Flexibility in NPD Projects: A Multiple-Case Study on Flexibility Dimensions, Innovation Type and Project Outcome

Flexibility in NPD Projects:

A Multiple-Case Study on Flexibility Dimensions,

Innovation Type and Project Outcome

By

Nathalie Fuchs

University of Groningen,

Faculty of Economics & Business,

MSc BA Strategic Innovation Management

Date: 15

of June, 2015

Supervisor: Wim Biemans

Co-assessor: Hans van der Bij

Nathalie Fuchs n.fuchs.1@student.rug.nl Student Number: s2020475

TABLE OF CONTENTS Abstract 3 Introduction 3 Theoretical Background 6 Organizational flexibility 6 Informational flexibility 7 Temporal flexibility 7 Design flexibility 7 Resource flexibility 8

Strategic decision-making flexibility 8

Methodology 10 Research design 10 Data collection 12 Data analysis 13 Research quality 14 Results 15 Mindset flexibility 16

Successful versus unsuccessful NPD projects 18 Radical versus incremental NPD projects 22 Main actors in project flexibility 27

Discussion 29

Successful versus unsuccessful NPD projects 30 Radical versus incremental NPD projects 31 Managerial implications 33 Limitations and avenues for future research 34

Conclusion 34

References 35

Flexibility in NPD Projects:

A Multiple-Case Study on Flexibility Dimensions,

Innovation Type and Project Outcome

ABSTRACT. Flexibility is argued to be the answer for NPD projects to the increasingly turbulent environment where continuously new technological and market information is provided. Yet, on the project level the theory is characterized by fragmentation and incompleteness. Therefore, by adopting a theory development approach, this paper presents the multiple-case study performed at a large Dutch consumer electronics firm to explore how flexibility is created in NPD projects. From the in-depth analysis it is proposed that successful projects and radical projects apply a larger number of flexibility dimensions than their counterparts. Moreover, mindset flexibility is suggested as a new dimension, and relations between particular flexibility dimensions and innovation type and project outcome are proposed. To conclude, directions for future academic research and managerial implications are presented.

INTRODUCTION

Consequently, by freezing designs early, firms are unable to incorporate the continuously arising new market and technology information.

The inability to deal with the external turbulence during development (i.e. the inflexibility), however, will mean that the firm in question will experience a significant competitive disadvantage in comparison to their flexible counterparts, as “sustainable competitive advantage lies in a firm’s ability to quickly adapt to the changing environment” (Calantone et al., 2003). Therefore, flexibility, as the ability to “embrace environmental turbulence rapidly adapting to new technological and market information that emerges over the course of a project” (Biazzo, 2009), is crucial for innovation performance, and thus for firm performance.

There is, accordingly, a growing body of literature on flexible NPD practices to manage the increased environmental dynamics. Much of the earlier work in this field focused on freezing the concept later to create more time to ‘react’ to new information (Iansiti, 1996; MacCormack et al., 2001; Verganti, 1999). Sanchez and Mahoney (1996) among others have focused their efforts on development flexibility in terms of (modular) design through for instance the standardization of interface specifications. Additionally, Biazzo (2009) introduced new dimensions related to the “degree of structuration in the process” (organizational flexibility), the intersection between problem formulation and problem solving activities (informational flexibility) and the task scheduling of the activities (temporal flexibility). More recently the NPD flexibility domain has been extended to resource flexibility (Tatikonda and Rosenthal, 2000), including the workforce as a source of flexibility (Kok and Ligthart, 2014), and strategic flexibility, which reflects the ability to identify environmental changes and to rapidly (de)commit resources in response to change (Kandemir and Acur, 2012; Yuan, Zhongfeng, and Yi, 2010).

(Buganza et al., 2009). Thirdly, many flexibility studies are of a quantitative or theoretical nature and are therefore still in need of more in-depth qualitative empirical research (e.g. Biazzo, 2009) to better understand the underlying mechanisms of project level flexibility. And lastly, most scholars focus either on radical or incremental innovation, thereby disregarding the potential difference in the role that flexibility plays in these types of innovation projects.

The preceding discussion shows that, despite the fact that the body on flexibility literature is growing and recognizes the importance of this field, to date, there is no exhaustive framework in place that explains how firms deal with flexibility in projects. Additionally, the current literature does not describe this phenomenon adequately, hence, a literature study will not be sufficient; a field study is needed to obtain the necessary qualitative data to explore how flexibility works on the project level. Therefore this paper aims to fill these gaps by answering the following research question: How do firms create flexibility in new product development

projects?

The primary contribution of this study is that it is the first to include the variety of established flexibility dimensions, and to explore them in the new product development project context, leading to a completer view of project flexibility with rich, qualitative case descriptions. Second, by including successful and unsuccessful, and radical and incremental projects, this research has brought a more nuanced view of the application of flexibility in projects. Third, this paper offers useful implications for managers engaging in new product development, including the main actors to involve in project flexibility. And lastly, opportunities are offered for future research to advance the field of project flexibility.

THEORETICAL BACKGROUND

To provide a better understanding of the current state of literature of flexibility on the project level, this section will provide a theoretical background which is organized according to the identified dimensions. The early notion of flexibility was established in the manufacturing literature, where flexibility is often considered as an “adaptive capability to respond to environmental uncertainty” (Gerwin 1993; Kandemir and Acur, 2012). Following Biazzo (2009) this study views flexibility within NPD projects as the ability to “embrace environmental turbulence rapidly adapting to new technological and market information that emerges over the course of a project”. To further specify, according to the Project Management Institute (PMI), a project is a temporary undertaking with a “defined beginning and end in time, and therefore defined scope and resources” (2015). Next to that, new product development concerns “the management of the disciplines involved in the development of new products” (Trott, 2012). Yet, as explained, most studies focus on firm level flexibility, leading to gaps in the NPD flexibility literature, especially on the project level. The remainder of this chapter will discuss the relevant flexibility dimensions found in recent literature.

