Bridging the gap: from failure to success.

(1)

Bridging the gap: from failure to

success.

Best practice study of Dutch companies in innovation

management

Name:

Ruud Egberts

Date:

25-03-2010

Institution: University of Groningen

(2)

Bridging the gap: from failure to success.

Best practice study of Dutch companies in innovation management

Author: Ruud Egberts

Student number: 1419625

Email address: r.h.egberts@student.rug.nl

Faculty of Management and

Organization

Master Strategy and Innovation

Supervisors: Dr. W. Dolfsma

Dr. H. Snijders

(3)

Acknowledgments

I started in the summer of 2008 with my internship at The Bridge business innovators. They gave me the opportunity to take a look into a business environment and how life was like as a consultant. I was very happy to be part of the team that made the innovation-monitor 08/09 to a success.

Now almost two years later my academic period has come to an end. The last hurdle to complete my master degree was also the biggest and hardest one. Since this research paper is about success factors it is a bit strange to conclude that writing my master thesis wasn’t much of a success at all. I’ve identified some success factors while writing my thesis:

- Make clear deadlines, over and over again and stick to them. - Pick a subject that has your interest

- Don’t postpone writing it, see it as work. - Seek help when almost going crazy.

I would like to thank my supervisors at the university of Groningen, my supervisor at the Bridge and the rest of the Bridge business innovators for their feedback.

I would especially like to thank my friends and family and in special my girlfriend Nicole for believing in me and supporting me during the journey with some ups and a lot of downs. A study is one thing, finishing it is even more difficult. But finally I did it.

Thanks everyone!

(4)

Summary

In this quantitative study the goal is to come to a general understanding of success factors of innovation management in Dutch services and manufacturing business. With help of three year of questionnaires from the Bridge business innovators, statistical analysis were made.

The intention of this study is to find similarities with the five main success factors in innovation management and the Dutch industry. The study is taken on the organizational level and is based on the five success factor model of Cooper (1995).

The results show a link between success factors described in the literature and the Dutch industry. Several general important success factors were generated: management involvement, portfolio management, and culture and climate.

Also a context specific dimension was added, namely service based companies and manufacturing based companies. Main result from this exploration was the difference in the formalization of the innovation process.

Keywords

(5)

1. Introduction

Management research confirms that innovative firms – those that are able to use innovation to improve their processes or to differentiate their products and services – outperform their competitors, measured in terms of market share, profitability, growth or market capitalization.

Reorganizing and reengineering the innovation processes and structures is an endeavour in which management can seek endless improvements. The first step in any such redesign activity is to first understand the critical success factors. Benchmarking other firms often can provide in this insight (Cooper, 1995). This research will broaden the current literature about success factors in innovation management. It can be used as a benchmark for firms whose focus is on creating a competitive advantage by improving their current innovation practices.

1.1 Background

There are numerous studies which have investigated the success factors for both new product and new service development. But, the majority of studies focuses on either services or products in stead of both. Also the main stream of research is focused on new product development. And if talking about services, research is mainly focusing on financial services.

Therefore it is important to have a mix of both product and service companies investigated, because many organizations manage a portfolio of a mixture of both tangible and service products (Slack et al, 2004). To determine the overall success factors of a company it is therefore important to include both service and manufacturing companies.

Besides, the majority of research is focused on the project level (Cooper & Kleinschmidt, 1995). These success factors concern for example product advantage, project synergy, and process related service. Successful projects are then compared to failed projects in order to distinguish the defining factors for success.

From the two points mentioned above it can be said that two points lack in current research; including both services and manufacturing companies in the research, so differences can be revealed. And second, there is need for a broader vision, or macro view of the determinants of success.

This broader, company level view, is different from the project level in two ways (Cooper & Kleinschmidt, 1995):

1. Success at the company level may be somewhat different than on a project level; 2. There may be factors that are important for success but are missed at the project level. Therefore this research will have the unit of analysis of a firm instead of a project and measures for performance are gauged at the company level.

(8)

I will shortly explain the origin of the data, and continue with the main questions that will be answered in this study. Second, the theory will be build upon the current literature studies at the company level success in innovation management focusing on the framework by Cooper & Kleinschmidt (1995). This framework will be evaluated and complemented with help of a literature study and data. Finally, conclusions, limitations and implications for further research will be drawn from the results.

1.1.1 Origin of the data

This research is build upon three years of study into the Dutch field of innovation management. With help from the Dutch consultancy company ‘The Bridge business innovators’ the data is retrieved among more than three hundred companies dealing with innovation issues. The most current issues are monitored each year and are analyzed and presented in an ‘innovation – monitor’.

The data behind the monitors of 2007, 2008 and 2009 are used to select the questions revealing the success factors in innovation management. The consultants at The Bridge put their experience to the test of the Dutch companies through an online questionnaire asking the companies how they feel about the issues currently playing a defining role in innovation management. The participation was open, but mainly companies within the relation network of The Bridge participated in this questionnaire. How the data is build up, its content, and limitations will be further explained in the research methodology section.

I will now continue with the main part of this research.

2. Design and theory

This section will discuss the research questions, the conceptual framework and will continue with the theory discussion.

2.1 Research questions

In this research the main goal is to find success factors for innovation management on an

organizational level. The focus will lay on the new product development (NPD) and new service development (NSD).

(9)

In the first part of this research is chosen for a dyadic approach: comparing success with failure (Montoya-Weiss & Calantone, 1994). In the first part I want to figure out the differences between successful and less successful Dutch companies. The question I want to answer is on which factors the company’s overall NPD and NSD performance is dependent. From this question the research

questions are set up. Since this research is narrowed by the origin of the sample, one can’t generalize this sample to a whole population. The respondents all are from companies active in the Netherlands. So the scope is restricted to companies in the Netherlands. The further distribution of the sample is explained in the methodology section.

In the second part I want to explore the differences between service based and manufacturing based companies, since literature suggests that they might differ in type of success factors (Griffin, 1997).

Cooper & Kleinschmidt (1995) have developed through years of research five building blocks that will serve as the framework for this research. Since it is impossible to include all possible factors, this model will serve as a starting point. Other relevant factors additional to the model will be discussed in the literature part.

