• No results found

The Influence of Power and Politics on the Sensemaking Process

N/A
N/A
Protected

Academic year: 2021

Share "The Influence of Power and Politics on the Sensemaking Process"

Copied!
46
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

________________________________________________________________

The Influence of Power and Politics on the Sensemaking Process

A polar type comparative study between two Agile teams

Master Thesis

________________________________________________________________

MSc Business Administration Change Management

Faculty of Economics and Business

University of Groningen

Semester 2, the academic year 2019 – 2020

Groningen, 22-06-2020

Rogier Emmelkamp

S3858928

Supervisor: dr. I. Maris-de Bresser

Co-assessor: dr. H. Bruns

(2)

2

ABSTRACT

Situational factors influence the sensemaking process. Research indicates that power and politics can influence the sensemaking process. However, our understanding of the effect of power and politics on the sensemaking process is still limited. This study aims to broaden our understanding of the influence of power and politics on the sensemaking process and how a shared understanding is established within Agile teams concerning their roles and the working method. An interview-based study is conducted at the public organization Dienst Uitvoering Onderwijs (DUO). DUO has started adopting Agile methods in 2016 and is still changing to become an Agile organization. Two Agile teams were studied, one storming and one performing, through a qualitative comparative research approach. Findings show that power and politics influence all three phases of sensemaking. Furthermore, power and politics are present differently in the two distinct stages of team development. Immature Agile teams engage more in the use of micro-behavioral tactics to influence interpretations, while mature Agile teams preferably influence each other rationally. This study shows that the interpretations phase is mostly influenced by power and politics. Lastly, both formal and informal macro-structural influence have a more considerable effect on the sensemaking process of employees. Power and politics should be used constructively to enhance the creation of a shared understanding within Agile teams. Managers are advised to steer the desired behavior through functional descriptions. Furthermore, managers are advised to give immature Agile teams more guidance by educating them and by sharing best practices among teams. These implications can enhance the development of Agile teams towards the establishment of a shared understanding.

(3)

3

1. INTRODUCTION

Organizations experience difficulties in adopting Agile methodologies (Yu & Petters, 2014; Vidgen & Wang, 2009). Agile working methods imply a rapid response by self-organizing autonomous teams to possibilities in the turbulent environment and towards changing customer demands (Denning, 2015). However, a lack of theoretical prescriptions on Agile methodologies causes many organizations to struggle in applying Agile practices (Dybå & Dingsøyr, 2008). Transforming into an Agile organization is more than solely adopting its practices, Agile is regarded as a mindset (Manen & Vliet, 2013). Successfully changing to Agile management requires a considerable shift in how people think and how they behave (Cockburn & Highsmiths, 2001), making people a crucial element for success (Bootla et al., 2015; Gandomani et al., 2014). Within an Agile transformation, self-organizing teams form the core (Chow & Cao, 2008; Hoda et al., 2012). Agile transformations force people to adopt new work practices while being faced with a new reality, which triggers employees to engage in sensemaking (Weick, 1995).

Sensemaking has been mentioned within Agile studies from a cognitive perspective before, concerning the creation of a shared understanding (Moe et al., 2010). Academic research shows that Agile practices play a role in establishing a shared understanding within teams (Yu & Peters, 2014) and that a shared understanding is essential for team effectiveness (Moe et al., 2010). According to Weick et al. (2005), sensemaking can help explain how Agile teams establish a shared understanding. However, how the sensemaking process unfolds within Agile teams, and the effect of power and politics in creating a shared understanding is under-researched.

When organizations change towards Agile management, employees will become part of multidisciplinary autonomous self-organizing Agile teams (Hoda et al., 2012). Emphasizes are placed on increased interactions with others and the environment, more autonomy, and increased participation, particularly in the decision-making process (Vidgen & Wang, 2009; Ashmos et al., 2002). In the light of self-organization established meaning is reshaped. However, since members of Agile teams have different areas of expertise, it is common that contradictory interpretations of the same event take place, leading to political struggle (Weick, 1995).

Sensemaking is defined as “the process through which individuals work to understand novel,

unexpected, or confusing events” (Maitlis & Christianson, 2014, p.58). Sensemaking is a perspective

(4)

4

as in how Weick originally intended it. To bring sensemaking literature forward, the sensemaking process of ‘perceiving cues, creating interpretations, and enactment’ are central within this study (Sandberg and Tsoukas, 2015; Maitlis & Christianson, 2014).

Besides looking at the whole sensemaking process, Weick et al. (2005) argue that the sensemaking literature has to take the influence of politics more into account. According to Weick, it is impossible that sensemaking occurs in isolation since situational factors form implications on the sensemaking process (Weick, 1995; Sandberg & Tsoukas, 2015). Of the situational factors, power and politics is regarded as a considerable factor (Sandberg & Tsoukas, 2015). People use power as a ‘sensemaking resource’ to shape the meaning of others (Maitlis & Christianson, 2014). There is a theoretical intuition that power and politics play a role in sensemaking, for example, through sensegiving (Maitlis & Lawrence, 2007; Sonenshein, 2010; Gioia & Chittipeddi, 1991). Sensegiving is defined as “attempts to influence others’ understanding of an issue” (Maitlis, 2005, p.21). But our understanding of the effect of power and politics on the sensemaking process is still limited (Sandberg & Tsoukas, 2015; Maitlis & Christianson, 2014; Weick et al., 2005). Therefore, more research is needed to explain how power and politics influence the sensemaking process.

This study aims to broaden our understanding of the influence of power and politics on the sensemaking process during Agile teams development, and how a shared understanding is established within teams concerning their roles and the working method. The central question in this research is:

“How do power and politics influence the sensemaking process of employees in Agile teams?” To

answer this research question, a comparative interview-based study was conducted at DUO, a governmental organization. DUO has started adopting Agile methods in 2016 and is still changing to become an Agile organization. Within this study, two Agile teams were studied that are at different stages of team development. The comparison between the two teams broadens our understanding of how power and politics influences the sensemaking process, while teams try to develop a shared understanding.

(5)

5

et al., 2010), and ultimately the success of the Agile transformation. Also, it provides insights in how people are influenced in their understanding and behavior, which could help managers to steer desired behavior within Agile teams.

The remaining of the paper is structured as follows. The next chapter describes the main theoretical concepts used in this study. Thereafter, the method section describes the research setting, data collection and analysis. Next, the findings are presented. Lastly, the discussion and conclusion will complete this study.