Organizational Flexibility

occurring in each stage. Hence, the more structured and standardized the milestones and activities are, the less flexibility resides in the organizational dimension ( Biazzo, 2009).

Informational Flexibility

The second dimension proposed by Biazzo (2009) is that the activities defined for each NPD stage can be categorized as product definition (problem formulation) or detailed design (problem solving) activities; the degree of intersection between problem formulation and problem solving activities reflects the flexibility in the informational dimension. Accordingly, it becomes clear that traditional (stage-gate) models of innovation, despite their inflexible organizational dimension, still can build in flexibility though the informational dimension (Biazzo, 2009). In other words, by allowing an incomplete product definition at a ‘gate’ and continuing with both problem formulation and problem solving activities, the intersection of these activities can create flexibility without depending on the organizational dimension.

Temporal Flexibility

The third and last dimension in Biazzo’s (2009) framework is the temporal dimension, which reflects the task scheduling of the tasks assigned, leading to for instance simultaneous performance of tasks. Thus, the distinction between the informational and temporal dimension can be explained by overlap and intersection: overlap, “the simultaneous carrying out of tasks by different groups of people”, relates to the temporal dimension, whereas the intersection of problem formulation and problem solving activities is represented by the informational dimension (Biazzo, 2009). To conclude on Biazzo’s framework, however, his framework of flexibility dimensions remains to be elaborated in “case studies with thick descriptions of effective NPD processes characterized by different types of flexibility” (Biazzo, 2009).

Design Flexibility

standardized components (Sanchez and Mahoney, 1996). Accordingly, modular product architectures enable quick development of product variations to respond to changing markets or technologies (MacCormack et al., 2001; Sanchez and Mahoney, 1996; Thomke, 1997) through the rapid creation of new product variations based on recombining new or existing modules, and accelerated learning by quickly leveraging product variations at reduced cost (Sanchez and Mahoney, 1996).

Resource Flexibility

Many scholars argue, however, that flexibility literature has been mostly addressing flexibility in structures and processes of NPD, while it can also be found in other dimensions (e.g. Kok and Ligthart, 2014), such as resource flexibility. On the firm level Yuan et al. (2010) defined resource flexibility to be determined by the inherent properties of the resources, while coordination flexibility reflects a firm’s capabilities to apply the resources (Sanchez, 1995; Sanchez, 1997). Tatikonda and Rosenthal (2000) investigated resource flexibility, “the flexibility in reallocation of project resources”, on the project level. Their results indicate a positive relationship between resource flexibility and project execution success, and contrary to their expectations, technological novelty of the project did not moderate the relationship. Typical resources included materials, financial, and personnel resources (Tatikonda and Rosenthal, 2000). Kok and Ligthart (2014) contribute to the resource flexibility literature by investigating on a firm level the effects of workforce flexibility. This type of flexibility is twofold: functional flexibility to deploy employees’ skills across a broader range of assignments, and numerical flexibility to adjust the working hours and amount of employees in line with the level of demand. In conclusion, resource flexibility could play an interesting role in NPD projects.

Strategic Decision-Making Flexibility

qualitative research must still reveal how these resources operate (Kandemir and Acur, 2012). In addition, the study of Kandemir and Acur (2012) measured strategic decision-making flexibility on the firm level, yet, the opportunity exist to explore it on the project level in this study.

To conclude, this chapter has clearly separated the identified dimensions to structure the theoretical field of flexibility and enhance data collection and analysis (table 1 below summarizes the dimensions); however, this distinction was not always that evident in the individual articles, let alone in the review of the theoretical field. In some cases the boundaries between different dimensions were ambiguous or overlap was found, which may also be the case in practice. For instance, strategic decision-making flexibility may require additional resources to maintain alternative paths, or the alternative decisions may take place on the product through the development of different module options. Despite this ambiguity, the review still enhanced the data collection and analyses and indicates that flexibility has become a recognized phenomenon within the NPD literature. However, on the project level, where environmental turbulence has the highest impact (Buganza et al., 2009), the state of art is still of an exploratory nature (Van Aken et al., 2012). Therefore a theory development approach (Van Aken et al., 2012; Yin, 2014), with the focus on exploration, has been deployed to establish a conceptual framework of flexibility dimensions within NPD projects, based on rich in-depth empirical data resulting from case study research. The next chapter will further elaborate on the research methodology of this study.

Flexibility dimensions Definitions

Organizational flexibility “the degree of structuration in the process design” (Biazzo, 2009) Informational flexibility “the classification of the development activities into problem

formulation and problem solving tasks” (Biazzo, 2009)

Temporal flexibility “the execution strategies of development tasks, which refers to the

task scheduling” (Biazzo, 2009)

Design flexibility “the extent to which a product concept can be kept open for change”

(MacCormack et al., 2001); “the incremental cost and time of modifying a design” (Thomke, 1996).

Resource flexibility “flexibility in reallocation of project resources” (Tatikonda and

Rosenthal, 2000)

Strategic decision- making flexibility

“a capability that enables firms to make effective strategic decisions by maintaining multiple simultaneous decision alternatives”

(Kandemir and Acur, 2012)

METHODOLOGY

The research design of this study is based on the explanatory paradigm, as its purpose is to develop new knowledge and to solve the current literature gaps on flexibility in NPD projects (Van Aken et al., 2012). Accordingly, a theory development approach was regarded most appropriate in this study as many companies engage in NPD activities and flexibility becomes increasingly important, meaning that it is a “generally recognized business phenomenon” (Van Aken et al., 2012; Yin, 2014). Additionally, a multiple-case study design has been selected because, as indicated before, the flexibility literature on the project level (the unit of analysis) is rather exploratory in nature (Van Aken et al., 2012), and characterized by fragmentation and incompleteness, resulting in a strong need for empirical evidence (e.g. Biazzo, 2009). The remainder of this chapter will elaborate on the selected research design, the data collection, the data analysis, and the measures taken to raise the quality of this study.