According to Cooper & Kleinschmidt (1995) these building blocks determine the performance of the company’s overall new product development.

These building blocks are:

1) Process: the firm’s new product development process and the specific activities within this

process.

2) Organization: the way the program is organized

3) Strategy: the firm’s total new product strategy

4) Culture: the firm’s internal culture and climate for innovation

5) Commitment of senior management: senior management involvement with and corporate

commitment to new product development.

The aim of this research is to look at comparison between the five defining factors in the model of Cooper and the factors that are discussed in the data of the innovation monitors.

The main question that will be answered in this research are:

1. Which of Coopers success factors of overall NPD can be applied to the Dutch service and

manufacturing industry ?

1.1 Sub question is if there are additional factors that could be added/excluded to/from the five factor

model.

The alterations in the model will result in a framework for the empirical analysis of the model, which will focus on the next two questions:

2. How do successful Dutch companies differ from unsuccessful companies in the performance factors

(10)

3. How do Dutch service based companies differ from manufacturing based companies in the

performance factors of innovation development?

2.2 Theory discussion

Innovation management often is about managing the process of innovation. This process is described in the beginning as a pool of ideas that needs to go through a funnel where the right ideas are selected. Next the selected ideas and concepts will be further developed and last will be commercialized. This is the well known ‘innovation development funnel’ by Clark & Wheelwright (1993). Many scholars have adapted this model and used it to describe the necessary steps that need to be taken to make innovation management seemingly successful. Portfolio management covers the process of innovation by being a framework for making decisions about the current projects in the organization (Cooper, 1999). This NPD or NSD process can be seen as one of the main influencers of the overall innovation performance of the organization. Cooper (1995) has combined several literature studies into five factors that the company’s overall new product performance depends on.

1) The new product development process; this process is often captured in the already named innovation development funnel. The activities that comprise the new product process are strongly associated with project outcomes. Correlated with success are for example (Cooper & Kleinschmidt 1995, p. 377): having strong market orientation, undertaking the marketing tasks in a quality fashion, doing the predevelopment activities well (the homework), and having sharp, early product definition before product development begins.

2) How the firm organizes for new products: this factor is referring to the success on new product development of cross functional teams. Larsson & Gobeli (1988) also note in this respect the

importance of an empowered project leader. A product champion is also often mentioned as a success factor (Cooper & Kleinschmidt, 1995).

3) New product strategy: ‘An explicit product innovation strategy enables management to plan for and to make available adequate resources for specific product developments’ (Cooper & Kleinschmidt, 1995 p. 377). Besides, having an explicit new product strategy results in a more successful new

product program (Cooper, 1984).

4) Culture and climate: A positive culture and climate for new product development is vital to successful product development, some examples are: support teamwork, encourage the employee submission of new product ideas (Cooper & Kleinschmidt, 1995)

5) Senior management’s involvement and corporate commitment: management’s role is very important for example in sending ‘clear messages from senior management about the role and importance of new product development’ (Cooper & Kleinschmidt, 1995).

(11)

project/product. But overlapping success factors exist in this literature study with the model of Cooper. On the firm level the following success factors were distinguished:

- Firm culture;

- Characteristics of the R&D team;

- The firm’s strategy towards innovation; and - Experience with innovation;

The first three also correspond with the model of Cooper. The last (experience with innovation) refers to a learning effect from previous endeavors. Faster introduction of innovations is one advantage coming from previous experience in the development of innovation projects.

Van der Panne & Beers (2003) see the management’s innovation style as a project level success factor. Management’s involvement is mostly on a firm level since management makes decisions concerning the whole firm and consequently the innovation projects. The innovation management style is parallel to the new product development process. Van der Panne & Beers (2003) mentioned splitting up projects in phases. Managing innovation in phases is again linked to the innovation development funnel in which each phase can be managed separately.

The innovation development funnel is referring to the ‘process’ factor Cooper describes. This will come forward in the next section.

2.2.1 The innovation development funnel

The innovation development funnel can be looked at from two aggregation levels, namely at the individual project level and the organizational level. An individual innovation project runs through different phases from the idea to the implementation of the product or service. Cooper (2001) describes a so called ‘stage gate system’ in which a (new product) project is separated in stages and between these stages gates are used to make go or kill decisions. A gate needs to be opened (by certain criteria being met) to continue to the next stage. Cooper (2001) distinguishes the new product process in six phases:

1) Discovery: pre-work designed to discover and uncover opportunities and generate ideas; 2) Scoping: a quick, preliminary investigation of the project- largely desk research;

3) Building the business case: a much more detailed investigation involving primary research- both market and technical- leading to a business case, including product and project definition, project justification, and a project plan;

4) Development: the actual detailed design and development of the new product, and the design of the operations or production process;

5) Testing and validation: tests or trials in the marketplace, lab, and plant to verify and validate the proposed new product and its marketing and production/operations;

(12)

At the organizational level Goffin & Mitchell (2005) and Tidd, et al. (2005) use the innovation funnel idea to discuss the different management phases of innovation within a company. Tidd et al (2005) have adopted a model of three main phases namely: search, select and implement, with the last phase being separated in four sub phases: acquire, execute, launch and sustain (see figure 1). The search and select phases are seen as strategic phases which shape the framework for innovation. The implement phase is where the ideas are actually converted into a product or service.

Goffin & Mitchell (2005) describe the organizational process of innovation in their innovation pentathlon framework (see figure 1).

Figure 1.Left: Innovation process model by Tidd et al. (2005) Right: Innovation Pentathlon Framework by Goffin &Mitchell (2005).

Both models show three main phases:

The first phase is the idea or search phase. The first challenge for a company in the innovation process is to widen the funnel and to explore as much new ideas and concepts as possible (Clark &

Wheelwright, 1993). It is important for the company to understand which factors create the environment of the company and so to create the right systems for gathering information from this environment. The new products or services can either come by customer needs (pull forces) or by internal drive within the company (push drivers) (Goffin & Mitchell, Tidd et al, 2005).

