A change process within an institutional context : the Dutch retail case

(1)

A Change Process within an Institutional

Context: the Dutch Retail Case

Name Amber Hovestad

Student number 6071147

Date 25 March 2016 - final version

Program MSc. in Business Administration - International Management track Institution University of Amsterdam

(2)

Statement of originality

This document is written by Student Amber Hovestad, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this

document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

Purpose – This paper is about change management in a multinational. Specifically, the focus

of the research is the change process towards the implementation of a software system. The influence of voice versus exit, trust and authority on the change process will be examined. Furthermore, the researcher looks at how the change process evolves from initiation towards implementation and to what extent all conditions, necessary for change, are present. Ten critical success factors have been identified as conditions for change implementation. Specifically, the research focuses on the unfreezing and moving stage, based on Lewin’s (1951) three stages. In addition, recommendations and propositions for the refreezing stage will be made. Unanticipated or emergent changes are taken into account. Although there is much literature on change management, there is a lack of case studies, especially involving voice versus exit, trust and authority. Moreover, no cases have been found where research is conducted as observer participant. Consequently, since most initiatives for change fail, there is need fore more case studies where change processes are being researched.

Method – Qualitative research has been conducted. Nineteen semi-structured interviews have

been completed. In addition, relevant documents have been analyzed to fill in missing pieces of information. The researcher attended meetings for observation.

Findings – Voice, trust and authority are interrelated, vital analytical constructs for

implementing change. Change implementation will not be successful without input from employees and top management support and involvement. Therefore, managers should encourage voice and discussion and empower their employees. Leadership appeared to be very important, although not initially taken into account. Leaders should create and promote interpersonal trust. Trust is fundamental in many ways: in relationships, cooperation,

(6)

Implications – Organizations should not underestimate the importance of voice, exit, trust and authority on the change process. Employees should be involved and listened to, in order to guarantee user participation. Voice and trust are very important constructs influencing the change process, but also with respect to many other organizational aspects. For successful change implementation, ideally, the ten critical success factors identified in this research should be met. Especially top management support and involvement and adequate project management.

Limitations – Although open access was agreed upon, in real time, access was not always arranged: possibly, not all information have been provided. In addition, people may have not told everything, which may have biased the results. Furthermore, a single case study may not be enough to generalize from. In future research, more research may be needed to provide further insight in change processes in multinationals. Finally, a discrete and limited time frame posed limitations on the scope and quantity of data collection.

Key words – Change management, Change process, Multinational, Voice, Exit, Trust, Authority

(7)

Introduction

The business environment is rapidly changing. Organizations are facing increasing

competition, expanding markets and higher customer expectations. To remain competitive, organizations must improve their business practices and procedures (Umble et. al, 2002). In order to improve, innovation is key. Nowadays, the use of data is becoming more and more important to improve performance. Data enables organizations to be more effective and efficient in coordinating and responding to customer demand, supply and production and to cope with high competition (Manyika et. al, 2011; McAfee and Brynjolfsson, 2012; Umble et. al, 2002). All kinds of software systems have become available for managing data.

IBM (2011) found that organizations are struggling to automate processes and to deal with large amounts of data, turning it into useful insights and actions. Innovation asks for change. Organizational change is about making “fundamental changes in how business is conducted in order to help cope with a new more challenging market environment” (Kotter, 1995, p. 59). In most cases, change is initiated when there is reason to believe that an organization is misaligned with the environment, which prevents an organization from

functioning as it should, to succeed. However, most initiatives for change, fail (Burnes, 2004; Kotter, 1996, 2006).

Much research about change management has been conducted. However, despite the fact that many organizations cope with change, little empirical research has been conducted about the change process involved. This research is about change management in a

multinational. Specifically, the change involves the implementation of a software system: which uses historical data to improve order-forecasting and capacity management. Therefore, the research focuses on this specific change project, where the influence of voice versus exit, trust and authority on the change process will be examined. The researcher looks at how a change evolves from initiation to implementation, thereby focusing on the unfreezing and

(8)

moving stage (Lewin, 1951). Besides planned change, potentially unplanned changes will be taken into account. The research question is: “What is the influence of voice versus exit, trust

and authority on the change process and which role do labor and norms and values play within an institutional context?”.

As conditions for realizing change, ten critical success factors have been identified, based on previous research on the implementation of software systems. Critical success factors include factors such as top management support, resources and project management (Armenakis and Bedeian, 1999; Nelson and Somers, 2001; Umble et. al, 2003; Wixom and Watson, 2001; Zhang et. al, 2005).

The next chapter discusses relevant literature, where the concept of organizational change will be introduced, with different perspectives on and types of organizational change. Change in multinationals will be illustrated with some case studies. The last paragraph is about the influence of (inter)national context on change. The second chapter is a justification for the chosen research, method and analysis. In the third chapter, the results will be

discussed for each analytical construct. In the fourth and fifth chapter, discussion and conclusion will follow.

(9)

1. Literature review

1.1 Organizational change

In this paragraph, organizational change will be defined and explained. Though there are many different approaches to organizational change, there is a general agreement that the two prominent ones are the planned and emergent approaches (Bamford and Forrester, 2003; Burnes, 2004; Stace and Dunphy, 2001). Lewin (1951) introduced three stages of planned change; Kotter identified eight stages of emergent change. In this paragraph previous theories and several types and motivations for the application of these types will be outlined.

There is a lot of literature on organizational change. At a general level, change is defined as “a phenomenon of time. It is the way people talk about the event in which

something appears to become, or turn into, something else, where the ‘something else’ is seen as a result or outcome” (Ford and Ford, 1994, in Weick and Quinn, 1999, p. 362). In

organizations, change is difference “in how an organization functions, who its members and leaders are, what form it takes, or how it allocates its resources” (Huber et. al, 1993, in Weick and Quinn, 1999, p. 363).

The goal is “to make fundamental changes in how business is conducted in order to help cope with a new more challenging market environment” (Kotter, 1995, p. 59). In most cases, change is initiated when an organization is perceived to be misaligned with the environment, which prevents an organization from functioning as it should to succeed. One of the major causes of misalignment is change in the environment at a technological and economic level.