Research Design

To answer the question on how firms create flexibility in NPD projects, this research followed a multiple-case study design and was structured by the steps for inducting theory from case studies by Eisenhardt (1989), which consists of the following steps: (1) getting started, (2) selecting cases, (3) crafting instruments and protocols, (4) entering the field, (5) analyzing data, (6) shaping hypotheses, (7) enfolding literature, and (8) reaching closure. This structure helped in guiding the research and more details of these steps are elaborated on in the following sections.

Since the focus is on NPD projects, a large consumer electronics company headquartered in the Netherlands was selected, where the research and development (R&D) departments contain a multitude of product development teams and engage both in incremental and radical NPD activities in multiple product categories. By having specified the population appropriately, extraneous variation can be reduced and generalizability extended (Eisenhardt, 1989) as the findings will regard large corporations operating in the consumer electronics business. Yet, as in exploratory research, the focus remained on theory building as opposed to generalizability (Strauss and Corbin, 2008).

Corbin, 2008; Yin, 2014). As this research focuses on flexibility in NPD projects, the analyses took place on the project level with the NPD project as the unit of analysis. Additionally, a distinction was made between successful and unsuccessful projects, and incremental and radical innovations, to better grasp the relationship and mechanisms between flexibility, project result, and innovation type. Moreover, to increase the likelihood of theory replication and extension, cases were also selected from different product categories within the consumer electronics business. Consequently, four types of projects emerged as the focus of this research: (1) incremental-unsuccessful, (2) incremental-successful, (3) unsuccessful, and (4) radical-successful. In total, twelve cases were selected including three of each type, as is summarized in table 2.

The actual case selection took place by contacting directly the project management office managers, i.e. the head of the project leaders, at different R&D locations both in the Netherlands and Belgium. These managers are well aware of the projects that took place recently and are able to distinguish the four project types in their portfolio. By taking into account the definitions used in case selection and engaging in good discussions with these managers, they could indicate which projects could be of added value to this research. The next step was to select the projects based on theoretical purposeful sampling (Strauss and Corbin, 2008) and recruit the informants for the data collection phase. With regards to the case selection, “variation is especially important in theory building because it increases the broadness of concepts and scope of the theory” (Strauss and Corbin, 2008).

The distinction between an unsuccessful and successful project was made by the company; each project had a business case connected to it, which is comprised of data about for instance estimated sales, profit margin, and project costs. The business case represents the argumentation to execute the project from a business perspective, thus if it was not met, or after launch clear that the project would never be able to meet the business case, then the project was considered unsuccessful. Radical innovations, however, are considered to need more time to get accepted by the market, which is reflected in the business case data.

practices”, which should be considered relative to the firm (Garcia and Calantone, 2002). Ultimately, the projects selected had a project duration of one to five years, and came from four different product categories, which are all related to the consumer electronics business and share similar project characteristics. Table 2 displays the different types of projects and the names that were assigned to them for the purpose of anonymity.

Projects Unsuccessful (U) Successful (S)

Incremental (I)

IU-1 Product category 1

1 2

IS-1 Product category 1

IU-2 Product category 2 IS-2 Product category 3

IU-3 Product category 3 IS-3 Product category 2

Radical (R)

RU-1 Product category 1

3 4

RS-1 Product category 4

RU-2 Product category 3 RS-2 Product category 1

RU-3 Product category 4 RS-3 Product category 3

Table 2. Case Selection

Data Collection

The primary source of data is the information gathered via the interviews. In recruiting interviewees the focus was on project managers as they have the most complete view throughout the project, take an important role in the project team, and are close to the action (as opposed to higher level management). After getting in contact with them, the project managers of the relevant projects were informed about the purpose of this research, and were guaranteed anonymity when participating. Albeit, in one case the project manager had left the company already, however, since the case was expected to be theoretically interesting, two key members of the project’s core team were interviewed together. So, in general, one interview was conducted per case and all interviews, except for one, took place in a face to face meeting with the official interviews lasting about an hour (the shortest was 50 minutes, and the longest 1.5 hours). All thirteen interviewees had been working for the company for over five years, and had experienced multiple NPD projects within their category.

the meaning of flexibility in that particular project, the type of flexibility dimensions that were applied, and their relative importance. As this study’s aim is to research the why and how of phenomena, open-ended questions and probing questions were actively used to gain more understanding. Moreover, to not limit the interviewee and his or her knowledge, the interview was semi-structured and was followed loosely in the sense that the interviewee was allowed to guide the content. In addition, all interviews were with consent taped and transcribed to allow for more detailed analysis. Thereafter, together with the secondary sources, case study summaries were created and were verified through an interview member check, and where needed adjusted in cooperation with the interviewees, also with the aim to raise the construct validity of this study (Koelsch, 2013).

Additionally, the main secondary sources consulted during this study were documentation and archival records (Yin, 2014), in order to provide additional or confirming information with regards to the cases selected. The company had granted access to their official project documentation storage location, where for instance milestone reports that contain the information upper management needed to progress the project to the next phase, project planning, project cost estimations, product price calculations, industrialization plans, and risk management documents could be accessed. Hence, all the documents needed to run a NPD project were stored in this database, and provided sufficient evidence in addition to the interviews. The different data sources enabled the creation of a ‘chain of evidence’ (Eisenhardt, 1989; Yin, 2014), and were stored and analyzed per case in the case study database. To conclude, data collection (and analysis) was discontinued at the point of theoretical saturation, in other words, when further collection and analysis “add little new to the conceptualization, though variations can always be discovered” (Strauss and Corbin, 2008).

Data Analysis

be concluded before continuing with further analysis in order to foster the independent replication logic through the cases (Yin, 2014). Thereafter, cross case analysis and pattern matching took place, facilitated by tables and graphs (Eisenhardt 1989). During this process it was important to continuously refine the insights based on all iterations. Additionally, to maintain consistency in the data it was essential to return to the original evidence during the analysis. Moreover, hypotheses were not defined prior to data collection and analysis; in order to develop the proposition the method of explanation building was used, including the use of rival explanations (Eisenhardt, 1989).

