Counteracting Corporate Amnesia: The Effective Management of Attitudes towards Lessons Learned Collection in Project Based Make-to-Order Organisations

(1)

Counteracting Corporate Amnesia: The Effective Management

of Attitudes towards Lessons Learned Collection in Project

Based Make-to-Order Organisations

Dyanne van de Wijdeven

Dr. Ying Yang

10-12-2018

Dr. Ing. Stefano Fazi

“Knowledge is power”

(2)

(3)

Degree:

DD-MSc. Technology & Operations Management

Supervisor & Assessor:

Dr. Ying Yang

Co-assessor:

Dr. ir. Stefano Fazi

Student name:

Dyanne van de Wijdeven

RUG Student number:

S3597881

NU Student number:

170243812

Newcastle University Business School

5 Barrack Road, Newcastle upon Tyne,

NE1 4SE

University of Groningen Faculty of Economics and

Business

Nettelbosje 2, 9747 AE Groningen

Siemens

(4)

ABSTRACT

Purpose This research aims to generate a greater understanding of how different cultures in organisations affect the attitude of cross-functional project members towards identification and sharing of Lessons Learned. This is done to be able to provide practical insights and solutions that can improve cross-functional project learning processes, which ultimately will decrease corporate amnesia.

Research methodology The research adopts a single case study, utilising three research methods: a questionnaire for departmental culture identification; semi-structured interviews to generate in-depth knowledge about different attitudes caused by diverse cultures; and an evaluation session with a lessons learned facilitator to review solutions on using the insights gained from interviews at the case company.

Findings The study clarifies how attitudes towards LL identification and sharing are influenced by clan and hierarchical cultures, demonstrating advantages and disadvantages for both cultures, and emphasising key aspects to consider when dealing with these cultures. After which solutions are provided that are applicable in practice, reducing the gap between high theoretical knowledge and low understanding in practice.

Practical implications The study provides applicable recommendations for project managers and facilitators on how to use the insights gained on clan and hierarchical cultures affecting attitudes towards LL. Improving the management of members from different cultures, regarding their attitude towards identification and sharing of lessons learned, decreasing corporate amnesia.

(5)

Acknowledgements

This thesis has taught me countless of valuable things, I could not have done it without the support of so many others. First of all, I want to thank my main supervisor, Dr. Ying Yang, for opening her door to me for continuous encouragement and assistance. Secondly, I would like to thank my second supervisor Dr. ir. Stefano Fazi, who helped me both during meetings as afterwards, with useful feedback and information on my topic.

Moreover, I could not have done my research without the help of Richard Gorton. He always found the time to send me all the information I needed and made sure I got enough input for both my questionnaire and interviews. Furthermore, I would like to thank everyone at Siemens who provided input for my research. Especially the ones I got to conduct interviews with, thank you for taking the time to help me with my study, even at the very last minute.

Additionally, I want to thank my family and friends for their unconditional support whenever I needed it. But also, their critical feedback that has improved my research so much, even though I might not always have looked forward to hearing I had to improve more.

(6)

List of tables and figures

Table 3.1 – Construct reliability questionnaire ... 25

Table 3.2 – Interviewee profiles ... 29

Table 3.3 - Coding tree ... 31

Table 4.1 - Attitude towards Lessons Learned Practice ... 33

Table 4.2 - Ranking LL importance interviewee 1 ... 34

Table 4.6 - Attitudes towards highest ranked Lessons Learned types ... 35

Table 4.7 - Attitudes towards each Lessons Learned type ... 37

Table 4.8 - Willingness to share ... 37

Table 4.9 - Commitment to project and learning ... 38

Table 4.10 - Summary of analysis of hypotheses ... 39

Table 5.1 – Comparison cultural influence for each LL type ... 43

Figure 2.1 – Conceptual Model ... 21

Figure 3.1 - Organisational culture composition ... 26

Figure 3.2 - Quality culture composition ... 26

Figure 3.3 – Facility Operations culture composition ... 26

Figure 3.4 - Engineering culture composition... 26

Figure 3.5 - IT Business support culture composition... 26

Figure 3.6 - Grouped organisational culture composition ... 26

Figure 3.7 - Perceived importance per LL type ... 27

List of abbreviations

LL

Lessons Learned

LLP

Lessons Learned Process

OCAI Organizational Culture Assessment Instrument

COC Clan Organisational Culture

(9)

1. Introduction

Corporate amnesia is the loss of corporate memory due to non-transferred skills or undocumented knowledge (Corporate amnesia, 2018). Two decades ago, Kransdorf (1998) already stated that learning from projects and successfully retaining and reusing what is learned can reduce corporate amnesia. However, a recent global survey conducted by Jive Software Inc. (2017) found that around 47% of UK knowledge workers, such as engineers and architects, believe that corporate amnesia is still a problem within their company. Additionally, the survey results indicated that the global workforce spends approximately 29% of the day searching for information to complete projects (2017). These results indicate that issues with corporate amnesia are still unresolved, causing inefficient processes and reduced performance in projects (Fugate et al., 2009).

(10)

One way to incorporate LL into projects is by implementing a Lessons Learned Process (LLP) into business practices. According to a study by Newell (2015), the LLP enables effective sharing of newly obtained knowledge, preventing organisations from wasting resources, repeating mistakes, and re-inventing the wheel. Because members from the different departments, involved in cross-functional teams, can have different departmental cultures (Ajmal and Koskinen, 2008), the LLP becomes complicated to manage. Department cultures directly affect the attitude of members towards what new knowledge is important to retain within their project team, affecting the effectiveness of the LLP (Huang and Newell, 2003). This is of particular importance for collection of the LL. The LL collection consists of two parts, the identification of LL that are important to share and the actual sharing of LL. In both steps the individual goes through an internal process of determining whether the lesson learned is worthwhile to share with others in the project team. It is therefore important to understand how culture affects this internal process that determines the attitude towards LL identification and sharing. So that measures can be taken to align attitudes of all team members, improving LL collection and the LLP, reducing corporate amnesia.

