NPD as a sustainable work process in a dynamic business environment

NPD as a sustainable workprocess in a dynamic businessenvironment

Tommy Olin1 and A.B. (Rami) Shani2

1Ericsson Microwave Systems AB, SE-431 84 Molndal, Sweden, and The Fenix Research Program,Chalmers University of Technology, SE-412 96 Goteborg, [email protected] Polytechnic University, Orfalea College of Business, Management Area, San LuisObispo, CA 93407, USA, and the Fenix Research Program, Stockholm School of Economics,SE-11383 Stockholm, [email protected]

New technologies are having a widespread impact on new product development (NPD)

projects. In the telecommunication industry, companies are forced to integrate ‘old’ and ‘new’

ways of working with the utilization of unknown technologies. There is an emerging need to

develop sustainable work systems. This paper investigates the complex relationship between

NPD and the sustainability of the work systems in which it is organized. A proposed

framework for analysis is followed by an examination of a specific NPD project at Ericsson.

Introduction

Tensions cannot be eliminated between sus-tainability, new product development pro-cesses and performance because of their verynature. Understanding the dynamics betweenthem is crucial for the long-term success of thefirm (Docherty et al., 2002; p. 221)

T he revolutions of new technologies arechanging the dynamics of the workplace.

Companies in the emerging telecommunicationbusiness environment are forced to continuouslyexplore the utilization of new and unknowntechnologies in their new product developmentunits. New organizational forms, processes, and

activities are being designed and implemented inorder to cope with the environmental andtechnological trends. At the same time, businessperformance and sustainability are not onlyaffected by the changes, but are also a source ofmajor concern at the individual, organizational,and societal levels.Sustainability, in the context of new product

development (NPD), is viewed as the organiza-tional ability to continuously regenerate re-sources, improve quality of work life, achieve ahigh degree of system flexibility that allows forcontinuous change and development of human,technological and work processes, and to im-prove business processes and outcomes (Dochertyet al., 2002). Thus, at a basic level, meeting theincreasing demands on NPD units to deliver

R&D Management 33, 1, 2003. r Blackwell Publishing Ltd, 2003. Published by Blackwell Publishing Ltd, 19600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

improved products at a faster rate and pacecreate major tension with the emerging need todevelop sustainable work systems.For many businesses and industries, NPD is

now the single most important factor for successor failure. The increased demands to cope withcomplex and rapid changes in the environmentfoster work intensity that is difficult to sustain(Kira, 2002). At the same time, social andtechnological subsystems within an organizationwill be affected when the environmental contextchanges (Pasmore, 1988). These changes must bemanaged to ensure that they will contribute to asustainable and successful NPD process (Love-lace et al., 2001). Thus, increasing our under-standing of the complex dynamics between NPDwork and sustainability is critical.The intent of this paper is to explore the

complex dynamics between NPD and sustain-ability. Following a brief review of the NPD andsustainability literature we advance a framework,the foundation of which can be found in strategicmanagement and sociotechnical system theories.The framework is utilized to describe andexamine the Asterix Project at Ericsson1. Thepaper concludes with a discussion that focuses onthe potential merit of the proposed framework,the creation of actionable knowledge, designingfor flexibility and sustainability in NPD units,dynamic capability in NPD, and managingsustainability in the NPD work systems.

NPD

The literature on NPD is vast and seems to beanchored in a variety of disciplines and theoreticalperspectives – strategy, organization theory anddesign, organization behavior, marketing, opera-tion management, sociology of organizations andengineering design (Krishnan and Ulrich, 2001).The NPD process has been characterized ascomplex to organize and manage. NPD isinherently an orderly and disorderly process. Infact, the management of NPD in R&D andengineering work organizations has been charac-terized as maintaining a balance between orderand disorder (Pasmore, 1988). A recent studycontrasted traditional vs. flexible approaches inNPD and advocated the transition to a flexibleNPD approach for developing products on ‘Inter-net time’. In this context, ‘internet time’ means

being able to respond to environmental futureevolutions in markets and/or technologies that areoccurring rapidly (Iansiti and MacCormack, 1997;MacCormack et al., 2001).The traditional R&D processes are highly

structured, with a variety of models that rangefrom specific stages to specific gates, all of whichrepresent a hierarchy of design decisions. Afuture product is designed, developed, transferredto production, and rolled out to the market inclearly articulated, sequential phases. Such pro-cesses usually begin with the identification ofusers’ needs and an assessment of the varioustechnological possibilities. Then, a detailed set ofproduct specifications is created and, onceapproved by senior management, set in stone.At that point, attention shifts to implementationas a functionally integrated team translates theconcepts into reality. If the up-front work hasbeen done correctly, inherently expensive changesto the specifications are kept to a minimum. Thenumber of engineering changes is often used as ameasure of a project’s effectiveness (Schilling andHill, 1998). Greater numbers of changes corre-spond to more inferior effort. Where technology,product features and competitive conditions arepredictable, or evolve slowly, a traditional devel-opment process works well.In contrast, the flexible NPD approach delays

any final commitments to design configuration.The concept phase and implementation phase,thus, overlap instead of following each othersequentially. By accepting the need for andreducing the cost of change, e.g. introducing alearning process to the implementation phase,companies are able to respond to new informationthat arises during the course of a new product’sdevelopment. Systematic changes in project defini-tion and basic direction are managed proactively,while designers begin this process with no preciseidea of how it will end. Adopting the flexibleapproach, and introducing a learning process tothe implementation phase, will probably haveimplications for techniques and tools used (tech-nological subsystem). This also will have an impacton the profile and competence of the engineers(social subsystem). The relations between theenvironmental context and the social and techno-logical subsystems, therefore, need to be studied inorder to get a more efficient NPD process.Silicon Valley firms that are presently driving

