How ambidextrous are gazelles?
Exploring managerial responses and practices
in relation to exploitation-exploration tensions
Author name: Arthur Alexander Snijder Student number: S11145803
Date submitted: 19 August 2016
MSC. Business Administation Amsterdam Business School
STATEMENT OF ORIGINALITY
This document is written by Student Arthur Snijder, 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.
TABLE OF CONTENTS
ABSTRACT 4
INTRODUCTION 4
LITERATURE REVIEW 5
DATA AND METHOD 15
RESEARCH PARADIGM 16
SAMPLING AND RESEARCH DESIGN 17
RESULTS 26
CONCLUSION 43
ABSTRACT
Identifying the need for more research on what drives growth in gazelle firms, this thesis explores how managers of gazelle firms respond to paradoxical tensions. A multiple-case study is used to apply paradox theory to both the level of the manager and the firm-level
INTRODUCTION
Fast-growing firms (known as ‘gazelles’) are responsible for a substantial share of GDP growth (Sims & O’Regan, 2006) and job creation (Birch, 1981). In addition, fast growth is linked to a significant increase in a firms’ survival chances (St-Pierre, Julien & Morin, 2010). However, academic literature is still divided on which factors are responsible for firm growth (Grundström, Sjöström, Uddenberg & Öhrwall Rönnbäck, 2012). Although an increasing body of research point to innovation as being an important driver of growth (Coad, 2009), it is still unclear how successful firms are managing the tensions underlying innovation. Ambidexterity literature identifies the most important innovation-related tension to be
exploitation-exploration (Andriopoulos & Lewis, 2009). which is why this tension will be the focal point of this thesis. Ambidexterity, referring to the capability of managing exploitation and exploration simultaneously, is a growing area within innovation management. Especially in uncertain environments ambidexterity “appears to be positively associated with increased firm innovation, better financial performance, and higher survival rates” (O’Reilly &
Tushman, 2013, p. 7). Considering that gazelle firms are commonly associated with complex environments (Birch & Medoff, 1994), they fit well in the theory of ambidexterity.
Ambidexterity literature has identified exploitation-exploration tensions (Birkinshaw & Gupta, 2013) as the key to innovation. By looking into the paradoxical tensions associated with exploitation and exploration, this thesis aims not only to contribute to an increased understanding of gazelle firms, but also to ambidexterity literature
LITERATURE REVIEW
GAZELLE FIRMS
Gazelle firms account for the majority of new jobs created in the economy (Birch, 1981; Acs et al., 2008). Birch (1981) estimated that even though only 4% of US companies can be characterized as being gazelle firms, these account for a staggering 70% of newly created jobs. Although the exact numbers have been disputed, most studies agree that gazelles worldwide have an enormous impact on the creation of new jobs (Lopez-Garcia & Puente, 2012). Not only gazelle firms were found to be highly impactful: SME’s, a group to which most gazelles belong (Henrekson & Johansson, 2010), have been found to contribute significantly to job- and value creation. SME’s are estimated to account for over half of the GDP in Western countries (Sims & O’Regan, 2006). Gazelles can be found in all sectors don’t necessarily have to be small or young firms (Grundström et al., 2012). These is no consensus in academic literature regarding the exact definition of a gazelle firm (Henrekson et al., 2010). The present thesis adopts the widely used criteria as introduced by Birch, Haggerty & Parsons (1995), who define gazelle firms as achieving a minimum average revenue growth rate of 20% during a minimum period of three years. The revenue in the base year should be at least $100.000,-.
Despite these insights, most studies have focused on large firms and neglected small, fast-growing firms (Lubatkin, Simsek & Veiga, 2006). A common misconception, which can at least partially explain this discrepancy, is the idea that findings related to large firms can be copied to smaller firms without dedicated theoretical or empirical research (Sims & O’Regan, 2006). This is not the case, because key differences exist between small and large firms. This is illustrated by Lubatkin et al. (2006), who state that top managers in smaller firms tend to be “closer to the firm’s operating core” (p.649) and are therefore more involved in the day-to-day implementation of strategies. Of the studies that have been conducted on SME’s, the majority has focused on survival-related factor (e.g. financing), instead of the growth process or the ways in which firms achieve a competitive advantage (Sims & O’Regan, 2006). This is remarkable, because firm growth has been shown to significantly increase the chances for a firm to survive (St-Pierre, Julien & Morin, 2010). The question of how firms grow is therefore still not clearly answered by academic literature. As pointed out by Grundström, Sjöström, Uddenberg & Öhrwall Rönnbäck (2012, p. 15), “more research is needed in order
observed fast growth are somewhat speculative at this stage”. With regard to the question how growth can be achieved, many studies point towards the general direction of innovation playing a crucial role (Coad, 2009). Especially in today’s dynamic world, the advice is that managers aiming for growth should place strong emphasis on innovation (Laforet, 2010). Whereas there is little dispute about the claim that innovation is important for growth, many questions remain to be answered with regard to management practices promoting innovation. As pointed out by leading ambidexterity scholars (e.g. Tushman, Smith & Binns, 2011), increasing innovative performance is easier said than done. It involves a trade-off with the short-term business, which many small and resource-constrained firms find hard to make. Illustrating this trade-off is the work of Hansen & Hamilton (2011), who looked into the factors distinguishing small firms that grow from those that don’t. They found that “growth depends primarily on exploiting opportunities to increase sales” (p. 284), requiring both the ability to recognize opportunities for new business (exploration) and the ability to exploit it successfully (exploitation). However, Henrekson & Johansson (2010) fail to elaborate on how these abilities are balanced. It makes intuitive sense to think that gazelles would benefit from ambidexterity, considering they are inherently complex and are characterized by instability and volatility (Birch & Medoff, 1994). This places high demands on leaders in terms of flexibly pursuing different strategies. Many mistakes are made within Gazelles, but large successes are enjoyed when things go well (Birch & Medoff, 1994). Due to the
dynamically changing environment, gazelles need to constantly adapt in order “to maintain, gain or retain competitive advantage” (Sims & O’Regan, 2006, p. 952). Whereas large and established companies are able to survive while over-emphasizing exploitation or
exploration, albeit with sub-optimal returns, young companies “necessarily have to balance exploration and a healthy amount of exploitation in order to survive and thrive” (Reeves & Harnoss, 2015). However, many question remain with regard to this balancing act have remained unanswered.
