFSC strategies help ODMs to save costs, improve

margins, augment their customer base and increase revenue. Despite that, it is in the ODM's

interest to recover as much of its design cost from customers in order to improve its own

financial metrics and offer low prices to customers. Therefore, when negotiating NREs, risk-

sharing or other costs with ODMs, the benefits to the ODM should be evaluated so that the OEM

does not subsidize its own competition. Analytical cost models can help shed light on the

ODM's gains through DFSC and should be used as a tool during the negotiation process.

6.3.1.2.5 Power and Negotiation

The OEM's power determines the outcome of the negotiation about distribution of costs

and benefits of DFSC. Power in a negotiation can come from many sources, some of which are-

> OEM's business as a percentage of the ODM's total revenue.

> Knowledge that the ODM would like to acquire.

124
 Volume of business either currently insourced or outsourced to a different provider that the
ODM desires.

> ODM's dependence on the OEM for key portions of the product.

> OEM's size and reputation in the industry.

> Degree of knowledge independence.

The above considerations affect how much the ODM may be willing to give up in the

negotiation. For instance, in order to acquire knowledge about certain technology, the ODM

may charge lower NREs. Smaller OEMs might be charged higher margins or fees than their

larger counterparts. The OEM's product portfolio might be of interest to the ODM for future

development. Some OEMs may even get deals just so the ODM may claim them as their

customer.

6.3.2 Business Observations and Suggestions

Design outsourcing is a risky strategy because it makes an OEM lose control over its

products' value proposition, making them homogenous and ultimately driving them out of the

supply chain. The business model of ODMs depends on multiple customers for the same product

and does not preclude them from becoming future competitors. This is especially the case as

products become commoditized and margins get thinner. The motivation to share the razor-thin

margins with the OEM becomes even lower. The opportunities to partner with downstream

suppliers like retailers or in case of mobile phones, wireless telephony service operators become

more lucrative. As seen in section 4.2.1, the threat of competition from ODMs is real.

Some strategies that companies may be able to use in order to retain their dominance at

least temporarily in the supply chain are as below:

125
1. Evaluate the product in order to determine the most complex portions where the OEM owns

the IP or standard or where future innovations may occur. By retaining the design for these

parts of the product in-house, they may be able to ensure that the ODM is dependent on them

in future.

2. Migrate as much of the functionality and performance of the product into the software

instead of hardware. Since ODMs don't yet have the ability to integrate software and

hardware, this may extend the temporary advantage. Burrows (2005) gives the example of

Apple Computers, which creates its own software - Mac Operating System, iPhoto, iTunes,

etc. This gives Apple an edge over competitors such as HP and Dell. This is because even if

Apple's hardware becomes commoditized, its software will need to be licensed and will

remain its strong IP. Cisco Systems also uses software to a large extent to retain its power.

3. Redefine the business proposition to identify areas of the business where they may have more

strengths. For example, IBM sold its PC division to Lenovo and now has a bigger focus on

its consulting business.

4. ODMs are focused on incremental innovation. Using their design resources, ODMs take

existing products and make them better by improving the different aspects of the same

product in order to attract newer customers. OEMs need to use the opportunity to develop

new innovations and technologies in-house. They should scout the market, investigate ideas

and drive innovations by transferring the cost savings from design outsourcing to R&D

focused on new innovations.

126
6.3.3 Being a good partner
I'm very loyal in a relationship. Any relationship. When I go out with my mom, I
don't look at other moms. I don't go, "Ooh, I wonder what her macaroniand
cheese tastes like."

- Gary Shandling

Ultimately, "supply chains are about people" (Byrnes 2005). All OEMs talk about

building trust and partnerships with their outsourcing service providers. On a day-to-day basis,

however, the relationship is mostly adversarial. The level of trust between most of these

"partners" is very low. Outsourcing, unlike traditional buyer-supplier relationships is more

intimate. The fates of the two companies are closely entwined in an outsourcing relationship and

thus, each entity needs to build trust and strong relationships with the other. Often, words such

as partnership and collaboration are thrown around whereas the spirit of partnership stays in the

upper echelons of management without really percolating down to the front-lines. Many actions

can have potentially damaging impacts on the relationship. Constant bashing of the ODM about

cost reductions, blatant disregard for ODM priorities, frequently changing business partners and

distributive negotiating to split the pie - all these actions by OEMs create ill-will. In addition,

many companies feel that keeping a service provider concerned about the future of the

relationship is the best way to manage it. Constant and subtle threats of canceling the agreement

may be effective over the short-term, however, will eventually result in failure (Lynch 2000).