The second phase is the selection or prioritization phase. The opportunities arising within the

environment must be selected according to certain criteria, set within the business strategy. The funnel so to say, must be ‘narrowed’ (Clark & Wheelwright, 1993). For the success of the company this is an essential phase, since the right decisions must be made, dealing with the uncertainty and risks the projects withhold. In this phase portfolio management is introduced in both models. Although portfolio management is meant to govern all the phases in the innovation funnel it is most present in the selection/prioritization phase. The reason for this is that the criteria set for the innovation projects are used to either decide to continue, stop or alter the project. This is the essence of portfolio

management; making decisions on individual projects based on strategic criteria (Cooper, 2005, Goffin & Mitchell, 2005, Tidd et al, 2005).

(13)

commercialization stage is the last in the implementation phase according to Goffin & Mitchell (2005), Tidd et al. (2005) see this phase in more detail, looking at it from the individual project level, which is similar to the NPD phases of Cooper (2001) described earlier. Implementation means in their case actually converting an idea in a new product, service or process and also implementing it.

First it is important to acquire the necessary knowledge resources, combining new and existing knowledge and finding an answer to the problem. Second is the execution of the project, which is the hart of the innovation process. Good project management, meaning considering all parties involved and dealing with all the information available to make the project go from idea to launch is very important. Third is preparing the market into which the new innovation will be launched. It involves collecting information about customer needs and feeding this into the product development process (Tidd et al, 2005). The last phase is about the adoption and use of the innovation in the long term- or revising the original idea and modifying it (re-innovation). This is the important learning phase where projects are evaluated looking for improvements within the process (Goffin & Mitchell, 2005).

2.2.2 Discussion of the models.

Looking at both models Tidd et al (2005) clearly take a learning approach. The goal of the innovation process is to learn from each process and feeding this information back into the process. Goffin & Mitchell also recognize this, but see it more as an end phase, evaluating the final innovation project, and consequently learning from it.

Clear difference between the two models is the role of the innovation strategy. Goffin & Mitchell see the innovation strategy as the shaping element, which influences all the other elements of the

pentathlon framework. This is in line with the literature of portfolio management (Cooper et al, 1999, Archer & Ghasemzadeh, 1999). This literature sees an important role for the innovation/ business strategy, since innovation projects must be aligned with the strategy. The reason for this is that decision making about projects will become clearer, making the right decisions in terms of the overall strategic course set out. Thus, the innovation strategy (as being part of the overall strategy) has the function of creating a framework for innovation within the company. Tidd et al, consider that the innovation strategy is formed along the way. They see it more as a part of the select phase within their model, where strategy making is enabled.

(14)

1997), and the rest of the innovations are clearly more incremental innovations, which can be fairly well managed.

Cooper (2005) expands the role of portfolio management in the new product development (NPD) literature from a project management tool to an organizational governance model that covers all the phases in the process of innovation management. This becomes clear from his definition of portfolio management which is acknowledged in literature as the most comprehensive one (Archer &

Ghasemzadeh, 1999, Artto, 2001)

2.3 Portfolio management

Cooper et al. (1999) defines portfolio management as: “a dynamic decision process, whereby a

business’s list of active new product (and R&D) projects is constantly updated and revised. In this process, new projects are evaluated, selected, and prioritized; existing projects may be accelerated, killed, or de-prioritized; and resources are allocated and reallocated to the active projects.” “The portfolio decision process is characterized by uncertain and changing information, dynamic opportunities, multiple goals and strategic considerations, interdependence among projects, and multiple decision makers and locations.”

“The portfolio decision process encompasses or overlaps a number of decision-making processes within the business, including periodic reviews of the total portfolio of all projects (looking at the entire set of projects, and comparing all projects against each other), making go/kill decisions on individual projects on an on-going basis (using gates or a stage-gate process), and developing a new product strategy for the business, complete with strategic resource allocation decisions” (In: Cooper et al., 1999, p.335).

2.3.1 Portfolio issues in the innovation process.

The key issues in portfolio management that must be considered are (Cooper, 1999, Goffin &Mitchell, 2005):

- Valuation criteria:

1. Each project should add value to the total portfolio;

2. Allocation of resources must be made clear throughout the organization; - Portfolio balance criteria:

(15)

4. The portfolio must fit and respond to the company’s strategic needs. Perfectly good projects may have to be delayed or aborted in favor of others more strategically important.

Since an organization’s innovation process starts with a strategy and follows with stating the criteria that projects must meet to, this will be discussed first.

2.4 Strategy

The relation between strategy and success is, according to Ernst (2002), still not well understood. The reason for this can be explained by the early research of Cooper (1984, ab)

According to Cooper the new product strategy is focused on the individual product level instead of the total new product program of the company. This narrow focus often resulted in the logical outcome prescriptions of a ‘market-driven, conservative new product program, one that emphasizes incremental innovations and product modifications’ (p.6). Incremental innovations are less risky and therefore will also have a higher rate of new product success. But this will only work on the short term. In order to be competitive in the long run more radical innovations will be necessary to survive. This short term vs. long term strategy can be translated into a portfolio model of a balance between long vs. short and high vs. low risk.

The research of Cooper (1984) makes a clear statement. If you want a successful strategy focused on the long run a more aggressive tactic is required focused on innovation. Cooper (1984a) distinguishes between a high impact program and high relative performance. Both are focused on a broader context instead of the narrow focus on the success of individual products. A high impact program is focused on a high percentage of sales by new products/services. High impact programs need a heavy R&D spending, and entering markets that are not necessarily synergistic with the firm’s marketing resources (Cooper, 1984). High relative performance programs are more balanced: one that features an

aggressive strategy but with a strong marketing orientation. This means a proactive stance, seeking products in high potential, broad markets.

(16)

disruption can result in a transformation on the respective level (company, industry or worldwide). “Incremental innovations enhance existing practices or make small improvements in products and processes” (Dess et al, 2005, p. 411). When the innovation is more incremental it is often a change in an existing product or service.

The first assumption that can be made resulting from the literature above is that, it can be expected that companies, when looking at a higher relative performance and high impact programs (with a high performance on making new products/services part of the revenue), will be aimed at a more radical approach or attitude (H1a) and will also give innovation a higher priority i.e. will invest more in innovation (H1b).

In line with Cooper this research will also focus on a broader context when talking about performance in innovation. The term ‘performance’ needs to be clarified in order to make a distinction between the best performers and the rest.