Related to the exploitating-exploration tension is the notion of the “success trap”. This happens when a firms achieves early favorable results by focusing on exploitation activities, causing it to focus even more on exploitation and thus losing out to the competition in the long run (Gupta et al., 2006). This makes it a relevant problem for gazelles, considering that many gazelles grow as a result of finding early success. The mechanism behind the success trap is that firms tend to apply solutions which have worked in the past, to problems which have arisen in a new and different context (Levinthal, 1992). It has been argued that one of the key ways in which gazelles can prevent themselves from falling victim to the success
trap, is by developing ambidextrous capabilities (Rosing et al., 2011). However, this is contradicted by the work of Sirmon et al. (2011), who state that ambidexterity doesn’t seem to become important before the maturity phase, as only exploration is necessary in the entrepreneurial phase and only exploitation is necessary in the growth phase. Other research suggests that gazelles are especially strong in managing the trade-off between exploitation of the current product/service portfolio, while simultaneously exploring new opportunities in order to remain competitive in the future (Sims & O’Regan, 2006; Eisenhardt & Martin, 2000). These examples make it clear that more research is needed on the topic of
ambidexterity in gazelle firms.
By exploring ambidexterity in gazelle firms, this study aims to contribute to the debate about what enables firms to grow so exceptionally fast. This can prove to be beneficial for both managers and policy makers alike, since they both benefit from strong firm growth: either in the form of value creation for themselves and their employees, or in the form of value creation for the economy. This study takes an explorative approach, introducing existing literature as a deductive element. This approach answers the call of O’Reilly & Tushman (2013) for more qualitative and in-depth studies, focusing on ambidexterity in practice (O’Reilly & Tushman, 2013). In addition, the studies that have been conducted on gazelles have largely been descriptive and “explore a limited number of characteristics of fast-growth firms” (Lopez-Garcia & Puente, 2012, p. 1030), allowing plenty of room for research from an explorative angle.
AMBIDEXTERITY
The term “organizational ambidexterity” was first used by Duncan (1976), introduced the idea that firms needed two different structures: one for initiating innovations and one for executing innovations with. The concept of ambidexterity started receiving wide attention after the work of March (1991), who focused on exploration and exploitation in the context of organizational learning. Since this publication, the terms “exploitation” and “exploration” have received increasing attention in relation to issues like “technological innovation, organization design, organizational adaptation, organizational learning, competitive advantage and, indeed, organizational survival” (Gupta et al., 2006, p. 693). As a result of this wide diffusion, some questions exist as to what “ambidexterity” exactly means.
Addressing this issue, Birkinshaw & Gupta (2013) argue that the concept of ambidexterity is theoretical in nature – considering that managers rarely use the term – and therefore propose
Ghoshal (1999), who introduced the idea that the existence of firms is justified because they are able to do things which markets cannot. Birkinshaw & Gupta (2013) interpret this as meaning that “firms exist to do difficult things, such as making trade-offs between short- and long-term demands and allocating scarce resources among competing priorities depending on multiple criteria” (p. 290). Noting the increasingly broad definition of ambidexterity that ambidexterity literature has been Birkinshaw & Gupta (2013) worry that ambidexterity is becoming a catchall term and is therefore losing meaning. The same concern is expressed by O’Reilly & Tushman (2013, p. 15), who observe that “ambidexterity has been applied to phenomena such as strategy, networks, new product development, technology, software development, intellectual capital and other topics that, while interesting and important, may have little to do with the practical tensions involved in how managers and organizations deal with exploration and exploitation”. O’Reilly & Tushman (2013) agree with Nosella,
Cantarello & Filippini (2012, p. 459), who advocate a return to the “original definition of the construct as a capability for resolving tensions”. In line with these authors, the present thesis will therefore view ambidexterity as a capability for solving tensions related to exploitation and exploration.
O’Reilly and Tushman (2004) argue that balancing the exploitation of existing capabilities with the exploration of new opportunities is one of the toughest managerial challenges. Companies that succeed in this balancing act are called ambidextrous (O’Reilly and Tushman, 2004). Traditional strategy research has often advocated that a firm should focus its activities on either exploration or exploitation to avoid being “stuck in the middle” (Porter, 1980). In this traditional view specialization is preferred over duality. However, recent work suggests that pursuing different strategies simultaneously might not only be viable, but possibly even preferred as way to achieve strategic ambidexterity (Markides & Oyson, 2010). For example, Zimmerman, Raisch & Birkinshaw (2015) state that firms are increasingly required to pursue “jointly desirable but competing objectives” (p. 1119). Examples of the dualities that firms often try to balance are: “efficiency and flexibility, alignment and adaptability, and exploitation and exploration” (Zimmerman et al, 2015, p. 1119). As stated earlier in this chapter, the present thesis focuses on the exploitation-exploration duality.
STRUCTURAL AMBIDEXTERITY
O’Reilly & Tushman (2004) found that ambidextrous companies share several characteristics. Most importantly, “they separate their new, exploratory units from their traditional, exploitative ones, allowing for different processes, structures and cultures” (p. 75). This emphasis on separation for achieving exploration and exploitation is known as structural ambidexterity. Other authors have argued against this form of ambidexterity, as it leaves some important issues unanswered (Raisch & Birkinshaw, 2008). For example, the issue of interunit coordination and integration poses a challenge when units are structurally separated. More specifically, integrating knowledge and leveraging existing capabilities might become difficult due to “a lack of direct transmission channels and the institutional distance caused by hierarchical coordination” (Raisch & Birkinshaw, 2008, p. 397).
Zimmerman et al. (2015) react to this objection by proposing the possibility of maintaining structural ambidexterity “through the development of a supportive organizational context within a particular organizational unit” (p. 1119).
The idea of structural ambidexterity stems from two problems that organizations face when trying to unite exploration and exploitation: first, a culture of incremental innovation “often creates an institutional hostility towards discontinuous innovation” (Papachroni, Heracleous & Paroutis, 2015, p. 73). And secondly, these different types of innovation are competing for the same scarce resources. In order to solve this issue, proponents of structural separation propose to create different units within the organization, each focusing solely on exploration or exploitation. Units focused on exploration are usually “small, decentralized with loose processes”, whereas exploitative units “are described as larger, more centralized and with tighter processes” (Papachroni et al., 2015, p. 74). Structural ambidexterity mainly entails separation on the unit-level, as there are usually several integration mechanisms in place at the organizational level. A key role is reserved for the top management team, which are supposed to function as the “corporate glue” by mediating any potential tensions between exploitation and exploration (O’Reilly & Tushman, 2004). Middle-management is also prescribed as playing a mediating role by transferring knowledge between the separate domains (Jansen, Vera & Crossan, 2009).