Earlier studies have focused on organisational culture and its influences on knowledge sharing (eg. Duffield and Whitty, 2015; Goh, 2002; Skyrme and Amidon, 1997; Davenport et al., 1998). However, these studies have not considered the effect of varying cultures at departmental level on LL collection and thus the LLP (Newell, 2015), which is believed to have a direct effect on performance of learning in cross-functional projects. Previous studies have assumed that everyone in the organisation shares the same culture, which often is not the case (Hofstede, 1998). Therefore, these studies fail to apply to cross-functional project learning, as these deal with people from different departmental cultural backgrounds. As such, this study focuses on influences from departmental level on attitudes towards LL identification and sharing in cross-functional projects. Furthermore, prior research has not provided in-depth insights into how certain culture types affect the attitude towards LL collection, they merely determined the correlations between the two (Suppiah and Singh Sandhu, 2011; Mueller, 2014). Leading to the first research question:

RQ1: How do different departmental cultures affect an individual’s attitude towards the identification and sharing of Lessons Learned?

(11)

altogether (Duffield and Whitty, 2015). To generate a greater understanding of how the findings of RQ1 affect business practice and how this can be implemented into the LLP management practices, it is essential to take LLP management in practice into consideration. Therefore, this research aims to understand the management of differing attitudes towards LL identification and sharing that occur in practice, and how the earlier findings of this study can improve LLP management.

RQ2: How can the in-depth understanding of departmental cultural influences on LL identification and sharing be incorporated into future Lessons Learned Processes to improve practical application and

management within a cross-functional setting?

This study is a single case study, in which both quantitative and qualitative methods are used: a questionnaire, interviews, and an evaluation session. First, the questionnaire is used for departmental and organisational culture identification, which then serves as building blocks for the interviews. Second, interviews are conducted with members from different culture compositions. The interviews create an understanding of the influences of the culture type on the attitude towards LL per group and how these attitudes affect the current LLP. Lastly, the insights obtained throughout the research are discussed during an evaluation session with an LLP facilitator to review current LLP management practices and suggested solutions for the incorporation of the new insights into the LLP. The evaluation session will ensure applicability of proposed implementations in practice.

(12)

2. Literature review

Organisations increasingly rely on cross-functional teams (Randel & Jaussi, 2003; Ehrhardt et al., 2014) to stay ahead of competition (Bishop, 1999). Cross-functional projects have features that make them very effective within these fast-paced market developments, such as rapid market response, the ability to perform non-routine projects, and knowledge creation (Pinto and Pinto, 1990). Especially within cross-functional projects more new knowledge can be created due to the mix of skills and qualities of the members collaborating, advancing organisational learning (Huang and Newell, 2003). However, the cross-functional project-specific features require the LLP to adapt in order to function in this form of collaboration. Cross-functional teams involves people from a variety of functional departments, thus possibly members from different cultures (Ajmal and Koskinen, 2008), with varying levels of project involvement these factors affect the LL identification and sharing attitude in different ways.

2.1 The Lessons Learned Process

Organisational learning arises from the management of knowledge, which entails the creation, development, retention, and application of knowledge in the workplace through sharing information (Soto-Acosta et al., 2014; Syed et al., 2018). As Syed at al. (2018, p. 2) state, “The knowledge-based economy is reflected in an increasing emphasis on the dissemination and use of knowledge as a source of competitiveness for organizations and countries”. As knowledge-based assets are often more difficult to imitate, they increase the chances of a long-term, sustainable, competitive advantage (Ghobadi and D’Ambra, 2012). The LLP is an essential part of organisational learning as it transitions newly identified knowledge into practical, valuable, and most importantly, reusable knowledge (Wiewiora et al., 2018).

The LLP consists of five steps, collection, prioritisation, documentation, communication, and assimilation (Dülgerler and Negri, 2016). Most aspects of the LLP have been researched extensively before, from describing the barriers and enablers of successful LL sessions to more specified research into different technological systems options (Newell et al., 2006; Swan, Scarbrough and Newell, 2010; Wiewiora et al., 2014). Recent studies have shown organisations still struggle with collecting LL (Newell, 2015; Besner and Hobbs, 2006), however, only few studies have delved into the underlying causes and reasoning of these barriers and enablers.

LL collection

(13)

identified, such as trust, organisational structures and reward systems (Jeon, Kim and Koh, 2011; Xue, Bradley and Liang, 2011; Tangaraja et al., 2015). However, one of the most important factors barring or enabling knowledge sharing is culture (Asrar-ul-Haq and Anwar, 2016; Fong Boh, Nguyen and Xu, 2013; Mueller, 2014). Syed et al. (2018) state that culture greatly affects the quality of knowledge sharing within and between projects due to possible misalignment of attitudes. A great deal of the issues relating to collection of LL are thus affected by the involvement of people, they are the ultimate decision-makers regarding what, how, and when LL are shared. LL collection consists of two parts, first identifying which LL are important to share and then the actual sharing of those LL identified as important. Especially culture is important in this collection stage because both parts are affected by the culture of the individual. This is because when a lesson is learned by an individual, this person goes through an internal decision-process regarding whether it is necessary to share this LL or not. Considering the great importance of the people-aspect for a successful LLP, this study focuses on gaining in-depth knowledge about the underlying reasoning of attitudes regarding LL collection. Since people share the culture of their department, everyone in a department has similar attitudes towards learning and will therefore be aligned, which can be managed easily. The difficulties arise in cross-cultural contexts with cross-functional projects. When dealing with different cultures, the LL attitudes are not aligned, and members will have different attitudes towards LL. Therefore, the types of LL studied for this research are narrowed down to those that are important for other cross-functional team members. These LL are project-specific rather than expertise-specific, entailing everything that should ensure project success, because expertise-specific LL are more focused on the expertise within a department. Such project-specific lessons are customer preferences, scheduling, communication, expectations, quality, risk, and status reporting. These groups were defined through thorough analyses of LL documents in an organisation with cross-functional, long-term, projects with make-to-order, servitisation and manufacturing characteristics. This is done to ensure the most important LL topics of this type of organisations are included. Thus, LL from experiences, that occurred while working on a project, that have consequences for others from other departments in the team.