2.5 Innovation performance

Innovation performance can be measured in many possible ways. It can, for example, be concerned with input management. Input management is about the resourcing of innovation activities. The construct research and development (R&D) intensity is often used as a global measure of input (Adams et al, 2006) but is also a measure which neglects the output side of the innovation activity. The simple conclusion of companies that invest more in research and development (and innovation activities in general) are more successful, is therefore doubtful.

On the output side innovation performance can also be measured. In this respect most researchers use total expenditure, expenditure expressed as a proportion of sales or revenues, and expenditure by item (Adams et al, 2006). Although these output indicators have its drawbacks, they can be easily

quantified and therefore be compared within a research population. Highly innovative firms are expected to have a higher percentage of sales from new and improved products and services (Rogers, 1998). The measure is often quantified in percentage of revenues coming from new products and/or services in the last 3 to 5 years (Cooper, 1984b, Griffin et al, 1996)

Griffin (1997) distinguishes between overall success of the company by measuring the ‘position in the industry’ (most successful, top 1/3, middle 1/3 or bottom 1/3) and market and financial success which is measured as percentage of revenues.

(17)

Portfolio management success is less based on quantitative outputs and more on qualitative. Cooper et al. (1999) have done their research asking managers about their opinions about there current portfolio management aspects. Cooper et al. (1999) identified the main drivers behind the performance metrics of portfolio management. They also found in earlier research that management perceptions and satisfaction metrics have an intermediate role in this. First there are four sets of variables that seem to drive the performance metrics and management perceptions and satisfaction metrics. These drivers include:

-How important portfolio management is perceived to be in the business;

-The reasons why portfolio management may be important; why more formal approaches are adopted in the organization;

-The nature of the portfolio management methods used by the business; for example how explicit and formal the method is;

-The specific portfolio models or tools used by the business; for example financial versus strategic methods.

The research also identified two main factors or dimension of portfolio management satisfaction and perception of their portfolio management methods, namely:

1)Overall quality rating; including having a realistic portfolio method, management would recommend their portfolio method to others, method is rated as excellent by management, method is used for making go/kill decisions, method is user friendly and understood and last the method makes the right decisions- is effective;

2): Management fit; including the portfolio method fits management’s decision-making style, management rate the method to be efficient (does not waste time) and method is understood by management.

So the success results are more based on the perception of the manager rather than on quantitative output. By discussing the relation between portfolio factors perceived important by managers and innovation success (measured as output) both qualitative and quantitative measures are combined. In this research success is measured on the basis of financial and relative (commercial) performance. The reason for this is that in the end the success of a project in the shape of a product or service is determined by its revenue.

2.6 Linking the projects to strategy

(18)

Selection frameworks helping the decision making about projects all start with stating that the strategy should be leading in the process (Archer & Ghasemzadeh, 1999, Cooper, 2005). This phase is also referred to as the ‘strategic considerations phase’. The strategy is used to setting aside funds

(envelopes or buckets of money) destined for different types of projects. This is also referred to as the top-down approach (Cooper, 2005). Strategic criteria can also be included in the go/kill decisions within the stage gate system. This means that every time a project wants to pass a gate it has to meet the strategic criteria. This is more bottom-up.

Cooper et al (1999) found in there research that benchmark companies, which are the companies that scored the best on the quality of the portfolio and the management fit, scored high on the two strategic metrics:

- having a portfolio of projects that are aligned with the business’s strategy, and - having a portfolio whose spending breakdown mirrors the business’s strategy and

strategic priorities.

Companies that use strategic criteria in the decision making process are considered to have better aligned projects. Better aligned projects are a strong guarantee for the success of a company. Our next assumption is that better performing companies use more explicit strategy criteria when assessing the value of their innovation projects.

H2: Better performing companies use more explicit strategic criteria when assessing innovation

projects in comparison to lower performing companies.

2.7 Role of top management

The role of top management in portfolio management is very important. Top management is required to set the direction of the company and holds the responsibility of monitoring the portfolio (Goffin & Mitchell, 2005). This is also one of the important influencers of a successful NPD and NSD process (Cooper, 1995). How to balance your portfolio is also very important (Cooper et al, 1999). Knowing what is going on, considering all projects together (which is the essence of portfolio management), and get rid of the focus of micromanaging projects will result in a better representation of the strategic focus of a company in the current projects (Clark & Wheelwright, 1992, Cooper et al, 1999). Clark & Wheelwright (1992) therefore promote the so called ‘aggregate project plan’ which can be an

(19)

decisions about which projects to continue and which to stop. Cooper (2002) often suggests using a matrix to make the current projects visible to the managers.

This formal list of projects is often the start of a portfolio management system. It enhances more balance among the projects, resources can be more easily distributed and strategic coherence is faster achieved (Cooper, 2002). Blichtfeldt & Eskerod (2008) mention some important failures concerning project portfolio management. Most importantly top management has the task of making sure

resources are devoted to the most important projects. What often is missing is a clear overview of how many projects are actually running in the company. Therefore projects unknown by the general management can take up resources. Blichtfeldt & Eskerod(2008) divide projects by enacted and not enacted project. Where the first are acknowledged by the management and the second the management is not aware of. Our next assumption is that having an overview of current projects enhances the success rate of the company, but that management has the responsibility to communicate this direction of innovation within the company.

H3a: Better performing companies are having a more up to date overview of current projects in the

pipeline.

Managerial tolerance for innovation is important to create a innovation climate within the organization. Saleh and Wang (In: Adams et al, 2002) describe this as consisting of three main components: risk-taking, proactiveness and persistent commitment to innovation. These include top management responsibility for innovation within the organization, including specifying and

communicating a direction for innovation. Griffin (1997) found that a higher percentage of the best performing respondents (75,9%) have a specific strategy that covers the total NPD program, while a lower percentage (58,8%) of the lower performers have an explicit strategy. The next assumption therefore is:

H3b: Management that sets and communicates an explicit direction of innovation (innovation

strategy) will result in better performing companies.