In order to truly be good partners, it is essential to build long-term strategic partnerships

with carefully handpicked outsourcing service providers and seeing them through their rough

times. For instance, Boeing works with its equipment suppliers during periods of slow growth.

The company gives low-priority improvement projects to them to provide them a low but steady

127
stream of revenue (Fine 1998). This is partnership in the true sense of the term and it ensures

that suppliers will always give Boeing disproportionate mindshare.

Trust is an important ingredient in successful outsourcing relationship management.

Linder (2004) says that trust is not built directly; it grows as a result of experiences.

Conventional tactics such as accountability, service levels and disciplined change control to

manage relationships, though important, fall short of true relationship management needs.

Linder proposes contract negotiation, performance track record, strategic governance and

personal relationships as the four pillars for intentional relationship management. Contract

negotiations are an opportunity to set the right priorities and have the right conversations.

Recognizing that the service provider deserves to make a fair margin can go a long way in

determining how the OEM is treated by the service provider. Strategic governance begins at the

upper levels of management and pervades throughout the organization. Establishing formal and

informal communication processes, enabling visibility through open-book policies go a long way

in building trust and confidence. Setting periodic reviews to go over milestones and

achievements are a way to document performance and remind both companies of their

interdependence. Finally, personalrelationships developed by working side-by-side and helping

each other out at each business level go a long way in establishing true and lasting partnerships.

6.3.4 Areas of Future Research

This thesis identified the motivations, strategic considerations in outsourcing design, the

different ways in which OEMs engage with design service providers. It introduced the different

DFSC strategies with examples of companies using them and evaluated the impact of design

outsourcing on the ability of the OEM to envision and implement them in their products. It

128
brought out the tradeoffs and how the two companies might view and evaluate them differently.

Since design outsourcing is a relatively new practice in the high-tech industry, much research

still needs to be done from the perspective of both, industry dynamics and OEM dynamics

perspectives.

6.3.4.1 Industry Dynamics

Further research needs to be done to understand the role of OEMs in the long run when

almost all traditionally core activities are outsourced. As EMS providers go to an ODM model,

they are taking on some products that are higher-end in comparison to PCs or cell phones. Thus,

they may encroach on the turf of some OEMs who may soon see their products be increasingly

commoditized.

The movement of EMS providers to an ODM model consolidates a lot of bargaining

power in the hands of a single entity. Further research into studying how this changes value-

chain dynamics and impacts OEMs' ability to protect their own interests is required.

Christensen (1997) wrote about how large companies focus on incremental innovation

and hence miss the boat on disruptive innovation. Since design outsourcing shifts the onus of

incremental innovation on ODMs, there may be some opportunity for research into what OEMs

can do to overcome barriers in finding the next disruptive technology.

6.3.4.2 OEM Dynamics

Those DFSC activities that are adversely impacted because of outsourcing are supplier

integration, product variety management, product obsolescence and postponement. While a lot

of literature exists on each of these strategies, some of the results may not be applicable when an

129
OEM no longer controls product design. Thus, further research can help address the problems

and provide recommendations about how best to retain control over these activities in a design

outsourcing scenario.

Incentive misalignment is a major issue in outsourcing. Design outsourcing creates one

more level at which incentives may be misaligned. Future research can help identify best

practices in the industry in aligning incentives. Specifically, research can help in identifying

incentive misalignments in one-stop-shop outsourcing where all functions are controlled by a

single service provider.

130
Bibliography
Amaral, J. (2005). Managing Director, Emeraldwise. LLC. Interview on March 22, 2005.

Amaral, J., Billington, C. & Tsay, A. (Nov 2004). Outsourcing Production without losing
control. Supply Chain Management Review. 44-52.