2.2 Organisational & Departmental cultures

(14)

Al Saifi (2015) has shown the importance of culture in organisations and confirmed its impact on organisational learning and its direct relatedness to organisational performance. According to De Long and Fahey (2000), there are multiple ways in which culture can influence behaviour. Culture acts as a mediator between individual and organisational knowledge, it creates the context for social interaction, which in turn affects sharing and applying knowledge. Furthermore, it shapes the processes with which new knowledge is created, legitimised and distributed. However, most importantly, cultures determine what knowledge is perceived valuable and how it should be managed (2000). Cultures can be developed when a group of people faces similar problems or tasks, leading to shared ideas, values and norms (Schwarz, 2016). Departmental cultures are examples of where cultures are established because the members share similar tasks and skills. These cultures can be different from the overall organisational culture, depending on the strength of the organisational culture (Brewer, 1993). If the organisational culture is strong, the departmental cultures will be more similar to the organisational culture, whereas if the departmental cultures are strong, they differ from the organisational culture. Often, the departmental cultures share the core values of the organisational culture, combined with group-specific values and attitudes unique to the departmental culture considered (Alvesson, 2012). Since culture provides the context for members to create, acquire, share, and manage knowledge (Kayworth & Leidner, 2003), the strongest culture determines the attitudes towards LL identification and sharing (Jones et al., 2006).

Within the LL context, implications arise when the LLP involves members from departments with strong and different cultures. They have different attitudes towards what knowledge should be managed and how all LL should be shared. Moreover, they will have different attitudes towards learning in general (Wiewiora et al., 2014), which shows the people-factor hindering the LLP.

Culture types in organisations

(15)

valuable framework for organisations that try to improve their performance by managing cultural influences.

Organisational and departmental cultures can have any one or a combination of the four culture types. Cameron and Quinn (2006) describe the culture types as follows. For organisations in which the clan culture is dominant, the leaders are perceived as warm and supportive, this culture revolves around teamwork, growth, and facilitating. If an organisation is mainly adhocracy dominated, they are future-oriented, entrepreneurial, innovative and risk taking. Both these cultures value flexibility, however the clan culture is more internally focused, whereas the adhocracy culture is more externally focused. Market dominated organisations are focused on winning, it is results oriented and entails a lot of negotiating and directing. A hierarchy focused organisation is characterised by formalised and structured processes, it emphasises stability and efficiency. These two cultures value stability and control over flexibility, furthermore market is similar to adhocracy that it is more externally focused whereas hierarchy is similar to clan, internally focused (Cameron and Quinn, 2006). These four types affect knowledge sharing differently (Suppiah and Singh Sandhu, 2011).

When an organisation contains multiple departments with different cultures, meaning their cultural compositions are different from each other, cross-functional projects will have to deal with these differences too. Considering the core differences of the cultures, aligning team members with different cultural and departmental backgrounds is challenging, especially when working in a cross-functional context. The departmental culture influences what members perceive as useful, important, and valid knowledge (De Long and Fahey, 2000). If these attitudes are not aligned with the rest of the project team, as well as with the organisation, knowledge will be lost in translation or not even retained in the first place. This has considerable implications for the LLP, as members that are not aligned do not have the same attitude towards the LL identification and sharing. Instinctively, people with similar values and norms face less difficulties with LL sharing than people with varying values and norms, as they are more aligned already. These implications are worth studying as LL identification and sharing can be improved by ensuring all team members are aligned from the beginning.

RQ1: How do different functional departmental cultures affect an individual’s attitude towards the identification and sharing of Lessons Learned?

(16)

theory-practice gap, preventing academic knowledge from being applied in theory-practice (Heisig et al., 2016). The issues relating to the lower understanding in practice was recognised both in literature (Duffield and Whitty, 2015) and in practice (Jive Software Inc., 2017), having caused organisations to fail to implement a successful LLP. Therefore, this research aims to generate a more in-depth understanding of what and why certain attitudes occur in practice, to provide practical and applicable managerial suggestions for LL sharing.

RQ2: How can the in-depth understanding of departmental cultural influences on LL identification and sharing be incorporated into future Lessons Learned Processes to improve practical application and

management within a cross-functional setting?

2.3 Lessons Learned collection issues in Cross-functional projects

The LLP is particularly important for long-term projects in make-to-order manufacturing-servitisation industries. For such organisations projects are at the core of their business (Hicks et al., 2000), therefore its performance affects the strength of their market position. These projects are one-off projects, dealing with a lot of new knowledge that also requires the team to work closely together on a high level with each other as well as with the customer (Holweg and Pil, 2001). Furthermore, due to the long length and the manufacturing aspects of the projects, it will incur job switches, part-time involvement of staff and other stakeholders. The servitisation aspect intensifies the level of customer contact, which increases the need for accurate knowledge management relating to customer requirements (Chirumalla, 2016). This makes it increasingly difficult to collect LL and have everyone aligned regarding identification and sharing of project-specific LL. Moreover, every client has its own preferences and requirements, increasing the importance of learning throughout the project. Many issues with cross-functional projects come from the varying perceptions of what LL are important, the differences in willingness to share LL, and the differences in project commitment. These are all affected by the cultural background of the individual, which will be discussed further below.

2.3.1 Implications of varying attitudes in cultures and culture strength

(17)

Cross-functional project teams involve members from different cultures and will therefore have to manage opposing views and attitudes towards LL identification and sharing. However, not every organisation has the same organisational and departmental culture balance. In organisations with a strong organisational culture the departmental cultures will be weaker and vice versa (Martin, 1992). If multiple departments have similar culture compositions similar to the organisational culture, it can be assumed the organisational culture is stronger than the departmental cultures. Therefore, a moderator was added to explain for the strength of the impact the culture types have. Hence,

H1: Firms with strong organisational cultures have greater alignment of the identification and sharing

of project-specific lessons learned.

2.3.2 LL identification: attitudes towards importance of project-specific Lessons Learned

Individuals from different departments have different practices and values they adhere to. As department members share the values and norms stressed by their departmental culture, their

perceived importance for each project-specific lesson learned can vary. The attitude towards the

importance of each lesson revolves around identifying whether a lesson is worthwhile to share with others, therefore focusing more on LL identification rather than LL sharing.

The clan culture tends to focus on personal development and teamwork, this could imply they perceive LL on communication as most important to share (Wiewiora et al., 2014). The adhocracy culture revolves around innovation, which might increase their LL focus on risk, as they try new things, and communication, to increase creativity (Cameron and Quinn, 2006). The hierarchy culture is focused on processes and efficiency. Therefore, it is likely they will focus on direct LL as status reporting and planning, with which they can improve efficiency. Lastly, the market culture is competition and transaction focused, this might shift their importance towards LL types that decrease costs, such as quality and risk levels. Hence,

H2: The type of departmental culture influences which project specific lessons learned are perceived as

important to share.