Now that I have discussed the strategic relation between portfolio management and innovation

(20)

2.8 Assessment criteria for innovation projects

The benefit of using a formal system like stage gate has been a proven success factor. In short, the stage gate is a system were a project is separated in phases from idea to implementation (see paragraph 2.2.1) and between each phase ‘gates’ are set which stand for criteria that an innovation must meet or else the project must stop. Cooper (2002) state that better performing companies have strict gate criteria for the projects. The strict criteria prevent projects to be continued while they shouldn’t be and consequently using resources unnecessarily. Management must therefore set formal and measurable criteria for projects. Eventually this should lead to the right projects that will be completed and resources more effectively used. The next statement therefore is that better performing companies will have formal and measurable criteria in place and consequently have the right number of projects in their portfolio. Better performing companies kill projects in order to use the resources for the right projects.

H4a Better performing companies have formal criteria in place for the assessment of innovation

criteria.

H4b Better performing companies kill bad projects and use the resources to boost the right projects.

2.9 Financial vs. strategic criteria.

In portfolio management Cooper et al. (1999) distinguishes several different portfolio methods. The most common are financial and strategic models.

-Financial models and financial indices; These range from ranking or selecting projects based on traditional Net Present Value (NPV), internal rate of return (IRR) and payback methods through to various financial ratios.

-Strategic models; Here, the selection of the portfolio of projects is largely driven by the strategy of the business. The business strategy decides the split of resources across different categories—for example, by types of projects, markets, or product lines—to create strategic buckets. And strategic considerations dominate the decision to do (or not do) certain R&D or new product projects.

(21)

move an outline idea to something with clearer shape and form, on which decisions about resource commitments can be made’ (Tidd et al, 2005, p. 369).

Spreadsheet calculation is an important part of the business case. But trusting on solely financial figures and to much calculating can result in rejecting good potential projects (Cooper et al, 1999). It is about knowing the whole business case and collecting as much information as possible to make a good decision about a project (Salomo et al, 2008)

What came from the research of Cooper et al (1999, 2002) (displayed in figure 30 appendix), concerning the popularity of financial measures, is that significantly more benchmark business (who performed the best) relied on three portfolio methods in conjunction - a financial method, a strategic approach, and a scoring model. There was a distinct link between performance results and the dominant portfolio method: First, although financial methods were the most popular, they yield the poorest performance results. They produce portfolios with poor value projects, too many projects for the resources available and gridlock in the pipeline. This suggests that although an NPV of a project can be good it doesn’t necessarily have to be a good project to continue/start with. Second, in contrast, strategic approaches (letting the business’s strategy decide resource allocation and even choice of projects) perform the best.

Last, scoring models produce positive performance and fare the best in terms of yielding a portfolio containing high value projects- profitable, high return projects with solid economic prospect.

The next assumption therefore is that better performing companies look at more than just the financial return of an innovation project, but look more at the total picture of it.

H5: Successful Dutch companies have a better balance between strategic and financial criteria

compared to less successful Dutch companies.

2.10 Active portfolio management

One of the main goals of portfolio management is to make sure that a project adds value to the portfolio, this means active managing the portfolio of projects in order to take the full potential from the available resources (Cooper, 1999). I have already discussed the issue of having an up to date overview of the current projects underway in the portfolio of the organization. Balancing the portfolio is one of the goals Cooper discusses. Cooper (2000) discusses in his research the main problems arising with portfolio management:

(22)

profitability (Cooper, 2003). Being ‘first to the market’ is known as an important competitive advantage, especially for new products and new services (Cooper, 2003, Brown & Eisenhardt, 1995, Andrew & Sirkin, 2005).

- Project comparison; as already discussed, having criteria set in the organization to evaluate the project is important to make decisions with the right information available. Cooper (2000) found out that projects must not be evaluated solely on these criteria, but must also be compared to each other in the portfolio. This will result in a more balanced portfolio. Cooper (2001) found that better performing companies have a more balanced portfolio (see figure one in appendix)

The two issues above all are a result of a lack of management to be active in prioritizing projects in the portfolio. Common problem is not stopping projects which are doomed to fail, since ‘the investment is already made’. Recognizing delay must result in action to do something about it. Often resources are waiting on each other to make the next move, causing enormous delay (Cooper, 2000). Speed-to-market is an often cited term when talking about success factors of any typical project. Andrew & Sirkin (2005) state that increasing speed to market can increase market share, because you introduce your product or service faster in comparison to your competitors. The increase in competition also cause for companies to introduce the new products faster and more efficient (Griffin, 1993). According to Keller (2004) an essential role is set aside for the product champion. Leadership capabilities of the project manager are essential for speed and successful implementation of the innovation in the market and were the only factor according to Keller (2004) to predict speed to

market in a five year during study.

The next hypotheses will be divided in four of sub hypotheses but all contribute to the same assumption that, better performing companies are more active in their portfolio management in comparison to lower performing companies.

H6a: Better performing companies introduce innovations quicker to the market in comparison to the

rest performing companies

H6b: Better performing companies are more active in looking for delay in projects and taking action

to it, in comparison to lower performing companies.

H6c: Better performing companies also compare the projects in the portfolio against each other and

do this more in comparison to the rest performing companies.

H6d: Better performing companies are more active in resource handling (making and keeping

resources available) for projects in comparison to the rest performing companies.

2.11 Teams

(23)

Four types of team structures relevant for innovation projects are distinguished (Clark & Wheelwright, 1992, Goffin & Mitchell, 2005):

- Functional

- Light weight cross functional - Heavy weight cross functional - Autonomous

Functional teams have members coming from only one function in an organization. Typical functions are R&D, Marketing, Finance, etc. A functional team has the advantage that its members all have similar goals and so little management time will be required to set up the team (Goffin & Mitchell, 2005). A functional team is more suitable for incremental innovation projects.

In a cross functional team each functional area assigns people to the innovation project. The project manager is ‘light weight’ in two aspects. First he/she is often a person who usually has little status or influence in the organization. Second, although they are responsible for coordinating the activities of the functional organization he/she still reports to their normal (functional) manager. (Clark & Wheelwright, 1992). So the main advantage over a functional structure is that there is a project manager coordinating all the functional team members. Besides cross functional also means more combination of expertise and knowledge (Goffin & Mitchell, 2005).

