1.1.1 – Rethinking the strategy with the BSC

6Focusing on the role of the BSC in strategic processes leads to the study of the complex relationships that link control and strategy and form the basis of strategic management control. The historical case studies of Chandler (1962) and, in particular, his descriptions of the implementation of a «decentralization with coordinated control» at Dupont de Nemours and General Motors in the early twentieth century, have revealed that control is part of an organizational structure which, itself, stems from a strategy. The founding works of Anthony (1965) situate management control between strategic planning and execution control. Management control can then be defined as «a process by which managers ensure that resources are obtained and used effectively and efficiently in achieving organizational goals» (Anthony, 1965). Later, Anthony (1988) will deepen the strategic dimension of management control defining it as «a process by which managers influence other members of the organization to implement the strategies». Since then, this question of the interconnections between control and strategy has been studied and discussed. Without doubt this is because of its complexity and the many ambiguities it allows (see Bouquin 2000, Denis 2003). For Bouquin (2000), research has explored new dimensions of the relationship between control and strategy and, through this enrichment, they discovered the complexity of the issue. As a result of the limitations in Anthony’s typology, and based notably on the work of Simons (1995), Bouquin then sketches an interesting hypothesis:

7

“In the face of ‘introverted’ reactive management control, (Bouquin, 1986), of surveillance, the best known and unfortunately the most taught, there exists a proactive, extroverted management control, which is a key part of the strategic emergence process “.

8In the same perspective, Denis (2003) observes that it was from criticisms of the traditional approach to strategic management control that a renewed thought of strategic control developed, resulting in four different visions of roles played by control as part of its articulation with the strategy (Denis, 2003, p. 3). Four areas of questioning thus question the role of control in the deployment of the strategy, the alignment of the variables in coherence with the strategy to focus the strategies, the management of the changes and, the exercise of a managerial vigilance on the merits of the strategy. Denis then conforms with several authors (Schreyögg and Steinmann, 1987, Prebble, 1992) showing a «critical» approach with regard to strategic control:

9

“These authors believe that the primary function of strategic control is to submit the assumptions, objectives and plans of the organization to continual criticism and challenge. It is therefore necessary to develop a capacity to conduct double-loop learnings and continually test the relevance of the ‘staging’ in effect within the organization.

10These evolutions of the conceptions of the control of strategic management refer more fundamentally to a change of perspective in the representations of the strategy. Several studies (Mintzberg, 1994, Mintzberg and al., 1998) describe two visions of strategy: the strategy considered as a plan; the strategy envisioned as a model—a theory—of action or, in other words, as a hypothesis. In this latter approach, the strategy is no longer seen as an abstract black box and reified. Far from being a priori, it emerges at the end of a process of collective deliberation:

11

“The strategy is a set of hypotheses about causes and their effects. The BSC needs to clarify the relations or assumptions between the objectives (and the indicators) in several areas so that these objectives are validated and used to guide the actions”.

12The strategy map of the BSC then appears as an essential factor in the construction of the strategy. Although observing business practices reveals that the BSC is not always accompanied by a strategy map (Othman, 2006; Bukh and Malmi, 2005), in the Kaplan and Norton model, however, it appears to be consubstantial with the BSC. If, in the first version of the BSC, especially in the initial presentation given to us by Kaplan and Norton (1992) in their first article published in the Harvard Business Review, the BSC appears at first glance as a strategic dashboard structured in four axes. In subsequent publications (Kaplan and Norton, 1996a, 1996b, 2001a, 2001b, 2004, 2006, etc.), the strategy map will rapidly take centre stage in the model, to the point where the two initiators of the BSC will devote an article to it (Kaplan and Norton, 2000) and a book (Kaplan and Norton, 2004). As noted by several authors (Cobbold and Lawrie, 2003; Trebucq, 2011; Choffel and Meyssonnier, 2005), the BSC thus experienced several stages in its evolution to the point of going beyond the stage of a simple tool of control of management, to be established, at least in the publications of its designers, like a strategic management device and this, thanks in particular to its strategy map.

13The «cause-effect» model which structures the stratey map raises however several questions regarding its fundamentals. Researchers criticize Kaplan and Norton for the lack of clarity on the meaning they attach to the concept of causality (Nørreklit, 2000). However, Bukh and Malmi (2005) remind us that, according to Kaplan and Norton, the strategy map is primarily a hypothetical schema and not a model of relationships established. The strategy, as captured in the BSC «cause-effect» model, is more conjectural than generic. In fact, it represents the assumptions made by the top-management on «the best thing to do». Presented in this way, it is more in the managerial virtues of the BSC and in its capacity to create meaning for a collective of actors that we must look for its major contribution and not in any aptitude to verify causal relations a posteriori. As a model of representations and hypotheses about a strategy that does not appear as «engraved in stone», the contributions of the BSC lie more in its potentialities of double-loop learning (Argyris and Schön, 1978), than in simple correction loops with respect to pre-established objectives. Thus, as a model of hypotheses, the underlying representation of the strategy, as it is contained in the strategy map, refers to the theories of action developed by Argyris and Schön (1978) in their model of learning. It reflects guiding values and basic paradigms that can be constantly questioned. As noted by Kaplan and Norton (1996a), the BSC allows simultaneous learning in simple loop if the strategy is applied as intended and double-loop learning when determining whether the strategy remains viable and competitive. The BSC thus makes it possible to confirm or refute the assumptions made at the time of the formulation of the initial strategy (Kaplan and Norton, 1996a).