2.3.3 LL sharing: willingness to share project-specific Lessons Learned

(18)

Since the culture types have different norms and values, their attitude on sharing lessons learned differs as well, such as when learning is perceived as making mistakes. Suppiah and Singh Sandhu (2011) studied the effects of the culture types on the willingness to share knowledge, their findings will be discussed below. Additionally, Egan (2008) found differences in attitude towards sharing LL between cultures, where an innovative or supportive culture has a positive relationship with sharing LL and a bureaucratic culture has a negative relationship with LL sharing. Hence,

H3: The type of departmental culture influences the level of willingness to share project specific lessons

learned differently.

The significant differences of the cultures affect the willingness to share in varying ways. Park and Lee (2014) found that information sharing is positively related to trust and the feeling of dependence in a team setting, LL will thus be shared more in such cultures. The clan culture is characterised by participation, involvement, a recognition for employees and a caring climate (Cameron and Quinn, 2006), all factors that affirm a culture of trust. Which Wiewiora et al. (2014) confirm in their research, they found that those values facilitate the sharing process. Nevertheless, they also state that a too strong clan culture can lead to members being concerned they will disturb the group harmony when speaking up. Moreover, because members from a clan culture are internally focused, they will be more reluctant to share information with external stakeholders. However, their research focused solely on project managers, rather than all team members. Lastly, Suppiah and Singh Sandhu (2011) found that a clan culture has a positive influence on information sharing. Therefore,

H3A: A Clan Culture will increase the willingness to share project-specific lessons learned within

the team.

Wiewiora, Smidt and Chang (2018) described cultures that support experimentation and risk-taking as characteristics enhancing LL sharing, characteristics that can be found in the adhocracy culture. These characteristics increase dialogue and evaluation, discussing and sharing LL with others. Moreover, Cameron and Quinn (2006) emphasise the centrality of continuous improvement in the adhocracy culture, which encourages LL sharing. Hence,

H3B: An Adhocracy Culture will increase the willingness to share project-specific lessons

learned within the team.

(19)

Furthermore, the Market culture is focused on competitiveness, in which knowledge is perceived as power, treating the exchange of knowledge as a transaction rather than sharing (Suppiah and Singh Sandhu, 2011). Therefore,

H3C: A Market Culture will decrease the willingness to share project-specific lessons learned within the

team.

The hierarchy culture revolves around strict procedures, clear lines of decision-making and control. Suppiah and Singh Sandhu (2011) claim the hierarchy culture is the least likely to encourage knowledge sharing, dealing with the lowest willingness to share knowledge. Naranjo-Valencia et al. (2016) support these findings, stating that this culture shows the most resistance to adaptation and change, underestimating the usefulness and importance of continuous improvement. Hence,

H3D: A Hierarchy Culture will decrease the willingness to share project-specific lessons learned

within the team.

2.3.4 LL sharing: project commitment to sharing Lessons Learned

Another factor affecting the sharing of LL, is project commitment. Cross-functional projects often consist of team members that deal with other responsibilities as well, they are not fully devoted to one project. Due to the long-term scope and make-to-order strategy of the projects this study focuses on, it will include team members that are only involved for short periods, for example when expert knowledge is required (Hoegl et al., 2004). Questions arise whether people will be motivated enough to commit time and resources into the learning aspect of projects when they are working on other projects simultaneously (Buvik and Tvedt, 2016). Evidently, cultures have great significance on people’s attitude towards project commitment. According to Cameron and Quinn (2006), members from different cultures have different perceptions of how to spend their time. Certain culture types promote collaboration and continuous improvement, whereas other cultures are short-term goal focused, decreasing the need to be committed to a level in which you share lessons learned. Hence,

H4: The type of departmental culture influences the level of commitment one feels for the project. The clan culture is based on commitment to teamwork and collective growth (Yazici, 2009), which indicates members from the clan culture are strongly committed to projects and to reach collective goals as well as growth through learning. They usually involve all stakeholders in decision-making processes, increasing the commitment of these employees. Thus,

H4A: A Clan Culture will increase the project commitment and therefore increase

(20)

Krajcsák(2018) states that due to the self-efficacy and professional commitment norms the adhocracy culture enforces, it demonstrates a high level of employee commitment. Furthermore, this culture revolves around innovation and experimenting (Cameron and Quinn, 2006), which on its own already requires commitment to the development and growth of ideas and projects. Hence,

H4B: An Adhocracy Culture will increase the commitment to the project and therefore increase

project-specific lessons learned sharing.

Short-term oriented cultures, such as the market culture, often underestimate the value of evaluating the project, focusing on new projects and immediate project deliverables rather than long-term results (Carillo et al. (2013). This urge to finish as fast as possible to start the next project, decreases the project commitment, they will not take the time to share LL. Furthermore, their external focus causes them to be more committed to external forces than internal ones (Cameron and Quinn, 2006). Hence,

H4C: A Market Culture will decrease the commitment to the project and therefore decrease

project-specific lessons learned sharing.

A hierarchy culture accommodates centralised decision-making, this causes employees to be less involved in the project and therefore be less committed to share LL (Suppiah and Singh Sandhu, 2011). Additionally, this culture focuses on strict procedures and tight control, requiring ample paperwork (Cameron and Quinn, 2006). This can discourage members to participate in anticipation of extra work. Thus,

H4D: A Hierarchy Culture will decrease the commitment to the project and therefore decrease

(21)

The three constructs discussed are the ones analysed in this study; the identification of importance of different project-specific LL types, the willingness to share project-specific LL, and the level of project commitment towards learning and sharing LL, see Figure 2.1. The aim of this study is to create understanding of how the departmental culture background of individuals affect these variables that affect LL identification and sharing. The different cultures are known to have great impact on the LLP, as identified in previous research, certain culture types lead to better performance than others. However, it is unknown what exactly drives the team members to act in certain ways, which has made it difficult to deal with these differences. This research aims to create a greater understanding of the underlying attitudes to provide suggestions that can be formalised into practical procedures. Aiming to defy the theory-practice gap currently apparent in knowledge management and with that reduce corporate amnesia.

(22)

3. Methodology

This chapter discusses the methodology and research methods used in this study. First, the research design is provided, discussing the three research methods used are: questionnaire, interviews, and evaluation session respectively. Then, the questionnaire is reviewed, discussing the unit of analysis, questionnaire development, data collection, reliability and validity, and questionnaire findings. The findings are discussed in this section as they are required for the interview development, which follows directly after, with the unit of analysis and data collection, interview development, data analysis method, and validity and reliability. Lastly, the evaluation session is discussed.