1.1.1 Stages of change

Many authors have defined stages of change. However, two of the most influential will be discussed: Lewin’s three stages and Kotters’ eight stages of change. Lewin (1951) defined

(10)

three stages of planned change: unfreeze, change and refreeze. Unfreeze is the stage where the status quo is being examined. The driving forces for change will be increased and resisting forces against change will be decreased. Change is the stage where movement will occur. Action is taken, changes are made and people throughout the organization will be involved. In order to realize change, people should be empowered to make decisions. Change should be made permanent and desired outcomes rewarded, to finally fulfill the refreezing stage. These stages are still used as a basis for organizational development. From the 1960s to the early 1980s, organizational development established itself as the dominant Western approach to organizational change (Benn et. al, 2014).

Most initiatives for change fail (Burnes, 2004; Kotter, 1996, 2006). Kotter (2006) states that too many managers do not realize that transformation is a process, which includes many stages, and is not an event that happens at once. The relevant stages build on each other and each stage has its own unique pitfalls to take into account. It is important to know and build on each stage to work towards a change implementation.

Kotter outlined eight stages in his 1996 book ‘Leading Change’: from establishing a sense of urgency, to communicating a vision, to changing a culture (Kotter, 2006). The eight stages of Kotter have similarities with the three stages of Lewin (1951). Lewin’s unfreeze corresponds with establishing a sense of urgency, creating a guiding coalition, developing a vision and communicating the change vision. Move or change is the stage where people are empowered to realize change, generate and share short wins and gains. Refreeze is the stage where new approaches are incorporated in the culture and throughout the organization. Kotter’s stages of organizational change, with the actions needed for success and the pitfalls of each stage are displayed in Appendix 1.

(11)

1.1.2 Types of change

There are several types of change. A distinction can be made, based on rhythm. Change can be episodic, discontinuous and infrequent, but also continuous, evolving and incremental (Weick and Quinn, 1999; Dunphy and Stace, 1988). The former is also known as ‘second-order, ‘radical’ and ‘transformational’, the latter as ‘first-order’.

Episodic change occurs in the order of unfreeze-transition-refreeze, whereas

continuous change follows freeze-rebalance-unfreeze. Episodic change occurs during periods of divergence, when organizations are moving away from their status quo because the

environment demands. Continuous change implies that organizations are emergent and self-organizing; change is constant and evolving (Weick and Quinn, 1999; Dunphy and Stace, 1988). Watzlawick et al (1974) and Bateson (1972) were the first to make a distinction between incremental and radical change, or ‘first-order’ and ‘second-order’ change. This distinction is still often used in research (Weick and Quinn, 1999, p. 363).

Radical or transformative change started to gain attention after changed environmental conditions demanded a different approach: in the mid to late 1970s and 1980s the

environment became more turbulent in many sectors in Western economies. Drucker (1969, in Dunphy and Stace, 1988) calls this the ‘age of discontinuity’. Incremental change fits well to an environment of stability and growth. Dunphy and Stace (1988) provide an argument for this approach, stating that an effective manager improves an organization by small, logical steps by acquiring environmental information over a longer period of time. This give the entire organization time to internalize and adjust.

Dunphy and Stace (1988) state the essential difference between incremental and transformative change is whether organizations are constantly learning and changing at all levels through the organizations or whether they learn and change periodically, with large-scale changes in either strategy, structure or processes (Kimberly and Quinn, 1984 in Dunphy

(12)

and Stace, 1988). Dunphy and Stace (1988) are proponents of a more contingent approach to change (Bourne et. al, 2003).

Some scenarios may create conditions where large-scale changes are needed, such as: environmental creep; organizational creep; diversification, mergers and acquisitions or entire shutdowns; reorganization of the industry and major technological breakthroughs. During an environmental creep the environment may be changing incrementally and hardly visible to managers. Eventually the impact of the change may require a major adjustment by the

organization. In the scenario of an organizational creep, the organization itself may move out of strategic alignment in an environment that may remain quite the same. In this situation, it is thus the organization creating a misalignment with the environment (Dunphy and Stace, 1988).

Diversification, acquisition, merger and shutdown involve major structural changes. Incrementally changing may not be realistic. Both industry re-organization and major

technological breakthroughs may have an impact on structures or costs and the organizations involved may need to transform their structures or systems radically and fast in order to respond adequately to the changing environmental circumstances (Dunphy and Stace, 1988).

1.1.3 One best way to manage change

A comprehensive change theory includes both incrementalist and transformational approaches to change. It depends on the circumstances, which approach to apply: when a situation is relatively stable, incremental change may be the most appropriate. Under

conditions of large economic developments, such as recession or discontinuity, a more radical approach may be needed. A key element in achieving success, is to choose the most

appropriate approach under the specific circumstances (Burnes, 2004; Dunphy and Stace, 1988; Huy and Mintzberg, 2003).

(13)

Huy and Mintzberg (2003) distinguish three types of change: dramatic, systematic and organic. Their main position is that neither type of change works well in isolation. Each type has its requirements and complements the other types in some way. To illustrate this position: dramatic and systematic change needs to be balanced by order or leadership in order to work. Both depend on acceptance and engagement throughout the organization. Furthermore, to realize organic change, it should be supported by leadership and be managed in a systematic way. The authors state that organizations should prevent dramatic change initiatives.

In conclusion, organizational change can be characterized by rhythm; episodic or continuous change, or by nature; planned versus emergent change. Based on scale; there is large-scale versus incremental, based on pace; rapid versus slow change, and so on. As Bamford and Forrester (2004, in Burnes, 2004, p. 899) concluded, there is no ‘one best way’ to manage change. Key for managers is to define clear goals on what they are trying to

achieve, the context in which their organization is operating and the strengths and weaknesses of the various approaches to change (Burnes, 2004; Dunphy and Stace, 1988; Huy and

Mintzberg, 2003).

1.2 Organizational change in multinationals

Brock and Birkinshaw (2004) identified six major changes facing multinationals: increasing levels of global integration, innovation by design, new network structures, outsourcing of major value-chain activities, ecommerce, and the backlash against globalization. Geppert et. al (2003) make a distinction between global and national contextual forces which have low context; like economic and technological, and high context; like institutional and cultural, dimensions. Change management processes at a subsidiary of a multinational are based upon both factors. The main finding of the authors was that the more globalized the strategies and structures of a multinational, the more it allows for and relies on national specific

(14)

characteristics to play a role in each national subsidiary within an organization’s portfolio. Burnes (2003) conducted two studies at multinationals involved in change: ‘ABC Engineering’ and ‘XYZ Construction’. In both organizations, there was need for large structural changes. Whereas the change in ABC failed, the change in XYZ was successful. ABC wanted to transform in a ‘Lean Enterprise’ but did not succeed in changing its organizational structure. The key problems identified, were because previous change initiatives had not been successful. To explain: the organization’s existing practices were inefficient, the trade unions experienced a lack of trust in the management and senior managers did not really understand the meaning of the new concept. Furthermore, the company lacked the required knowledge and skills to implement the change initiative and make it a success. Above all, the management team did not work as a team and the managing director lacked trust in his team (Burnes, 2003, p. 634).