Barnes, E., Dai, J., Deng, S., Down, D., Goh, M., Lau, H. & Sharafali, M. (2000). Electronics
Manufacturing Service Industry. TLI-AP Report, Georgia Institute of Technology.

Brown, A., Lee, H. & Petrakian, R. (2000). Xilinx improves its Semiconductor Supply Chain
using Product and Process Postponement. Interfaces. 30:4. 65-80.

Brown, S. & Venkatesh, V. (2003). Bringing Non-Adopters Along: The Challenge Facing the PC
Industry. Communications of the A CM 46:4. 76-80.

Burrows, P. (March 2005). Commentary: Apple's Blueprint for Genius. Business Week Online.
(Retrieved April 20, 2005 from
http://www.businessweek.com/magazine/content/05 12/b3925608.htm)

Byrnes, J. (2005). In-class lecture in ESD.261. Case Studies in Logistics and Supply Chain
Management. Massachusetts Institute of Technology.

Carbone, J. (2003). ODMs offer Design Expertise, Quicker Time-to-Market. Purchasing.com.

Reed Business Information. (Retrieved March 15, 2005 from
http://www.purchasing.com/article/CA269147.html)

Carbone, J. (2005). Worldwide Outsourcing Rises. Purchasing.com,Reed Business Information,

(Retrieved October 29, 2004 from http://www.purchasing.com/article/CA501253.html).

Cargille, B., Bliss, R. (2002). How Supply Chain Analysis Enhances Product Design. Supply
Chain Management Review. 5:5. 64-74.

Cargille, B. (2004). Interviews on November 11, 2004 and November 22, 2004.

Cargille, B, Amaral, J. (2005) Interviews on March 22, 2005 and April 7, 2005.

Chuang, W. & Yang, H. (2004). Design Chain Collaboration: A Strategic View. International
JournalofElectronic Business Management. 2:2. 117-121.

Christensen, C. (1997). The Innovators' Dilemma: When New Technologies Cause Great Firms
to Fail. Cambridge, MA. Harvard Business School Press.

Cisco Systems, Inc. Product Stewardship Program. (Retrieved April 4, 2005 from
http://www.cisco.com/en/US/about/ac227/ac228/ac23 1/about cisco product design.html).

131
Collins, R & Bechler, K. (1999), Outsourcing in the chemical and automotive industries: Choice
or competitive imperative? Journalof Supply Chain Management. 35:4; 4-12.

Coker, B. (2004). The ODM threat to EMS. Circuits Assembly. (Retrieved January 20, 2005
from www.circuitsassembly.com).

Costlow, T. (Sep 2002). Outsourcing takes hold in design. Design News. Boston. 57:17, S8-S 11.

Dapiran, P. (1992) Benetton - Global Logistics in Action. InternationalJournalofPhysical

Distribution & Logistics Management. Bradford.22:6, 7-12.

DeNeufville, R. & Pirnar, A. (1999). A dynamic technology strategy for Xerox to respond to the
threat of high-tech commoditization. InternationalJournalof Technology Management.
Geneva. 17:1,2; 73-93.

Deavers, K. (1997), Outsourcing: A corporate competitive strategy, not a search for low wages.
JournalofLabor Research, 18:4, 503-519.

Dillon, P. (1994). Salvageability by Design. IEEE Spectrum. New York. 31:8. 18-21.

Engardio, P. & Einhorn, B. (March 2005). Outsourcing Innovation. Business Week. Special
Report. 84-94.

Erhart, J. (2004). Director Worldwide Channel Sales, Broadcom Corporation. Interview

December 20, 2004.

Feitzinger, E. & Lee, H. (1997). Mass customization at Hewlett-Packard: the power of

postponement. HarvardBusiness Review, 75:1, 116-121.

Fine, C. (1998). Clockspeed: Winning industry control in the age of temporary advantage.
Cambridge, MA: Perseus Books.