3.1 Research Design

The research is designed as a single case study, this is done to step away from the quantitative studies done before on this topic. A single case study examines the data more in-depth as it is mainly analysed qualitatively, providing more and better possibilities to gain a greater understanding of the topic in a practical, real life, context (Dul and Hak, 2007). Methodological triangulation is used, combining a questionnaire, interviews, and an evaluation session. Such methodological triangulation can offer a better understanding of the correlation between theory and empirical findings, as it facilitates integration of both quantitative and qualitative findings (Östlund et al., 2011).

First, a questionnaire is distributed among employees of departments involved in cross-functional projects. Respondents are asked to answer questions on the culture within their department to determine differences between departments. In case of strong differences, this proves the existence of different departmental cultures, subcultures. Furthermore, respondents are asked to rank the importance of project-specific LL, the willingness to share knowledge with others and the level of project commitment to sharing LL. Previous research already established the correlations between the different cultures and the constructs, therefore this is not a topic of interest in this study. The main aim of this questionnaire is the measurement of culture compositions within departments. The questionnaire data is used to plot the culture compositions per department, after which the departments are analysed for any similarities and differences among them. To determine the strength of the organisational culture, departments with similar cultures as the organisational culture are grouped together. The different cultural compositions and the culture types are then analysed in relation to the attitude to sharing and identifying LL constructs. Which then functions as building blocks for the interview construction.

(23)

the practical side of the research into consideration, providing clearer insights into real-life situations. This is achieved through semi-structured interviews with people from different culture composition groups. The main interest is on the general view on the LLP, the level of commitment to projects and learning in projects, the willingness to share and their own input, and the ranking of importance of different LL types. The predominant objective is to discover the underlying reasoning behind the different attitudes to create a clear overview, with which project managers can work to align cross-functional team members.

Third, an evaluation session is held with the LL facilitator of the case company, reviewing the current LL practice and the use of insights in culture in management of the LLP in cross-functional settings. This is done to improve applicability of insights in practice. During this session possible solutions to improve LL identification and sharing, regarding the management of different attitudes of departments, were evaluated to gain insights in the practical aspects of the LLP management.

3.2 Questionnaire

The questionnaire is meant as a building block for the interviews, focusing on getting the base knowledge used for in-depth analyses of cultures and their attitudes towards the LLP. The questionnaire aims to verify the existence of different culture compositions within the organisation. After which the possible cultures can be analysed and a course of action can be determined for the interviews.

3.2.1 Unit of Analysis questionnaire

(24)

cross-functional projects are the units analysed as they are studied on their culture and attitudes towards sharing project-specific LL in the LLP of cross-functional projects.

3.2.2 Questionnaire development

The questionnaire is divided into three sections, (1) demographics, (2) the culture analysis, (3) LL identification and sharing attitudes, which can be found in Appendix I. Of these, the second part is the main aim focus. The first part establishes what department respondents are from and whether they have been part of projects before or not. The parts related to LL sharing attitudes are added merely to provide a complete picture of the situation and theory-testing, rather than in-depth analysis.

Often a holistic approach is preferred for analyses of culture immersing in a culture, however this is incredibly time consuming (Gregory, 1983). Especially in this case, where the aim is to distinguish different cultures, it would be too time consuming to immerse in all departments and observe the cultures. Therefore, a quantitative approach, OCAI, is used to illustrate culture compositions and conduct comparisons among them. To ensure the questionnaire measures culture rather than climate, scenario analysis procedures are used. Such procedures require respondents to report the extent to which written scenarios are indicative of their own department’s culture. Cameron and Quinn (2006) defined six content dimensions that identify aspects of organisations/departments that reflect key values and assumptions. These dimensions ask about cultures indirectly, such that respondents use their underlying archetypal framework, uncovering the underlying organisational/departmental culture (Mitroff and Kilmann, 1978). Not only are the content dimensions well defined, the OCAI also provides the possibility to measure multiple cultural aspects at once, it determines cultural strength, congruence and type. Furthermore, the OCAI has been successfully used in research relating to knowledge management and projects (Gray and Densten, 2005; Fong and Kwok, 2009).

(25)

3.2.3 Data collection questionnaire

The questionnaire software Qualtrics is used for development and distribution of the questionnaire and collection of the data afterwards. It is distributed by email to employees of all departments involved in projects at the case company. The questionnaire received 55 responses, of which 37 were finished. Of these 37, one was dismissed for careless response. Leaving 36 responses for departmental culture measurement. Four departments had five or more responses, therefore these were used for further analysis of departmental cultures, leaving 21 responses for analysis. To measure and depict organisational and departmental cultures, the OCAI was used, the standardised instrument developed by Cameron and Quinn (2006). Departments with similar cultures were grouped together for the descriptive analysis to facilitate comparison with different cultures.

For the LL attitude validation analysis, all respondents with previous experience in projects were regarded for the LL collection, as this part focuses on the general effect of culture types on attitudes. Of all respondents, five had no project experience and were therefore discarded. One of the 31 showed careless response in the last section of the questionnaire and was therefore discarded, leaving 30.

3.2.4 Reliability and Validity of questionnaire

The reliability defines the extent to which instruments measure constructs consistently. Various studies have tested the reliability of the OCAI (Quinn & Spreitzer, 1991; Denison and Spreitzer, 1991), with different sample sizes and industry focus, all with a Cronbach’s Alpha of over .7. This suggests there is sufficient evidence to assume the instrument is reliable irrespective of industry or sample size. Analysis of questionnaire data shows that all but the sharing knowledge construct is reliable, see Table 3.1.

Constructs Reliability (Cronbach’s apha)

Sharing knowledge .616 Project commitment .783 Clan .803 Adhocracy .773 Market .824 Hierarchy .735

Table 3.1 – Construct reliability questionnaire

(26)

constructs were found unreliable (α=.616), discarding findings for this construct. Since the usable results show consistency with previous research, it will be assumed that the other correlations between the cultures and willingness to share and project commitment still hold. This, however, still does not explain the underlying reasoning this research is aiming for.

3.2.5 Questionnaire findings

The data was analysed in two ways, the OCAI culture plots and statistical/descriptive analysis. First, the organisational culture was plotted with the use of all respondents of the questionnaire, see Figure 2. Afterwards the departmental cultures were measured and plotted for four departments; Engineering, IT Business support, Facility operations, and Quality, see Figures 3.1-3.6.