As in the case of XYZ, the structure needed to change because there were problems between the head office and the regional offices within the organization, concerning

communication and poor work relations. Later on, there was an initiative for restructuring the organization into a more teamwork-based, process-focused organization. The management sought outside assistance since he believed that XYZ lacked the skills to plan and implement the change. Eventually, the new structure was launched (Burnes, 2003).

The development and implementation of the new structure at XYZ is a case of planned change, rather than emergent change. The planned change consisted of five stages, similar to Lewin’s three stages model (1951). The change process started with gathering information on the shortcomings of the current structure and trying to determine the degree of readiness to change. Since it also served to prepare the whole workforce for change, it was the beginning of the unfreezing stage (Burnes, 2003).

(15)

choosing the most appropriate and supported one. In stage three there was a follow-up with communication training for employees and in stage four the implementation started. In the fifth stage the refreezing stage took place. Managers were expected to acknowledge the new structure and were prepared to make further changes to reinforce the new structure. These last developments also seem very much like Lewin’s three stages model (Burnes, 2004, p. 898). Burnes evaluates the degree of change based on Kotter’s eight stages. In the case of ABC, the management wanted to implement a large structural change, however, management change was also needed because of shortcomings in their skills and competencies to be able to achieve such change. In the case of XYZ, the management was aware of the fact that the organization was not capable of implementing the change, and sought help outside.

Eventually, most of the workforce was involved in the process and a new, effective structure was launched (Burnes, 2003, p. 635).

The learning from both cases is that change requires the appropriate skills and competencies of the organization’s management. Furthermore, the types of change do not have to be mutually exclusive. In the case of XYZ, both emergent and planned change were part of an integrated program. Therefore, emergent and planned change can complement each other (Burnes, 2004).

1.3 The influence of (inter)national context on change

In this paragraph, the main contextual factors and context perspectives will be outlined. The case study of Bamford and Forrester (2003) illustrates the influence of both internal and external contextual influences on the process of change. To choose the most appropriate type of change, the particular contextual circumstances must be taken into account. Burnes (2004) notes that it is only by understanding the context within which change takes place, that the most beneficial types of change can be introduced in the most effective way.

(16)

There are ‘high context’ and ‘low context’ perspectives in the literature on national context. High context perspectives pays high attention to national contexts. Low context perspectives are insensitive to specific national contexts and see corporations structured by their environment, which in most cases consists of markets. High context perspectives, on the other hand, see organizations as deeply enrooted in institutions and socially embedded into their national contexts (Geppert, 2003).

Bamford and Forrester (2003) conducted a case study at an international

manufacturing organization. Over the years, the organization tried to improve by a couple of initiatives. Both external and internal forces affected the change initiatives and processes. One key learning of the research of Bamford and Forrester (2003) is that previous experiences can influence the way change initiatives are received within the organization. In a relatively stable economy employees may not see reason for change, while employees in a less stable economy subject to a great deal of socio-economic changes in the preceding years more likely will. These differences may result in people less or more willing to participate in change initiatives (Bamford and Forester, 2003).

Contextual factors may influence strategy formulation and implementation. For

example, the societal context, as part of the institutional context, involves informal institutions like national culture, norms and values. In this case study, the influence of ‘voice versus exit’, ‘trust’ and ‘authority’ on the change process will be studied within an institutional context. On an institutional level, labor will be taken into account; on a societal level, norms and values. Berry et. al (2010) identify several dimensions of cross-national distance. The cultural dimension, with respect to norms and values, is defined by differences in attitudes toward authority, trust, individuality and importance of work and family (Berry et. al, 2010; Hofstede 1980; Whitley, 1992). Labor is organized around the particular representation rights in a country and the presence and organization of labor unions. Representation rights of

(17)

employees differ among countries, but also among industries and organizations. The degree of coercion is an important factor, concerning unilateral employer action, collective bargaining and law. Means to establish employee representation include establishing representation rights contractually, such as collective agreements, but also law or interventions by the state.

An institutional context with weak representation rights often lacks internal means to represent employees, excluding employees from organizational decision-making. In that case, largely organized labor may enable employees to mobilize collective action against

managerial decisions. An institutional context with strong representation rights implies strong internal channels, giving employees a voice in the organizations’ decision-making, protected by law (Aguilera and Jackson, 2003).

Furthermore, employees may influence decision-making by means of four different strategies, by either external control or internal participation. Strategies of external control are most effective when the management of an organization has decided to make the decision-making its exclusive privilege. Collective action may result in democratization of the decision-making process. Central in internal participation is consensus and cooperation in making and implementing decisions.

The extent, to which employees may consider leaving an organization, partially depends on the way labor is organized. This, in turn, may influence the extent to which people exercise voice. ‘Voice’ refers to trying to improve conditions by discussing issues with a supervisor or colleagues, taking action to solve problems, by suggesting solutions or seeking for external help, like help from a labor union. Voice is considered to be active and

constructive, whereas exit is considered active but destructive in terms of employee-organization relationships (Farrell et. al, 1988, p. 601). ‘Exit’ refers to leaving or thinking about leaving an organization, but also quitting, looking for a transfer or a different job opportunity.

(18)

Besides the way labor is organized, the skills of employees and the degree of investment in these skills are important factors influencing the extent to which employees favor voice over exit or vice versa. When the skills of employees are transferrable across firms or an organization does not invest (enough) in the skills of its employees, employees may favor exit over voice. However, when skills are firm specific, employees have a greater dependence on the firm. This makes their incentive to leave less likely to exist. Therefore, investing in firm-specific skills creates incentives to exercise voice, for example in the way these skills are formed and deployed but also in sharing comments and opinions (Aguilera and Jackson, 2003).