Contrary to structural ambidexterity, contextual ambidexterity is not about structural
separation of units. Instead, it argues that ambidexterity is best achieved “by building a set of processes or systems that enable and encourage individuals to make their own judgments about how to divide their time between conflicting demands for alignment and adaptability” (Gibson & Birkinshaw, 2004, p. 210). This means that ambidexterity is achieved within a unit, in which individuals are encouraged to make judgments of their own regarding the way they should spend their time: either on exploitation or similar to exploration (Gibson & Birkinshaw, 2004). This requires temporal separation at the individual level, as people switch between exploitation- and exploration-related activities. Contextual ambidexterity can also entail temporal separation on an organizational level, meaning that the organization as a whole emphasizes exploration or exploitation at different points in time.
PARADOX THEORY
Apparent in both the structural and contextual approach to ambidexterity is the assumption of an inherent conflict between the activities related to exploration and those related to
exploitation (Papachroni, 2015). Looking to provide a way out of this dichotomy, an
increasing body of researchers suggests looking through the lens of paradox theory (Lewis & Smith, 2014). These researchers see in paradox theory the potential to “take us (…) toward potential synthesis of poles forming a duality, or transcendence of tensions, as well as emphasizing a longitudinal, dynamic, and productive interrelationship between poles” (Papachroni et al., 2015, p. 72). A paradox is defined as “contradictory yet interrelated elements that exist simultaneously and persist over time” (Lewis & Smith, 2014, p. 134). The traditionally dominant approaches to ambidexterity and the view from paradox theory. Paradox theory is proposed as “a valuable approach for research on how tensions of ambidexterity are perceived and managed simultaneously at the individual level” (Papachroni et al., 2015, p. 84). Instead of the duality approach of “either/or”, paradox theory suggests opening the door to “both/and” thinking. In this view, exploration and exploitation are seen being complementary rather than irreconcilable. An illustrative example of the application of paradox theory is found in the work of Farjoun (2010), who poses that stability was
traditionally associated with rigidity and inflexibility, and considered contradictory to adaptation. However, paradox theory would allow one to realize that adaptation might actually be a viable way of achieving stability and continuity (Farjoun, 2010). Papachroni et al. (2015) argue that paradox theory provides a way to combine elements from structural, contextual and individual ambidexterity in a more holistic approach to ambidexterity
research. It is perceived by many as a welcome alternative is, because traditional
ambidexterity studies have offered only limited insight in the processes involved (Nosella et al., 2012). Paradox theory is quickly gaining a larger following, and is expected to become and increasingly area of ambidexterity research (Smith & Lewis, 2011, p. 398).
ORGANIZATIONAL TENSIONS
Adopting a paradox perspective makes it possible to understand why managing the
exploitation-exploration tension is such a daunting task, as research has shown that the choice between two alternatives can have a paralyzing effect on both groups and individuals (Smith & Berg, 1987). When confronted with a paradox, people often react defensively and take an either/or approach. This leads them to choose one alternative over the other (Vince & Broussine, 1996), eventually causing demand for the other alternative to resurface. As the pulling force of the opposing pole increases, a downward spiral known as a “vicious cycle” is started (Lewis & Smith, 2014). For this reason, managers looking to drive organizational performance experience have been shown to strongly benefit from the ability to recognize and address paradoxical tensions in time (Lewis et al., 2014; Smith, 2014). Even though there are strong indications that cognition and sense-making may underlie this ability, very few studies have been done connecting individual manager’s thinking to ambidexterity (Ingram, Lewis, Andriopoulos & Gotsi, 2009). In one of the earliest studies exploring this link, Gibson & Birkenshaw (2004) state that ambidextrous thinking is necessary for dealing with
innovation-related tensions. In addition, Smith & Tushman (2005) point to paradoxical cognition as important for dealing with tensions related to pursuing both exploration and exploitation. Important to note here is that a tension being present does not necessarily mean that managers are aware of them: tensions can go unnoticed for years (Gilbert, 2005). This is problematic, because only after becoming aware of a tension can it be adequately managed (Miron-Spektor et al., 2011).
Elaborating on this, Lewis & Smith (2014, p. 133) argue that “from a paradox
perspective, the construction of paradox emanates from actor’s responses to tensions”. Smith (2014) and Lewis & Smith (2014) provide two general categories in which these responses can be divided: defensive and strategic. According to Lewis & Smith (2014), defensive responses can be cognitive, behavioral or institutional in nature and “serve to temporarily avoid or reduce the negative affect raised by tensions” (p. 135). These responses have in common that they tend to avoid risk and conflict by promoting consistency and simplicity
competing forces” (Lewis & Smith, 2014, p. 135). Instead of denying or avoiding the tension, a strategic response seeks to “embrace, cope with, and thrive through tensions” (Lewis & Smith, 2014, p. 135). This is the result of an actor being able to transcend the paradox, and thus being aware of the potential synergy of addressing both poles (Lewis, 2000).
Because exploration-exploitation tensions concern core organizational issues, the way in which these in which tensions are managed is considered to have great impact on an organization’s performance and survival chances (Jay, 2013; Lewis & Tushman, 2011). Associated leadership practices can be divided into two main categories: differentiation and integration (Andriopoulos & Lewis, 2009; Smith, 2014). Although these approaches have been recognized as being complementary since the work of Lawrence & Lorsch (1967), ambidexterity scholars have thus far mostly focused on only one of the two leadership practices (Raisch et al., 2009). Differentiation is commonly advocated by proponents of structural ambidexterity, who argue that exploitation and exploration efforts should be kept separate. In this context, differentiation is generally achieved by temporal or spatial
segregation of exploration and exploitation activities (Puranam et al., 2006). This means, respectively, that the same unit is used “at different times for either exploitation or exploration” (Andriopoulos & Lewis, 2009, p. 697), or that different units are established with the focus on only one of the domains. Smith (2014, p. 1616) adds that differentiation tactics include “allocating domain-specific roles, comparing domains to raise novel distinctions, and seeking information about domains independently”.