Figure 3.1 - Organisational culture composition Figure 3.2 - Quality culture composition

Figure 3.3 – Facility Operations culture composition Figure 3.4 - Engineering culture composition

Figure 3.5 - IT Business support culture composition Figure 3.6 - Grouped organisational culture

(27)

The first thing to note is the similarity in shape of most departments, and how similar this is to the organisational culture. This shows the organisational culture in this organisation is relatively strong, it influences the cultures within the departments. However, the IT business support department figure indicates this department has a distinct subculture from the organisational culture, with hierarchy being the strongest influence. In general, it can be said that the clan culture is valued most in the organisation as it is high for every department. The prominence of the clan culture corresponds with the characteristics and goals of the organisation, namely, individual development, through LL sessions, and tradition (Cameron and Quinn, 2006). Furthermore, the influence of hierarchy can be explained by its need for reliable output, since the projects are often large in terms of money and scope, which requires standardised rules and procedures. Since the clan and hierarchy culture are found to be the most significant cultural influences for all departments, further research will focus on these two culture types. The determination of cultures aids in the development of the interviews, as it shapes the interview strategy regarding unit of analysis and background information. The departments with similar cultures (Engineering, Quality, Facility Operations) are grouped together and called Clan Organisational Culture (COC). For the IT business support department Hierarchical Subculture (HS) will be used.

Figure 3.7 - Perceived importance per LL type

The ranking of importance of project-specific LL, Figure 3.7, has resulted in some new insights. It shows the COC values most LL equally and a lot, and values communication more than HS, whereas task

1.5 2 2.5 3 3.5 4 4.5 5 Impo rt ance

Perceived importance project-specific LL per

department culture

(28)

responsibility, quality levels, risk levels and planning are perceived more important by the HS. Possibly caused by the difference in perceived authority between the two cultures and the hierarchy focus on process. Furthermore, risk, quality and planning are all ranked high for both groups, most likely also due to the nature of projects the organisation is involved in. Lastly, LL on shared language are valued the least by all. This could be caused by the strong influence of the clan culture, this culture is built around having a shared language which employees might take for granted.

Project commitment shows a positive correlation with Adhocracy (.417 with significance of .014) and a negative correlation with hierarchy (-.483 with significance of .004), which confirms earlier findings in literature (Krajcsák

,

2018; Pinho, Rodrigues, and Dibb, 2014). Furthermore, when comparing means, the COC shows higher project commitment (mean = 4.5) than the HS (mean = 4). This is consistent with previous research as well, as hierarchy is negatively correlated to project commitment, and clan positively (Cameron and Quinn, 2006). Since both groups have a combination of culture types, they all have some influence, the level of influence depends on the strength of each culture type within the composition.

Regarding the willingness to share, the two culture groups do not show much difference, with a mean of 4.46 for COC and 4.33 for HS. Only willingness to share success stories is found to be positively relating to clan culture (.474 with a significance of .029). However, due to the small sample size and low Cronbach’s alpha the measurements are likely to have been ineffective. As discussed in section 2.3.3, literature has found that both clan and adhocracy have a positive influence on the willingness to share, whereas hierarchy and market are negatively correlated (Suppiah and Singh Sandhu, 2011; Wiewiora et al., 2018). The interviews will be used to study these relationships and gain more in-depth knowledge on this.

(29)

3.3 Interviews

The interviews will be conducted to gain a better understanding of the practical aspects of the LL identification and sharing process. Opposed to questionnaires, interviews have the advantage of flexibility, the questions can be adapted and changed during the interview (Alshenqeeti, 2014). This flexibility aids in the explorative aspect of this study, since the aim is to identify why and how certain cultures affect the LL identification and sharing attitude.

3.3.1 Unit of Analysis and data collection interviews

From analysis of the questionnaire, participants are selected based on: (1) being a member of either the COC or the HS, (2) have been part of cross-functional projects at least once, (3) and have experience with the LLP. The interviews were conducted with two employees from the COC group, and two from the HS group, see Table 3.2.

Interviewee Department Culture Duration

1 Engineer Clan Organisational Culture 45 minutes 2 IT Business support Hierarchical Subculture 35

3 IT Business support Hierarchical Subculture 40 4 Facility Operations Clan Organisational Culture 35 Table 3.2 – Interviewee profiles

The respondents were interviewed, face-to-face, for approximately 40 minutes, using a semi-structured interview guide, see Appendix II. All interviews were recorded, after approval of the interviewees, for transcribing purposes.

3.3.2 Interview development

The interview guide is developed to answer RQ1 about how the culture types affect the attitudes of the two culture groups on both LL identification, perceived importance of each LL, and LL sharing, willingness to share LL and commitment to project learning.

(30)

commitment is reviewed, concerning the maximum effort they would put into learning and why. After which the willingness to share is discussed, relating to their experiences regarding their own input into LL sessions and the difference in attitude between sharing within the department and within the cross-functional project team. These questions are followed by the ranking of importance of each LL type, the ones perceived most and least important in the questionnaire findings will be discussed to identify the different attitudes towards which ones are important. After which all LL types are discussed regarding their importance within projects, this will give more insight into LL identification.

3.3.3 Data analysis method interviews

All interviews were recorded, after which they were transcribed. To simplify the analysis for the RQ1 interviews, answers were systematised using a coding tree, which organises statements extracted from the interviews into the constructs and topics (Basit, 2003). Statements were then analysed to determine the nature of influence of the attitude. This was done by cross-checking interviewee answers with culture type identification indicators that describe each culture, see appendix III. This clarified which culture type and/or organisational factor influenced the attitude, while maintaining neutrality of the researcher. Once all cultural influences of the attitudes were determined, the statements were linked back to their culture group to facilitate comparison between the two groups. An overview of the coding tree is given in the Table 3.3 below.

Coding tree

Construct Topic Description

1. General attitude towards LLP

Learning an

investment or expense

All statements relating to learning in projects and its perception of being a burden or benefit 2. Project learning

commitment

Balance learning vs. project deliverables

Comments and comparisons relating to the importance of project tasks and learning tasks. Maximum effort

committed to learning

Answers explaining what actions are undertaken regarding learning in projects.

Formal/Informal Statements referring to the approach taken to commitments, formal and/or informal. 3. Willingness to share

LL

Input

project learning

Descriptions on their own experience and history of sharing within project and why.