Fine, C., Vardan, R., Pethick, R. & El-Hout, J. (2002). Rapid-Response capability in Value-chain
design. MIT Sloan Management Review. 43:2. 69-75.

Fine, C. & Whitney, D. (1996). Is the Make-Buy Decision Process a Core-Competence? Working
Paper,MassachusettsInstitute of Technology, Cambridge, MA

Grant, P. (2000). Outsourced Knowledge: Knowledge transfer and strategic implications of

design outsourcing. MassachusettsInstitute of Technology. (Retrieved March 30, 3005 from
http://hdl.handle.net/1721.1/9278)

Hadaya, P, Lefebvre, E. & Leger, P. (2000). Roles and strategies of contract manufacturers in the
telecommunications industry. IEEE EMS InternationalEngineeringManagement
Conference. 458-463.

Huckman, R. & Pisano, G. (2004). Flextronics International Limited. HarvardBusiness School

Case Study #9-604-063.

132
Invest Japan. Corporate R&D Strategies in the Age of Advanced Communications. Invest
Japan,Special Feature3, 8-12. (Retrieved March 24, 2005 from
http://www.usajapan.org/PDF/corporate rd strategies.pdf)

Lakenan, B., Boyd, D. & Frey, E. (2001). Why Cisco fell: Outsourcing and its Perils? Strategy
& Business, No. 24, 54-65

Langberg, M. (July 30, 2004). Three Little Initials are Changing Electronics. Mercury News
(Retrieved April 3, 2005 from http://www.siliconvalley.com/mld/siliconvalley/927 9 9 7 1.htm)

Laseter, T. & Ramdas, K. (2002). Product Types and Supplier Roles in Product Development:
An Exploratory Analysis. IEEE Transactionson EngineeringManagement. 49:2. 107-118.

Lee, H. (1993). Designfor Supply Chain Management: Concepts and Examples, Perspectivesin
Operations Management: Essays in Honor ofElwood S. Buffa (ed. R. Sarin). Boston, MA.
Kluwer Academic Publishers.

Lee, H., Tang, C. (1997). Modeling the costs and benefits of delayed product differentiation.
ManagementScience. 43:1, 40-53.

Linder, J. (2004). OutsourcingforRadical Change. New York, NY. American Management

Association.

Lyons, T., Krachenberg, A. & Henke, J. (1990). Mixed Motive Marriages: What's Next for
Buyer-Supplier Relations. Sloan Management Review. 31, 3.

Lynch, C. (2000). Logistics Outsourcing- A Management Guide. Oak Brook, IL. Council of
Logistics Management.

Mangun, D. & Thurston, D. (2002). Incorporating Component Reuse, Remanufacture and

Recycle into Product Portfolio Design. IEEE Transactionson EngineeringManagement.
49:4. 479-490.

Mortimer, J. (2001). Suppliers' Extension or Contingent Business Interruption Insurance.

(Retrieved March 5, 2005 from http://www.converium.com/2037.asp)

Narayanan, V, Raman A, (2004). Aligning Incentives in Supply Chains. HarvardBusiness

Review. Boston. 82:11, 94.

New Environmentalism.org. Industry Ecology - Company Spotlight, Hewlett-Packard.

(Retrieved April 5, 2005 from
http://www.newenvironmentalism.org/ecology profile.cfm?ID=8)

Normille, D. (2004). These Slim Margins are not by Design. ElectronicBusiness Online. Reed
Electronics Group.

Novak, S. & Eppinger, S. (2001). Sourcing by Design: Product Complexity and the Supply
Chain. Management Science. 47:1. 189-204.

133
Petersen, K., Handfield, R. & Ragatz, G. (2003). A model of supplier integration into new
product development. Journalof ProductInnovation Management. 20; 284-299.

Quinn, J. (2000). Outsourcing Innovation: The New Engine of Growth. Sloan Management
Review. 41:4. 13-28

Ragatz, G., Handfield, R. & Scannell, T. (1996). Success factors for integrating supplier into new
product development. JournalofProduct Innovation Management. 14:3, 190-202.

Robertson, D. & Ulrich, K. (1998). Platform Product Development. MIT Sloan Management
Review. 39:4, 19-31.