much of the innovation in software and related

Tommy Olin and A.B. (Rami)Shani

2 R&D Management 33, 1, 2003 r Blackwell Publishing Ltd 2003

computer technology seem to have nurtured analternative NPD process. The visions for newproducts develop through scouting and searchbehavior by many organizational players – notsimply the vision of a single entrepreneur orexecutive. ‘It is the constant search and exchangeof ideas in forums throughout the organization andacross organizations that makes innovation ‘every-one’s’ business in the Valley’ (Delbecq and Weiss,2000). There are simple processes for approval offeasibility studies and slack resources available tosupport such studies. For many of the leadingcompanies, the approval of feasibility studies is avery non-bureaucratic review process that en-courages and enables, rather than complicatesand obstructs. Firms have learned that it isimportant to support exploration of new ideas.Furthermore, some CEOs emphasize that it isimportant to allow and invest in the productdevelopment ideas that are contrary to thecompanies’ current visions (Delbecq and Weiss,2000). How will the NPD processes and theoutcome be affected when the management systemis changed to make innovation ‘everyone’s’ busi-ness? In order to secure the long-term businessprofitability, we need to know more about therelations between the management system, theNPD phases and processes, and how these changeswill affect the outcome and business result.Recently, there have been several literature

review articles on product development (Brownand Eisenhardt, 1995; Griffin and Hauser, 1996;Balachandra and Friar, 1997; Krishnan andUlrich, 2001), each focusing on a different aspect:Brown and Eisenhardt (1995) propose a compre-hensive model of factors affecting the success ofproduct development projects; Griffin and Hau-ser (1996) explore the R&D-marketing interface;Balachandra and Friar (1997) adopt a contingentapproach; while Krishnan and Ulrich (2001)present a broad review, encompassing marketing,operations management, and engineering design,which focuses on product development projectwithin a single firm. However, most authors donot make any distinction between implementa-tion projects and innovation projects, i.e. adevelopment project where the prerequisites areknown versus projects where conditions areunknown and need to be explored during theactual product development process.Influenced by Brown and Eisenhardt (1995),

Verona (1999), in a recent study, advocates a

resource-based view of product development.According to the resource-based view of the firm,the presence of different organizational capabil-ities positively affects the outcomes of theproduct development process. As such, theresearchers proposed a resource-based model ofproduct development in which agents influencetechnological, external integrative, internal inte-grative and marketing capabilities which, in turn,affect both process efficiency and product effec-tiveness. Based on Verona’s reasoning, a frame-work that could add to our understanding of howto design NPD processes to respond better tonew, competitive situations, should considerresources, capabilities, and competencies as cri-tical elements.During the 1990s, considerable attention has

also been given to knowledge management andknowledge creation during the NPD process.Improving knowledge management, in order tointegrate fragmented and distributed knowledgesources, has become a critical ingredient forsuccess in the knowledge creation process. Re-cently, Nonaka and Konno (1998) introduced theJapanese concept of ba, which roughly translatesinto ‘space’, to argue that knowledge creation isembedded in ba.

Ba can be thought of as a shared space foremerging relationships. This space can be physi-cal (e.g., office, dispersed business space), virtual(e.g., e-mail, teleconference), mental (e.g., sharedexperiences, ideas, ideals), or any combination ofthem. What differentiates ba from ordinaryhuman interaction is the concept of knowledgecreation. Ba provides a platform for advancingindividual and/or collective knowledge. It is fromsuch a platform that a transcendental perspectiveintegrates all information needed (p.40, italics inoriginal).

Yet, the increasing emphasis on knowledgecreation as an integral part of NPD seems toincrease work intensity.

Intensity and sustainability in NPD worksystems

Our review of the literature reveals that, whileincreasing attention is being given to the role ofthe context, design configurations, resources,

Sustaining NPD in a dynamic environment

r Blackwell Publishing Ltd 2003 R&D Management 33, 1, 2003 3

knowledge management and physical space, littleis given to sustainability. NPD work has beencharacterized as ‘an intensive work’ that tends todrain resources (Lewis et al., 2002). Intensivework systems have major consequences at theindividual, unit and organizational levels. At theindividual level, work intensity emerges from animbalance between an individual’s resources andwork demands, the individual’s needs and workopportunities. This eventually leads to the con-sumption of psychological and physical re-sources, the potential to work and derivehappiness from the work. The emotional andpsychological erosion is a process that leads frominitial exhaustion to cynicism and detachmentfrom work and ultimately ends up in ineffective-ness (Maslach and Leiter, 1997).Consequently, the erosion at the individual

level leads to serious negative consequences at thework group, unit and organizational levels. Thenegative consequences of work intensity at theorganizational level relates to both an indivi-dual’s behavior and actions deriving from his/herexhaustion, as well as to the collective downwardspiral of a social system. Thus, rather thanstriving to fulfill their primary tasks and goals,individuals and groups tend to turn inward andconcentrate on defenses meant to collectivelyconstrain experiences and anxieties. Kira (2002)argues, for example, that in intensive worksystems, productive and creative ways to operateare replaced by rituals and irrational norms.While the understanding of complex emotionaland cognitive relations between people and theirwork is beyond the scope of this manuscript, it iscritical to note that imbalances must be addressedto achieve a sustainable work system (Kompierand Cooper, 1999).In the context of NPD work, we view sustain-

ability as the firm’s strategic and design choices,about the continuous investment in the develop-ment of resources (i.e., human capital), designconfigurations, processes and outcomes that yieldlong-term success. Therefore, there is a need todevelop a framework that takes into account thestrategic nature of NPD, the design configura-tions of the work, its outcomes and sustainability.Strategic management, coupled with sociotechni-cal system thinking, provides a basic languageand analytical framework that can advance theexploration of the relationship between NPD andsustainability.