2. LITERATURE REVIEW

Agile management

There is no widely agreed definition for Agile, but Conforto et al. (2016) define Agile as: “the

project team’s ability to quickly change the project plan as a response to customer or stakeholders needs, market or technology demands in order to achieve better project and product performance in an innovative and dynamic project environment” (p. 667). Agile is an umbrella concept for several

management methods like Kanban, Lean and Scrum (Denning, 2016). Agile methods have four main elements in common, also called the Agile ideology: “(1) Work is performed by self-organizing teams,

(2) work is focused directly on meeting customers’ needs, (3) interaction with the customer is central, and (4) work proceeds in an iterative fashion and progress towards fulfilling the needs of customer is assessed at every stage” (Denning, 2015, p.11). According to Denning (2016), benefits of successfully

implemented Agile methods are better responsiveness to consumer demands, more employee engagement, fast implementation of innovation and increased consumer contentment. However, firms must be careful not to fall back into the old command-and-control way of working to be successful (Denning, 2016).

(6)

6

Peters (2014) argue that specific Agile practices are important to establish a shared understanding, which enhances collaboration. While these studies indicate the importance of establishing a shared understanding for team effectiveness, they do not research how this shared understanding unfolds.

The team development model by Tuckman (1965) can bring insight in how teams develop towards establishing a shared understanding. The model is elaborated below. In this study, the phases of team development function as ‘sensemaking episodes’ of how the sensemaking process unfolds.

Team development

By functioning as a common language, the team development model of Tuckman has proven its value to discuss and explore group dynamics (Rickards & Moger, 2000). According to Tuckman and Jensen (1977), teams develop through five phases: “forming, storming, norming, performing and adjourning” (p.436). While the group moves through the team development phases, members are concerned with “resolving both interpersonal relationships and task activities” (Miller, 2003, p.122). The way people act, behave and relate to each other, determines the structure of the group and its current development stage (Tuckman, 1965). Teams move through the following five phases.

Forming. During this phase a group is formed. This phase is characteristed by task activity

orientation. Members establish interpersonal and task boundaries by testing each other to determine their position in the group and the direction of the group. Also, they establish relationships with others and pre-existing standards (Tuckman, 1965).

Storming. This phase is specified by “conflict and polarization around interpersonal issues”

(Tuckman, 1965, p. 396). Team members respond emotional and they resist group influence and the formation of a group structure. Also, they combat against the accomplishment of the task (Tuckman, 1965).

Norming. This phase starts when the group develops coherence, and members find a

commitment to their tasks. Individual opionions are conveyed and different characteristics are tolerated. During this phase, shared meaning and the most adequate form of working is established (Neuman & Wright, 1999). Tuckman (1965) argues that an entity is formed, team feeling arises, and conflicts over tasks are evaded in favour of the group.

Performing. Tuckman (1965) argues that in this phase the group members support each other

and “functional role relatedness” (p.390) starts to emerge. The group becomes ‘a problem-solving instrument” (p.387). Group cohesion and identiy makes it possible for roles to function as instruments to help others and enhance task activities in favour of the group.

Adjourning. The final phase, added to the original model by Tuckman and Jensen (1977),

(7)

7

In this study, a storming team and a performing team are compared. The transition towards Agile methodology requires employees to work differently. Teams try to make sense of the new reality as they develop through the phases of team development (Tuckman, 1965). They have to abandon old habits and adopt an Agile mindset (Manen & Vliet, 2013). These changes trigger the sensemaking process of individuals (Weick, 1995). Sensemaking literature will be used to help understand how individuals make sense of Agile methodologies.

Sensemaking

Sensemaking was firstly introduced in the organizational context by Karl Weick. Weick (1979) argues that a group becomes organized through the convergence of their cognitive cause maps (i.e. the sense made). Sensemaking can be understood as a ‘process’ that is ongoing and fundamentally social, whereby members of an organization interpret the environment through interaction with each other (Maitlis, 2005). For a person to engage in sensemaking, it needs to be activated. According to Maitlis (2005), sensemaking is activated when “members confront events, issues, and actions that are

somehow surprising or confusing” (p. 21). This interruption forces the person to make sense of the

situation to restore its ongoing activity (Weick, 2001). Moreover, sensemaking is a process that happens retrospectively, as Weick (1995) argues “people can know what they are doing only after they

have done it” (p.26).

Within the literature, the sensemaking perspective has been given different meanings throughout time as it is primarily studied from two different viewpoints (Maitlis & Christianson, 2014). From a cognitivist perspective, sensemaking is focused on interpretations and it results in the establishment of a shared mental model. Through the interactive formation of convergent cause maps, sense is made of a prevalent situation in a person’s mind (Sandberg & Tsoukas, 2015). From a constructivist perspective, the outcome of sensemaking is “actionable intersubjectivity” (Sandberg & Tsoukas, 2015, p.9). This is constructed through language (Sandberg & Tsoukas, 2015). This position regards sensemaking as a social process that develops through enacting the world. It develops through the language used between socially embedded actors, while “meaning is negotiated, contested, and

mutually co-constructed” (Maitlis & Christianson, 2014, p.66). Since this study is focused on teams, and

(8)

8

study (Daft & Weick, 1984; Weick, 1995; Sandberg and Tsoukas, 2015; Maitlis & Christianson, 2014). The sensemaking process is elaborated below.

Perceiving cues. Perceiving cues is the first phase of the sensemaking process. This phase

involves: “bracketing, noticing, and extracting cues from our lived experience of the interrupted

situation” (Weick, 1995, p.35). These cues, which are issues, events, or situation, trigger sensemaking

when meaning is ambiguous. Ambiguous cues interrupt people’s ongoing activity when there is a discrepancy between expectation and reality (Maitlis & Christianson, 2014). This can be the case when something unexpected occurs or through the non-occurrence of something expected. The feeling an actor perceives is described as: “that something is not quite right, but you can’t put your finger on it” (Weick & Sutcliffe, 2007, p31). The experience of discrepancy or violation should be great enough for an actor, subjectively, to trigger sensemaking (Maitlis & Christianson, 2014). Within this study, one type of triggering event is especially relevant, and that is an organizational change towards Agile management. Major planned organizational change is regarded as a type of event that affects most of an organization’s activities (Sandberg & Tsoukas, 2015). Organizational change initiatives force employees into significant sensemaking exertion since it interrupts the usual ways of achieving things within organizations (Sandberg & Tsoukas, 2015; Maitlis & Christianson, 2014). More specifically, actors have to make sense of how they should perform their work as prescribed by the organizational change initiative.

Creating interpretations. Creating interpretations is the second phase of the sensemaking

process. The perceived cues trigger a need for understanding, and this stage involves: “fleshing out the

initial sense generated in the creation process [i.e. perceiving cues] and developing it into a more complete and narratively organized sense of the interrupted situation” (Sandberg & Tsoukas, 2015,

p.14). In this stage, collective effort is undertaken to make sense of events or issues, and for intersubjective meaning to be constructed (Maitlis & Christianson, 2014). Balogun & Johnson (2005) argue that actors use a variety of forms within the social processes of interactions to create interpretations. The following forms are used by actors “their everyday experiences of the actions and

behaviors of others, and the stories, gossip, jokes, conversations and discussions they share with their peers about these experiences, shape their interpretations of what they should be doing” (Balogun &

Johnson, 2005, p.1574). The newly established sense made of the perceived cues, lead to actions (i.e. enactment).