In contrast, integration tactics emphasize the importance of combining exploitation and exploration as an answer to tensions. This is associated with a paradoxical mindset (Papachroni et al., 2015), which emphasizes social and cultural approaches like
“organizational vision, project norms for improvisation, and socialized identities as practical artists” (Andriopoulos & Lewis, 2009, p. 708). Integration tactics are therefore often
advocated by proponents of contextual ambidexterity. In general, differentiation and
integration tactics have been observed to affect the following categories within organizations (Smith, 2014): resource allocation (e.g. human resources, financial resources and time), organizational design (e.g. processes, structure and metrics), and product design (e.g. technology and target market). Finally, Smith (2014) found that a clear pattern started to show after observing full sets of leaders’ decisions; some leaders consistently made decisions supporting only exploration or exploitation, consequently neglecting the other domain. Other leaders were able to constantly switch between the domains; a capability fundamental to ambidextrous leadership (Smith, 2014). This allows for the categorization of a leader’s
decision pattern as being explorative, exploitative, or ambidextrous. Because the leader has been identified as having crucial influence on resource allocation and organizational structure (Lavie, D., Stettner, U. & Tushman, 2010), this will be the main subject of investigation. For a researcher “setting the stage for paradox theory and research”, a key initial step is “clearly articulating the focal paradox or paradoxes of inquiry” (Lewis & Smith, 2014, p. 134). In order to articulate these focal paradoxes, the point of departure is the definition of ambidexterity as the capability of solving tensions related to exploitation and exploration (Nosella et al., 2012; O’Reilly et al., 2013; Birkinshaw & Gupta, 2013). This gives rise to the question which exact exploitation-exploration tensions this definition is referring to. Because this subject is still relatively new, only a few studies have focused exploitation-exploration tensions in practice (O’Reilly & Tushman, 2013). Studies that did were mostly theoretical in nature (Martini et al., 2013), focused on senior leader characteristics (Smith & Tushman, 2005) or did not address “the dynamic processes underlying organizational ambidexterity” (Raisch et al., 2009, p. 693). A notable exception is the widely cited work of Andriopoulos and Lewis (2009). By researching exemplar cases of innovative SME’s, specifically looking at exploitation-exploration tensions, Andriopoulos & Lewis (2009) managed to find three paradoxes to be strongly present in all firms: strategic intent, customer orientation, and personal drivers (figure 1). These tensions will form the starting point of this research paper.
Figure 1: Key tensions according to Andriopoulos & Lewis (2009)
The first paradox, strategic intent, refers to the opposing goals associated with exploitation and exploration. These are “profit emphasis” and “breakthrough emphasis” respectively (Andriopoulos & Lewis, 2009). The paradox exists because exploitation is associated with the pursuit of slow and steady increases in revenue and profit (March, 1991) and is
2006). On the other hand, exploration aims for the realization of breakthrough innovations (March, 1991) and, due to “the potential for extreme profits or extreme losses” (Taylor & Greve, 2006, p.723), is therefore associated with high variance in performance.
The second paradox, customer orientation, stems from seemingly contradictory ways of dealing with client demands. On one end there is “tight coupling”, which was found to be the case when informants “demonstrated great client respect, valuing their insights into their market and competition, and seeking to ensure their satisfaction and loyalty” (Andriopoulos & Lewis, 2009, p. 705). This stems from the notion that the client is the expert in its field and that their current needs should be satisfied as much as possible. Tight coupling is therefore characterized by client – and market-driven constraints. In contrast, “loose coupling” is characterized by a disparagement of current client needs, and instead emphasizes the
exploration of new trends and emerging technologies. This is paired with the notion that the client’s demands are not always in its own best interest, causing the firm to stretch project boundaries as well as the client’s comfort zone (Andriopoulos & Lewis, 2009).
The third paradox, personal drivers, consists of discipline on one side and passion on the other. Discipline mainly concerns the degree to which “control, accountability, and structure” (Andriopoulos & Lewis, 2009, p. 706) are geared towards execution and efficiency. This is expressed, for example, in the form of tight control mechanisms and standardized tasks. In contrast, passion is concerned with the emotional and creative side of the people working at the firm. Passion was found to be strongly linked to a high degree of freedom, challenging tasks and intrinsic motivation. These factors were found to be vital to creative expression (Andriopoulos & Lewis, 2009), which in line with earlier research on creativity in firms (Brown & Duguid, 2001; Amabile et al., 2005).
Following the advice of Lewis & Smith (2014) to clearly specify the focal paradoxes at the beginning stages of the research, I chose to take the three paradoxes as described by Andriopoulos & Lewis as the focal paradoxes. The two types responses (i.e. defensive and strategic; Lewis & Smith, 2014) were included to shed light on how managers framed each tension mentally. The category management practice was included to depict the practical managerial actions and measures which were taken in reaction to specific tensions. These were divided into actions emphasizing only one side of the duality or both at the same time.
DATA AND METHOD
RESEARCH PARADIGM
A paradigm can be defined as the “basic belief system or worldview that guides the investigator” (Guba and Lincoln, 1994, p. 105), thus constituting a framework for the researcher to work in. Paradigms not only concern methodology, but also consist of the elements of ontology and epistemology (Guba & Lincoln, 1994). Ontology is explained by Saunders et al. (2009) as being concerned with the nature of reality. A central question in ontology is: “What assumptions do we make about the way in which the world works?” (Saunders, 2009, p. 110). Epistemology, on the other hand, is about “the researcher’s view regarding what constitutes acceptable knowledge” (Saunders, 2009, p. 119). Healy & Perry (2000) identify four main paradigms: Positivism, critical theory, constructivism and realism (Figure 1).
Considering these different paradigms, realism is most suitable for approaching the research question of the present thesis. Also known as post-positivism (Guba & Lincoln 1994), realism accepts “that there is a real world to discover even if it is only imperfectly and probabilistically apprehensible” (Healy & Perry, 2000, p. 125). This thesis investigates multiple the response and, importantly, also the practice displayed by managers in reaction to paradoxical tensions. This fits in the realism paradigm, because “a perception for realists is a window on to reality from which a picture of reality can be triangulated with other
perceptions” (Perry et al. 1999, p. 16). These different perceptions are provided by the informants that were interviewed for this thesis, which is possible due to of the multiple case study approach. It has also been noted that case studies fit especially well in the realism paradigm (Healy & Perry, 2000).