In a recent study, MacCormack et al. (2001)argue that the field lacks a holistic, integrative,theoretically based framework. Their study of 29NPD projects illustrated the need for a guidingconceptual framework. Shani and Sena (2001)conducted a study that focused on sustainabilityissues of NPD projects in a software developmentfirm. They recognized the need for interdisciplin-ary conceptual integration of contextual factors,management system factors, the NPD processand outcomes, and system sustainability. Theholistic foundation of sociotechnical systemthinking, coupled with a strategic-based view ofthe firm, provide an arena for an interdisciplinarytheoretical integration around the theme ofsustainability of NPD work systems.

Towards an alternative framework

MacCormack et al. (2001) argue that, given thecomplex nature of the flexible NPD process, aguiding framework that allows for a more flexibledevelopment process is needed. The difficulty toquantify costs and benefits of new technologies,the emerging notions of ‘Internet time’ and thesystem-wide implications of the new technologies,indicate that standard approaches may not beappropriate for guiding and appraising thepotential of new work forms in NPD. This paperproposes an approach that integrates two theore-tical traditions: sociotechnical system theory andthe economic/strategic resource-based view theo-ry. The continuous challenge around resourcesand resource allocation lends itself to the utiliza-tion of the economic/strategic resource-basedview theory. The sociotechnical system theorywas chosen due to its emphasis on the balancebetween technological and social elements ofwork in the context of the environmental systems.This approach recognizes and integrates thestrategic and sociotechnical character of theNPD process, and applies the principles ofsociotechnical system and strategic analyses tokey decisions.The sociotechnical system theory addresses

issues that are associated with the mapping ofenvironmental forces and dominant players whoinfluence change (the institutional theorists’ em-phasis), the mapping of resource patterns, changeand evolution of organizational forms, processesand sustainability (the organizational transforma-

Tommy Olin and A.B. (Rami)Shani

4 R&D Management 33, 1, 2003 r Blackwell Publishing Ltd 2003

tion and strategy theorists’ emphasis) and the NPDprocess and outcomes (the management andorganization sciences’ emphasis). At the most basiclevel, the sociotechnical systems perspective con-siders every organization to be made up of a socialsubsystem (all that is about humans in the contextof work) using the tools, techniques and knowledge(the technological subsystem) to produce a productor service valued by the business environmentalcontext (Pasmore, 1988; Taylor and Felton, 1993;Trist, 1982). The degree to which the design of thetechnological subsystem, social subsystem, and theenvironmental context are integrated determineshow successful and competitive the organizationwill be (Pava, 1986; Shani et al., 1992). Thus, whileevery organization is perceived as a sociotechnicalsystem, not every organization is designed usingsociotechnical system design principles, methods,processes and philosophy (Hanna, 1988; Adler andDocherty, 1998). The flexibility that is built intothis perspective is one of the key reasons that theeconomic performance of firms designed accordingto sociotechnical system design principles aresignificantly better than comparable organizationsof conventional design (Hanna, 1988; Van Eijnat-ten, 1994; Pasmore, 2001).The framework advanced in this paper identi-

fies five clusters that affect system sustainabilityin NPD work environment. The framework isespecially useful when one is attempting tounderstand why NPD efforts result in specificoutcomes, and to plan changes that lead toimproved results. Figure 1 illustrates systemsustainability as an outcome that is influencedby how well the five clusters balance (or fit) withone another. It also depicts the cause-and-effectchain that produces business sustainability. Weuse the term sustainability to encompass business

results and system flexibility, as well as regenera-tion and development of human capital.The business environmental context cluster is

comprised of elements and forces in the marketplace in which the firm competes. The socialsubsystem cluster refers to the persons who workin the organization and all that is human abouttheir presence. As such, individual attitudes andbeliefs, competencies and skills, relationshipsbetween group members, relationships betweensupervisors and subordinates, relationships be-tween groups, cultures, traditions, past experi-ences, assumptions, values, rites, rituals, workhabits and practices, and emergent role systemsare all an integral part of the social cluster. Thetechnological subsystem cluster refers to the tools,techniques, devices, artifacts, technological meth-ods, configurations, procedures and knowledgeused by the organizational members to acquireinput, transform input into output, and provideoutput or services to clients or customers. Themanagement systems cluster refers to managerialprocesses that attempt to link the environmental,technological, and social subsystems, such as taskfocused teams and integration driven design. Inthe context of NPD, a task-focused team isdefined as a number of persons, relieved fromtheir ordinary duties, working together on aspecific task. Integration driven design is definedas a process where the development work isdriven by the objective of integrating the wholeproduct as early as possible, rather than trying tofinish the development of all parts before puttingthe whole product together. The managementsystem provides the context within which NPDefforts are designed, developed, and carried out.The NPD phases, processes and outcomes clusterrefers to project phases and product quality, etc.System sustainability refers to system flexibility(human, technological and work processes),sustainability of business results (i.e. traditionaleconomic performance indicators), reproduction,regeneration and development of human re-sources, quality of working life, and businesschange processes for renewal and learning.

Research method

The preceding overview of some of the mainstudies and results in the field of NPD indicatesthe importance of both a continuous investiga-

EnvironmentalContext

Social Subsystem

TechnologicalSubsystem

Management Systems

NPD phases,processes and

outcomes

SystemSustainability

Figure 1. NPD on ‘Internet time’: towards an alter-native roadmap.