Enactment. The third phase in the ongoing sensemaking cycle is enactment. This phase

involves: “acting on a more complete sense made of the interrupted situation, in order to see to what

extent it restores the interrupted activity” (Sandberg & Tsoukas, 2015, p.14). Action is an essential part

of sensemaking. According to Weick (1988): “Cognition lies in the path of action. Action precedes

(9)

9

learn about the situation and cues through action (Weick, 1988). Besides, action can also examine beliefs and understandings established by prior sensemaking. In other words, action provides feedback to the actor (Maitlis & Christianson, 2014; Rudolph et al., 2009). Lastly, the environment for sensemaking is shaped by actions, since an understanding is gathered from an actors action. Maitlis and Christianson (2014) argue that situations that brought about sensemaking can change due to the actions that people take. People take action in an attempt to look for explanations, however, actions can also lead to changed situations. As a result, the situation that initially activated sensemaking may have been altered by the taken action (Maitlis & Christianson, 2014). Therefore, the sensemaking process should be regarded as an ongoing cyclical process, which may appear continuous, but consist of several series of distinct episodes (Weick, 2012).

Within this study, it is taken into account that situational factors can influence the sensemaking process (Sandberg & Tsoukas, 2015; Weick, 1995). Of the situational factors that influence sensemaking, power and politics are regarded as a considerable one (Sandberg & Tsoukas, 2015). People use power or politics as ‘sensemaking resources’ to shape the meaning of others (Maitlis & Christianson, 2014). The power and politics resource is elaborated in the next paragraph.

Power and Politics

Power and politics play a prominent role in organizations. Many scholars have characterized organizations as political systems since groups and individuals seek different interests and they compete to shape decisions to their benefit (Mintzberg et al., 1998; Pettigrew, 1987; Pfeffer, 1981). Central in the idea of power and politics is that there are differences between individuals regarding their degree of influence. A common definition of politics is: “those activities taken within

organisations to acquire, develop and use power and other resources to obtain one’s preferred outcome in a situation where there is uncertainty or descensus about choices.” (Pfeffer, 1981, p.7). While power

can be explained as: “the possession of position and/or resources.” (Burnes, 2014, p.249).

The lack of studies that incorporate power in sensemaking can be explained by the focus of researchers on the outcomes, instead of the sensemaking process (Sandberg & Tsoukas, 2015). Research is needed since sensemaking theory explains how meaning is constructed, and power can influence the creation of understandings by influencing the sensemaking process. The use of politics and power is specifically evident in organizational change, as various actors will try to influence the sensemaking process in their interest (Balogun & Johnson, 2004; Burnes, 2014).

(10)

10

understand how power is used within organizations, attention must be paid to both forms (Krackhardt & Porter, 1985).

Who holds power, is a central question in power literature. According to Brass and Burkhard (1993), people with formal authority or in a central informal position within an organization have macro-structural power. People in formal positions have power regardless of the type of active influence tactics (micro) they use (Brass & Burkhardt, 1993). Since power in legitimate positions is recognized and accepted by both the superior and subordinate (Madison, Allen, Porter, Renwick, & Mayes, 1980). Furthermore, the informal power of people is dependent on their access to resources or their control over resources (Brass & Burkhardt, 1993). Informal power is established when people have a central position in informal network channels. Being central in those network channels is a form of power since one can exercise more influence on the social construction of reality (Brass & Burkhardt, 1993; Burkhardt, 1991; Rice & Aydin, 1991).

Active use of macro-structural power can be regarded as conceptually related to sensegiving, as it is often studied from the active influence of a manager on the sense made by employees (Gioia, Chittipeddi, 1991; Sonenshein, 2010; Maitlis & Lawrence, 2007). Gioia and Chittipeddi (1991) found that managers engage in sensegiving to influence employees during strategic organizational change, as they address the importance of sensegiving for the overall change process effectiveness. Sonenshein (2010) found that managers use strategic (interwoven) narratives as a sensegiving tool to influence meaning. Furthermore, Maitlis and Lawrence (2007) found that sensegiving is enabled when stakeholders and leaders have issue-related expertise and/or legitimacy, besides the opportunity has to be there.

Micro-behavioral power refers to the political tactics used by individuals. Groups without formal authority, or informal position of power within the organization, are more dependent on the active use of micro-behavioral tactics to influence each other’s meaning (Brass & Burkhardt, 1993). Besides, political strain is created as conflicting interpretations over the same event are common, due to different areas of expertise (Weick et al., 2005). Agile teams do not possess formal authority, and probably no informal authority either. However, Agile teams are multidisciplinary and autonomous, it is therefore interesting to see how employees will influence each other’s sensemaking process through the use of micro-behavioral tactics.

Kipnis and Schmidt (1988) prescribe six categories of micro-behavioral tactics that people can actively use: “assertiveness, ingratiation, rationality, exchange, upward appeal, and coalition

formation” (as cited in Brass & Burkhardt, 1993, p.447). The first category, assertiveness refers to

tactics as “demanding compliance, ordering, and setting deadlines, as well as nagging and expressing

(11)

11

others feel valued, and acting humble or friendly (Brass & Burkhardt, 1993). Thirdly, rational tactics involve convincing through the use of reason, compromise to influence others, and the use of logic. According to Pfeffer (1981), rationality is highly appraised and socially acknowledged and accepted as a mean to influence people. The fourth category is exchange, which refers to receiving favour in return of helping others (Brass & Burkhardt, 1993). Fifth, the upward appeal tactic entails behavioral effort to gain superior support. Lastly, coalition formation consists of initiating efforts to form alliances with others (Kipnis et al., 1980).

Conceptual framework

The illustrated framework (Fig. 1) represents the concepts elaborated above. Sensemaking has been regarded as a process consisting of three continuous cyclic phases (Weick, 1995). Through these phases, employees make sense of the Agile methodology, while trying to establish a shared understanding within autonomous self-organizing teams. Sensemaking is an active process (Maitlis, 2005), therefore, it needs to be activated through a sensemaking trigger. In this case, the trigger is the Agile transformation, since it can be described as an ambiguous event which ultimately leads to changes in the way of working by employees (Maitlis & Christianson, 2014). Therefore, the starting assumption is that Agile transformation activates the sensemaking process of individuals within Agile teams.

(12)

12

members of the two teams (one storming, one performing) can be influenced by perceived macro-structural power of the organizations steering the change (e.g. Gioia, Chittipeddi, 1991; Sonenshein, 2010; Maitlis & Lawrence, 2007).