1.4 Theoretical framework

The purpose of this research is to look at how a change process evolves from initiation towards implementation and to what extent all critical conditions, necessary for change, are present. Specifically, the research focuses on the unfreezing and moving stage, based on Lewin’s (1951) three stages of planned change. However, also unanticipated or emergent changes are taken into account.

The influence of ‘voice versus exit’, ‘trust’ and ‘authority’ on this change process will be examined. At the same time, the institutional contextual factors ‘labor’ and ‘norms and values’ are expected to have an influence on these analytical constructs. The next paragraph discusses the propositions. In addition, ten critical success factors have been identified as conditions for the desired change implementation. These will be discussed in chapter 2.7. The following visualizes the conceptual framework:

(19)

1.5 Propositions

In The Netherlands, labor is organized around the principles of the welfare state.

Characteristic is the Calvinist tradition with a high level of care for employees. Managers and labor unions have a commitment to employee security (Whitley, 1998). Therefore, the

expectation is that employees have a high degree of ‘voice’ within Ahold (1.1). A high degree

of voice enables employees to communicate their needs. This will most likely positively influence user participation and involvement, which is an important criterion for future change implementation (Pinto and Slevin, 1987; Umble et. al, 2003).

However, also the (non-) transferability of skills influences voice versus exit. As mentioned before, when skills of employees are transferrable across firms or an organization does not invest (enough) in the skills of its employees, employees may favor exit over voice. However, when these skills are firm specific, employees have a greater dependence on the firm. This makes their incentive to exit less likely to exist. Investing in firm-specific skills creates incentive to exercise voice. Since Ahold is known as a good employer in the Netherlands (Van Middelaar, 2015), offering opportunities for (personal) development, the

(20)

Regarding the influence of voice on implementation, a high degree of voice assists in communicating employees’ needs. Consequently, user participation will occur. User

participation is an important criterion for change implementation (Pinto and Slevin, 1987; Umble et. al, 2003). Therefore, employees’ high degree of voice is expected to positively

affect the change process (1.3).

Trust in work relations may be influenced by the Dutch norms and values. La Porta et. al (1997) state that the degree of trust in large organizations also depends on national

institutions. Although the Dutch are individualistic, people highly value moral accountability, which is considered an individual responsibility (Olie and Van Iterson, 1992). The

assumption is that the Dutch norms and values have a positive effect on goodwill trust among co-workers (2.1). However, since there is a lower emphasis on personal career and

excellence, trust in the competences of co-workers may be lower because people may seem less result- and performance driven, and therefore potentially less capable to behave or perform as expected. Therefore, the assumption is that the Dutch norms and values have a

relatively negative effect on competence trust (2.2).

People can directly affect the success or failure of implementing change (Brooks, 1975). In particular, both technical and interpersonal skills of the people involved may have a major influence on the outcome of a project (Barquin and Edelstein, 1997; Wixom and Watson, 2001). Therefore, data has been collected about competence trust, referring to the perceived skills of colleagues. Consequently, the assumption is that the expected low degree

of competence trust decreases the overall team skills, having a negative effect on the change process (2.3). Besides team skills, project management and user participation and

involvement are essential in implementing change. All are related to people and their skills, central in the process of implementing change (Umble et. al, 2003; Wixom and Watson, 2001).

(21)

In addition, data has been collected about the relationships between colleagues within a team or department and the relationships with superiors. Besides the assumed advantage of high competence trust, a high degree of goodwill trust is assumed to enhance cooperation and consequently the interpersonal team’s skills. This may also positively affect the

implementation, because only a highly skilled and competent team can identify the

requirements of complex projects and solve potential problems on the way: a high degree of

goodwill trust positively effects the change process (2.4) (Umble et. al, 2003; Wixom and

Watson, 2001).

The reserved policy of governmental and financial institutions seem to contribute to managerial discretion and the cooperative attitude from the existing labor unions towards capital and management furthermore reinforces the idea of the Dutch firm as nexus of interests, involving all stakeholders. Employees can protect their interests in relation to decision-making by means of both external control and internal participation. The expectation

is that employees at Ahold do experience a relatively high granted authority and are actively involved in decision-making (3.1).

Furthermore, authority is assumed to have a positive effect on the change process

(3.2), in this case, a system. Barki and Hartwick (1994) found that users who participate in

the change process are likely to develop beliefs that a new system is good, important, and personally relevant. Give people part of the responsibilities, to think, discuss and decide. Through participation, users may be able to influence the design of a new system, satisfying their needs. Participation through discussion, autonomy and empowerment may develop feelings of ownership. They may develop a better understanding of the new system and how it can help them in their job, increasing chances of change implementation (Barki and Hartwick, 1994, p. 72).

(22)

The conceptual research question is: “What is the influence of voice versus exit, trust and authority on the change process and which role do labor and culture play within an institutional context?”

(23)

2. Research and methodology

This study examines the influence of voice versus exit, trust and authority on the change process. The study consists of a single case study. This chapter will discuss the research method, the unit of analysis, the specific research and institutional context, the analytical constructs, different kind of measures, analyses and the data used in this study.

2.1 Research design

For this research, a single case study has been chosen as the most appropriate study because it can provide a rich understanding of a phenomenon (Bamford and Forrester, 2003; Burnes, 2003, 2004; Yin, 1994). A single case study is used for the exploratory research’s objective to discover information about the topic, which is not clearly understood (Searcy and Elkhawas, 2012; Straus and Corbin, 1990).

Some other advantages of a single case study are that it enables researches to develop grounded theories that are both practical and relevant and causal relationships can be made because of the higher level of validity due to longer-term observations. Furthermore, a single case study provides a holistic view of a phenomenon in a real world setting. However, limitations may be that the reliability and external validity of quantitative research methodologies may be higher (Bamford and Forrester, 2003).

Denzin and Lincoln (1998) and Robson (1993) also state that qualitative research can capture the real world context including events taking place. It enables functions such as to capture the essence of events, such as a change process and how it unfolds. The more sources of evidence collected, the higher the accuracy and validity of findings (in Bamford and

Forrester, 2003; Burnes, 2003, 2004). Therefore, data will be collected mainly by the methods of observing meetings and interviewing both managers and lower-level employees.

(24)

2.2 Unit of analysis

The Dutch multinational retailer, Ahold, has been chosen as the unit of analysis. Here, the opportunity existed of being embedded in the organization while conducting research as an observer participant. Specifically, the unit of analysis is the change project. The scope of the research was the ecommerce department of Albert Heijn, referred to as ‘Online’, where the researcher had been working for already half a year.