SAMPLING AND RESEARCH DESIGN
In order to be able to replicate findings across cases, thus improving its reliability and generalizability (Baxter & Jack, 2008), I decided to use a multiple-case study approach. This was a parallel study, meaning that “the cases are all happening and being studied
concurrently” (Thomas, 2011, p. 517). Considering that the choice between descriptive, exploratory and explanatory research is mainly based on “the richness of the rival
propositions in theories related to topic of the study” (Yin, 1994, p. 2), and the lack of such propositions in ambidexterity research in fast-growing firms, the exploratory design makes the most sense for this study. Given the explorative nature of thesis I chose to conduct qualitative interview, as these lend themselves well for explorative research (Cooper &
likely choice “where it is necessary for you to understand the reasons for the decisions that your research participants have taken, or to understand the reasons for their attitudes and opinions”. One particular benefit of the qualitative interview is that it allows the researcher to ask the interviewee to further explain or build on a response (Saunders et al., 2009). This is important when the researcher is looking to “understand the meanings that participants ascribe to various phenomena” (p. 324). Although some researchers suggest combining qualitative and quantitative methods, “it is likely that quantitative methods and qualitative methods will eventually answer questions that do not easily come together to provide a single, well-integrated picture of the situation” (Patton, 1990, pp. 464-5). Finally, this study has an embedded design, meaning that multiple levels of analysis are included in the (Yin, 1984). In the this thesis these levels are the individual and the organization as a whole. In order to find gazelle firms to include in the research, I decided to use several recent on-line lists, ranking gazelle from highest to lowest growth. These lists display the results of annual competitions, which are held by different firms to award the firm which achieved the highest growth. I approached the firms on these lists by e-mail, following the lists from top to bottom (i.e. descending from highest growth rate to lowest). In doing so, I tried contacting the founder/CEO’s directly when possible. My aim was to include at least two people from every firm in the sample as a form of triangulation (Patton, 1999), one of which had to be the founder and a C-level organizational member. Triangulation means that different sources of data, such as interviews, observations and documentation, are combined in order to find mutual confirmation (Gibbert & Ruigrok, 2010), and is one of the most important ways to ensure validity (Yin, 1994).
I first developed a standard template for the e-mail based on the checklist from Saunders et al. (2009, p. 183), including suggestions like writing in a professional tone and clear presentation. Each e-mail was personalized and clearly stated my intensions, without going into detail about the research itself. In accordance with Saunders et al. (2009) I promised the firms to send them a personalized management summary, including recommendations based on my findings, after the thesis would be done. They were also ensured that their data was strictly confidential and would be handles with care. Case study research is not suitable for achieving statistical generalizability to a population (Lee, 2003; Yin, 1994), because “the case study, like the experiment, does not represent a ‘sample’” (Yin, 2009, p.15). The goal should instead be to “expand and generalize theories” (p. 15), also known as analytic generalization (Yin, 2009). This is why I chose cases based on theoretical, rather than statistical considerations (Glaser & Strauss, 1967). In order to standardize the
measurements and ensure theoretical generalizability of the findings, I included only firms which achieved growth from organic development (St-Jean, Julien & Audet, 2008). This means that firms which grew as a result of mergers or acquisitions were excluded. The other criterion was the gazelle definition of Birch et al. (1995), of at least 20% average annual revenue growth over three consecutive years, with a revenue of $100.000,- in the base year. No other criteria were included, as gazelle firms do not have other unique or defining characteristics that could be taken into account (Cassia, Cogliati, Minola & Paleari, 2016). After trying to recruit participants from the lists I found on-line, and finding limited success, I decided to start looking for contacts in my own network as well. In the end I found five informants willing to participate through my own network and five informants through e-mail approach, coming from seven different firms in total. With regard to the number of cases, Eisenhardt (1989) recommends to aim for minimum of four and a maximum of ten cases in total: with less than four cases it becomes difficult to find sufficient empirical grounding, whereas a total exceeding ten can make the volume and complexity of the data difficult to cope with (Eisenhardt, 1989).
Two important notes have to be made with regard to the case selection. First, one of the firms included in the research is not a gazelle (Firm B). Even though it is growing at a rapid rate, it was founded too recently (2013) and the revenue in the base year was too low to conform to the gazelle definition of Birch et al. (1995). The reason I decided to include this case is because this allows for deviant-case analysis. This means incorporating cases into the analysis which “do not fit with the theoretical framework or developing hypotheses” (Gibbert & Ruigrok, 2010, p. 714). The idea behind including a deviant case is that it can function as the exception that confirms the rule, depending on the way in which the case differs from the rest of the sample.
The second note concerns a case that was removed from the sample after the data was collected. This is because, while analyzing the data, I realized that this case was substantially less rich in information than the rest. This was likely because this was the only interview conducted by phone, whereas the other interviews were all conducted in person at the firm’s office. The phone interview was also shorter than the rest. The interviews that were used for this thesis lasted 50 minutes on average.
Table 1. Informants
6. INTERVIEWS
Semi-structured interviews were used, based on the interview guide approach as outlined by Patton (2002). This method is not only beneficial to the comprehensiveness of the data when compared to unstructured interviews, it also enables a systematic handling of the different participants’ interviews and allows for flexibility, while preventing logical gaps at the same time (Patton, 2002). Each interview was audio-recorded and transcribed verbatim. An audio record is more reliable than human memory, allows for detailed transcription and can be replayed in order to improve the transcript (Silverman, 2013). Notes were made during the interview to help with probing for important issues without interrupting the participant. Furthermore, as advocated by Yin (1994), a database was created containing the case studies and all accompanying materials. In addition to the semi-structured interviews, strategic documents were collected for analysis in order to provide an extra source of data. This is another means of triangulation, which allowed for the corroboration of data across sets and reduced the potential for bias (Bowen, 2009). Although the most important part of
the interview was semi-structured, I included 11 structured questions in the beginning of the interview. These were included to gain background relevant information about the cases.
DATA ANALYSIS
Due to the wide variety of qualitative data, there is no single standardized procedure for qualitative data analysis (Saunders et al. 2009). It is, however, possible to discern three main activities used in qualitative analyses: (1) summarizing of meaning, (2) categorization of meaning, and (3) structuring of meaning using narrative. These activities can be conducted on their own or in conjunction with each other (Saunders et al., 2009). Summarizing is useful when the researcher is looking to become more intimately acquainted with the main themes emerging from the data. This can help with the identification of issues or relationships requiring closer examination in later stages. Summarizing was not deemed necessary for this analysis, as I had already conducted repeated readings of the interview transcripts and archival material to become “intimately familiar with each case as a stand-alone entity” (Eisenhardt, 1989, p. 540). This was done in order to allow the unique themes and characteristics of the individual cases to emerge and speed up the subsequent cross-case analysis (Eisenhardt, 1989).