14Through its methodology, the BSC offers the opportunity for collective reflection on elements such as vision, mission, values or positioning. It thus provides a favourable framework for a process of collective construction of the strategy. In the same perspective, the collective conception of the strategy map and the causal relationships that structure it, the mechanisms of deliberation on the definition of objectives and the choice of indicators, can allow a collective of actors to interact and to gather together around a shared model of performance. The intersubjective communication process that emerges from these interactions can produce shared representations and stimulate organizational learning. The BSC can therefore be likened to a boundary object (Star and Griesemer, 1989) around which actors belonging to several social worlds (directors, functional managers, operational managers, etc.) interact, share representations and engage collectively in the construction of the strategy and a common performance model. By studying the deployment of the BSC in four entities of the same group, Hansen and Mouritsen (2005) show that the BSC is a boundary object in that it is sufficiently plastic to respond to specific contexts and organizational issues specific to each of the four cases, while maintaining its identity across different contexts of use. The authors reveal in particular that the BSC does not present itself as a predefined and normative model, but that the nature of the organizational problems specific to each studied unit explains the idiosyncratic nature of its use and the representations constructed by the actors. In the studied group, the strategic management control is the result of a learning process in which the BSC is an instrument for reformulating the strategy. This is no longer given a priori or simply admitted as a «black box» (Hansen and Mouritsen, 2005), but it is part of a reformulation by a group of actors. However, for such a process of collective construction of the strategy to work, the leaders must create favourable conditions. In particular, they must involve an enlarged group of subordinates in the experience of the BSC by stimulating dialogue and debate during the different phases of the design and deployment of the device. This interactive approach of the BSC can therefore be presented as a lever of organizational learning.

1.1.2 – The BSC, learning lever

15Organizational learning translates into collective learning and changes behaviour within the organization (Huber, 1991). In general, the issue of organizational learning refers to the acquisition processes and transformation of knowledge in the organization. It should be noted that these processes can occur at a triple, individual, group or organizational level. It is at the level of the organization in its overall strategic dimension that interests us in this study. Thus for Crossan and al. (1999, p. 522), organizational learning can be conceived as an essential means for achieving a strategic renewal of a business. This is thus how Kaplan and Norton (1998), see things when they use the Argyris and Schön model (1978) and are interested in the role of the BSC in double-loop learning. For Argyris and Schön (1978), organizational learning is the result of a process by which actors detect errors and act to correct them. It can be divided into two levels: adaptive learning in a single loop that produces minor changes in behaviour and does not result in significant change in values; the double-loop learning or generative learning that leads the organization to change the paradigm, which amounts to drastically revising the «theory of action» reflecting its guiding values and its founding paradigms. Figure 1 shows the model of learning defined by Argyris and Schön (1978) by locating the role that the BSC can play in the learning processes in single-loop and double-loop.

Fig. 1

The role of the BSC in organizational learning processes

The role of the BSC in organizational learning processes

16As defined by Kaplan and Norton (2001a), the strategy can be considered as a model of hypotheses on cause-and-effect relationships. This translates into a strategy map which, based on the Argyris and Schön (1978) model, can be considered the organization’s «action theory». This «theory of action» refers to the basic paradigms and guiding values on which «action strategies» are based. A primary learning in simple loop, consists of when comparing goals and targets to results. In case of deviation, it is then necessary to make the necessary corrections. In the words of Kaplan and Norton (2001a), this involves achieving strategic alignment and «translating the strategy in actions» (Kaplan and Norton, 2001a, 2006). But a second type of learning, in «double-loop», consists of going back to the premises of the strategy (Schreyögg and Steinmann, 1987) by questioning the strategy map, that is to say the guiding values that form the basis of the «strategies in action» on which the objectives and targets of the BSC are defined. This would be more of a learning process of (re-) thinking strategy (strategic thinking) than a strategic alignment process. The way to consider the measurement or diagnosis of results, whether it be a purely cybernetic approach in the context of a diagnostic control or an interactive approach, seems essential here.

17As Kloot (1997) points out,

18

“The use of control devices such as feedback loops and budget discrepancies is designed to detect operational problems and the information provided by these control systems is used within common operational paradigms to correct operational problems. Hypotheses and policies are usually not questioned, resulting in simple loop or adaptive learning “.

19In contrast,

20

“Management control systems can also play a critical role in detecting and solving problems caused by environmental change, resulting in paradigmatic change (double-loop learning or generative learning)“.