Input

department learning

Descriptions on their own experience and history of sharing within department and why.

Proactive/Reactive Mentions on future improvement actions. 4. Attitudes towards

project-specific LL types

Ranking All explanations of importance of types of LL and its ranking.

Communication with customer

(31)

Communication within project team

Comments on importance/unimportance of, and experiences with, LL on interactions and

communication within the project team, and why. Task responsibility

Comments on importance/unimportance of, and experiences with, LL on task responsibility, and why.

Shared language

Comments on importance/unimportance of, and experiences with, LL on having a shared language, and why.

Status reporting Comments on importance/unimportance of, and experiences with, LL on status reporting, and why. Quality levels

Comments on importance/unimportance of, and experiences with, LL on quality levels of products and services, and why.

Risk levels Comments on importance/unimportance of, and experiences with, LL on risk levels, and why. Planning Comments on importance/unimportance of, and

experiences with, LL on planning, and why. Clarification roles and

responsibilities

Comments on importance/unimportance of, and experiences with, LL on clarification of roles and responsibilities, and why.

Table 3.3 - Coding tree

3.3.4 Validity and Reliability of interviews

Lincoln and Guba (1985) state that there cannot be validity without reliability, therefore demonstration of validity suffices for interviews. The focus for qualitative research lies on generalisability and credibility. These factors can be enforced through different approaches in methodology (Creswell and Miller, 2000), of which three of them are used for this study; triangulation, peer review, and member checking.

Due to the combination of different data sources, triangulation enforces generalisation and trustworthiness when one confirms the other (Bekhet and Zausnziewski, 2012). First the different culture compositions and its effect on the dependent variables are determined with the questionnaire. After which these outcomes are tested during interviews, combined with more in-depth data collection for the underlying reasoning behind these attitudes. Lastly, findings are discussed during an evaluation session, gaining insights from a different point of view.

(32)

According to Lincoln and Guba (1985), the most essential technique to establish credibility, is by taking the data and interpretations back to participants. Since this study aims to develop practical solutions for the LLP, the LL facilitator is used for such member checking. Data and interpretations of other interviews are discussed, combined with the suggested solutions for improvement. The outcomes of that meeting are used to improve findings and solutions, increasing credibility and thus validity.

3.4 Evaluation

The evaluation session is aimed to both answer RQ2, how the new insights from this study can be applied in practice to improve the LL collection in cross-functional projects, and to validate findings of the questionnaire and interviews. The session is scheduled with the LLP facilitator, the one in charge of managing the LLP, this person leads the meetings and has the overview of how members from different departments behave differently. It is therefore the logical choice to discuss the findings and improvement suggestions with this person.

The evaluation session has a discussion setup consisting of three parts, (1) the current LLP management of cross-functional projects, (2) prevailing issues regarding departmental cultures and sharing LL in cross-functional projects, (3) evaluation of the usability of new insights gained from this study and proposed improvements. In the first part, the current practice is discussed to establish an understanding of the current level of management. In the second part, the prevailing issues are discussed to review departmental culture influences and if the LLP facilitator recognises these issues. Thirdly, the insights gained from this study are discussed and evaluated to identify how these can be used in practice together with other proposed improvements to increase alignment.

(33)

4. Results

The following chapter discusses the cultural influences of the COC and HS and its effect on attitudes towards identification and sharing of LL, determined on the basis of the interviews conducted at the case company. First, usability of the data is confirmed, after which results on the following constructs are provided: general view on the LLP, attitude towards importance of each type of LL, willingness to share LL, and commitment to project learning. Followed by a hypotheses summary and the results of the LL facilitator evaluation session.

4.1 Interview findings

The culture compositions of the two groups were validated as explained in section 3.3.3. Which demonstrates that Interviewees 1 and 4 of the COC are indeed mostly influenced by the clan culture with a strong hierarchy second influence. For interviewees 2 and 3 of the HS, it confirms the influences are both mostly hierarchy with a strong clan second influence. This therefore confirms both of the culture compositions as determined in section 3.2.5. In this case study, the culture types adhocracy and market are barely apparent in any department and are therefore regarded to be negligible. Consequently, hypotheses related to these cultures are not analysed.

4.1.1 General view on the LLP

The first question discussed the general attitude towards LL practices, of which both groups perceive the practices to be an investment rather than an expense. However, from the responses it could be drawn that the attitude of the COC revolves around personal and organisational development, whereas the attitude of the HS focuses on task efficiency, see Table 4.1.

General view on LLP

Topic discussed Clan Organisational Culture attitude Hierarchy Subculture attitude

Investment/expense LL practices

Investment:

Important for development of organisation and the team. Building onto experience is key.

Investment:

Makes everything more efficient.

It improves the business.

Table 4.1 - Attitude towards Lessons Learned Practice

4.1.2 LL identification: attitude towards importance of each type of LL

(34)

barely have issues with it, see Tables 4.2 - 4.5. Furthermore, both in the questionnaire and during the interviews, shared language was ranked the lowest importance by respondents of all culture groups.

Ranking of LL importance

Table 4.2 - Ranking LL importance interviewee 1

Interviewee 1: Clan Organisation Culture 1 Communication with the customer 2 Risk levels

3 Quality levels 4 Planning

5 Task responsibility

6 Clarification of roles and responsibilities 7 Communication within project team 8 Status reporting

9 Shared language/vocabulary

Table 4 .3 - Ranking LL importance inte rvie wee 4

Interviewee 4: Clan Organisation Culture 1 Risk levels

2 Communication with the customer 3 Communication within project team 4 Quality levels

5 Clarification of roles and responsibilities 6 Task responsibility

7 Planning 8 Status reporting

Interviewee 2: Hierarchical Subculture 1 Task responsibility

2 Planning 3 Quality levels

4 Communication within project team 5 Clarification of roles and responsibilities 6 Communication with the customer 7 Risk levels

8 Status reporting

Interviewee 3: Hierarchical Subculture 1 Quality levels

2 Planning

3 Communication with the customer 4 Task responsibility

5 Risk levels 6 Status reporting

7 Communication within project team 8 Clarification of roles and responsibilities 9 Shared language/vocabulary

(35)

Attitude towards the LL types ranked most important for each culture group

Clan Organisational Culture

attitude Attitude influences Hierarchy Subculture attitude Attitude influences

The customers are central to our business, therefore LL on communication with the customer are most

important. Clan influence - Loyalty - High commitment The make-to-order nature of product

and services cause customer to be involved from beginning.