Categorization of data – which was a key part of my analytical strategy – involves two steps. The first is to develop categories, the second is “attaching these categories to meaningful chunks of data” (Saunders et al., 2009, p. 492). These categories can initially be established based on data or on a theoretical framework, which was another important part of the analytical process. In both cases the categories are strongly influenced by the research question, which means that it is possible that two researchers come up with different
categories based on the same data (Dey, 1993). Naming the categories can be done by using terms appearing in the data, ‘in vivo’ codes (i.e. words literally used by interviewees), or existing literature (Strauss & Corbin, 2008). No matter how the categories are formed, they should always serve the purpose of providing the researcher with a clear framework for further analysis (Saunders et al., 2009). According to Dey (1993), all categories must have two aspects: “an internal aspect – they must be meaningful in relation to the data – and an external aspect – they must be meaningful in relation to the other categories” (p. 96-97). After the initial categories have been developed, the researcher can start unitizing the data. This means that units of data – differing in size from a single word to a whole paragraph – are assigned to the appropriate categories (Saunders et al., 2009). Categories can be subject to change as the analysis progresses (Strauss & Corbin, 2008).
In line with the suggestions of Suanders et al. (2009) outlined above, I adopted an analytical method called template analysis (King, 2004). The essence of template analysis is “that the researcher produces a list of codes (‘template’) representing themes identified in
their textual data” (King, 2004, p. 256). This procedure combines deductive and inductive elements, in the sense that “codes can be predetermined and then amended or added to as data are collected and analyzed” (Saunders et al., 2009). Template analysis lends itself to be used within the realism paradigm, in which the researcher “is concerned with ‘discovering’ underlying causes of human action” (King, 2004, p. 256). Template analysis resembles the grounded theory approach, but is different in the sense that template analysis allows for prior specification of categories whereas grounded theory is almost exclusively inductive. For my initial template I used the interview guide as the framework. Considering that the interview guide was based on the research question and the literature reviews, this ensured the same basis for the initial template
The categories for the initial template were derived directly from the interview guide. Considering that the interview guide was based on the research question and the literature review, this ensured the same theoretical grounding for the template. Making use of the freedom that template analysis provides (King, 2004), I assigned a second code to all management practices. See figure 4 for the initial template. This second code signals which organizational area was affected by the corresponding management practice (e.g. resource allocation, organizational design or product design).
No quantitative coding was done, because template analysis emphasizes “flexible and
pragmatic use of coding” (King, 2004, p. 256). This makes it problematic to assume that “the frequency of a code in a particular text corresponds to its salience” (King, 2004, p. 256). Coding used in template analysis can be descriptive or interpretative, and is characterized by a hierarchical organization allowing for analysis at different levels of specificity. “Higher order codes” indicate a general theme or direction, whereas “lower order codes” are used for indicating details and finer distinctions (King, 2004). After establishing the initial template, therewith concluding the deductive part of the research process, I started coding the data using qualitative data analysis software ‘Dedoose’. After an initial round of coding using the initial template, I decided to remodel the template based on new insights. Restructuring is considered an integral part of template analysis (King, 2004), and allows for the
RESULTS
INDIVIDUAL CASE ANALYSIS
With regard to the presentation of the data, template analysis lends itself to several different approaches. The main decision to be made is whether to start with an analysis of all
individual cases first, or immediately jump to a discussion about the main themes while using cases as illustration (King, 2004). I chose to start with a presentation of all individual cases, giving the reader “a good grasp of the perspectives of individual participants” (King, 2004, p. 267). I argue that this is highly valuable, given that the complex nature of the subject makes it important to understand each case’s unique dynamics. In addition, this method tackles the potential problem of a wide chasm separating the conclusions from the data (Eisenhardt, 1989).
Using categories as displayed on the second template, each case was analyzed in detail. The results can be found below.
CROSS-CASE ANALYSIS
The first goals of the cross-case analysis was to establish whether the firms included in the sample tended to focus on a specific side of each paradox. If this was found to be the case, the next step would be to go back to the data and see how this can be explained. The idea here is to see whether certain finding can possibly be generalized to theory. As stated by Yin (2009, p. 15): “Case studies, like experiments, are generalizable to theoretical propositions and not to populations or universes. In this sense, the case study, like the experiment, does not represent a ‘sample’, and, in doing a case study, your goal will be to expand and generalize theories (analytic generalization) and not to enumerate frequencies (statistical generalization)”.
The first paradox, strategic intent, concerns the trade-off between profit and breakthroughs innovations. What has interesting to note with regard to this paradox is that several informants indicate that they have recently shifted from their focus from
breakthrough to profit. In all cases this was the result of a breakthrough innovation having caught on in the market, after which resources previously spent on breakthrough innovations could now be spent on efficiency-related measures
“Now that we have a system [i.e. realized a breakthrough] we are looking how to expand further. We’re constantly tuning our business model to become more efficient”
– CEO Firm A
Also interesting to note is that many firms strongly emphasizing profit indicate that their product or service is so strong that they don’t have to worry about the success trap.
“Increasing profit is what it’s all about. We want to realize this with the business we currently have, because there is still so much left to gain. This is why we’re not looking for radically new ways to make money” – CEO Firm E
The only firms with a strong focus on breakthrough innovation both have good reasons for it. One is the Firm B, the deviant case, which is still in the process of actually realizing their breakthrough. The other is firm C, which aims to make the development of radically new software solutions its core business.
The way in which case firms approach the second paradox, customer orientation, seemed to depend largely on the type of market in which the firm operates. Firm 2 is highly restricted in in freedom towards customer orders, because clients know exactly what they want and will be unhappy if delivered something else. In contrast, Firm C, D and F not that they have lots of room to experiment. They all appoint this to the situation being that they have much more product knowledge that the customer. The customer expects them to take them by the hand, thus leaving room for the firm to fill in the project as they desire. The firm only firm able to truly transcend the paradox between tight and loose coupling was Firm D. Both the CEO and the Account Director of Firm D gave strong indications that they emphasized both sides of the paradox at the same time. This is done by iterative testing with clients, continuously processing feedback, and even varying their approach depending on the size of the client.
“After the learning process at a small customer is completed, we go to the bigger customer
knowing exactly what we have to offer” – CEO Firm D
This is in strong contrast to Firm A, which is the only firm committed to delivering exactly what the customer wants.