Sustaining NPD in a dynamic environment

r Blackwell Publishing Ltd 2003 R&D Management 33, 1, 2003 5

tion and the generation of new actionable knowl-edge in the NPD environment. Taking our pointof departure in the current studies reported in theliterature, we followed up by an in-depthexamination of a specific NPD project atEricsson.According to Yin (1994), case studies are

preferred when how or why questions are beingposed, when the investigator has little controlover the events, and when the focus is oncontemporary phenomena with limited real-lifecontext. In the first place, our study aims toanswer the questions of how sustainability can beobtained in strained NPD projects, and becausewe cannot expect to have any control over theevents in an industrial NPD project, we consid-ered the case study approach most suitable forour study. As one of the authors was one of theproject managers of the examined project, thestudy can also be viewed as a self-ethnographystudy (Alvesson, 1999).A research approach was used where the

generation of new knowledge and improvedunderstanding take place by comparing observeddata with the existing body of knowledge. As ouraim is to eventually develop an integrativeframework, a detailed definition of hypothesesto be tested makes little sense at the outset of thestudy. To provide a better understanding of thebackground, we aimed at describing the environ-mental context, social subsystem, technologicalsubsystem and the management system throughqualitative data. The data is based on in-depthinterviews, documents and literature. Projectdocumentation (process descriptions, project re-ports, project specification and assignment speci-fications), in combination with interviews, wasalso used to capture the NPD process and itsoutcomes. Interviews, inquiries and comparisonto praxis/normal cases were used to describe thesystem sustainability.The gathering of data was limited to qualitative

data. In connection with the conclusion of theproject, eleven, randomly selected project mem-bers were interviewed and asked to complete acomprehensive, written protocol. The data wereanalyzed by the researchers and summarized in apreliminary report. The result from these inter-views and inquiries were then used as a catalystfor discussions at workshops during the comingyear where academics and practitioners shareddata, developed a shared interpretation of the

data, and provided additional insight. Theseworkshops resulted in a refined understandingof the case and its contribution to sustainabilityfor the organization. In the next section, we willprovide a description of the Asterix Project atEricsson while utilizing the proposed, compre-hensive framework.

The asterix case

Environmental context

The telecom industry has been changing rapidlyduring the past decade. Although the market isgrowing very rapidly, the competition among thesuppliers is extremely high. Every move made byone supplier is met by a countermove from theothers. Two of the major suppliers are Ericssonand Nokia.Nokia’s announcement that they would present

the world’s smallest radio base station for GSM(Global System for Mobile Communications)2 ata telecom fair was the trigger for Ericsson to startthe development of a similar product. A radiobase station is the part in the mobile system thatconnects, via radio signals through the air, to theuser’s mobile phone. It is also connected to aswitch that transfers the call either to anotherradio base station (if the call is made to/fromanother mobile phone), or to the stationaryphone network (if the call is made to/from astationary phone). Nokia’s product, the microbase station, measures less than 35 l. and weighsless than 40 kg, with only one transceiver at 5-watt output power. Ericsson’s response to Nokiawas to build, as quickly as possible, a demon-strator to exhibit at the same fair. The demon-strator had to be smaller and lighter, and featuresomething that Nokia’s micro base station didnot have. Ericsson chose to equip their microbase station with two transceivers instead of one(i.e. serving twice as many simultaneously calls),but with less output power per transceiver (i.e.shorter range).The new concept, micro base stations for

GSM, created, surprisingly, great interest fromleading telecom operators, which forced Ericssonto immediately start an NPD project. The workhad to begin from scratch, since the initialdemonstrator did not meet all the requirementsof GSM standard or legal requirements for radio

Tommy Olin and A.B. (Rami)Shani

6 R&D Management 33, 1, 2003 r Blackwell Publishing Ltd 2003

equipment. Matching Nokia’s date for productrelease implied a ‘mission impossible’. A normaldevelopment process would result in a time delayof half a year. To be able to finish on time, thedevelopment team at Ericsson had to find newways of working to reduce lead-time.

Social subsystem

Almost all human resources at the department fordevelopment of GSM base stations at Ericssonwere spent working on existent products orproducts soon to be released. Therefore, a mainproject manager, with experience only in pre-studies and feasibility-studies, was appointed. Adivision at another company within the Ericssongroup was invited to participate in the develop-ment project. This division had previous experi-ence of base station development for the Japanesemarket, as well as experience in designingcompact outdoor equipment.The invited division had to create a new

organizational unit, recruit new personnel, learna new system (GSM), and collaborate with a newcompany culture. They also grew out of theirpremises; hence a new office was needed.Although this proved to be a real challenge, itcreated a pioneer spirit, making it fun andexciting to work in the project while contributingto the quality of work life. The project team wasthen distributed to three different Ericssoncompanies at five locations, 150–900 km apart.The various companies and long distances madematters such as communication, informationflow, and working towards the same goal critical.

Technological subsystem

All documents and drawings were stored in acommon database located at one of the project’sgeographical locations. For fast communicationand distribution of documents and information,an electronic mail system called Memo was used.Drawings were not used for description of themechanical and industrial design. Instead, three-dimensional models were built in computers andsaved as numerical files. These files were thentransferred to the toolmaker for downloadinginto his numerically controlled machines. Theonly drawings made up were the place drawings

used at the manufacturing plant for assemblingthe final product.Ericsson uses a project management method

called PROPS3 for NPD projects. This modelprovides a common project view and terminologywithin the company. There is also a stage-gatesystem called the Tollgate model, which ensuresthat projects are initiated and accomplished in abusiness-oriented manner instead of a technical-oriented manner, i.e., it is not just feasible todevelop the product, but it will also generateprofit. PROPS also include many importantactivities and concepts for a quality system, forexample, configuration management, audits andreviews.

Management systems

The assignment was to develop a small, light, andindustrially designed radio base station. Themicro base station should complement existingmacro base stations, which were large and heavy,and required a specially built footprint to beinstalled. The micro base station should be easyfor one person to install on a wall or a pole. Themain focus was to reduce site costs to ensurecustomer profitability of micro cell networks.Some ordinary requirements of the macro basestation functionality, such as output power andnumber of channels, were relaxed. On the otherhand, the new product should be less than 33 l.and weigh less that 30 kg, compared to a macrobase station of 400 l. and 200 kg.