Moreover, micro-behavioral tactics can be used as sensemaking resources (Maitlis & Christianson, 2014) to affect the sensemaking process of individuals that actively try to make sense of the working method and their roles. The absence of formal authority in democratic Agile teams leave members dependent on the active use of micro-behavioral tactics (Brass & Burkhardt, 1993). Since different areas of expertise can lead to contradictory interpretations, causing political struggle (Weick, 1995). Lastly, informal macro-factors can also influence the sensemaking process. Employees with more access to or control over resources can more effectively influence others (Brass & Burkhardt, 1993; Brass & Krackhardt, 2012). However, it is assumed that this kind of informal macro-structural influence is low within democratic autonomous Agile teams. Nevertheless, Agile teams may perceive informal influence, as they are more in contact with their direct environment both inside and outside the organization (Vidgen & Wang, 2009).

This study examines how the process described above plays out in two stages of Tuckman's (1965) team development model, storming and performing. During the storming stage, it is expected that power and politics play a more prominent role compared to the performing stage. Since a storming team has not yet established a shared understanding of Agile methodologies (Tuckman, 1965). Therefore, it is assumed that employees try to actively make sense of the Agile methodology and develop and negotiate meaning through power. In contrast, in the performing phase, employees will have a shared understanding of Agile methodologies (Tuckman, 1965). Therefore, it is assumed that power and politics are used less to influence the sensemaking process.

Since sensemaking is a continuous process (Weick, 1988), the sensemaking outcome is regarded as temporary (i.e. a snapshot). Meaning can be reinterpreted when a new sensemaking cycle begins. These two stages, storming and performing, are the two sensemaking episodes (snapshots) of this study. Furthermore, a comparison between the sensemaking process of employees within the two teams is conducted. The framework is the theory, which is examined in practice during fieldwork. The method section elaborates the empirical research set-up of this study.

3. METHODOLOGY

Research design

(13)

13

framework was formed through the extensive gathering of research, and this framework guided the data collection.

This research is an interview-based study. Due to the COVID-19 pandemic, it was unfortunately not possible to conduct a single-case study type of research. However, an interview-based study provides comprehensible insights and data to study phenomena and broaden its understanding (Edmondson & McManus, 2007). Interviews provide the researcher with the opportunity to gather detailed data and dig deeper into specific topics. Therefore, an interview-based study is suitable to obtain qualitative data on how power and politics influence the sensemaking process of employees.

This study aims to compare two self-organizing Agile teams that are adopting Agile practices within the same organization. The units of analysis are two self-organizing teams within the same organization, who have to make sense of the Agile methodology. Therefore, this study can be regarded as a comparative interview-based study.

This study focusses on two extremes of team development stages (Tuckman, 1965), storming and performing. Therefore, this research is regarded as a “polar type” theoretical sampling approach (Eisenhardt & Graebner, 2007). According to Eisenhardt and Graebner (2007), the polar type approach makes (contradicting) patterns more visible “of the central constructs, relationships, and logic of the

focal phenomenon” (p.27). This study aims to understand how power and politics influence the

sensemaking process in these two extreme phases of team development. By comparing findings of these teams in two different stages (i.e. episodes), insight will be gained about what happens when teams develop (regarding power and politics and sensemaking).

Research setting

The research was conducted at the Dienst Uitvoering Onderwijs (hereafter: DUO). DUO is a public organization that handles many tasks related to education in The Netherlands and is responsible for conducting various regulations and laws. DUO is well known for providing student finance. DUO also organizes national exams and recognizes diplomas, among other things. Throughout the years, DUO has organized many of its processes digitally. However, to better meet customer demands, DUO started to introduce Agile practices in 2016. The organizational change was not implemented company-wide at once, but has been rolled out in several core processes in a sequential way and is still ongoing.

(14)

14

customer demands were neglected. Through the implementation of Agile methods at DUO, products are now iteratively built by teams and adapted to continuous customer feedback.

Within DUO, multiple departments and teams have adopted Agile practices in the last four years. Some teams have adopted the Agile methodology quite successfully and can be regarded as norming or even performing teams (Tuckman, 1965). While other teams struggle in adopting Agile methods and can be regarded as forming or storming teams. The presence of teams that adopted the Agile methods differently makes DUO an appropriate organization for a comparative polar-type of research.

The study is focused on two Agile DevOps teams. DevOps stands for a combination of two disciplines that form a team, namely ‘development’ (Dev) and ‘operations’ (Ops). A DevOps team is responsible for building software. Within a DevOps team, the following disciplines are present: functional designers, developers, testers and operation engineers. A DevOps team is assisted by a Product Owner, who is responsible for translating and securing customer preferences and maintaining customer contact.

To select the appropriate teams for this study, they were first pre-selected through self-evaluation. Multiple teams were invited to provide examples of why they fitted a particular stage of Tuckman. Subsequently, a representative of each team was shortly interviewed to confirm that the team met the characteristics of the specific stage of Tuckman (1965). Out of the selection, Team A and Team B were considered to be the most suitable for this polar-type study. Team A, from the department of ‘Zakelijke Dienstverlening’, was formed one year ago and is regarded as the storming team. Team B, from the department of ‘Registers & Exams’, was formed two years ago, and is regarded as the performing team within this study.

Data collection

Data collection was done through semi-structured interviews. All respondents were asked the same pre-formulated open question, to ensure all topic were adequately addressed. Open questions were used to encourage respondents to share their understanding while being flexible to dig deeper into specific issues. This resulted in in-depth information which is presented in the results section. The interview protocol used can be found in Appendix A. In total, 14 employees were interviewed. An interview took approximately 75 minutes and was conducted through online video-calls. All members of both teams were interviewed, more details of respondents can be found in Table 1.

(15)

15

function, time within the team, previous experience and questions related to macro-structural influence. The interview continued with questions related to the DUO way of working, the Agile practices the teams have to adopt. An example question is: “In your experience, what were challenges

in the working method?”, this question should convey insight into the cues employees perceived,

things that triggered them to engage in sensemaking. Also: “How did you get to know things about the

DUO way of working?” This question aimed to explain how employees dealt with perceived cues

regarding the working method.

Moreover, questions were asked to see how employees interpret situations, for example:

“What is your interpretation of a self-organizing team?” Questions that aimed to explain how

employees actively used power and how their interpretations were affected by power are: “Can you

explain how you attained your current role within the team?” And: “Which influence did others have on how you fulfil your role?”

All interviews were recorded for analysis. All respondents of DUO were Dutch. Therefore, all interviews were conducted in Dutch to ensure validity during data collection.