Research Quality

To conclude, during this study the most important quality criteria can be considered the degree of controllability, validity, and reliability (Van Aken et al., 2012). Controllability enables others to replicate the study and is prerequisite for judging reliability and validity (Van Aken et al., 2012). Through transparent and precise reporting it has been aimed to maximize controllability. With regards to reliability and (construct, internal, and external) validity, the most common and relevant tactics have been gathered from Yin (2014), Eisenhardt (1989), and Van Aken et al. (2012) and summarized in table 3. These tactics have been followed during this study and have been highlighted in the previous sections of this chapter.

Tests Caste study tactic Research phase

Construct validity

 Multiple sources of evidence

 Chain of evidence

 Key informants review draft case study report

Data collection Data collection Composition

Internal validity  Pattern matching

 Explanation building  Rival explanations  Logic models Data analysis Data analysis Data analysis Data analysis External validity  Specified population

 Replication logic in multiple-case studies

Research design Research design

Reliability  Case study protocol

 Case study database

 Input from multiple investigators

Data collection Data collection Research design

RESULTS

In this section will be elaborated on the insights that emerged from the data. With regards to flexibility, all interviewees clearly recognized the phenomenon of flexibility in NPD projects. For instance, the development engineer of RU-1 described it as “being able to adjust to the circumstances; a team that adjusts its focus”. The project manager of the IU-2 project described it as “when the road you have planned to go from point A to point B, somewhere faces a roadblock or is simply not quick enough, means that you need to look for alternative possibilities”. In the end they all describe it as the ability to adjust during the project to changing circumstances. The most mentioned source of change is new market information, or as the interviewee of IS-2 puts it, “to react on the changing needs of marketing”. Yet, the R&D organization is also considered a source of turbulence because of new technological information through for instance product testing or new technological knowledge.

In each project flexibility was regarded important, even in the unsuccessful projects. “It was very important, also in this project. We do not get immediately the right specifications; this is what you should make. The specifications you build up together and there you should also be flexible to incorporate changes” (IU-3). The fact that it is difficult, if not impossible, to establish the right specifications from the start is also illustrated by the comment of one of the interviewees: “It is an illusion that you can plan a project to such a degree that everything is set upfront; that will not work”.

To discover how flexibility is created in new product development projects both incremental and radical, and successful and unsuccessful cases have been studied. With regards to how to react in a flexible manner to the turbulence, the dimensions identified prior to the data collection and analysis (see table 1 for the overview), were all recognized in the cases selected for this study. In other words, these are dimensions that are relevant to projects, whether applied flexibly or not. Additionally, no differences were found across the different product categories the projects were selected from, so that analysis is left out here.

Including one newly identified dimension, a total of seven flexibility dimensions on the project level were analyzed among twelve NPD projects. Table 4 and 5 provide a high level overview of the cases and the flexibility dimensions; ‘+’ indicates that the dimension was applied in a flexible manner during the project, and ‘-’ the opposite. Hence, the plus indicates also a higher degree of flexibility in a particular dimension than the minus does. Important to note, however, is that the application does not accurately indicate the strength or the importance of that particular dimension during the project, as that would have required extra quantitative measures, which is out of the scope of this research. Notwithstanding these notes, the tables still provide interesting insights and enable comparison among the cases.

More interesting, however, is the fact that a new dimension has been identified and how the different flexibility dimensions link to the different project characteristics. Therefore, for the sake of theory development, this chapter will elaborate on these novel insights. Firstly, light will be shed on mindset flexibility, the newly found flexibility dimension. Secondly, the link between specific flexibility dimensions and NPD project (un)successfulness will be elaborated on. And thirdly, the radical or incremental nature of the innovation under development and its connection to the different flexibility dimensions will be discussed. And lastly, attention will be given to the main actors involved in project flexibility.

Mindset Flexibility

emerged from the data analysis, which refers to a mindset that is able to deal with the turbulence during a project. The first aspect important to mindset flexibility is the belief of the team and its individual members that they are working on a meaningful product and project. Only then the project team is willing to make the extra mile, to act differently and be flexible. Second, the team needs to work together towards one goal, which means giving and taking, instead of standing at opposite ends of each other. The project manager of IS-2 explains this clearly:

“If you have a role in quality, it is always very easy to say: that is not good. But obviously you do not help [the company] moving forward, so you must also try to cooperate; so what is then good? In that sense I believe it is important that you as a project team work towards one goal and it is also very important for the success of the project”

And lastly, it is also a matter of being able and willing to switch quickly as a team and individual. A more traditional approach to react to new information would be to add the resources it would normally cost to implement the changes, resulting in a delay or more costs. The flexible approach, however, looks at where the opportunities (so in the other flexibility dimensions) are to still make it in time, quality, etcetera. The IS-3 project manager says the following about this: “it is how you deal with it … you can make out of everything a problem, which is a mindset. But you can also say, everything is an opportunity, let’s make [the product] better.”

The mindset flexibility dimension was regarded very important, and will therefore also be elaborated more on in the next section. The strength of the dimension among the cases varied, at

RU-1, however, it was clearly evident throughout the project. “At a certain point it also got very

natural to us, if nothing happened on a day it was fine, but then probably something will go wrong tomorrow, and then we should be ready to react on it. That happened so often, it was actually a way of life; every moment something changed, and we were running for it.” To conclude, an example of the application of the mindset flexibility dimension is the RU-2 project, where the project was urged to start, while the team perceived it as an inferior product and a project where they had to solve the problems created elsewhere. In the following fragment of the

RU-2 project, the project manager indicates the need for mindset flexibility and how the lack of

“People did not want to participate, partly because they did not see it happening, partly because they were hiding behind other projects, and partly because they were hiding behind procedures. If you then have to start a project, or take over one, it becomes very difficult. So that was an example where we were not flexible, where people simply hid. Higher management really had to step in there, and gradually you saw movement in the team during the project, but even then still very slowly.”