Integration driven development (IDD) consti-tuted the basis for the approach used in theAsterix project. Rather than trying to describeevery interface in advance, and attempting todevelop every part or sub product independent ofeach other, the approach was based on anongoing effort to integrate all parts as quicklyas possible. The alternative approach of describ-ing accurately every interface of a system inadvance was considered impossible. The IDDapproach aimed to discover errors and mistakesas early as possible, allowing ample time forcorrections. As such, the IDD approach providedsmall victories at regular intervals; victoriescreating a positive atmosphere and go-aheadspirit that kept up the pace in the project.

The system alert was a forum where, above all,technical decisions were made. It consisted of

Sustaining NPD in a dynamic environment

r Blackwell Publishing Ltd 2003 R&D Management 33, 1, 2003 7

members from each subproject and subsystemwithin the project. The system alert was not aplace for long discussions, rather, for shortpresentations and decisions. The system alerthad frequent meetings and when the intensity washigh, meetings were held almost daily. One of theobjectives by the system alert was to illustrate thedegree of management interest. The system alertwas not dependent on people being physically inone place, but was conducted via telephonemeetings.

The project newspaper was a one-page informa-tional letter containing the latest news of interestfor the project. It included problems or progressin the development, information and reportsfrom the market, competitors, production orany test or verification performed within theproject. It could also include general informationabout new routines, or a reminder of what hadbeen agreed upon. The project newspaper waspublished by the project management and servedas a mouthpiece.

The stand-up meetings once a week were, asidefrom the project newspaper, the most importantforum for keeping the project members informedabout the project. The average meeting lasted aquarter of an hour and the project manager heldthe meeting, often inviting people from outsidethe project.

A common set of values was defined early in theexecution phase, as the need for a set of values tobe shared by every project member becameobvious. A set of shared values was compiledthrough discussions and reviews at seminars.Keeping the project-shared values valid requireda continuous process throughout the wholeproject. The values included the following:

� Openness and respect are prestige words withinAsterix;

� If one sub project fails, we have all failed;� Never disgrace a colleague;� Active information sharing and informationseeking;

� The receiver of a delivery is the one who ownsthe problem if the delivery gets delayed;

� Deliver on what you have agreed;� It is fun to work in the Asterix project.

The set of values created the context forcollaborative work. By making every problemvisible to everyone, all members of the projectrealized the challenge and tried to help each other

as needed. As such, no one became the scapegoat.It was clear to everyone in the project that a herodoes not just mind his own business, but issomeone who helps others. This openness made itfun to work, even if there were difficult problemsto deal with.

NPD phases, processes and outcomes

At Ericsson, every project starts with a pre-study,followed by a feasibility study to evaluatedifferent alternatives and to guarantee the busi-ness profitability for the product. This usuallyimplies parallel work of different implementationalternatives. Asterix took a short cut and chose togo for only one alternative from the beginning, tosave lead-time, which turned out to be verysuccessful. Two months before the official start ofthe execution phase of the project, the first draftwas good enough to convince the sponsor toallow an expense of several million SEK in die-cast tools.Early in the definition phase, a dozen

senior designers met for three days to draw thefirst draft. It was extremely important as thework for designing the die cast tools had to beready early. Their mission was to present a firstdraft of the micro base station. The outcome wasa paper and plastic 1:1 model of the final productthat showed the proportions of the microbase station, including positions and sizes of allparts.To give everyone in the project a good picture

of the final product, the outcome from thegathering was used to visualize the final productand overall goal. The mock-up made it very easyto explain thoughts and suggestions. Severalplastic mock-ups were built and distributedthroughout the project. Having mock-ups at eachsite made it easy to discuss problems andsolutions over the phone. The mock-ups alsomade it possible for every new member of theteam to quickly get a good picture of the finalproduct. The plastic mock-up also made theproject more understandable for people outsidethe project, and gave the project higher prioritywithin the company. Also, almost all design workcould start in advance, as every designer knew thesize, functionality, and performance of his/herpart of the product.

Tommy Olin and A.B. (Rami)Shani

8 R&D Management 33, 1, 2003 r Blackwell Publishing Ltd 2003

Timing, attraction, and trust are three wordsthat describe the recruitment process. The shutdown of another project provided some initialresources to launch the new project. The fact thata new organizational unit was built made theproject attractive, as it gave opportunities forpeople to embark on a new endeavor. A neworganization also offered a pioneer atmosphere,which attracted many people. The managementput a lot of trust in all project members, eventhose newly graduated from university. The bestexample of this was the hardware verification subproject. The manager, recruited from anotherEricsson company, had vast experience. A secondperson, recruited internally, had experience frombuilding and managing verification labs. Exclud-ing these two persons, all others were recruited asuniversity graduates.The Asterix project can be characterized as a

project in which an accelerated-pace of integra-tion took on new meaning. From the verybeginning, the plan was to have several mile-stones where progress of the project was checkedby an integration of all available parts for thetime being (see Figure 2). The drive for anaccelerated pace of integration increased workintensity to a new level.For example, to achieve TTM (Time To

Market), the first castings had to be deliveredseven months after the start. Through the normalprocess, it would take six months to design the diecast tools and another six months to producethem; hence, a new process had to be found.Instead of a single major delivery date, severaldelivery dates were set. An exception to normalpractice was that as little as possible should befixed at every delivery. Instead of having anoutput drive, by trying to deliver as much aspossible at every stage, an input pull was used byletting the receiver define what was necessary todeliver.

As the toolmaker was very skillful, he quicklyunderstood the needs, shared his ideas, andhelped the team avoid many mistakes. This closeand collaborative way of working continued untilthe time when the cast tool had to be hardened.When the first casting was delivered only a fewweeks later than planned, the president of thefoundry said:

I really didn’t believe in this. I didn’t think thatit was possible to design and produce a die casttool of this size in just seven months. This mustbe the world record.