Data Analysis

(16)

16

Burkhardt, 1993). An example of deductive code is: “experienced ambiguity” which is related to the first phase of sensemaking (perceived cues). Also, deductive codes related to power like: “coalition forming” and “rationality” were applied. Through iterative analysis, new concepts emerged. The inductive codes like “absence of supervision”, “experienced trust”, “perceived autonomy to organize work” were found during coding. Lastly, axial codes were assigned to codes that referred to the same concepts. For example, the deductive codes, “perceived ambiguity” and the inductive code “lack of supervision” were grouped into the category “ambiguity concerning responsibilities”. This category was selectively coded as ‘perceived cues’, referring to the first phase of sensemaking.

All selective codes for both teams are related to the three phases of sensemaking. Within the selective category, axial codes can be found that refer to Power and Politics. The within-case analysis was done through the analysis of the data to identify the main aspects and patterns per team. Furthermore, cross-case analyses were done to identify commonalities and variations between the major identified themes per team. Since this study is interested in the use of power in the three phases of sensemaking, and the comparison between a storming and performing team, the selective categories for both teams are the same. In contrast, themes within those categories can vary. The codebook is attached in Appendix B. The results chapter provides more insight into the codes that were used and the findings of this study.

4. RESULTS

This chapter describes the findings on how the sensemaking process of the DUO Agile Way of Working (after: WoW) unfolds in two DevOps teams, and how this process is influenced by power and politics. Results are presented, following the three different phases of sensemaking. For each phase, the influence of power and politics is elaborated. It should be noted that the phases of sensemaking are highly interrelated, since sensemaking is regarded as a process. Besides, sensemaking is a retrospective phenomenon and develops through a continuous cycle (Weick, 1995). Since respondents explain their experiences of the Agile transformation reflectively, not all findings are tied to the team’s current stage of team development. It could be, for example, that specific findings refer back to earlier phases of the team’s development.

The chapter starts with the findings of the storming team. Secondly, the sensemaking process of the performing team is described. Finally, a comparison is made between the two teams. Respondents are indicated as RP#, the number (#) is in correspondence with the concerned respondent displayed in Table 1.

(17)

17

To engage in sensemaking, employees first need to be triggered. Below the three most mentioned cues are presented. Thereafter, perceived power is explained.

Ambiguity regarding the working method. The data shows that all members of team A

perceived ambiguity regarding the DUO WoW: “For me, it is hard to tell what it is, or what it is not

(RP4).” In line with this comment, RP2 also indicates that it is hard to tell what the DUO WoW entails:

“Principles are very ambiguous. I would like to see that more clearly.” These examples indicate that members of team A perceived ambiguity regarding the adoption of the working method. The perceived ambiguity triggered employees to engage in sensemaking.

Perceived freedom. The majority of team A perceived much freedom to fulfil tasks and roles.

As RP1 explained: “Everyone has the same tasks in the scrum team. I have never been told how to fulfil

my role. I also think that everyone is free to determine that themselves.” However, the data indicates

that not all employees expected to have this much freedom. The data shows that some prefer more structure: “Some love to determine how to organize that freedom, but we also have people on the

team, including myself, who prefer rules and lists that you have to comply with (RP2).” The tension

between freedom and structure resulted in a discrepancy of expectations, which triggered employees to engage in sensemaking.

Perceived discrepancy regarding the working method. Despite the perceived ambiguity,

employees had to make sense of the working method. However, data indicates that employees perceived a discrepancy in viewpoint over the working method: “Not all of us had the same Agile

working approach in mind as we would all expect (RP2).” RP7 gives the same explanation: “I think it is for the whole team a little bit unclear [what the working method should be]. One half of the team does it one way, the other half of the team does it another way, that causes some discussions.” The perceived

discrepancy triggered employees to engage in sensemaking.

Influence on perceived cues. The majority of employees indicated that the Product Owner

(after: PO) was partly responsible for the perceived ambiguity regarding the working method. Interview data indicates that the PO displayed assertive behavior by ordening the team to be more reliable in their velocity. As RP7 indicated: “The Product Owner would ideally tell the business at which

date certain things are finished. But that is not in line with how we work. Sometimes you can see some friction between that.” Furthermore, RP2 mentioned: “If the PO asks the team to be more reliable in their output, that is not nice to hear.” The PO has no formal authority over employees. However, the

(18)

18

Interpretations

Interpretations arose through action. The data shows that employees formed their

interpretation by just doing it. These actions varied, as employees engaged in discussing over the way of working and participated in team meetings, but also took action to develop their roles. These findings are more elaborated in the enactment part of the results section.

Team’s responsibility and roles. Members of team A indicated that they are responsible as a

team for determining which tasks are performed (8/8), dividing tasks (7/8), making work agreements (7/8), and team work (4/8). Tasks within the team are collectively divided among the different functional disciplines. The data indicates that this process happens through rational consultation, and that consensus was quickly reached. RP3 explained: “[work is divided] just by mutual agreement.

Sometimes one person picks up his preferred task, other times the other[..]. We have not experienced any problems with that yet.” Since each member has their specific functional expertise within a

multidisciplinary team, tasks can be easily divided.

Moreover, employees interpret their roles and responsibility mostly functional-based. As indicated by a tester in the team: “In principle, I make sure that I have tested something well (RP8).” RP2 also refers to his function: “I think it has come a bit along the way by also including the job

description.” Employees feel responsible for their functional domain. However, interview data

indicates that most members (4/8) of team A lacked responsibility beyond their functional domain. As RP1 explained: “People take less initiative to do things beyond their function [..] if I ask something by

email, almost all team members are passive, and some make unserious remarks [..] We could be a bit more proactive within the team.”

Working method. While team A mostly agreed on their team’s responsibilities, the data

indicates that they have no shared understanding of how to fulfil those responsibilities. RP2 explained:

“What is very frustrating is when you choose a working method, and after a month someone says that they disagree, and sums up a list of arguments why, which the other half of the team do not experience at all. This results in discussions between people who have a completely different point of view, and who cannot find overlap in their perspectives.” (RP2)

The lack of a shared understanding of the working method could be due to already established interpretations, as some members (5/8) worked in previous DevOps teams before: “When RP14 and I

joined this team, this team was working differently than we used to do. [..] For us, it was challenging to find our way to adapt to the new working method (RP5).”

Moreover, members experienced much trust to convey their opinion. However, the data indicates that this did not automatically lead to acceptance: “During retrospectives [meetings] matters

(19)

19

received, but it is always conveyed (RP2).” The trust to convey issues, but the lack of accepting other

people’s opinion could explain the absence of a shared understanding of the working method. Since team A is democratically responsible for making decisions, things cannot be forced. Failure to overcome the perceived discrepancy through interaction resulted in ongoing discussions.