Advantages include established relationships and open access. Furthermore, three days a week was spent as ‘observer participant’, embedded in the organization. An observer participant may see the change process evolve over time. It is worth noting that the change project started March 2015, but somehow there was decided not to proceed. At that moment the researcher was not yet involved. However, later on that same year, in December 2015, the change project continued. From that moment on, the researcher was involved throughout the process as observer participant.

Research has been conducted to find out why the change project was delayed and which unanticipated changes emerged. The part of the change project around March 2015 will be regarded as the ‘unfreezing’ stage, when it all started. The part from December 2015 on will be regarded as the ‘moving’ stage, where actual changes were made.

2.3 Research context

As a retailer, Ahold operates in a very dynamic and competitive environment. Bamford and Forrester (2003) found that external factors affect change initiatives and processes. Within the organization, there are several initiatives for changes. One of the changes is the

implementation of a software system, which is the focus of this research. The software system uses historical data to forecast orders. It will improve current order-forecasting and capacity management because its forecasts are more accurate. Consequently, it will enhance

(25)

efficiency. The software system is in line with a new strategy, towards more data-driven decision-making in order to improve overall managerial decision-making.

2.4 Institutional context: The Netherlands

The research at Ahold is conducted within the specific institutional context of the

Netherlands. Specifically, the influence of labor and norms and values on voice versus exit, trust and authority will be examined. In the case of Ahold, change is initiated because both technological and economic changes demand adaptation. As stated before, more and more organizations are incorporating data. In order to cope with competition and remain the market leader in retail, Ahold is heading towards more fact-based decision-making.

The Netherlands is an example of a combination of business systems, as defined by Whitley (2000). The Netherlands “combines a corporatist state with an elaborate system of technical education and strong legal constraints for the market for corporate control, but which also has a banking system that is more Anglo-Saxon than collaborative in many respects” (Iterson and Olie, 1992, in Whitley, 2000, p. 859).

Whitley (1998) describes the institutional context of the Netherlands by emphasizing the characteristics of institutions and historical social developments. The environment is shaped by three features: “the system of authority relations, the system for establishing trust and obligation relations between exchange partners, and, finally, the structure and policies of political and bureaucratic state elites” (Whitley, 1998, p. 105).

Characteristic for the Dutch business system is the number of large multinationals. The Dutch business system has a very strong international orientation, is mostly distributed in sectors and includes a strong focus on trade and services. The Dutch are known for their developed ‘welfare state’, with a high level of care placed on employees. This is further reinforced by the Calvinist tradition in The Netherlands, which emphasizes “thrift, sobriety

(26)

and a realistic attitude” (Whitley, 1998, p. 107).

Furthermore, Dutch people are individualistic, but consider moral accountability an important individual responsibility. “The emphasis is on conformity rather than on individual excellence, which helps to explain the low emphasis on individual performance, the relative lack of a careerist attitude among managers, and the unpretentious leadership style”.

The market organization is organized around short-term and instable market relations. Also, the present hierarchy changes often and little financial linkages exist between

organizations. Most of these are defined in terms of control. Law, in contrast to most other countries, does not strictly forbid collusion between organizations (Iterson and Olie, 1992). The degree of managerial discretion from owners is high in The Netherlands. Family-owned organizations and state-Family-owned organizations are relatively rare. There is considerable separation of ownership from control, resulting in high managerial discretion, which is further facilitated by the permanent executives, the relation of trust with the supervisory board and the minor influence of shareholders (Whitley, 1998). In addition, shareholders and employees have a fair degree of influence: the management is required to take the interests of all involved parties into account. There is a strong emphasis on compromise and consensus-building in the Netherlands. Decision-making is based on consensus, managers apply a non-authoritarian leadership style (Iterson and Olie, 1992).

Concerning the influence of institutional factors, the reserved policy of both

governmental and financial institutions seems to contribute to managerial discretion as well. Existing labor unions cooperate with organization’s management. The unions focus on broad issues: they consider all social and economic issues in Dutch society, which is called a tradition of ‘broad-unionism’ (Iterson and Olie, 1992). The main focus of the unions,

however, is to work out collective labor agreements, the Dutch call ‘CAOs’: these are legally established agreements that represent the outcome of negotiations between labor and

(27)

management (Whitley, 1998, p. 112). Furthermore, both are committed to employee security. This prevents large-scale dismissals of employees. Furthermore, “both the unions and the Enterprise Council, which handles most matters related to work and working conditions in

2.5 Method and data collection

The research is conducted by means of qualitative data collection techniques at the ecommerce department of Albert Heijn, Royal Ahold. The ecommerce department is the scope of the research. Data was gathered by interviewing and documentary research, using meeting notes, a blue print and a management presentation.

In total, twenty-two interviews were conducted. Three in addition to the documents for the first measurement, nineteen interviews for the second measurement. The interviews were in a private one-on-one, but very informal setting. The interviews were semi-structured so no important topics or questions would be missed, but there was still room for free

conversation. Each interview started with an introduction and the question whether it was approved to record the conversation. Participants were encouraged to speak freely. The informal nature and established relationships were expected to support in the acquisition of as much in-depth knowledge as possible.

In order to safe guard the interviewees’ anonymity, no further explanation about their roles or departments will be given. The purpose of using multiple sources of data is to strengthen research findings, as the evidence is triangulated. Explanations will enable

understanding of the phenomenon and theory will be tested (Farquhar, 2012).The purpose of the first measurement is to identify pitfalls caused by emergent changes in the unfreezing and moving stages of the change process and the extent to which these can be explained by the presence or absence of voice, trust and authority.

(28)

Table 1: First measurement Documents Blueprint Managerial presentation Meeting notes Table 1.1: documents Interviews Time

Project manager Ahold 58 Minutes

Project team lead 47 Minutes

Project manager vendor 70 Minutes

Table 1.2: interviews

The second measurement is based on semi-structured informal interviews and meeting

observations. Fifteen people at all levels were interviewed. Specifically, those people who are expected to work with the new software system in the nearby future and who are now

involved in the testing process: at Market Intelligence, Operations, Finance and Commerce, but also the Management Team.

Interviewees were selected by non-probability sampling, since the sample was based on a selection by the management and interviewer. The sampling procedure for both

measurements was, therefore, purposive (Farquhar, 2012). Open access was arranged.