The Asterix project managed to keep the timeschedule, succeeding to keep the critical goal ofTTM (Time To Market). Ericsson was one of thefirst telecom suppliers to include a micro basestation in their product portfolio, which led to alarge market share. This, in combination withrelatively high margins in the beginning, led to abreak-even point during the first year. After that,the product continued to do well in the marketand the same product unit has successfullylaunched its successors.Regeneration and development of human

resources were considerable. Being that a neworganization was built, and all resources hadto be recruited from other organizations, intern-ally or externally, almost all project memberswere unfamiliar with either the company cultureor the area of base stations. Many of theengineers also came directly from university,without prior experience in the industry or froma development project. However, the Asterixproject succeeded to couple a few experiencedengineers with the less experienced engineers ineach team.Despite the fact that most of the project

members were either very inexperienced orunfamiliar with the type of products, they gained

First die-cast cabinet delivered to the project Start 3 month 6 m. 9 m. 11 m.12 m. 15 m. 17 m.

(a) (b) (c) (d) (e) (f) (GA)First draw-up of the product First micro base station delivered

to a customer (a) Mechanical model (d) Performance model (b) Electrical model (e) Pre-serial model (c) Functional model (f) Serial model

Figure 2. Time Schedule for developing the new micro base station.

Sustaining NPD in a dynamic environment

r Blackwell Publishing Ltd 2003 R&D Management 33, 1, 2003 9

substantial knowledge in a relatively shortertimeframe than what was common in thecompany. The best example is the HW verifica-tion team. Not only did they manage to build anew verification lab and verify the product ontime, but they also improved the verificationroutines gradually and have, to date, successfullyreduced the verification time by one third. Theteam questioned the established routines andintroduced more efficient internal processes andnew routines for handling external partners, suchas certification houses.A substantial number of project participants

have been promoted to positions outside theproduct unit, thereby diffusing experience,knowledge, tools and methods to other parts ofthe company. For example, the new HW verifica-tion model was introduced as a business changeprocess for renewal and learning in many otherorganizations within Ericsson.

Discussion

All organizations can be understood from asociotechnical systems perspective, but not allorganizations are designed using the sociotechni-cal systems theory or design principles. Theproject members of the Asterix project were notfamiliar with STS theory. However, they usedmany components of the theory to controlchanges in the social and technological subsys-tems affected by changes in the environmentalcontext. While analyzing the Asterix case, theSTS theory provided lenses through which newinsights were generated.Managing NPD in an environment in which

future evolutions are hard to predict, and inwhich evolutions occur rapidly, presents manychallenges. As such, the emerging NPD workplace and dynamics must be of a special naturethat is based on a delicate balance between orderand disorder, intensity and sustainable work.Many issues can be discussed in this last sectionof the manuscript due to the richness of the case.Yet, due to space limitations, we have chosen tofocus on five issues: the relevance of the proposedalternative framework; actionable knowledgecreation practices; designing for flexibility inNPD systems; dynamic capabilities in the NPDprocess, and managing for sustainability in NPDwork systems.

Relevance of the proposed alternativeframework

The most obvious challenge is whether theproposed framework makes a contribution tothe existing body of knowledge and practice onflexible NPD process. MacCormack et al. (2001)argue that the field lacks a holistic, integrative,theoretically based framework. Shani and Sena(2001) recognized the need for interdisciplinary,conceptual integration of contextual factors,management system factors, the NPD processand outcomes, and system sustainability.The description of the Asterix NPD project

illustrated some of the interdisciplinary integra-tion issues around sustainability. For example, thequick mapping of the environmental factors, therealization of the competitive threat, the fastassembly of resources, and the development ofintegration mechanisms are a good illustration ofthe interdisciplinary nature of integration. Insteadof letting changes in the environmental contextalone affect the social and technological subsys-tems in an uncontrollable way, the managementteam gradually changed the NPD process to affectthe system in a controlled manner. Furthermore,the broad spectrum of the categorization schemeof the STS-based framework allowed the emer-gence of specific critical variables (i.e., feasibilitystudy, visualization, and system flexibility) and ofthe road map to examine the Asterix project in aholistic fashion.From the case description, we can see that the

development was more of an architectural in-novation (Henderson and Clark, 1990) than amodular innovation, i.e. the different parts of theproduct were not affected very much, but werecombined and integrated in a totally new way.Architectural innovation may cause problems forestablished companies, due to the way knowledgeis organized in these organizations (Hendersonand Clark, 1990).

An established organization setting out tobuild new architectural knowledge mustchange its orientation from one of refinementwithin a stable architecture to one of activesearch for new solutions a constantly changingcontext. As long as the dominant designremains stable, an organization can segmentand specialize its knowledge and reply onstandard operating procedures to design and

Tommy Olin and A.B. (Rami)Shani

10 R&D Management 33, 1, 2003 r Blackwell Publishing Ltd 2003

develop products. Architectural innovation, incontrast, places a premium on exploration indesign and the assimilation of new knowledge.

According to Henderson and Clark (1990), afundamentally different approach, in which ex-ploration and knowledge creation is prioritized, isneeded to succeed with architectural innovation.Perhaps the fact that the established GSMorganization was unable to develop the newproduct, and a new unit had to be built up,actually facilitated the work as the new unit did nothave any preconceived opinions nor work routines.

Actionable knowledge creation

The concept of usable or actionable knowledge isknowledge that can be used to produce actionwhile, at the same time, contributing to a theoryof action (Argyris et al., 1985, p. ix). Actionableknowledge creation may very well be done whilesolving practical problems. It seems to be an areathat is attracting practitioners and scholars alike(Adler and Shani, 2001). In the context of NPD,and the increasing pressures to develop new waysof thinking about shortening development cycles,actionable knowledge creation needs to berevisited. Knowledge resides within the minds ofindividuals. Enabling mechanisms, such as ba, arelikely to foster knowledge creation and thetransformation of the knowledge created intoaction (actionable knowledge). The mock-up,made from the outcome of the early gatheringof senior designers, contributed substantially toform ba as it helped everyone build a shared spacefor emerging relationships and knowledge crea-tion (Nonaka and Konno, 1998). The proposedalternative framework provides a roadmap thathighlights the importance of knowledge manage-ment as a key and integral management systemprocess. Furthermore, the example from Ericssonillustrated that the emphasis on the creation ofactionable knowledge helped the project moveforward at a high and unprecedented speed.