Power regarding roles and responsibilities. The data indicates that formal, informal and

behavioral power influenced interpretations. Formal, macro-structural power was perceived through formal job descriptions. Since people with authority form functional descriptions, this can be regarded as macro-structural influence. To make sense of the perceived freedom and ambiguity, employees looked at the functional descriptions for guidance:

“A senior [in his functional domain], is expected to tackle the somewhat more difficult jobs, and to have an overview of what happens with the other teams within DUO. Also, to prevent the juniors from falling into traps that you have already experienced a couple of times.” (RP3)

Some employees (3/8) explicitly mentioned that in order to be or become a senior, they had to display certain behavior. RP2 explained: “They have set up a kind of feature package that a senior developer

should actually do. [..] they steered a lot on cross-team work [..] that is my goal. I want to proceed to the next salary scale.” These examples indicate that the formal job description influenced how people

interpret their role and how they behave.

Moreover, the data indicates that members of team A were informally influenced by their PO in the interpretation of the team’s responsibilities. As RP2 explains: “one suggestion that was put

forward by our PO was to ask other teams to see if they had work for us to perform.” RP3 also

mentioned this informal influence: “Currently we are looking for more work to perform, so that we still

have work in a few weeks. Then it is very easy to say that as a DevOps team, you should be self-organizing [and look for work yourself].” This informal influence was described as demanding

(assertive) behavior and resulted in a broader interpretation of responsibilities within the team. Interview data illustrated that some employees (4/8) explicitly mentioned this type of responsibility.

Lastly, micro-behavioral power was used by members of team A to influence each other’s interpretations. The data indicated that rationality is the most used micro-behavioral tactic (8/8). Employees actively tried to convince each other through reason: “In general, it is just convincing with

arguments, so explaining why the chosen solution is best according to that person (RP2).” The data

indicates that rationality tactics were most effective in functional discussions, as people tend to trust the expertise of others. The functional level of expertise also resulted in the acceptance of rational influence: “The opinion of a senior [..] is often an important opinion within the team. A senior’s opinion

is also adopted more often.” Rational tactics were useful in functional, tasks related discussions.

(20)

20

weeks to finish it.” The interview data indicates that assertive behavior could be displayed when task

activities are not going smoothly: “When things do not go smoothly, I want to be left alone so that I

can work.[..] I also notice that people start to make comments, and the attitude in the team deteriorates (RP2).”

Power regarding the working method. Rationality is the most used influential tactic in

discussions over the work method (8/8). However, employees failed to establish a shared understanding rationally. Therefore, members engaged in other micro-behavioral tactics. Data indicates that coalition forming is a common tactic used within team A (4/8). RP5 explained: “At the

beginning, person X and I tried to convince others of the work method we used in our previous team.”

The absence of creating a shared understanding could be due to the lack of reaching a majority within the democratic Agile team. However, the data indicates that people get tired of the ongoing discussions, which results in assertive tactics (4/8). RP3 indicated: “You notice that people sometimes

get tired of meetings.” RP7 also indicated frustration: “Sometimes I notice that people outvoice others, they just let it go and go along.”

Enactment

Actions and interpretations of the working method. Through action, people learned about

Agile methodologies. As RP7 mentioned: “In the beginning, you have to figure it out for yourself. You

just go with the flow, you look at what the rest does, and you try to participate. Gradually you develop an idea of what you should be doing.” Although people learned about Agile through action, this did not

result in a collective interpretation. This can be explained by not succeeding in influencing each other’s interpretation, which resulted in not accepting the chosen direction. As RP2 earlier indicated: “What

is very frustrating is when you choose a working method, and after a month someone says that they disagree, and sums up a list of arguments why [..]”

Role development and responsibilities. The data shows that members of team A developed

their roles through action. As noted by RP8: “In my first team, the role of FO was primarily to translate

customer wishes into a functional design. Our FO does a lot more [..]. He aligns things with others, more analytical work, he discusses things much more with customers.” RP4 also indicates how he developed

his role: “At the time our scrum master left the team, so that role had to be fulfilled. [..] So I stood up

and filled in that space.” These are examples of how members actively developed their responsibilities

beyond their functional domain.

However, the data shows that only some employees (4/8) had a desire to develop their role. As RP7 explained: “I am not very ambitious. I am happy with things quickly[..] I would prefer more

(21)

21

These findings indicate that the development of roles and responsibilities depends on the individual desire to take action.

Power in the enactment phase. Since the members of team A did not establish a shared

understanding of the working method, they actively looked for guidance. The team choose to seek the help of an expert. Through the micro-behavior upward appeal, the team intended to establish a shared understanding: “sometimes it seems relevant to me that an expert in that area says how it is actually

meant so that people can come on the same page on certain issues (RP2).” The data indicates the

preference for an expert’s opinion to stop the ongoing discussions.

Moreover, data indicates that to develop your role, you also need to engage in coalition forming. As RP4 indicated: “I was thinking about the role of Scrum Master, and I asked some colleagues

what they thought of it, they supported it. That support helped me to decide to take on that role.”

Furthermore, RP1 explained: “When I introduced something new in the team, I perceived a bit of

resistance first. [..] But after they saw the benefits, the team accepted my improvements more easily.”

These examples show that obtaining a role within the team requires the acceptance of a coalition.

Sensemaking process of Team B: Performing team

Perceived Cues

Ambiguity concerning the working method. The data shows that the majority of respondents

perceived ambiguity regarding the working method. As RP11 explained: “When I arrived at DUO, they

were already working Agile. However, it was very unclear to me what the working method should be for the team.” RP12 also mentioned: “That was one thing that I found very difficult when I just started here, one of my first questions was, hey what is your way of working?”. These examples show that

employees perceived ambiguity in the working method, which triggered them to engage in sensemaking.

Ambiguity concerning responsibilities. Employees perceived ambiguity in responsibilities. The

majority of team B had experienced some form of authority in their previous function: “At my previous

employer, everything was much more structured and stricter. [..] if you wanted to improve something, you had to ask your supervisor (RP9).” RP11 also explained: “At my previous employer, I experienced the presence of a supervisor who had some vision of the composition of the team, how people should develop [..] and what we should improve.”

Moreover, the data indicates ambiguity among disciplines: “It was unclear who was

responsible for maintaining backlog items, the team was waiting for the PO, and the PO was too busy and delegated it to the team, resulting in a mismatch of expectations (RP11).” Their previous

(22)

22

ambiguity. Besides, who was responsible for certain tasks was not clear. These ambiguities triggered employees to engage in sensemaking.

Perceived freedom. Data indicates that the employees perceived much freedom to determine

things. As RP12 indicated: “We are not very stuck in standards and guidelines.” RP9 explained: “I think

the most surprising perhaps was the reasonably low expectation of my role.[..]What is expected of my functional domain is quite minimal, I have much freedom.” These examples indicate that employees

perceived a lot of freedom to determine the work method and their role.

Interpretations

Interpretations arose through action. Data indicated that members of team B made sense of

the working methods and their roles through action. Actions varied from discussing things with colleagues, participating within meetings and actions to resolve ambiguities and roles. This is more elaborated in the enactment section.