Successful versus Unsuccessful NPD Projects

The purpose of this study is to explore flexibility dimensions on the project level, but also investigate into the potential differences in flexibility practices between successful and unsuccessful projects. As described earlier, the (un)successfulness of a project is defined as our informants did, thus, based on the business case set up for the project. Based on the data analysis, with the high level overview illustrated in table 4, successful projects seemed to apply more types of flexibility dimensions than unsuccessful projects. In general, this does not necessarily mean that these projects are more flexible (as it does not accurately reflect the strength of the dimensions); however it does at least indicate a larger variety in the flexibility dimensions applied during a project. Moreover, three dimensions appear to be more strongly related to the project outcome. Specifically, design flexibility and mindset flexibility were found more at the successful projects, and informational flexibility was more evident among the unsuccessful projects. These dimensions and their connection to project outcome will be elaborated on below.

Design flexibility. One of the dimensions identified to be related to successful NPD

projects is design flexibility, which can be defined as the degree that the product concept remains open for change based on the characteristics of the product itself during the course of a project (MacCormack et al., 2001). The successful cases in this study showed relatively more design flexibility in comparison to the unsuccessful cases. The following statement from the project manager of the successful IS-3 project clearly illustrates the dimension in practice.

Project RS-1 RS-2 RS-3 IS-1 IS-2 IS-3

Innovation Radical Radical Radical Incremental Incremental Incremental Result Successful Successful Successful Successful Successful Successful Organizational flexibility

+

+

-

-

-

-

+

-

-

-

+

+

+

-

+

+

-

+

-

+

+

+

+

+

+

-

+

+

+

+

-

+

-

+

+

+

+

+

+

+

Table 4. Successful versus Unsuccessful NPD Projects

Project RU-1 RU-2 RU-3 IU-1 IU-2 IU-3

Innovation Radical Radical Radical Incremental Incremental Incremental Result Unsuccessful Unsuccessful Unsuccessful Unsuccessful Unsuccessful Unsuccessful Organizational flexibility

-

-

+

-

-

+

+

+

-

-

+

+

-

+

+

+

+

-

-

-

-

+

+

+

+

-

+

-

+

-

-

+

+

-

+

-

+

design flexibility more product variation can be achieved, leading to potentially more pieces to be produced and sold; and the higher the volumes, the higher the revenues for the supplier as well, and thus, the more interesting the case also becomes for suppliers. How design flexibility aids in industrializing the product is clearly illustrated in the IS-2 project, which is a good example where industrialization was facilitated and costs reduced:

“We always have a problem; we only sell 2000 pieces a year, then we have a successful project. But the problem, if you then want to industrialize, you will never reach the right prices; [the suppliers] look at it as just a few pieces. So, to reach a certain degree of mass production we should try to combine different project that contain a standard element. Then you simply talk about higher volumes, maybe 50.000, 60.000 pieces, and then the suppliers become much more interested. It makes the production process much easier.”

Second, design flexibility enables accelerated development since changes can be implemented at a higher pace and the team can work in parallel on different parts of the product. During the RS-2 project special module teams were set up to independently work on different problems in different modules, without hampering each other; status reports provided even updates on these different modules. And development can be shortened because when building upon existing architectures, one does not have to develop it from scratch. For instance the project manager of the IU-2 project about it: “we did have to develop complete new platforms, … but the architecture was already known”; “we also made changes in the design, there we have changed some parts on the outside of the product in such a way that we would have maximum impact on the customer with a minimum amount of resources.” The same was reported in their presentations towards upper management: “Maximum reuse of existing [previous model] modules”.

Mindset flexibility. Not only is it identified as a new flexibility dimension in this study,

“In the formal attitude which we see in general, and we get contradicting or disappointing test results, which were not anticipated or combined with a backup plan, then most people are inclined to say: we have to do it again, it costs that amount of time, and that is it.”

On the other hand, when flexibility is found in the mindset of the team, it takes less energy to react to the turbulence during a project. For instance, when new customer preferences emerge during the course of project and it is decided to include it in the scope, the team could decide to extend the project planning with one year, because such changes normally take that amount of time. With mindset flexibility, however, these assumptions are challenged, and people search ways to deal with the new information without having the impact on the project that it would normally have in for instance costs or time. “When things are not delivered in the normal time frames or if there are setbacks, then people are willing to do the things and find different ways to fix it” (IU-2).

Informational flexibility. The most evident dimension among the unsuccessful projects

was the informational dimension. According to Biazzo (2009) that entails the interaction between project definition and project design activities during a project, hence the more interaction, the more flexible the project. By the interviewees this was regarded as returning to the definition of project scope and requirements. The project manager of RS-2 describes it as follows: “I have an original problem definition, I call it scope; so what are you going to deliver with which requirements. A few times we had to step back because the ideas were not good enough, new problems, new activities …”. An example of such continuous change in requirements is the IU-3 project. There the product was above the required price level, so the team lowered the product requirements to achieve the target price; the project manager about the case:

“At that moment we have tried all options already, from the most luxurious version to the cheapest version, and actually we should have, and some stakeholders also said it, stopped the project, but we continued. So we took the first version again, the luxurious version, … , but there we made the mistake, we accepted the high price. And now it is visible that we sell very few appliances.”

RS-1 project this was all settled, “there was no switching [between problem formulation and

problem solving] I believe. It simply was very much anchored from the beginning of the project on. This is what we can do, these are the benefits, those are the risks, and stakeholders are supporting, ‘go’!” In the opposite situation the starting requirements are unclear; meaning that if new information and turbulence emerges there is less guidance into which direction the project should adapt. The project manager of the RS-2 project about the beginning of the project: “there were very large gaps in terms of requirements, deliverables, and knowledge”; and the quality project leader of RU-1 about it: “With flexibility it is convenient to have set good requirements. So that you can say, this is in and the rest is out.” One of the interviewees explained this phenomenon very clearly:

“You need to keep the possibilities open, that is flexibility, but it should not distract you from your goal. You have to take care that if you have made the decision to go to the left that you stay walking to the left. Never forget that you can always receive information along the road to discover that the right direction would have been better.”