LL on planning are important because good planning should result in good projects. If it is not planned correctly, it will be inefficient. Hierarchy influence - Smooth-flowing processes - Efficiency - Reliability

LL on risk are important because if there is a risk, it can hurt the team and the organization.

Clan influence

- Sense of we - Internal focus

Task responsibility lessons

should be shared because often this is where it goes wrong, if it is not clear who should do it, it does not get done. It increases efficiency if it is done right.

Hierarchy influence

- Efficiency - Bureaucracy

practices

Issues with quality affect either the organisation or both customer and organisation.

Clan influence

- Sense of we - Corporate

commitment

Quality is the core focus of

Siemens, if there is a quality issue it will impact everyone.

Clan influence

- Corporate commitment - Sense of we Issues with planning affects

someone somewhere. Clan influence _{- Sense of we} - Loyalty

Table 4.6 - Attitudes towards highest ranked Lessons Learned types

When analysing attitudes towards each LL type separately, the cultural influence becomes clear, see culture type and influencing concepts in Table 4.7. For HS 6/9 LL and for COC 7/9 LL types show a direct influence of their main culture on the attitude towards the type of LL. Additionally, 2/9 for HS and 1/9 for COC has influences of the second most influential culture. Lastly, communication with the customer is mostly affected by the make-to-order and servitisation aspects of the organisation rather than culture.

Attitude for each LL type and their main cultural influence

Lesson Learned Type Clan organisational culture attitude Hierarchy subculture attitude

Communication with customer

Customer is centre of attention. Not much interaction, but the varying requirements makes it necessary to share.

(36)

Clan influence - Loyalty - High commitment Hierarchy influence - Centralisation - Bureaucracy practices Communication within project team

Definitely important, but not a lot of issues with it, therefore less

important.

Need for good communication to have control, if there are issues with communication, people might start doing something else.

Clan

- Use of informal communication - Teamwork

- Shared language

Hierarchy

- Need for control

- No feeling of shared goals and values

- Weak loyalty - Weak commitment

Task responsibility

Everyone knows who does what. If this is not clear, nothing will get done.

Clan

- Teamwork - Consensus

Hierarchy

- Aim for efficiency - Authority practices

Shared language

Never an issue, we use the same terminology everywhere, so least important.

Never an issue, we all speak the same language, we all know what the abbreviations mean.

Clan - Shared language - Consensus - Sense of we Clan - Shared language - Consensus - Sense of we Statusreporting

Not a lot of issues with it. Not a lot of issues with it.

Hierarchy

- Standardised rules and procedures - Documentation

Hierarchy

- Standardised rules and procedures

- Documentation

Quality levels

It will hurt the organisation if the

quality is off. Core focus of Siemens, it has direct impact on organisation.

Clan - Corporate commitment - Sense of we Clan - Corporate commitment - Sense of we Risk levels

This hits Siemens, it hits us. With good planning and

communication, risk is mitigated.

Clan

- Sense of we - Internal focus - Loyalty

Hierarchy

- Stability and control - Reliability

(37)

Clan - Sense of we - Loyalty Hierarchy - Smooth-flowing processes - Efficiency - Reliability

Clarification roles and responsibilities

This is important because it affects planning, quality, risk and customer communication.

If not done correctly it will affect the efficiency and reduce the output.

Clan - Teamwork - Sense of we - Corporate commitment Hierarchy - Product focus - Efficiency

Table 4.7 - Attitudes towards each Lessons Learned type

4.1.3 LL sharing: willingness to share LL

The willingness to share LL was discussed to identify underlying reasons for sharing or not sharing LL with other project members. First the interviewees were asked to explain what input they have provided in the past in both team and department setting, after which the reasoning for doing so was described. The COC shows a high willingness to share in both the project team as well as in the department, expressing the developmental benefits it has for others in the team and organisation. The HS demonstrates to be less willing to share LL within project teams compared to within their department, in which they share everything. Caused by a feeling of unappreciation for their input by departments, as well as a lack in understanding of necessity. Furthermore, the COC appears to be more proactively involved and willing to improve to share more, whereas the HS display willingness to share with others but do not seek improvement actively, see Table 4.8.

Willingness to share LL

Topics discussed Clan Organisational Culture attitude Hierarchy Subculture attitude

Input provided into project learning

Share everything, can only benefit others and the team.

Most powerful way of learning is to involve all stakeholders.

In a project the team is the centre of all learning.

Only if it has a big impact, it will be shared, smaller things tend stay within the department. Never shared with whole departments or multiple people, only sometimes directly with one. Input does not always seem to be adopted, especially for high-pressure projects.

Input provided in department

Have monthly meetings Discuss LL with everyone in the department

Proactive/Reactive

Passionate and actively involved to

improve ways to share. Reactive, value learning but are reluctant to share, because they do not see the value in it often.

(38)

4.1.4 LL sharing: commitment to project learning

The balance of commitment towards learning and project deliverables was reviewed, during which both groups stated to be committed to project learning but to a lower extent than the direct project deliverables. To gain a better understanding of the level of commitment towards learning, the maximum effort committed to learning was reviewed. This has led to greater insights as the answers between the two groups vary substantially. Motivated by the feeling of the project being a team effort, the COC provides both formal input at LL workshops and shares LL informally during the project. The HS on the other hand, is committed to learning mostly informally within their department, see Table 4.9. Only for considerable issues the department would discuss it with others outside of their department. However, they do articulate the ambition to become more involved, of which they were of the opinion it would make them more committed to the project team learning.

Commitment to project learning and the team

Topics discussed Clan organisational culture attitude Hierarchy subculture attitude

Balance Learning and project deliverables

Definitely committed to the LL aspects, but less than project deliverables, especially when project pressure increases.

Always keeping it in mind whenever something is learned

Focus on LL is more off the cuff, not the main focus.

Maximum effort committed to learning

Always provides input during the LL workshops, no collection during project but does share LL informally.

Discuss within department then bring it across to other departments from department perspective. Would like to be involved more.

Formal/ informal Both formal and informal Informal

Table 4.9 - Commitment to project and learning

4.1.5 Overview hypotheses and findings

Counteracting Corporate Amnesia: The Effective Management of Attitudes towards Lessons Learned Collection in Project Based Make-to-Order Organisations