Team’s responsibility and roles. Members of team B indicate that they are responsible as a

team for picking up and dividing tasks (6/6), making work agreements (6/6), contributing to teamwork (6/6), and for team work (5/6):

“A team that devises and implements its own work processes. That has the freedom to think for themselves about they want to organize their work, and how to solve problems. That can devise initiatives together, and is also free to do so.” (RP11)

Moreover, the data indicates that employees interpreted their role referring to their functional domain. However, the majority (5/6) also explicitly mentioned their responsibilities within the team:

“I see it as my role to test things properly and give guidance to others (RP14).” RP9 also mentioned: “That I know things about the application, how it functions, and that I can solve issues when we encounter them.” These examples indicate that despite a shared understanding of the team’s

responsibilities, employees also have a feeling of responsibility beyond their functional domain.

Working method. People interpret the working method mostly in terms of the Agile practices

(i.e. meetings) they conduct each week. However, interview data indicates that regardless of the perceived ambiguity, everyone is on board with how things are done within the team: “I think the

process does not take that much effort anymore. We no longer want to do things completely different, everyone is now on board with how we do things, decide things, improve things (RP9).” The data shows

that there is a shared understanding within the team on how things should be done and improved. As explained by RP11: “The flow, how we perform things in our team goes very smooth. Also, we can

perform our work very effectively.”

(23)

23

things, but everyone also listens very well to each other.” Data indicates that all matters are discussed

openly:

“We try to agree on certain process improvement in meeting. But this always happens in consultation. [..] Even if something can be implemented without resistance, it is still discussed, and it is then determined that there is no resistance.” (RP9)

The high degree of trust in each other’s opinion, and the space for everyone to share their opinion can explain the shared understanding that has emerged within team B.

Power regarding roles and responsibilities. The data indicates that formal, informal and

behavioral power influenced interpretations. Interview data shows that macro-structural influence was perceived through formal functional descriptions, which steered behavior and the interpretation of employees. RP14 explained: “As a senior, you are expected to you that you develop yourself beyond

your own domain.” Besides, reaching seniority is also connected to an increase in salary, imposing an

incentive for employees to display behavior according to the functional description.

Moreover, the data indicates the informal macro-structural influence of the PO on the interpretation of the team’s responsibility. The majority of team B indicated the facilitating influence of their PO:

“We thought something had to be built, and there was someone who asked the PO’s opinion during the meeting. He responded that we were a self-steering team, and we had to decide for ourselves. If we thought it was necessary, we should do it.” (RP12)

Despite the informal power of the PO, the PO allows the members of team B to make their own decisions. This type of influence could be regarded as the micro-behavioral tactic ingratiation, as the PO acts humble and lets the team make important decisions.

Lastly, the data indicates that interpretations of roles are guided by the influence of the expectations of others. As illustrated by RP12: “When the Scrum Master is not present, people

automatically look at me for taking over those tasks. Sometimes they think of that too easily. [..] But I have no problems with fulfilling that role.” RP11 also explained this type of influence: “I think that the team expects me to organize improvement initiatives and to initiate them.“ This type of influence can

best be explained by the micro-behavior ingratiation, as expectations result in a feeling of importance by an employee fulfilling a specific task. Besides, the data shows no evidence that responsibilities or tasks are demanded by assertive behavior, which could explain formed expectations.

Power regarding the working method. The data shows only evidence for micro-behavioral

power in discussions over the working method. All members of team B mostly influence each other rationally. RP12 explained: “If I notice an issue, I will always discuss it with my colleagues [..] I always

(24)

24

received[..] If I have an idea, I will submit it and discuss it to align with my team. You often do not even have to go further than that. It is quite easy.”

Since the team is democratic, coalition forming is also important. However, this happens very organically: “I always try to involve people in improvements initiatives etc. It is not that I am going to

say what we should do as a team, you need to have support (RP11).” Data did not indicate that

coalitions were formed to steer decisions within the team. Moreover, data did not indicate the presence of other micro-behavioral tactics or macro-behavioral influences within the team. This can be explained by the success of influencing each other rationally, as data indicates this was enough to establish a shared understanding.

Enactment

interpretations developed through action. The data shows that employees developed

interpretations of the working method through action. As RP11 explained: “The theory of scrum is not

very extensive [..] You just have to experience it, so to speak.” Also, RP9 explained: “in the beginning you just go along with the team and experience it.”

Development of roles and responsibilities. The data indicates that roles develop through

action. The active development of roles is undertaken by the majority of employees (5/6). However, role development depends on the actions taken by the individual: “When you enter the team, you will

get information about the different functions in the team, then it is up to the person to develop himself within the team (RP11).” RP12 also explained: “You should take the initiative yourself.” RP13 was

passive in his role development, due to inexperience in his functional domain: “My attitude is more to

test things properly first, before developing your role beyond your core tasks.”

Moreover, members of team B took action to restore ambiguities in the working method and responsibilities. To restore the ambiguities in the working method, RP9 took action to clear the gap in the absence of customer feedback:

“I am mainly the one within the team that guarantees how the applications work and what kind of impact certain changes will have on the applications. Or what things we could improve. Actually, I am kind of the customer. People sometimes say that I am the customer.”

RP11 also took action to restore a gap in responsibility: “The organization [of backlog items] is now a

bit more my responsibility. I try to involve the team a bit more to coordinate issues. That was something that we did not do well in the beginning.” These examples indicate the actions members of team B

took to restore the interrupted situation within the team.

Power in the enactment phase. The data only shows evidence for the existence of power

(25)

25

Ops task myself.” Also, RP9: “To give the testers some more freedom and space, me and a couple of colleagues performed some tests once.” However, it should be noted that being active in other

functional domains is prescribed by the organization through macro-structural influence. DUO stimulates employees to develop beyond their functional domain, which could explain development in other functional domains.

A comparison between the storming and performing team

Perceived cues

Both teams perceived ambiguity regarding the working method. This triggered them to engage in sensemaking, as they actively tried to make sense of the working method. However, the storming team also experiences a discrepancy among members on how the working method should be conducted, which can make it problematic to establish a shared understanding. Moreover, both teams perceive much freedom to determine their roles and how to operate as a team. The perceived freedom is in line with being a decentralized autonomous Agile team. However, employees of both teams did not expected this high degree of freedom, which triggered them to engage in sensemaking. Finally, the performing team experienced some ambiguities regarding responsibilities. This was not mentioned in the storming team, which is probably due to the difference in the maturity phase.

Furthermore, the perceived discrepancy within the storming team was partly due to informal macro-structural influence of the PO. The PO influenced established interpretations of the working method through assertive behavior. In contrast, the performing team did not notice any influence on cues. This could be a coincidence but is probably due to differences in displayed behavior by both PO’s.