Designing for flexibility in NPD systems

Designing for flexibility is important when theenvironment is highly turbulent and detailedproduct specifications are impossible to obtainat early phases. As product specifications may

only be feasible at later product design stages, itwould seem, in these cases, that feedback plan-ning is necessary. Verganti (1997) argues that afirm can plan ahead (i.e. feed-forward) to beflexible, such that late innovations and changescan be incorporated with minimal cost and timeimpact. Planned flexibility, then, is a synergy offeedback and feed-forward techniques.In the case description of the Asterix NPD

project, we can see how the integration drivendevelopment (as a part of the management systemin the proposed alternative framework) entails adesign process that not only discovers errors, butalso aims at late changes in the developmentprocess, as the product has to be designed in away that allows integration of all parts beforethey are completed.The close collaboration with the die-cast tool-

maker contributed to a new development processfor the mechanical unit. Instead of completing thedesign work before starting the development ofthe tool (such that a detailed product specifica-tion is set in stone), they seized a more flexibleapproach, often finalizing the design at laterstages. By delaying decisions, the risk of having tochange the decision decreases as the knowledgeneeded to make the decision probably increaseswith time.

Dynamic capabilities in the NPD process

Teece et al. (1997) point to whether a firm’scompetitive advantage is eroded depends on thestability of market demands, and the ease ofreplicability (expanding internally) and imitat-ability (replication by competitors). In short,identifying new opportunities, and organizingeffectively and efficiently to embrace them, arefundamental to private wealth creation.The example of the micro base station is an

illustration of the importance of business intelli-gence and the ability to adopt competitors’development work. In combination with theability to build up a new development unit, wecan see how Ericsson gained business successfrom being one of the first telecom suppliers toinclude the micro base station in their productportfolio. Keeping up with competitors and earlyintroductions of new product concepts may beone result of dynamic capabilities that contributeto sustainability of the NPD.

Sustaining NPD in a dynamic environment

r Blackwell Publishing Ltd 2003 R&D Management 33, 1, 2003 11

The lack of a heavy weight project manager,which according to Brown and Eisenhardt,(1995), correlates to successful projects, did notseem to have much influence. Perhaps there is adifference between implementation projects andinnovation projects, where the former is moredependent on a powerful leader and the latter ismore dependent on curiosity and the ability tofind new solutions to many problems.

Managing for sustainability in NPD worksystems

The managerial challenge rests in the design andmanagement of NPD such that sustainability canbe achieved. This requires a large-scale involve-ment of the knowledge workers from the outset.At the Asterix project, the recruitment processstressed the expectations of total involvement inthe work and participation in decision-making.Individuals were provided with opportunities andplatforms to share experiences and to gain fromthe experience of others. People were givencontrol of the design process and continuousimprovement of the process.We can see that the Asterix project and its

outcome contributed substantially in several waysto the long-term sustainability for the GSMorganization within Ericsson. Business results,i.e. traditional economic performance indicators,increasing a system’s capability and flexibility todevelop new products, regeneration and develop-ment of human resources, quality of working lifefor the employees and business change processesfor renewal and learning are some areas that wereinfluenced by the project and its outcome.Project members achieved a high quality of

work life largely due to the frontier spirit,openness, regeneration and development ofknowledge. The success of meeting all thefrequent milestones strengthened the team’s mor-ale. The team’s strength and the conception thatevery milestone should be used as an opportunityfor celebration and social activities contributedconsiderably to the quality of work life.The most obvious contribution to the long-

term sustainability of the GSM organization wasthe HW verification sub project’s remodeling oftheir work. This subproject consisted of only twoexperienced and some non-experienced engineers.Having ideas and curiosity allowed them tosuccessfully remodel the HW verification process

and reduce the lead-time considerably. Not onlywere their own activities affected, but also theactivities carried out by third party certificationhouses.

Conclusion

The quest to improve the NPD process continuesto gain momentum in the fields of engineeringdesign, marketing and management. The vastbody of literature that has been published onproduct development and design documents theinterest. For example, the most recent review ofresearch articles is based on over two hundredcitations (Krishnan and Ulrich, 2001). Yet, whilethe cross-functional product development re-search is increasing, the field lacks a coherentcomprehensive guiding framework. This manu-script makes a modest attempt at such a frame-work. The proposed framework is anchored in asociotechnical system theory and economic/stra-tegic resource-based view theory.The nature of the proposed framework pro-

vides an opportunity to examine current NPDpractices, as well as to identify new researchopportunities. The proposed framework wasutilized to examine the Ericsson’s Asterix NPDproject and sustainability. A careful review of theliterature reveals that no academic research canbe found on the issue of sustainability in thecontext of NPD, nor about the tensions betweenthe emerging NPD processes and sustainability.Yet, sustainability and organizational mechan-isms that foster sustainability may be criticalfactors in explaining the long-term success inNPD. The lack of research on sustainability, andthe possible interplay or cause-and-effect rela-tionships between NPD and sustainability, mayreflect an inherent difficulty in empirical measure-ment and testing. However, it is perceived as anarea that can contribute substantially to ourunderstanding of NPD and the firm’s long-termperformance.

References

Adler, N. and Docherty, P. (1998) Bringing business

into STS theory and practice. Human Relations, 51,

3, 319–345.