However, this was no guarantee for one hundred percent involvement throughout the process of the research. The purpose of the second measurement was to research to what extent people have voice, trust and authority and how this relates to the change process in general, with an emphasis on the moving stage.

(29)

Table 2: Second measurement

Department Interviewees Interviews

Management team 3 Senior director operations:

interviewed twice

Finance 2

Market intelligence 4 Project team lead: interviewed

twice

Operations/Improvement 4 Manager operations:

interviewed twice

Commerce 3

Total 16 19

Table 2: interviews conducted. Each interview lasted 45-60 minutes

Appendix 2 shows the interview protocol, including interview questions. Topics for the management interviews were almost similar regarding the topics, however, questions differed to some extent. In the first measurement, questions differed to the extent that they were asked to fill in missing pieces from the document reviews. Appendix 3 shows the observational protocol.

2.6 Data analysis

The first measurement is qualitative content analyses, used to analyze documents (Hsieh and Shannon, 2005). Documents about the project have been reviewed: a blueprint, meeting notes and a management presentation to identify emergent changes in the change process.

Qualitative content analysis focuses on communication with attention to the

contextual meaning of the texts in the documents involved (Kondracki and Wellman, 2002; McTavis and Pirro,1990). It can be defined as “a research method for the subjective

interpretation of the context of text data through the systematic classification process of coding and identifying themes or patterns” (Hsieh and Shannon, 2005, p. 1278).

Documents have been collected by asking people involved in the project to share information with the researcher. Multiple sources were used: a blueprint, meeting notes and a

(30)

management presentation. Since open access was arranged and the involved people agreed on sharing information, there may be assumed that these documents were sufficient for analysis. In addition, semi-structured interviews were conducted with key persons involved.

The analysis started with reading all documents to be able to achieve understanding of the project, its main objectives and the people and parties involved. Data was then read again, to be able to acquire more in depth knowledge. Codes have been derived by highlighting words and sentences, which seemed to capture the key thoughts and concepts (Hsieh and Shannon, 2005).

Next, notes of impressions, thoughts and a first analysis have been made. Missing information was written down in the form of interview questions, to be able to fill in the missing pieces later on, by data retrieved from the interviews. During the interviews, the researcher made notes. Interviews were then transcribed. Both notes at transcriptions were coded with Atlas.ti, a program for qualitative data analysis. Just like with the documents, the analysis started with reading all documents (Hsieh and Shannon, 2005). Codes were based on the assumed most important topics and most of the time consisted of one or a few sentences. Codes were used to identify themes and patterns, within and between codes. Again, codes were labeled and categorized (see Appendix 4).

Most labels were based on either the relevant analytical constructs or aspects of these analytical constructs. This resulted in the initial coding scheme. Codes were gathered in categories to be able to organize codes into meaningful clusters of information.

Categorization has the purpose of generating analytical constructs, which are supposed to measure what the research intends to measure (Weber, 1990).

(31)

2.7 Analytical constructs 2.7.1 Labor, norms and values

There are two institutional factors, labor and norms and values, taken into account. Most information was gathered by reviewing existing literature, such as Whitley (1998).

2.7.2 Voice versus exit

Voice refers to the extent to which people speak up, for example whenever they disagree with decisions or when they are trying to improve conditions, taking action to solve problems or suggesting solutions. Exit refers to the extent to which people consider leaving the

organization. Exit is influenced by the extent to which skills are firm specific and by the extent to which an organization invests in skills. People may have a higher incentive for leaving when they perceive an organization to invest insufficiently in their skills. Voice is being researched by asking people whether they think they have voice, whether they speak up whenever they disagree with decisions and whether people listen to them. Information on voice and exit was gathered by asking interview questions like: “Do you have a voice within Ahold?” and “Do you ever consider leaving the organization?”. Results about voice are based on the researcher’s interpretation and classified in a ‘low’, ‘moderate’ or ‘high’ degree of voice, based on the overall responses. The extent to which people disagree with decisions have been classified in ‘never’, ‘sometimes’ or ‘often’. Opinions about investment in skills were classified in three categories: ‘low’, ‘moderate’ and ‘high’.

2.7.3 Trust

Trust is about ‘goodwill’ and ‘competence’ trust, where goodwill refers to attributions regarding the intention of a party to behave in a trustworthy manor and competence refers to

(32)

attributions regarding the ability of a party to behave or perform as expected. Therefore, two-way questions were asked, like: “To what extent do you trust your colleagues to have positive expectations regarding your motivations?”, and the other way around: “To what extent do you trust your colleagues to behave in a trustworthy manner?”. Opinions about investment in skills were classified in three categories: ‘low’, ‘moderate’ and ‘high’.

2.7.4 Authority

Authority is defined by autonomy, discussion, empowerment and power relations. Herbert Simon defined authority as the right to select actions affecting part or the whole of an organization. ‘Autonomy’ means that people are autonomous bounded entities that express their preferences, opinions and ideas (Schwartz, 2006).

During the interviews, questions were asked like “Who makes decisions?” and about the extent to which people participate in decision-making and are allowed to make their own decisions. A management interview question was: “To what extent do people have

autonomy?”. ‘Discussion’ refers to the activity in which people talk about things and openly discuss their considerations. An example of an interview question asked to a manager is: “To what extent does the management encourages discussion?”. ‘Empowerment’ is the extent to which the management allows its workforce to have input into the planning or implementation of a change, or to provide support and ease the transition during a change (Caldwell et. al, 2008; Ugboro, 2006). ‘Power relations’ is about relationships with colleagues, their superior and hierarchy within teams, departments and the organization as a whole. It seems to relate to authority and factors of implementing change, like top management support.

Identified codes are, for example: ‘Hierarchy’, ‘Relationship with colleagues’ and ‘Relationship with superior’. Classifications regarding the level of autonomy, discussion and

(33)

empowerment are ‘low’, ‘moderate’ and ‘high’. Relationships are described in terms of ‘not good’, ‘ sufficient’, ‘good’ and ‘very good’.