The final paradox, personal drivers, concerns the way – and degree to which people are motivated for their work and driven to deliver certain results. Indicative of many cases was a high emphasis on passion. There was not a single case which didn’t emphasize the
importance of passion. Also interesting was the notion that discipline follows automatically when people are passionate.
“If I really like something, I will be willing to put a lot of time in it. And that is an important
part of what discipline means to me: putting a lot of time into something. This is why it can be difficult to tell engineers to stop working on a project that they enjoy, because I think they can better spend their time elsewhere” – CTO Firm C
“Discipline and passion go together extremely well. If you have passion for something, you’re going to be disciplined because you want to make the best out of it” – Account Director Firm D
Interestingly, little management practices were indicated to be used by managers to actually promote passion. The main way in which this was done, was by hiring passionate people. There were no firms, except deviant case Firm A, which promoted certain higher goals or
values. This makes the deviant case come to its intended use, because here it is the exception confirming the rule. Another trend that was clearly visible across all cases was the
importance of freedom. All managers gave their employees room to shape their work to a large degree. Another similarity appeared in relation to the boundaries of this freedom, as managers across cases emphasized the importance of setting priorities.
In general, few firms showed the ability to excel in both sides of a paradox. This is in line with earlier research, which has stated that managing exploration-exploitation tensions is a daunting task.
CONCLUSION AND LIMITATIONS
From the cross-case analysis it can be concluded that the gazelles from the sample were mostly unable to excel in both sides of a paradox. The only example was Firm D, which was able to combine tight and loose coupling in a remarkable manner. With regard to the strategic intent paradox, the cases showed a relation between the maturity stage of the firm and the focus being on either breakthroughs or profit. Many cases who adopted a profit focus had successfully launched a breakthrough innovation and had therefore switched focus. The customer orientation paradox was shown to be linked to the same goal: companies pursuing a profit focus were more inclined to emphasize tight coupling, with loose coupling being emphasized by firms aiming for breakthroughs. Lastly, the personal driver paradox showed a clear emphasis on the importance of passion across cases. Interesting to see was that despite this indicated importance, firms took little measures to actually promote passion. Future research is need to look further into the ways in which passion can be promoted in gazelle firms. Another promising area for future research is to look into the individual characteristics displayed by managers who were able to emphasize both sides of the paradox at once. An interesting question is whether these leaders possess certain skills allowing them to be ambidextrous, and, perhaps the most important questions, whether or these skills can be taught.
REFERENCES
• Acs, Z., Parsons, W., Tracy, S., & United States. (2008). High-impact firms: Gazelles
revisited. Washington, D.C.: SBA Office of Advocacy.
• Amabile, T. M., Barsade, S. G., Mueller & J. S., Staw, B. M. (2005). Affect and creativity at work. Administrative Science Quarterly, 50, 367–403.
• Birch, D. & David, L. (1981). Who Creates Jobs? The Public Interest, 65, 3-14. • Birch, D. & Medoff, J. (1994). “Gazelles.” In Lewis C. Solomon and Alec R.
Levenson (eds.), Labor Markets, Employment Policy, and Job Creation. Boulder: Westview Press.
• Birch, D. L., Haggerty, A., & Parsons, W. (1995). Who’s creating jobs? Boston: Cognetics Inc.
• Brown, J. S. & Duguid, P. (2001). Creativity versus structure: A useful tension. MIT
Sloan Management Review, 42, 93–94.
• Burns, P. & Harrison, J. (1996). “Growth”, in P. Burns and J. Dewurst (eds.), Small
Business and Entrepreneur ship (2nd ed.). Basingstoke: Macmillan Press.
• Cassia, L., Cogliati, G., Minola, T. & Paleari S. (2016, forthcoming). ‘Are hyper-growth firms inherently different? Preliminary evidence from a sample of fast-grown European SMEs. International Journal of Entrepreneurial Venturing
• Coad, A (2009). The Growth of Firms: A Survey of Theories and Empirical Evidence. Cheltenham: Edward Elgar.
• Cooper, D. R. & Schindler, P. S. (2008). Business Research Methods (10th ed.). Boston: McGraw-Hill.
• Denzin, N. K., & Lincoln, Y. S. (1994). Handbook of qualitative research. Thousand Oaks & London: Sage.
• Dey, I. (1993) Qualitative Data Analysis. London: Routledge.
• Eisenhardt, K. (1989). Building Theories from Case Study Research. The Academy of
Management Review, 14(4), 532-550.
• Eisenhardt, K. M., & Martin, J. A. (2000). Dynamic capabilities: What are they?
Strategic Management Journal, 21, 1105-1121.
• Gilbert, C. G. (2005). Unbundling the structure of inertia: Resource versus routine rigidity. Academy of Management Journal, 48, 741-763.
• Glaser, B., & Strauss, A. (1967). The discovery of grounded theory: Strategies of
qualitative research. London: Wiedenfeld and Nicholson.
• Grundström, C., Sjöström, R., Uddenberg, A. & Öhrwall Rönnbäck, A. (2012). Fast-growing SME’s and the role of innovation. International Journal of Innovation
Management, 16 3), 1-19.
• Guba, E. G. & Lincoln, Y. S. (1994). “Competing paradigms in qualitative research”, in Denzin, N. K. and Lincoln, Y. S. (eds.). Handbook of Qualitative Research, Thousand Oaks: Sage.
• Healy, M. & Perry, C. (2000). Comprehensive criteria to judge validity and reliability of qualitative research within the realist paradigm, Qualitative Marker Research: An
International Journal, 3(3), 118-126.
• Jarzabkowski, P., Le, J. K., & Van de Ven, A. H. (2013). Responding to competing strategic demands: How organizing, belonging, and performing paradoxes co-evolve.
Strategic Organization, 11, 245–280.
• Jay, J. (2013). Navigating paradox as a mechanism of change and innovation in hybrid organizations. Academy of Management Journal, 56, 137–159.
• King, N. (2004). ‘Using interviews in qualitative research’, in Cassell, C. & Symon, G. (eds). Essential Guide to Qualitative Methods in Organizational Research. London: Sage.
• Laforet, S (2010). Managing Brands - A Contemporary Perspective. Maidenhead: McGraw-Hill.
• Lavie, D., Stettner, U. & Tushman, M. L. (2010). Exploration and Exploitation within and across organizations. Academy of Management Annals, 4, 109-155.