The heavy weight project manager has, in contrast to the light weight manager, direct access and full responsibility over the team and the project. The manager is often a senior manager with much formal authority within the organization.

The autonomous team structure has individuals from different functional areas who are formally assigned, dedicated, and co- located to the project team. The project leader has full control over the resources (Clark & Wheelwright, 1992).

Brown & Eisenhardt (1995) have managed to summarize the NPD literature and come with a comprehensive model of the product development process. At the heart of their model is the project team since “they are the people who actually do the work of the product development” (p. 367). From their research study all literature streams agree on the facts that cross functional teams are critical to process performance. The research of Cooper & Kleinschmidt (1995) shows that cross functional teams show the most successful results in new products. Page (1993) also found that the majority of the respondents used a multi disciplinary team. Cross functional project teams improve inter-functional communication and co-operation which in turn promote success (Ernst, 2002).

Goffin & Mitchell (2005) nuance these results. They state that cross-functional teams should not be used for every type of project. There should be a right balance between the strategy and the available resources. It takes time to develop a good project manager and so organizations do not have the luxury of being able to assign a top project manager to every project (Goffin & Mitchell, 2005). A

(24)

Goffin & Mitchell (2005) also describe the importance of looking at the different roles needed in the team and tune this to the specific project. The next assumption is that a contingency approach will have the best results in terms of the team members that should be chosen, and therefore determining the structure of the team. The contingency approach refers to the fact that a team should be composed out of people with different needed functions, competences and roles and that these factors are dependent on the type and phase of the project.

H7: Better performing Dutch companies maintain a contingency approach in the compilation of the

right team members.

2.12 Culture and climate

Cooper & Kleinschmidt discuss a couple of things that are suggesting to have a positive contribution to the new product development process in a company. Two of them will be discussed here: range of freedom and idea generation.

A positive culture and climate for new product development is vital to successful product

development, according to Cooper & Kleinschmidt (1995). An open culture can be important for an organization to stimulate creativity and idea generation (Amabile, 1998). Cooper & Kleinschmidt (1995) found that better performing companies have a more entrepreneurial climate. This climate also included idea suggestion schemes for employees. A climate in which innovation is supported and idea suggestion from employees is appreciated and reward is regarded as stimuli for success. Voss (1985) found that an innovation friendly climate together with risk-taking behavior are occasionally identified as being relevant to success. The next hypothesis therefore is:

H8a: Best performing companies show more signs of having a culture aimed at innovation in

comparison to the rest performing companies.

H8b: Best performing companies pay more attention to idea generation and valuation in comparison

to the rest performing companies.

(25)

approach their work heightens their intrinsic motivation. One type of mismanagement of freedom is not to communicate the direction of the way there are heading (Amabile, 1998).

Goffin & Mitchell (2005) say that the role of the innovation manager is dependent on the degree of innovation within the project, this is supported by the research of Barczak & Wilemon, (1989, In: Goffin & Mitchell). They found out that the main difference between the normal operating team leader and an innovation project leader is the amount of creativity it is emphasizing in the daily team work. Again said this was not correlated to success but to the amount of innovativeness of the project. Radical projects require more freedom to operate within the current setting to make it a success (Goffin & Mitchell, 2005). It is therefore more obvious that innovation project managers will have more freedom in operation in comparison to normal project managers.

The next assumption therefore is that more successful companies won’t differ in the amount of freedom given to innovation projects managers in comparison to less successful companies, but that all companies will give innovation project managers more freedom in comparison to normal project managers.

H8c: Both the best and rest performing group will give innovation project managers more freedom

than normal project managers.

3. Manufacturing (goods) vs. service based companies

This part will be about the differences between manufacturing goods and services. The goal of this part is to explore if differences exist between service and manufacturing based companies. Griffin (1997) stressed the need for more context specific studies. He stresses that differences between

services and manufacturing goods companies exist and also show difference in success factors for new product/service development.

3.1 Innovativeness in service vs. manufacturing based companies

In the beginning of this paper I addressed the issue of the differences in innovation management, and in particular portfolio management, between NPD and NSD. I suggested that an innovation program which must have a high impact on the future company revenue will result in a more radical approach towards innovation management (Cooper, 1984). John and Storey (1998) argue that there is an overall lack of radical innovation in services. Griffin (1997) reported in his study that service firms report that 24,1% of revenues came from new services introduced in the last five years and that 21,7% of

(26)

the new product or service (Avlonitis et al, 2001). These needs remain latent, due to the offerings are mostly intangible in nature.

From these arguments it would be logical to expect that service companies aim at more incremental innovations in comparison to manufacturing companies. It will be first tested if service companies aim at a more incremental approach in comparison to manufacturing companies (H8a). Second it will be tested if manufacture based companies have a higher percentage from new products in comparison to service based companies (H8b).

3.2 Formality of the process

I already suggested that in services a stage gate and funnel way of maintaining your innovation process is more difficult due to the ad hoc characteristic of a service. The main idea of the funnel is to separate the phases and make go/kill decisions in order to evaluate each phase in the process (Cooper, 2005, Goffin & Mitchell, 2005, Tidd et al, 2005). This is difficult in services since the process is less linear, due to customer involvement in the process. There is simply more tendency to trust on intuition and simply go ahead with the project, even though it’s better to kill the project (Dolfsma, 2004). Griffin (1997) already showed that service based companies maintain much less a formal system such as stage gate, but the use of it could be a great help for prioritizing projects, also in service companies (De Brentani, 1989). De Brentani (2001) showed that financial services with a more formal system of selecting and prioritizing projects performed the best. Portfolio management in service companies has been introduced by implementing best practice of the NPD research into service companies. It has been shown that service-based organizations are at a lower level of maturity with respect to individual new product development processes (Griffin, 1997).

The portfolio issues that are already discussed will be tested again but now for the differences between service-based and manufacture-based companies. This issues concern: strategic guidance, overview of projects and formal assessment criteria. The goal of theses analyses is to find if there are differences between manufacturing companies and service based companies in general.