2.7.5 The change process

The change process refers to the first two stages of planned change, as defined by Lewin (1951): ‘unfreezing’ and ‘moving’. In this research, unfreezing refers to the stage in March 2015, when the project was initiated and started. Moving refers to the stage where actual changes occurred. After a small delay, caused by unanticipated changes, the project continued in December 2015. From this moment on, the moving stage is in progress. The research focuses on the change process itself: change is not realized yet, but there is movement. Both planned and unplanned changes are taken into account during the stages. Especially, what is the influence of voice, trust and authority on the entire process? For example, did unplanned changes occurred and caused delay because managers did not listen to their employees?

2.8 Critical success factors for change

Ten critical success factors have been indentified as conditions for successful change implementation: Top management support and involvement; clear understanding of goals, expectations and deliverables; excellent project management; project team; resources; user participation and involvement; user education and training; reengineering business processes; system and technology and quality vendor. These factors are summarized in Table 1, on the next page:

(34)

Table 1: Ten critical success factors for change

Critical Success Factor Description Reference

1. Top management support and involvement

Strong leadership, commitment and participation.

Barki & Hartwick, (1994); Van Dalen, (2015); Zhang et al. (2005); Nelson & Somers (2001); Umble et al. (2003). 2. Clear understanding of

goals, expectations and deliverables

The added value for all stakeholders, like customers, employees or suppliers, should be communicated.

Van Dalen, (2015); Nelson & Somers (2001); Umble et al. (2003).

3. Excellent project management

Define clear definitions of objectives, including the development of a work and resource plan, and tracking project progress.

Nelson & Somers (2001);Umble et al. (2003); Zhang et al. (2005).

4. Project team A skilled team, responsible for creating a project plan, overall schedule, assigning responsibilities and acquiring all necessary resources.

Barki & Hartwick, (1994); Nelson & Somers (2001).

5. Resources Facilitative conditions, such as time and budget.

Barki & Hartwick, (1994); Nelson & Somers (2001). 6. User participation and

involvement

Participation in the system development and implementation processes by representatives of the target user groups.

Barki & Hartwick, (1994); Zhang et al. (2005)

7. User education and training

Training in the software systems, maintain contact with all users and monitor the use of and potential problems with the new system.

Nelson & Somers (2001); Umble et al. (2003); Zhang et al. (2005)

8. Reengineering business processes

New software system may change the way the organization operates, which will prevent chaos and resistance.

Nelson & Somers (2001); Umble et al. (2003); Zhang et al. (2005).

(35)

work. Be committed to the using the new system, so people will work with it instead of around it.

(1994); Van Dalen, (2015); Umble et al. (2003); Zhang et al. (2005)

10. Quality vendor A vendor who has qualified consultants, who understand business processes and procedures, who possess good interpersonal skills and are good in working with people.

Zhang et al. (2005)

(36)

3. Results

3.1 Voice versus exit

The expectation was that employees would have a high degree of voice. Instead, the results show that people have a moderate degree of voice within Ahold. Most people sometimes disagree with

decisions. However, in most cases they would speak up, by challenging a decision or raising questions.

The proposition that employees favor voice over exit seems to be confirmed. People speak up, although they are more careful whenever it gets political, like a manager said: “People do speak up, certainly when they do not agree. However, they do not speak up in every occasion. Whenever it gets a bit political, people are more careful” (K).

Although people sometimes consider leaving the organization, this is not due to

dissatisfaction. Leaving is related to growth and development, consequently: the investment of the organization in employees’ skills. More than half of the respondents think the degree of investment in skills is ‘moderate’ or two even consider it ‘high’. However, there are also people who would like to be offered more possibilities.

People find their skills important and relevant for Ahold. Most mentioned are communication and analytical skills, which are not Ahold specific. Although the degree of investment in skills is considered fairly well, it may be improved. Currently, learning is mostly “learning by doing” and “learning on the job” (A; F; J).

The degree of investment in skills enhances incentives for leaving. The fact that most people speak up and are willing to learn and develop is likely to increase user participation and involvement in the project. People stated to be looking forward to the change project and using the new software system. User participation and involvement is especially important in the unfreezing stage, as it can help in forming a powerful group of people embracing the change and convincing others to support the change effort. Furthermore, people can facilitate change in creating and communicating a vision about the desired end-results: “people come up with a lot of ideas” (K).

(37)

training necessary for implementing and adequately using the system. This will increase the chances for change implementation. However, if people are really empowered to realize change, which is part of the moving stage, is harder to say when looking at voice. This will be discussed in paragraph 3.3, about authority, which includes autonomy, discussion, empowerment and power relations.

In December 2015 people seem to be eager to start the project and believe the project will be a success. Furthermore, people state to be willing to actively participate and to be looking forward to the implementation of the software system. More importantly, people seem to believe in the change and state to be actively involved in decision-making. Therefore, current prospects for change implementation are good.

However, there has been a delay in 2015. This delay was signaled in June by several people, but still occurred. Concerns implied that timings were too ambitious. Furthermore, there were problems with budget, which had been requested internally, but too late. Voicing concerns regarding the project, were ignored. This contradicts with the degree of voice and participation in decision-making people state to experience and seems to explain why the project dissolved at the end of the unfreezing stage (see Appendix 5: Results and quotations for the analytical construct ‘voice versus exit’).

3.2 Trust

The results are different than expected. The expectation was that the Dutch norms and values would have a relatively negative effect on competence trust and a positive effect on goodwill trust. However, people think colleagues are positive regarding both their competences and goodwill. Furthermore, competence trust in colleagues was somewhat higher than goodwill trust, in contrast to the expectations.

Remarkably, trust decreases when distance between colleagues increases. There is more trust in colleagues within a team than in colleagues outside the team or between different departments: “My expectations regarding the competences and goodwill of my colleagues are positive regarding the people in my team” (G).

(38)

At the same time, people do not doubt each other’s competencies, but do believe that some people are sometimes too focused on their own role and department, forgetting the bigger picture: “People are sometimes too much focused on their role and department: they forget the bigger picture. The problem’s origin lies in both competences and knowledge” (I). However, there is no direct reason to assume that this is typical for the Dutch, with respect to their norms and values. People focusing too much on their own role and department may be something that occurs particularly in large organizations.

People emphasize openness and keeping each other accountable for decisions, which is typical for the Dutch norms and values. Interviewees are very positive in their expectations regarding the trust of colleagues in them. Explanations vary from considering them self to be very open or to be highly motivated.