• Lawrence, P. R. & Lorsch, J. W. (1967). Differentiation and integration in complex organizations. Administrative Science Quarterly, 12(1), 1–47.
• Lee, A. S. (2003). Generalizing generalizability in information systems research.
Information Systems Research, 14, 221-243.
• Lewis, M. W. (2000). Exploring paradox: Toward a more comprehensive guide.
Academy of Management Review, 25, 760-776.
• Lewis, M. W., Smith, W. K. (2014). Paradox as a metatheoretical perspective: sharpening the focus and widening the scope. The Journal of Applied Behavioral
• Lopez-Garcia, P. & Puente, S. (2012). What makes a high-growth firm? A dynamic probit analysis using Spanish firm-level data. Small Business Economics, 39(4), 1029-1041.
• Lubatkin, M. H., Simsek, Z., & Veiga, J. F. (2006). Ambidexterity and Performance in Small-to Medium-Sized Firms: The Pivotal Role of Top Management Team Behavioral Integration. Journal of Management, 32(5), 646-672.
• March, J.G. (1991). Exploration and exploitation in organizational learning.
Organization Science, 2, 71-87.
• Martini, A., Laugen, B. T., Gastaldi, L., & Corso, M. (2013). Continuous innovation: towards a paradoxical, ambidextrous combination of exploration and exploitation.
International Journal of Technology Management, 61(1): 1-22.
• Miron-Spektor, E., Gino, F., & Argote, L. (2011). Paradoxical frames and creative sparks: Enhancing individual creativity through conflict and integration.
Organizational Behavior and Human Decision Processes, 216, 216-240.
• Nosella, A., Cantarello, S., & Filippini, R. (2012). The intellectual structure of organizational ambidexterity: A bibliographic investigation into the state of the art.
Strategic Organization, 10(4): 450-465.
• O'Reilly, C. A. & Tushman, M. L. (2011). Organizational ambidexterity in action: How managers explore and exploit. California Management Review, 53(4): 5-22. • Papachroni, A., Heracleous, L. & Paroutis, S. (2015). Organizational ambidexterity
through the lens of paradox theory: building a novel research agenda. The Journal of
Applied Behavioral Science, 51(1), 71-93.
• Papachroni, A., Heracleous, L. & Paroutis, S. (2016). In pursuit of ambidexterity: Managerial reactions to innovation-efficiency tensions. Human relations, 69(9), 1791-1822.
• Patton, M. Q. (1990). Qualitative Evaluation and Research Methods. Newbury Park: Sage.
• Patton, M. Q. (1999). Enhancing the quality and credibility of qualitative analysis.
Health Services Research, 34(5), 1189-1208.
• Patton, M. Q. (2002). Qualitative Research and Evaluation Methods. Newbury Park: Sage.
• Perry, C, Riege, A. & L. Brown (1999). Realism’s role among scientific paradigms in marketing research, Irish Management Journal, 12(2), 16-23.
• Puranam, P., Singh, H. & Zollo, M. (2006). Organizing for innovation: Managing the coordination-autonomy dilemma in technology acquisitions. Academy of Management
Journal, 49, 263–280.
• Raisch, S., Birkinshaw, J., Probst, G., & Tushman, M. L. (2009). Organizational ambidexterity: Balancing exploitation and exploration for sustained performance.
Organization Science, 20(4): 685-695.
• Reeves, M. & Harnoss, J. (2015). Don’t Let Your Company Get Trapped By Success.
Harvard Business Review. Retrieved from: https://hbr.org/2015/11/dont-let-your-company-get-trapped-by-success.
• Rothbauer, P. M. (2008). ‘Triangulation’, In Given, L. M. (eds.). The SAGE
Encyclopedia of Qualitative Research Methods. Thousand Oaks: Sage.
• Saunders, M., Lewis, P. & Thornhill, A. (2009). Research Methods for Business
Students (5th ed.). Prentice Hall: Financial Times.
• Silverman, D. (2005). Doing qualitative research. London: Sage.
• Sims, M. A., & O’Regan, N. (2006). In search of gazelles using a research DNA model. Technovation, 26(8), 943-954.
• Smith, K., & Berg, D. (1987). Paradoxes of group life. San Francisco: Josey-Bass. • Smith, W. (2014). Dynamic decision making: A model of senior leaders managing
strategy paradoxes. Academy of Management Journal, 57, 1592-1623
• Smith, W. K. & Tushman, M. L. (2005). Managing strategic contradictions: a top management model for managing innovation streams. Organization Science 16(5), 522-536.
• Smith, W. K., & Lewis, M. W. (2011). Toward a Theory of Paradox: a Dynamic Equilibrium Model of Organizing. Academy of Management Review, 36(2), 381–403. • Sobh, R. & Perry, C. (2006). Research design and data analysis in realism research.
European Journal of Marketing 40(11), 1194-1209.
• St-Jean, E., Julien, P. A. & Audet, J. (2008). Factors associated with growth changes in “gazelles”. Journal of Enterprising Culture, 16(2), 161-188.
• St-Pierre, J, Julien, P. & Morin, M. (2010). The effect of age and size on SME
financial and economic performance. Journal of Small Business & Entrepreneurship,
23(2), 287–306.
• Strauss, A. & Corbin, J. (2008) Basics of Qualitative Research (3rd ed.). Thousand Oaks: Sage.
• Taylor, A. & Greve, H. R. (2006). Superman or the Fantastic Four? Knowledge combination and experience in innovative teams. Academy of Management Journal,
49, 723–740.
• Thomas, G. (2011). A Typology for the Case Study in Social Science Following a Review of Definition, Discourse, and Structure. Qualitative Inquiry, 17(6), 511-521. • Tushman, M. L, Smith, W. K., & Binns, A. (2011). The ambidextrous CEO. Harvard
business review, 89(6), 74-80.
• Vince, R., & Broussine, M. (1996). Paradox, defense and attachment: Accessing and working with emotions and relations underlying organizational change. Organization
Studies, 17, 1–21.
• Yin, R. (1994). Case study research: Design and Methods. London: Sage.
• Yin, R. (2003). Case Study Research: Design and Methods (3rd ed.). Thousand Oaks: Sage.
• Yin, R. (2009). Case Study Research: Design and Methods, (4th ed.) Thousand Oaks: Sage.
• Zimmerman, A., Raisch, S. & Birkinshaw, J. (2015). How is ambidexterity initiated? The emergent charter definition process. Organization Science, 26(4), 119-1139