(27)

4. Research methodology

4.1 Research type

Data retrieved from the surveys done by The Bridge business innovators is used to perform secondary analysis. A survey is useful for a descriptive study; describing the comparable data across subsets of the chosen sample so that similarities and differences can be found. (Cooper & Schindler, 2008, p. 215). Besides in order to find out if a relationship exists between the variables described in the literature and the empirical dataset, the research could also be described as an explanatory research (Cooper & Schindler, 2008). This research can therefore be described as the combination between a descriptive and an explanatory research; its goal is not solely to describe the relationship between innovations success and the factors that influence this success but also to see if there exists an actual relation between the two i.e. do the factors as described also result in a higher innovation success rate. The analysis performed here is referred to as a secondary analysis. Secondary analysis is in its basics, performing a new analysis over an existing dataset with help of more sophisticated statistical

measurements with the goal to test hypotheses and answer questions in a more comprehensive and succinct manner than in the original report (Hakim, 1982).

4.2 Research design

It is important to describe the origin of the data, to explain how the database was realized. This is especially important to describe the reliability and validity of the database and consequently the conclusions drawn from the data. The innovation monitors were not build around a theoretical model, but were based on the daily work of the consultants working at the Bridge. The theoretical foundations are first described in the previous theoretical sections. The questions asked in the surveys will also be founded by questions used in the theory. This will be explained further on. First the survey set-up and questions will be discussed.

4.3 Type of research strategy

According to Cooper & Schindler (2008) four main research designs exists; qualitative research, observational research, surveys (quantitative) and experiments. This research can be classified as a survey. A survey is defined as; “a measurement process used to collect information during a highly structured interview – sometimes with a human interviewer and other times without” (p.215). The goal of the survey is to see if a relationship exists between subjects. These subjects should be made

comparable to each other in other to find similarities and differences; in this case find differences between successful and less successful innovators.

(28)

4.3.1 The innovation monitor

In the innovation monitor the research population consists of companies active in the Netherlands who are involved in innovation projects. The Bridge business innovators have chosen to approach

managers, board members and higher management to participate in the survey. The reason for

choosing higher management levels is that the people active on this level have a better overview of the innovation activities within their organization. A sample was taken from the whole population, but not randomly. A randomly chosen research population is preferable to prevent bias in the population. According to Baarda & de Goede (2000) a randomly chosen population occurs when all research units are arbitrary chosen, so all units of the population have equal chance to be chosen in the research population.

Main reason is that the sample of the innovation monitor consists of mainly connections of the Bridge, therefore the sample not being randomly chosen. Exceptions on this rule are the people interested in innovation who where free to participate in the survey, through the website.

Because the selection of the sample was not random, the data can be considered to contain a certain form of bias. This withholds implications for the validity of the data. The external validity of research findings is the data’s ability to be generalized across persons, settings, and times (Cooper & Schindler, 2008, p. 289). In this case this research cannot be generalized across the whole population of

companies active in the Netherlands, but is restricted to the research units. The extent to which the research population (the participants in the survey) are a reflection of the whole population is important to consider and will be discussed in more detail further on.

4.4 Data collection method

The data was collected by approaching the list of known connections of the Bridge by email. Both the monitor’s pending for participants has been done in May/ June of 2007, 2008 and 2009. Persons who wanted to participate where able to fill in the survey online. With help of a software program

(Netquestionaires) unique links were created for each person approached as a relation of the Bridge. There was also a general link available for persons who wanted to participate, through the website of the Bridge. The characteristics of both monitors are shown in the table below

Table 1

Innovation monitor 07/08 Innovation monitor 08/09 Innovation monitor 09/10

Respondents: 418 (complete surveys) Respondents: 565 (complete surveys) Respondents: 405(complete surveys)

Relations of the bridge: 316 Relations of the Bridge: 448 Relations of the Bridge: 306

(29)

Both monitor surveys consist of multiple choice answers and open ended questions. The questions relevant for this research are lined up in the appendix, and will also be separately discussed in the following parts.

4.5 Preparation of the data

The statistical program SPSS 15.0 is used in order to analyze the data. The data was retrieved from the Netquestionaires server.

The rough data files of both monitors has been analyzed and controlled on the following aspects: - Only complete surveys are allowed in the data; incomplete surveys were left out of the database. - The database are controlled for inconsistencies; this was done by making cross tabs along the

variables in order to see if there were any inconsistencies. For example a person who filled in the same answers throughout the whole survey was left out.

- In some cases the survey can contain possible answers which aren’t useful for analysis. In our case these are the answer possibilities; unknown, not applicable, etc. These answers will be left out (Huizingh, 2006). This is done by selecting them as ‘missing’ variables so SPSS will not calculate them in the analysis.

- Open ended questions were analyzed and if necessary were categorized in existing categories or else were left out.

- Because the tests are based on the innovation output, the respondents that didn’t enter the required data to calculate their innovation output number were deleted. More explicit explanation of the innovation output variables will be discussed further on.

4.6 Population description

The next section will give an image about the distribution of the respondents. This image will give an idea about the reflection of the total population the data withholds. The characteristics are:

type of industry, revenue, number of employees and position of the respondent.

Table 2

Revenue in euro’s Percentage (07/08) Percentage (08/09) Percentage (09/10) Average overall %

Less than 10 million 17,22 17,76 20,49 18,49

10 – 50 million 13,88 13,14 12,35 13,12

50 – 100 million 9,33 7,64 8,64 8,54

100 – 500 million 23,44 23,45 16,79 21,23

500 million – 1 billion 9,09 8,53 8,15 8,59

More than 1 billion 21,53 24,16 22,47 22,72

(30)

Table 3

Number of employees (FTE) Percentage (07/08) Percentage (08/09) Percentage (09/10) Average overall %

Less than 50 employees 17,70 20,43 21,48 19,87

50 – 250 employees 16,27 15,63 16,05 15,98

250 – 500 employees 13,16 9,77 10,86 11,26

500 – 2500 employees 22,25 20,78 20,25 21,09

More than 2500 employees 29,67 32,68 30,62 30,99

Unknown 0,96 0,71 0,74 0,80

Table two until four gives an impression about the distribution of the population in all the three year of innovation monitors. The last column of each table shows the average of the three years. If we

compare these figures with the actual figures distributed by the Central Bureau of Statistics (CBS) we can see if the results of this paper can be generalized over the whole population of Dutch companies.