Since the relatively high degree of competence trust has been confirmed, overall team skills, project management and user participation and involvement are expected to be quite high as well, which helped establishing the unfreezing stage. One project manager of the software vendor stated that the direct people involved in the change, do possess the right knowledge and skills. He appreciates the fact that one of the key people at Ahold involved in the project supports and works for the project. Consequently, essential aspects of the

unfreezing stage have been fulfilled. Furthermore, the project team and future users participated in trainings, which corresponds with actual moving.

Critical in the moving stage, is to have a highly skilled and competent project team that can identify the requirements and complexity and is able to solve potential problems. However, during the first phases of the project in 2015, two project managers were replaced. Although in one case people clearly out spoke that the new project manager was an

improvement, overall it may have slowed down the process, since a new project manager needs to be assigned and needs to master the details of the job. Consequently, this may have

(39)

had a negative effect on the change process.

It is unknown whether this was caused by a low degree of either goodwill or competence trust in the specific project managers. Nevertheless, the overall trust in

colleagues is good and directly involved people are skilled and trained, which is expected to have a positive influence on the eventual implementation (see Appendix 6: Results and quotations for the analytical construct ‘trust’).

3.3 Authority

On an organizational level, all interviewees agree on the hierarchical nature. However, managers are approachable and relationships with superiors are good. Relationships with direct colleagues are very good. The accessibility and approachability of both employees and managers facilitates cooperation, but communication seems to be insufficient. Furthermore, results show that authority should be enhanced although most people believe to have autonomy. Further questions showed there is need for more empowerment by the management. Sometimes people try to avoid slow decision-making and hierarchy by informally connecting with colleagues or departments. Although people have respect and obedience for authority, discussion is being highly encouraged.

Although most people believe to have autonomy regarding the decisions they make within their day-to-day work and experience sufficient authority, especially within their team; both managers and lower-level employees agree on the fact that more autonomy would be appropriate. The organization seems to be working on that. To the question “To what extent do people have autonomy?” a manager answers:

“Medium. I would say a 6.5, because of certain governance structures. There is the freedom to have ideas and to execute these ideas, but it is also hard to get approval. People do not really have autonomy. But the intention is there” (K)

(40)

Managers point out that they sometimes take authority (G) or commend it (J), whenever they think it is required. Some lower-level employees state that hierarchy is inherent to a large organization and experience it as “nothing bad” (D). Nevertheless, managers emphasize that they want to be open and approachable. Most people think the online department is the least hierarchic and most young and informal department within the organization.

Discussion is being highly encouraged. If you do not agree with decisions: bring your arguments and ask questions. Communication between departments is sometimes difficult, because of other subcultures and different ways of working. Interviewee (G) believes there should be more mutual insight and comprehension regarding processes. “There is little knowledge of each other’s processes, which does not help our mutual trust” (I). Obviously, departments need to work together within the organization. It seems that cooperation should be improved (A; G; I). Especially, within the change project, since the system contains a consensus aspect. It is important that there are clear agreements on how the consensus aspect is going to be designed.

However, some interviewees stated that they have question marks whether the marketing department is properly informed and involved. One respondent stated to have doubts whether the marketing department will agree on some of the agreements (being) made. Furthermore, some people would like to have more autonomy and some managers agree on this. Therefore, the current degree of authority may have a negative effect on the change process.

More importantly, top management must support the change project and the people involved. People need to be empowered to make their own decisions and to realize optimal user participation and involvement in the change project. Although the organization is considered hierarchical, within teams and departments people are accessible and approachable.

(41)

This will most likely positively influence change implementation (see Appendix 7: Results and quotations for the analytical construct ‘authority’).

(42)

4. Discussion

In line with Geppert et. al (2003): both global and national contextual forces affect change management. In this case, low context forces of economic and technological nature pushed the organization to change: in order to cope with competition and a continuously changing market environment, Ahold needed to adapt. The change initiative was the implementation of a software system. Furthermore, high context dimensions of institutional and cultural nature, in terms of norms and values, played a role in explaining the degree and nature of voice, trust and authority and their influence on the change process.

4.1 Voice versus exit

The results show that most people sometimes disagree with decisions. Whenever that occurs, most people would challenge a decision or ask questions. This is in line with the Dutch norms and values. Overall, people do speak up. However, they may remain silent when a subject is political.

This may be explained by the fact that employees do not always feel comfortable with expressing their opinions because their manager may not appreciate it (Gao et. al, 2011). Therefore, employees experience a moderate degree of voice, resulting in a negative influence on the change process. This is remarkable, because it would have been more likely for people to experience greater voice within the context of Ahold and The Netherlands.

Employee input can be very important in enhancing organizational functioning. In coping with challenges, employees have been recognized as invaluable in providing comments and recommendations for solving critical problems (Gao et. al, 2011). When it comes to the project, the delay in 2015 was signaled by several people, but still occurred. Concerns were regarding too ambitious timings. Furthermore, the budget had been requested too late. However, these concerns were ignored.

(43)

This contradicts with the degree of voice and participation in decision-making people

state to experience. However, it may be explained by the fact that people on some occasions

or regarding certain subjects, may be more careful with speaking up. The degree of voice is not as high as expected, especially with respect to the project. Moreover, the degree of voice was disappointing in the unfreezing stage of the project.

When it comes to skills, the most mentioned are communication and analytical skills. Skills like communication and analytical skills are not Ahold specific per se. This seems to confirm theory: according to Aguilera and Jackson (2003), the more firm-specific someone’s skills are, the less likely that person is willing to leave the organization because they have a greater dependence on the organization. Employees at Ahold favor voice over exit.

On the other hand, results show that opinions about the organization’s degree of investment in skills differ. When looking at theory, the moderate degree of investment in skills would imply a relatively higher incentive to leave: Hoon Lee et. al (2003) found that organizational commitment to an employee’s personal and professional growth in the form of trainings and development, resulted in higher commitment towards the organization, higher job satisfaction and less intention to leave. Gutteridge et. al (1993) found that organizational skill development efforts not only enhanced employee skills, but also employee retention. Nevertheless, employees at Ahold favor voice over exit.

Furthermore, investing in firm-specific skills may create incentives to exercise voice, which would positively influence change. The overall impression is that employees at Ahold do consider leaving the organization from time to time. This seem to be in line with literature, since the skills are not Ahold-specific per se which makes them less dependent on the

organization. Consequently, most employees’ positions are comfortable and safe enough to think about leaving from time to time.

A change process within an institutional context : the Dutch retail case