Rothstein, A. (March 1998). Outsourcing: An Accelerating Global Trend in Engineering. ASME

EngineeringManagement Journal.10. 11.

Sanchez, R. (1996). Strategic product creation, managing new interactions of technology,

markets and organizations. European ManagementJournal, 14:2, 121-138.

Schweber, B. (April 2003). Making it under contract. EDN, Reed Electronics Group. 67-73.
(Retrieved January 20, 2005 from (http://www.edn.com/contents/images/289985.pdf)

Simchi-Levi, D., Kaminsky, P. & Simchi-Levi, E. (2003). Designing and managing the supply
chain: Concepts, strategiesand case studies. New York, NY. McGraw-Hill/Irwin.

Solectron Corporation. (Retrieved January 20, 2005 from

www.solectron.com/about/glossary.html)

Sturgeon. (2002). Exploring the Risks of Value Chain Modularity: Electronics Outsourcing
During the Industry Cycle of 1992-2002. IndustrialPerformance Center, MIT.

Swaminathan, J. (2001). Enabling Customization using Standardized Operations. California

ManagementReview. 43:3. 125-135.

Swaminathan, J., Lee, H. (2003) Design for Postponement. OR/MS Handbook on Supply Chain
Management: Design, Coordinationand Operations. Elvesier Publishers. 199-228

Tae-Gyu, (Nov 2004). HP accused of stealing computer design. The Korea Times. (Retrieved
November 5,2004 from
http://times.hankooki.com/lpage/200411/kt2004110318222710230.htm)

Taylor, B. (2002). Planned obsolescence, The Tech Way. Recycling Today, Editor'sFocus -
Computer industry recycling and waste management - Editorial. (Retrieved March 22, 2005
from http://www.findarticles.com/p/articles/mi mOKWH/is 9 40/ai 92724907)

The Economist. (1994). Furnishing the world. The Economist: (Nov 1994) 19, 101-102.

The Economist. (2000). Have factory, will travel. The Economist, London: Feb, 2000. 354:8157,
61-63.

134
The Economist (2004). Battling for the Palm of your hand. The Economist, London: May 2004.
371:8373. 79.

The Economist. (2005). Moving On. The Economist, London: Jan 2005. 374:8409, 9-11.

Ulrich, K. (1995). The Role of Product Architecture in the Manufacturing Firm. Research
Policy. 24. 419-440.

Ulrich, K. & Ellison, D. (2004). Beyond Make-Buy: Internalization and Integration of Design
and Production. Working Paper, The Wharton School, Department of Operations and
Information Management, 1998, 2004. (To appear in Production and Operations
Management 2005). (Retrieved April 1, 2005 from
http://opim.wharton.upenn.edu/-ulrich/publications.html)

Ulrich, K., Randall, T., Fisher, M. & Reibstein, D. (1999). Managing Product Variety: A Study
of the Bicycle Industry. Managing Product Variety. Teck Ho and Chris Tang (editors).,
Kluwer Academic Publishers.

Vande Vate, J. & Ward, A. (2005). The Ins and Outs of Outsourcing. FrontlineSolutions
(Retrieved March 18, 2005 from
http://www.frontlinetoday.com/frontline/article/articleDetail.isp?id=145742)

Van Hoek, R. (1998). Logistics and Virtual Integration, Postponement, Outsourcing and the
Flow of Information. InternationalJournalof Physical Distributionand Logistics
Management. Bradford. 28:7. 508.

Venkatesh, S. & Swaminathan, J. (2002). Managing Product Variety through Postponement:

Concept and Applications. The Practiceof Supply Chain Management: Where Theory and
Applications Converge. Edited by Terry Harrison, Hau Lee and John Neale, Kluwer
Publishers. 139-156.

Zhu, Z., Hsu, K. & Lillie, J. (2001). Outsourcing - A Strategic Move: The Process and
Ingredients for Success. Management Decision. 39:5/5. 373-378.

Zinn, W. & Bowersox, D. (1998). Planning physical distribution with the principles of
postponement. JournalofBusiness Logistics. 9:2, 117-136.