Table 4

Number of employees CBS average percentage (over three years)

Average percentage innovation monitors (over three years)

Less than 50 employees 98,15 % 19,87 %

50 – 250 employees 1,66 % 15,98 %

250 – 500 employees 0,20 % 11,26 %

More than 500 employees 0,19 % 52,08 %

Based on table 20 in appendix. Source CBS Statline.

It is clear that looking at the number of employees the data set is quite different from the actual population of Dutch companies. The bigger companies are significant more represented in the dataset in comparison to the actual Dutch population. 0,19% of the Dutch companies employ more than 500 employees; the data set holds 52,08 % of companies with more than 500 employees.

The data set is therefore more focused on bigger companies. The simplest explanation for this is that large companies often are more involved in innovation practices, simply because there are more financial resources available. Therefore this population is not a real reflection of the actual population of Dutch companies.

(31)

The following classification is based on own interpretation of the CBS classification in appendix table 19 and the distribution given in the innovation monitors in table 5.

Table 5

Sector Average percentage CBS

(over three years)

Average percentage innovation monitors (over three years)

Agriculture and fishing 12,39 % 3,9 %

Industry and extraction of minerals 6,16 % 21,62 %

Public utilities 0,07 % 3,92 %

Building industry 11,41 % 8,49 %

Consumer goods and trade (retail and

wholesale) 21,02 % 4,45 %

Catering industry 4,68 % 5,98 %

Transport, storage and communication 3,70 % 9,26 %

Financial services 2,12% 7,7 %

Business services and consumer rentals 23,91% 21,86 %

Public administration, government and

education 2,37 % 5,68 %

Public health industry, cultural,

environmental and other non-profit. 12,19% 1,68 %

The most clear differences in the table between the CBS statistics and the innovation monitors is the percentages of industry, consumer goods, and public health, cultural environment and other non-profit industries. This table shows that there are clear differences between the real population and the population of the innovation monitors.

Table 6

Position of the respondents Percentage (07/08) Percentage (08/09) Percentage (09/10) Average overall % CEO/general manager 17,94 17,23 16,30 17,16

Business unit manager 13,88 15,99 13,58 14,48

Innovation manager/ Business

development manager 22,49 20,60 21,48

21,52

Financial director 0,72 1,60 1,23 1,18

(32)

Marketing manager 9,33 8,70 5,19 7,74

ICT - manager 2,15 2,13 1,23 1,84

Sales manager 4,55 3,91 4,20 4,22

Project manager 10,05 10,66 11,36 10,69

Other 7,42 10,12 15,80 11,11

This table shows that the people that filled in the monitor are mainly coming from higher management and innovation related departments (63,3%). Since this research is focused on innovation issues in which higher management often takes the most important decisions and portfolio management being a higher management responsibility (Cooper, 1999) the high percentage of higher management

respondents gives the data more credibility.

4.7 Method of data analysis

The data analysis method is dependent on the problem statement of the study (Baarda & de Goede, 2000). In this case I want to discover if there is a difference between groups of companies that are distinguished by matter of innovation success. How theses groups were formulated will be discussed shortly.

4.7.1 Statistical tests

The goal of this research is to find differences between two groups: successful vs. unsuccessful companies and manufacturing vs. service companies. The T-test is used for describing differences between two groups (Huizingh, 2006). In this case the average of the sample is subject to the test of the other average. This is known as the one – sample T-test (Huizingh, 2006). The T-test is used to determine if the average of the two groups match each other.

The zero hypotheses for each test is the statement that both groups match in their average value. The alternative hypothesis states that both groups differ from each other.

I will now continue first with explaining the innovation success criteria. Then the different success groups will be defined. Then the analysis of the conceptual model will follow.

4.8 Success outcome variables.

(33)

In line with previous research, the respondents are divided in two groups; the best vs. the rest (Cooper, 1999; Griffin, 1997).

The two dimensions1 overall success and financial success distinguish the best from the rest. Overall success was measured by questions one and two.

Table 7

Factor 1 ‘Overall success’ Correlation Factor 2 ‘Financial success’ Correlation

Better innovator in comparison to the competition

0,57 Percentage of revenue

coming from new products/services

0,78

Innovation program performance improvement

0,75 Percentage of revenue

coming from new customer segments

0,79

If the respondent was considered to be a better innovator in comparison to the competition and the innovation performance remained equal or became better the overall performance was categorized as the ‘best’.

Financial success was measured by questions three and four in the appendix. These questions had an chronbach alpha of α 0,78. The questions were first combined based on expected values, and were given relative values in SPSS (see appendix). The respondents who scored above the mean of the total population were considered as the best performers. Consequently the two dimensions (financial and overall success) were combined resulting in two groups: the best vs. the rest.

Table 8

Monitor 07/08 Monitor 08/09 Monitor 09/10

% Best performing group 20,2 % 18,4 % 20,8 %

% Rest performing group 79,8 % 81,6 % 79,2 %

4.9 Analysing the questions

In this section the hypotheses stated in the literature part will be discussed. The data analysis method is dependent on the problem statement of the study (Baarda & de Goede, 2000).

The formulation of the tests is through stating hypotheses. In statistics the null hypothesis (H0) is the hypothesis that has no effect: it is the negation of the scientific hypothesis. In this case the null hypothesis will be:

H0: The performance groups perform equally on factor A. The alternative hypothesis will then be:

1_{The two dimensions were derived from a factor analysis. This analysis distinguished two factors with in each}

Bridging the gap: from failure to success.

Bridging the gap: from failure to

success.

Best practice study of Dutch companies in innovation

management

Name:

Ruud Egberts

Date:

25-03-2010

Institution: University of Groningen

Bridging the gap: from failure to success.

Best practice study of Dutch companies in innovation management

Author: Ruud Egberts

Student number: 1419625

Email address: r.h.egberts@student.rug.nl

Faculty of Management and

Organization

Master Strategy and Innovation

Supervisors: Dr. W. Dolfsma

Dr. H. Snijders

Acknowledgments

Summary

Table of contents

1. Introduction

2. Design and theory

3. Manufacturing (goods) vs. service based companies

4. Research methodology