Radical versus Incremental NPD Projects

The potential differences between flexibility in radical and incremental innovation projects also have been studied in this research to better understand project level flexibility. The innovation developed during the project was weighed against its novelty compared to existing practices (Schilling, 2010), resulting into the classification of radical and incremental NPD projects. Looking at table 5, some dimensions were applied more in either radical or incremental project, which was supported by the data. In general, it was found that radical NPD projects apply a larger number of flexibility dimensions during the project than incremental projects apply. With regards to incremental innovation, it seems more typical to apply design flexibility; whereas radical innovation appears to be typically characterized by organizational, informational, and resource flexibility. The remainder of this section will discuss these dimensions and their connection to innovation type.

Design flexibility. Not only is design flexibility considered taking place more in

Project IU-1 IU-2 IU-3 IS-1 IS-2 IS-3 Innovation Incremental Incremental Incremental Incremental Incremental Incremental

Result Unsuccessful Unsuccessful Unsuccessful Successful Successful Successful Organizational flexibility

-

-

-

-

-

-

-

+

-

-

+

+

+

-

+

+

-

+

-

+

+

+

-

+

-

-

+

+

-

+

-

+

-

+

-

+

-

+

+

+

Table 5. Incremental versus Radical NPD projects

Project RU-1 RU-2 RU-3 RS-1 RS-2 RS-3

Innovation Radical Radical Radical Radical Radical Radical

Result Unsuccessful Unsuccessful Unsuccessful Successful Successful Successful Organizational flexibility

-

-

+

+

+

+

+

+

-

+

+

-

+

+

+

+

-

-

-

-

+

+

+

+

+

+

+

-

+

-

+

+

+

-

+

+

+

+

“Flexibility in the sense, that if you want you can make all sorts of derivatives, so different colors, versions, without innovating in many areas. What we have tried with [IS-3] is to actually build three different appliances, which we can make in all kinds of colors, with different attachments, and different types of packaging. And we have, what we call, a suitcase, a small box where you can put in the appliance. And between all those ingredients, which stand in the closet so to say, the [marketing manager] can pick what he or she wants. That is thinking in modularity. For example, for England, France, and Germany, we have very specific types that only go to that country, with many attachments, or a special color, or something similar. For China we developed three new products in three months, so only altering the colors. That sort of flexibility is now built in the appliance.”

With regards to radical NPD projects, design flexibility is considered less relevant. As radical projects are a first generation of new products, modules are not yet available. The project manager described the situation during RS-3: “What was the case here, and what is often the case, is that you do the module development almost parallel, or slightly delayed, with the product, so you cannot make use of the modules. Then it would already be, what we use to call, a second generation, a derivative.” Consequently, all elements in the product are very much interrelated. Or as the development engineer of the radical RU-1 project described it: “because everything has interaction with everything, an extension of the function will affect all elements”

Organizational flexibility. The case selection took place at a company where a default

process has been defined for innovation projects. So, in principle, NPD projects will follow this process, which is highly structured by means of a multitude of clearly defined milestones where for example officially resources are committed and concepts are frozen. Additionally the stages between the milestones contain designated activities and deliverables for the project. And in general the process is followed in most projects, however, innovation, and especially radical innovation contains a degree of uncertainty, and therefore there remains a degree of flexibility in the process related to “the degree of structuration in the process design” (Biazzo, 2009).

activity and deliverable was assessed to consciously decide to execute or not, or alter the activity. Additionally, radical projects often receive more management attention due to their importance for the future of the business. This was also true for the cases studied here; this is what the project manager of the RU-3 expressed about the process: “Eventually it got the shape, which was steered [by management], to combine everything: proposition development, building the technology, developing testing methodologies, everything combined, which is quite exceptional”. This also can be found back in their critical path planning, where many of the key activities were altered because of being combined with other activities. To conclude, the following fragment exemplifies how organizational flexibility took place during RS-2:

“Another thing is maybe not necessarily the team, but more the process. A few

times we had to change the development process as it is defined, for example strengthened requirements management, adjusted our change control, changed the entire way of working to transform requirements to failure mode and effects analyses to control plans to the factories, and we developed complete methodologies for availability planning of parts. I do not know whether that fits in team flexibility, but it does fit in process flexibility; so a few times we adjusted the process as it is defined in default to the specific requirements of this project.”

Informational flexibility. From the analysis it becomes clear that informational flexibility

“Yes, [the requirements] were set well in the beginning. It regards a product series which is relatively known already. So known grounds, thus [requirement changes] were less needed.”

“Everything was very clear, but they were clearly question marks! So absolutely unclear, because we had no reference material at all to set the initial requirements”

And secondly, due to the novel nature of this type of NPD project, more turbulence can be expected in terms of both market and technological information. For instance, during consumer testing for the RU-1 project it was discovered that certain plastics in the product corroded when in touch with particular cleaning detergents used by some of their potential customers. The milestone report after the test also reported many changes at the product level as a consequence of the consumer testing. To conclude, in different words, many interviewees supported the following statement made by one of the interviewees with regards to informational flexibility: “I believe that the degree of flexibility needed is dependent of how precise your problem definition is, and on the knowledge you have about it in the beginning, so about the scope.”

Resource flexibility. “The flexible reallocation of project resources”, in for instances

materials and personnel (Tatikonda and Rosenthal, 2000), is another dimension that seems to be more typical for radical than for incremental projects. The novelty of radical projects suggests that the scope of the project may change more (in part due to informational flexibility), which resulted in a changed need for resources in the cases studied here. This is for instance what the

RS-2 project experienced: “With [that scope change] the throughput time and cost level of the

evaluation and learnings). As a result they had to, among others, throw away and replace molds worth hundreds of thousands of euros.