Interpretations

Both teams formed interpretations through action. To make sense of the ambiguities, employees engaged in discussions, participated in meetings and developed their roles. Employees of the storming team referred to roles mostly functional-based, while employees of the performing team also indicated their responsibilities. This could be explained by the difference in the development of the two teams. Besides, members of the performing team influence each other in their role by expectations. Expectations were formed after members displayed certain behaviors through action. Furthermore, in both teams, rational behavior is displayed to divide tasks. However, in the storming team, assertive behavior is also displayed when things do not run smoothly.

(26)

26

performing team mostly referred to the influence as ingratiation behavior since they were stimulated in developing their own decision-making process.

The performing team established a shared understanding of the working method, while the storming team did not. The perceived ambiguity resulted in multiple interpretations, while some members of the storming team had already established interpretations due to experience in other DevOps teams. Besides, the difference could be explained by the lack of acceptance or trust in each other’s opinions. Members of the performing team had much trust in each other’s expertise and opinion, while members of the storming team only showed trust in functional expertise.

The difference in the establishment of a shared understanding is also reflected in the use of micro-behavioral tactics. The performing team only used rational tactics in discussions over the working method. While the storming team engaged in more micro-influential tactics, as they failed to influence each other’s meaning rationally. Data shows that coalition forming and assertive behavior is also displayed within the storming team, but this did not result in a shared understanding either.

Enactment

In both teams people develop interpretations through action. However, in the storming team actions did not result in a shared understanding regarding the working method. A possible explanation is that employees in the storming team failed to influence each other. To establish a shared understanding of the working method, the storming team asked the help of an expert. The upward appeal tactic was not perceived in the performing team, which could be explained by their success of influencing each other and establishing a shared understanding.

Moreover, role development is more actively undertaken within the performing team. The difference in stages of team development can explain the difference in the desire to develop a person’s role and responsibilities. Besides, results show that in the storming team, employees needed the acceptance of a coalition to develop their role and responsibilities. Furthermore, the performing team actively developed their roles to restore ambiguities in responsibilities and in the working method. The data of the storming team did not indicate this type of role development.

5. DISCUSSION

This study aims to broaden our understanding of the influence of power and politics on the sensemaking process during Agile teams development, and how a shared understanding is established within teams concerning their roles and the working method. The research question of this study was:

“How do power and politics influence the sensemaking process of employees in Agile teams?”

(27)

27

team development. Third, teams that do not establish a shared understanding do engage more in the use of micro-behavioral power tactics to influence each other’s meaning.

The main findings are interpreted through the use of existing academic literature. Furthermore, managerial and theoretical implications are addressed. Thereafter, limitations are discussed, and future research directions are suggested. Finally, the conclusion answers the research question of this study.

First of all, results show that power and politics play a role in the perceived cue phase. The perceived ambiguity regarding the working method in both teams could be partly explained by the lack of a theoretical base of Agile methodologies (Dybå & Dingsøyr, 2008; Yu & Peters, 2014). However, it is remarkable that there is an absence of (continuous) macro-structural influence of the organization on how Agile working methods should be perceived. This could also be explained as a lack of organizational sensegiving. Previous research confirms that sensegiving in organizational change can be an important tool to influence meaning (Sonenshein, 2010) and can be essential for the overall change process effectiveness (Gioia & Chittipeddi, 1991). Furthermore, according to researchers (Porter, Renwick, & Mayes, 1980; Brass & Burkhardt, 1993) macro-structural influence is regarded as acceptable by employees, independent of the micro-behavioral tactics used. The absence of organizational influence in the perceived cue phase could be explained by the emphasis of Agile management on self-organization and empowered teams (Hoda et al., 2012; Nerur et al., 2005), which restricts the interference of managers and give space for organic development.

(28)

28

Proposition 1: informal macro-structural influence can increase the ambiguity perceived within immature Agile teams when assertive behavior is displayed.

Regarding interpretations, this study shows that four factors influence how interpretations are formed. First, this study shows that interpretations are developed in both teams through action by employees participating in meetings, discussions and tasks. This finding is in line with Weick (1988), who defines sensemaking as an ongoing process, of which actions (i.e. enactment) can result in new cues and interpretations. Through action, the performing team developed interpretations that resulted in a shared understanding of the working method. This is in line with findings of Yu and Peters (2014), who claim that participating in Agile practices (i.e. meetings) contribute to the establishment of a shared understanding of the working method. However, findings show that action has not led to a shared understanding within the storming team. This could be explained by Tuckman (1965), who claims that teams in the storming phase resist group structure and influence. Besides, findings indicate that employees of the storming team were unsuccessful in influencing each other’s meaning.

Proposition 2A: Immature Agile teams do not establish a shared understanding of Agile methodologies through action

Secondly, this study shows that both teams use micro-behavioral tactics to influence interpretation of the working method. This finding is in line with previous researchers (Balogun & Johnson, 2004), who claim that employees engage in the use of power to influence each other’s interpretations. Besides, political strain is the outcome of conflicting interpretation due to different areas of expertise (Weick et al., 2005), which is a characteristic of multidisciplinary Agile teams (Hoda et al., 2012). Furthermore, since employees within Agile teams do not have formal or informal authority, they are dependent on the use of micro-behavioral tactics to influence each other (Brass & Burkhardt, 1993). However, results show that members of the performing team only use rational behavior to influence each other’s interpretation. This finding is in line with Pfeffer (1981), who claims that rational tactics are socially accepted and acknowledged as a mean to influence others. Besides, findings indicate members of the performing team have much trust in each other’s opinions and expertise. Tuckman (1965) claims that teams in the performing phase avoid conflicts and energy is conveyed in favour of the team, which could explain the trust in each other and the preference for only rational tactics.

Proposition 2B: Interpretations of employees in mature Agile teams are formed through rational influence.

Referenties

GERELATEERDE DOCUMENTEN

For instance, in lexical decision, the match boundary represents the amount of accumulated evidence to give a “word” response, and the non-match boundary represents

Next, taking a sociomaterial perspective, it seems that the type of technology used in this phase coheres both with the type of sensemaking (individual or collective) and the role

C2: The term might cause some trouble in the certificate and specifically, when the customer wants a product that is not in the portfolio and immensely demands

With these objectives the study seeks to answer the research question How do power and politics influence the dynamic interplay between sensemaking and sensegiving of

So, even the most stable institutions have to be seen as dynamic, as even those institutions need to be constantly reaffirmed (Weber & Glynn, 2006). When you

This study established as well that the agent’s sensegiving, by means of change agent behavior, influenced the extent to which the sensemaking of a recipient led to a

There were no pronounced differences between excluded and included cases (Appendix 2). All included cases met all inclusion criteria. Similar ratings based on received EWOM

This thesis conducted its research at an innovative technology company who recently implemented an organizational innovation, which is described below. In order to gain a