135
A Appendix A - Abbreviations
Abbreviations
AIO All-in-One

AVL Approved Supplier List

BOM Bill of Material

CM Contract Manufacturer (Same as EMS in this thesis)

CEM Contract Electronics Manufacturing (Same as CM or EMS)

CHAI Cisco Hosting Applications Initiative

DFE Design for Environment

DFL Design for Logistics

DFM Design for Manufacturing

DFR Design for Recycling

DFSC Design for Supply Chain

EMS Electronics Manufacturing Services (Same as CM)

EOL End-of-Life

FGI Finished Goods Inventory

HVAC Heating, Ventilation and Air-Conditioning

IC Integrated-circuit Chip

IP Intellectual Property

JDM Joint Design Manufacturer

LTB Last Time Buy

MFD Multi-Function Device

136
MOQ Minimum Order Quantity

MPQ Minimum Pack Quantity

NDA Non-Disclosure Agreement

NRE Non-Recurring Expenses

OBS Obsolete

ODM Original Design Manufacturer

OEM Original Equipment Manufacturer

R&D Research and Development

RF Radio Frequency

RFP Request for Proposal

SKU Stock-Keeping Unit

SME Small & Medium Enterprises

WIP Work-in-Process Inventory

137
B Appendix B - ODM Design
Market Share

A MATTER OF
Technology companies in the U.S. have long hired manufacturers in Taiwan and other countries to
make their products. Now an increasing number of tech products are being designed offshore, too.
PDAs NOTEBOOK DIGITAL MOBILE NETWORKING
PCs CAJMERAS PHONES EQUIPMENT
Da-"
OMMW70% 0 WU4M30% ouSLM:20% m Nascent
Persona" Taiwanese Top nds Handsels by CommUni
compabnies Ike 1,okono seconD4ier Cations
assistants esig and and Canon brands li ke,
(PDAs) and manufac- PhIllips and R nds lile
pocket PCs ture most Sieens m
tse many caberfas but and aedruim
twuent,
e ovi ng rov buy operators
technolges So even 'They also controlm any er Ioes
he colbtedeselynIrm st lte TMobileand AT&T
big b&like He*Wlt- patents for indtstdal Tai=a=edn~t edrt3
Ahirn Wireless are based on
Packard aind palmOne aid mechanical design. got inlo thermarket desins by outside in-ouse. But they're
colaborate vith Asian DellHewlett-Packarc quickly HP and Kodak firms. Motorda. Nokia, havygoutsiders
spiers on de sipn. Sony and een Appte collaborat doelly on Sony-Ericsson and design more low-ed
Ovhribrad~s use theirV usedesgs by cortract design with Asian others arm out design geanddeveopkey
venosatim- mantiactules rnanufacturers. oflowendrmoddel soltware.
LEADING DEStGNERS LEADING DESIGNERS LEADING DESIGNERS LEADING DESIGNERS LEADING DESIGNERS
HTC (AXve), Coiel, artaCompa(above), San Prmer (above), Corpa Lcelfon (above), Gemek (above),Delta,
Inventec ston, Awstek.FIC Ake Primnax Flextronics, BetQa CyberTAN, Huawei
Source: Engardio, P., Einhorn, B. (March 2005). Outsourcing Innovation. Business Week. Special Report. 84-94.

138
C Appendix C - OEM R&D
Budgets

Companiesarefamirngout thedesignot morenew proocts-

SALES VY MG&AL-MMJ 6JFACTWI

WNKC4 DZSIN MO MAKEPftVUCM SOID
UNE GHtOPANKS ePAND AMn

6o

30

bCdW
%0wqe0mt*S1"

.. As research anddeveopment budgetsaccourtifora sraerpercentageofsa es

ec5O EUIWMACKAIN UCUW?
INOSOIA ERIC 2A4 20
NDUTKL~
1 S 21
2!
I 1_j B

104/N
2
Z7
0 0 0 -01 0 0 9 - 0": "

Source: Engardio, P., Einhorn, B. (March 2005). Outsourcing Innovation. Business Week. Special Report. 84-94

139