Another aspect typically found at radical projects is that they receive significant management attention. These projects have high expectations as they are typically larger projects and are supposed to bring new opportunities to the firm. Receiving this attention often means that there is more room for flexibility in terms of the resourcing of these projects. Therefore many interviewees emphasize the importance of proper stakeholder management during the NPD projects, and in particular with regards to radical innovation. One of the project managers about stakeholder management and resource flexibility: “Without the trust [of the stakeholders] we would have never received the budget to add these resources, thus, without that the project would not have become such a success.” To conclude, the following fragment describes how the project manager of the RS-1 approached the combination of having higher project risk and the importance of the project for the business (which later has been captured in a tool for repetitive use):

“I identified all the key risks on the basis of the parameters I used to do the project quicker; so these are the potential risks you run by doing it that quickly. It resulted in an impressive list of risk, which I put on a one-slider and aligned with the stakeholders. There I said to them, if you want to go that fast, that is possible, and what we can achieve with that is two or three million extra profit. But if things go wrong then it will cost this. And they said ‘go’”.

“Stakeholder management; when you do such a project and something goes wrong and you need for example 700.000 dollar or euro, and you have not prepared for it, it will completely fail, guaranteed.”

Main Actors in Project Flexibility

included in the project, since a project is a temporary set-up that may cross functional boundaries (PMI, 2015), and therefore also need to contribute in a flexible manner to achieve the necessary flexibility during the project. In general, flexibility should reside in the entire value chain that a project crosses; however, in the cases studied here, especially upper management and supplier flexibility were argued to be of significant importance, which will be detailed in the next sections.

Management. The results discussed before already hint on the importance of flexibility

from the side of management, which also needs to display flexibility. The success and the progress of a NPD project are in addition to the project team itself, partly dependent on the management level above the project, for example by approving the project to progress to the next phase. Moreover, flexibility from upper management is often needed to enable the flexibility dimensions mentioned before. For example, in case of an extra need for resources in the project, management approval may be needed; this is also what the project manager of RS-3 experienced: “For the project we made a lot of development costs that year, so you must get the necessary approvals in time”. During the RU-1 project less flexibility was experienced from the management side, as becomes clear from the following fragment:

“[The management team] was a lot less flexible than we as a project team, for instance about the launch date, etcetera. That you can also see as flexibility, that everybody should give and take. Giving was not a lot the case at that time, everybody insisted, ‘this is mandatory, this should be done’.”

Suppliers. Another important stakeholder involved during the project was the supplier,

Additionally, it is also a matter of giving and taking in the relationship with the supplier(s); the supplier also has to earn his bread so to say. To manage this interdependence, all interviewees engaged in ‘supplier development’, to build and strengthen the relationship and trust, which are essential to this type of flexibility. For instance during the RS-3 project, the project manager spend two months at the supplier, with on and off other team members joining him. To conclude, the following fragment is a good example of supplier flexibility where the supplier helped out during the IU-3 project.

“If you look at our external suppliers … they really have to be flexible. You could see that at our mold supplier. We were running behind schedule, still busy with our drawings, while we should have started with the molds already. What they have done for us, is to already buy a significant piece of iron, which brings along quite some risks. And whatever was agreed upfront, small changes … were also incorporated in between, where they indicated upfront, these are the costs approximately, ‘do you agree’? ‘Yes ok’; and then the process also went very fast, and we got prices as indicated. It is very important that external suppliers go along [with the project].

DISCUSSION

Successful versus Unsuccessful NPD Projects

Looking at the current state of flexibility literature, flexibility has been portrayed as advantageous in any case (e.g. Biazzo, 2009; MacCormack et al., 2001), yet, more nuances seem to be needed based on the theory developed in this research. Indeed, the successful cases studied here showed a larger number of flexibility mechanisms than their unsuccessful counterparts. However, some dimensions were identified to be more characteristic for the different types of cases. This distinction could not be found in prior studies, and therefore opens interesting opportunities for future research. The propositions derived from the analysis between successful and unsuccessful NPD projects are described below.

P1: The number of flexibility dimensions applied is positively associated with

NPD project performance

P2: Design flexibility is positively associated with NPD project performance P3: Mindset flexibility is positively associated with NPD project performance P4: Informational flexibility is negatively associated with NPD project

performance

Two flexibility dimensions are proposed to be positively associated with NPD project performance. First, design flexibility was found more typical among the successful projects; many of the advantages of design flexibility were recognized by previous studies: the minimization of system interdependencies enhances flexibility (Thomke, 1997), and it enables concurrent and simultaneous development of components (Sanchez, 1996), and a “flexible design may be used to leverage product variation” (Sanchez, 1996) The important industrialization argument, however, has not been discussed yet in relation with design flexibility in previous literature. Second, the newly identified dimension of mindset flexibility is argued to be related to successful projects as well. The state of literature to date has not reported any similar dimension, whilst the flexibility of the mindset, as the ability to deal with the turbulence during a project, provides the motivation and ability to process and act on the continuous flow of new information.

information during the project improves performance (e.g. MacCormack et al., 2001). According to Biazzo’s (2009) framework most cases studies here would be classified as highly structured and low in informational flexibility, resulting in ‘structured rigidity’. However, the cases that did demonstrate higher levels of informational flexibility would then be considered as engaging in ‘structured flexibility’ (Biazzo, 2009), yet, this flexibility was not perceived as such, but rather as required reactive flexibility to the new information because requirements were set poorly in the beginning of the project. Consequently, this view may be more nuanced, as this study proposes that if requirements are set poorly in the beginning, informational flexibility may not be beneficial. To conclude, the propositions with regards to NPD project performance are summarized in figure 1.