Adler, N. and Shani, A.B. (Rami) (2001) In search of

an alternative framework for the creation of action-

Tommy Olin and A.B. (Rami)Shani

12 R&D Management 33, 1, 2003 r Blackwell Publishing Ltd 2003

able knowledge: table-tennis research at Ericsson. In

Woodman, R. and Pasmore, W. (eds), Research in

Organizational Change and Development, 13. JAI

Publications, pp. 43–79.

Alvesson, M. (1999). Methodology for Close Up

Studies – Struggling with Closeness and Closure,

Institute of Economic Research Working Paper

Series, Lund University, Sweden, given as a keynote

speech at Higher Education Close Up Conference,

University of Central Lancashire, Preston on 6–8

July 1998.

Argyris, C., Putnam, R. and Smith, D. (1985) Action

Science. San Francisco CA: Jossey-Bass.

Balachandra, R. and Friar, J.H. (1997) Factors for

success in R&D projects and new product innova-

tion: A contextual framework. IEEE Transactions on

Engineering Management, 44, (August), 32–56.

Brown, S.L. and Eisenhardt, K.M. (1995) Product

development: past research, present findings, and

future directions. Academy of Management Review,

20, 2, 343–378.

Delbecq, A.L. and Weiss, J. (2000) The business culture

of Silicon Valley: a turn-of-the-century reflection.

Journal of Management Inquiry, 9, 1, 37–44.

Docherty, P., Forslin, J. and Shani, A.B. (Rami) (2002)

Creating Sustainable Work Systems. London: Rou-

tledge.

Griffin, A. and Hauser, J.R. (1996) Integrating R&D

and Marketing: a review and analysis of the

literature. Journal Product Innovation Management,

13, 3, 191–215.

Hanna, D.P. (1988) Designing Organizations for High

Performance. Reading MA: Addison-Wesley.

Henderson, R.M. and Clark, K.B. (1990) Architectural

Innovation: the Reconfiguration of Existing Product

Technologies and the Failure of Established Firms.

Administrative Science Quarterly, 35, March, 9–30.

Iansiti, M. and MacCormack, A. (1997) Developing

product on Internet time. Harvard Business Review,

September–October, 109–117.

Kira, M. (2002) Moving from consuming to regen-

erative work. In Docherty, P., Forslin, J. and Shani,

A.B. (Rami) (eds), Creating sustainable work systems.

London: Routledge, pp. 29–39.

Kompier, M. and Cooper, C. (eds), (1999) Preventing

Stress, Improving Productivity: European Case Stu-

dies in the Workplace. London: Routledge.

Krishnan, V. and Ulrich, K.T. (2001) Product devel-

opment decisions: a review of the literature.Manage-

ment Science, 47, 1, 1–21.

Lewis, M.W., Welsh, A.A., Dehler, G.E. and Green,

S.G. (2002) Product development tensions: exploring

contrasting styles of project management. Academy

of Management Journal, 45, 3, 546–564.

Lovelace, K., Shapiro, D.L. and Weingart (2001)

Maximizing cross-functional new product teams’

innovativeness and constraint adherence: a conflict

communications perspective. Academy of Manage-

ment Journal, 47, 4, 779–793.

MacCormack, A., Verganti, R. and Iansiti, M. (2001)

Developing products on ‘internet time’: the anatomy

of a flexible development process. Management

Science, 47, 1, 133–150.

Maslach, C. and Leiter, M.P. (1997) The Truth about

Burnout. San Francisco: Jossey-Bass.

Nonaka, I. and Konno, N. (1998) On the concept of

‘‘ba ’’: building a foundation for knowledge creation.

California Management Review, 40, 3, 40–54.

Pasmore, W.A. (1988) Designing Effective Organiza-

tions: the Sociotechnical System Perspective. New

York: Wiley.

Pasmore, W.A. (2001) Action research in the work-

place: the sociotechnical perspective. In Reason, P.

and Bradbury, H. (eds), Handbook of Action

Research: Participative Inquiry and Practice. Lon-

don: Sage.

Pava, C. (1986) Managing New Office Technology: an

Organization Strategy. New York: Free Press.

Schilling, M.A. and Hill, C.W.L. (1998) Managing the

new product development process. The Academy of

Management Executive, 12, 3, 67–81.

Shani, A.B. (Rami), Grant, R. and Krishnan, R. (1992)

Advanced manufacturing systems and organiza-

tional choice: a sociotechnical system approach.

California Management Review, 34, 4, 91–111.

Shani, A.B. (Rami) and Sena, J. (2001) Knowledge

management and new product development: Learn-

ing from a software development firm. Proceeding of

the 3rd International Conference on Practical

Aspects of Knowledge Management. Basel, Switzer-

land, October 2000, pp. 19-1–19-6.

Taylor, J.C. and Felton, D.F. (1993) Performance by

Design: Sociotechnical Systems in North America.

Englewood Cliffs NJ: Prentice-Hall.

Teece, D., Pisano, G. and Shuen, A. (1997) Dynamic

capabilities and strategic management. Strategic

Management Journal, 18, 7, 509–533.

Trist, E.L. (1982) The evolution of sociotechnical

systems. In Van de Ven, A.H. and Joyce, W.F.

(eds), Perspectives in Organization Design. New

York: Wiley, pp. 19–75.

Van Eijnatten, F.M. (1994) Redesigning sociotechnical

systems design: concepts and methods. Journal of

Applied Behavioral Science, 22, 3, 201–222.

Verganti, R. (1997) Leveraging systemic learning to

manage the early phases of product innovation

projects. R&D Management, 27, 4, 377–392.

Verona, G. (1999) A resource-based view of product

development. Academy of Management Review, 24,

1, 132–142.

Yin, R.K. (1994) Case Study Research: Design and

Methods. Thousand Oaks CA: Sage.

Sustaining NPD in a dynamic environment

r Blackwell Publishing Ltd 2003 R&D Management 33, 1, 2003 13