Figure 1. Propositions Related to NPD Project Performance

Radical versus Incremental NPD Projects

This study is one of the first to include both incremental and radical innovations on the project level in the research scope. On a higher level it is proposed NPD project innovativeness is positively associated with the number of flexibility dimensions applied. As radical innovations typically experience more turbulence, and most studies so far have focused on turbulent environments (e.g. Biazzo, 2009; Iansiti, 1995; Buganz et al., 2009), this proposition is in line with most research performed in this field. However, a direct comparison between incremental and radical projects remained absent till now and turbulent environments may not necessarily be associated with radical innovation only. This study offers more nuance to this matter; it is proposed that design flexibility is typical for incremental NPD projects, and organizational, informational, and resource flexibility to be more characteristic for radical NPD projects. These propositions with regards to flexibility and innovation type will be further elaborated on below.

NPD Project Performance

Unsuccessful Successful Design flexibility

P5: NPD project innovativeness is positively associated with the number of

flexibility dimensions applied

P6: NPD project innovativeness is negatively associated with design flexibility P7: NPD project innovativeness is positively associated with organizational

flexibility

P8: NPD project innovativeness is positively associated with informational

flexibility

P9: NPD project innovativeness is positively associated with resource flexibility

From the data analyses it is proposed that NPD project innovativeness is negatively associated with design flexibility, hence, the more radical the project, the less applicable design flexibility becomes. Considering Schilling’s (2010) definition of incremental innovation, it is a logical consequence that design flexibility plays a more significant role for projects of that type, as it facilitates product variation (Sanchez, 1996). This study is, however, the first to make this distinction between innovation types in terms of flexibility on the project level. Consequently, it is logical for more radical projects to seek for other ways of mitigating the risk of changes during the project (Thomke, 1997).

Figure 2. Propositions Related to NPD Project Innovativeness

Managerial Implications

Next to the theoretical contributions this paper also offers managerial implications which are twofold. The first set of implications relates to the application of the different flexibility dimensions in practice. So far, flexibility scholars have preached flexibility in new product development, suggesting it is of added value in any case. This may be true, yet, also more nuanced according to this study. By identifying the dimensions that are related to the innovation type (incremental or radical) or the project outcome (successful or unsuccessful), managers will be able to better judge and guide the flexibility needs of a particular project. As a result resources can be spent more effectively and efficiently to react on the environmental turbulence, and increase the chances of project success.

Secondly, in terms of project level flexibility the project dynamics have not received the appropriate attention yet. According to the PMI (2015) a project is “unique in that it is not a routine operation, but a specific set of operations designed to accomplish a singular goal. So a project team often includes people who don’t usually work together – sometimes from different organizations and across multiple geographies.” Accordingly, it has been become clear in this study that project level flexibility was not only a matter of concern of the project team in the cases studied here; important roles were also played by the relevant managers in the layer above the projects who have decision rights concerning the project, and the suppliers who are needed to develop and industrialize the product. Moreover, to a certain degree flexibility is needed from all involved stakeholders in the value chain that the project covers, with some stakeholders playing a more significant role than others in certain projects. This has important implications, as it is apparently not only a matter of how to deal with flexibility, but also with whom. This study may support management in involving the right parties and to enable flexibility from their side.

Limitations and Avenues for Future Research

Just like any other research, this study has its limitations. First, in order to connect the flexibility dimensions to project outcome, data was collected from projects that had recently reached closure. Future scholars, however, may want to collect real-time data and see the flexibility dimensions in practice. Second, measures were taken to increase the rigor and quality of this research, yet, future studies may include multiple informants from different functions as opposed to single key informants. Third, project outcome has been defined from the firm’s point of view; future research may want to take a customer perspective to define the success of the projects, as flexibility provides the ability to adjust to their needs and preferences. Fourth, the definition of flexibility applied in this study includes the turbulence of both technological and market information, however, the differences in how projects deal with the types of information provide interesting opportunities. Fifth, this study investigated to what degree the identified flexibility dimensions took place in different project types, yet, more attention could be paid to whether and how these flexibility dimensions are applied differently among the cases. Sixth, during data collection, interviewees often referred to project management practices in their replies. Hence, it would be of added value to connect these two fields of theory more in the future. And lastly, the propositions offered remain untested in this study; therefore future studies are invited to quantitatively measure the strength of the flexibility dimensions more accurately and link the flexibility dimensions with NPD project performance and innovativeness.

CONCLUSION

REFERENCES

Biazzo, S. (2009). Flexibility, structuration, and simultaneity in new product development. Journal of Product Innovation Management, 26 (3), 336-353.

Buganza, T., Dell’Era, C., Verganti, R. (2009). Exploring the Relationships Between Product Development and Environmental Turbulence: The Case of Mobile TLC Services.

Journal of product Innovation Management, 26, 308-321.

Calantone, R., Garcia, R., & Dröge, C. (2003). The Effects of Environmental Turbulence on new Product Development Strategy Planning. Product Development & Management

Association, 20, 90-103.

Corbin, J., & Strauss, A. (2008). Basics of qualitative research (3rd edition). Sage publications.

Eisenhardt, K.M. (1989). Building theories from case study research. Academy of

Management Review, 14 (4), 532-550.

Garcia, R., & Calantone, R. (2002). A critical look at technological innovation typology and innovativeness terminology: a literature review. Journal of product innovation management, 19(2), 110-132.

Gerwin, D. (1993). Manufacturing Flexibility: A Strategic Perspective. Management

Science, 39 (4), 395-410.

Iansiti, M. (1995). Shooting the rapids: managing product development in turbulent environments. The Journal of Product Innovation Management, 13 (5), 457-458.

Kandemir, D., & Acur, N. (2012). Examining proactive strategic decision‐making flexibility in new product development. Journal of Product Innovation Management, 29 (4), 608-622.

Koelsch, L. E. (2013). Reconceptualizing the member check interview. International

Journal of Qualitative Methods, 12, 168-179.

Kok, R. A., & Ligthart, P. E. (2014). Differentiating major and incremental new product development: The effects of functional and numerical workforce flexibility. Journal of Product

Innovation Management, 31 (1), 30-42.