21In particular, the transition from adaptive learning to generative learning requires a collective effort within the organization that can be stimulated by interactive control levers in the sense of Simons (1995). Thus, the BSC mobilized in an interactive approach, could foster generative learning when the strategy map is the product of interactive exchanges between members of the organization. Under these circumstances, learning can be described as organizational insofar as the construction process of the BSC leads to the transfer of knowledge and information between individuals or groups of individuals. It is therefore necessary to construct a shared representation of the strategy and the underlying assumptions that are at the foundation of its performance model.

22In this generative process, individual learning is no longer sufficient. It is therefore necessary to better understand how teams learn and how to create structures and networks to share learning experiences within organizations (Marquardt and Reynolds, 1994). Several authors describe organizational learning through four sets of constructs and processes (Levitt and March 1988, Huber 1991, Kloot 1997): the acquisition and distribution of knowledge in which information from multiple sources is shared, leading to new information and understanding (Huber, 1991, p. 90); interpretation, defined as «the process by which information takes on meaning or as the process of translating events and developing an understanding and shared conceptual schemas» (Daft and Weick, 1984, p. 294-296); organizational memory, the process by which knowledge is stored for future use, involving the accumulation and maintenance of organizational experiences (Kloot, 1997, p. 57). Huber (1991) details these constructs and the sub-processes by which they proceed (Fig. 2).

23Huber’s (1991) model captures the knowledge production and transformation processes that underlie organizational learning. In the same perspective, several conceptual frameworks have come to illuminate these learning processes. The framework proposed by Crossan and al. (1999), or 4I’s model, distinguishes four processes: intuition, interpretation, integration and institutionalization. But it is mainly the knowledge conversion model (Nonaka, 1994; Nonaka and Takeuchi, 1997) which provides a particularly enlightening vision for knowledge management. At the heart of the model, the interrelationships between tacit knowledge and explicit knowledge over interactions between individuals are crucial for the formation of new knowledge (Nonaka, 1994, p. 15). The conversion of knowledge proceeds from four dynamic processes: socialization, externalization, combination and internalization. If the knowledge conversion model thus offers particularly fertile research perspectives for management control (Bollecker, 2002), we deliberately delimited our conceptual framework to the double-loop learning model developed by Argyris and Schön (1978), enriched and deepened by the notion of generative learning developed in the work of Huber (1991) and Kloot (1997). At the heart of the research programme that guided this longitudinal study, we wanted to delve deeper into a very interesting question, advanced but nevertheless little documented by Kaplan and Norton: that of the role of the BSC in the process of learning in double-loop. That is why we focused on this concept. Huber’s (1991) model allowed us to enter the «black box» of double-loop learning to observe in a more operational way the processes of acquisition, distribution, interpretation and memorization of knowledge, related to the design of a BSC.

24The design of the BSC, using an interactive approach can thus favour acquisition and distribution processes. This could involve, in particular, experimental learning sub-processes, in which the actors of the organization analyze return on cause-effect relationships formulated as assumptions in the strategy map. As noted by Huber (1991, p. 91), «an approach to fostering organizational learning is to increase the accuracy of feedback on cause and effect relationships between actions and results». Another aims to ensure the collection and analysis of these feedbacks. Huber (1991, p. 92) notes that a self-appreciation based on the interaction and participation of the members of the organization can also foster the acquisition of knowledge and thus organizational learning. He also emphasizes the importance of performance monitoring. The BSC may also generate interpretive processes, particularly in the context of construction of the strategy during which the actors develop a collective representation around a shared model of a strategy map and the definition of the constructs, giving rise to measures of performance. Finally, the strategy map of the BSC, could therefore constitute the cognitive map and the conceptual framework on which a shared performance model is based. In doing so, they build meaning, develop a translation process and produce a shared understanding and common schema conceptualizations (see Daft and Weick, 1984, pp. 294-296, cited by Huber, 1991, p. 102). For Kaplan and Norton (1996a), the BSC then appears as a representation of the shared vision of the strategy. The design of the BSC and, in particular, the collective formalization of the strategy map and its materialization in the form of objectives and measures, such as the recurrent use of the BSC during interactive piloting, would allow the construction of an organizational memory. As specified by Simons (1995), interactive control processes are by definition collective and can thus foster the emergence of generative learning stimulated by dialogue and debate. But, if the use of the BSC as an interactive piloting tool has already been subject to analyses and descriptions, a minority of research is focused on the learning processes that emerged during the design phase of the BSC. So it is precisely to this privileged moment of building meaning and collective learning that our research pays special attention. For Kloot (1997), the learning processes generated during the design phase of management control systems are crucial to accompany change and ensure the survival of the organization. The observation in our two case studies of the abandonment of the BSC a few years after its conception for example, provided us with an opportunity to observe to what extent the learnings developed during the design of the BSC could subsist and incarnate in practices and the implementation of new tools.

1.2.1 – A processual approach based on intervention research

26Studying the role of the BSC in strategic and learning processes requires a dynamic observation approach. The mobilized methodology is part of a processual approach consisting of describing, analyzing and explaining the what, the why and the how a sequence of individual actions, all based on the assumption that social reality is not a stable condition, but instead is a dynamic process (Pettigrew, 1997). We adopt a processual approach in order to understand the dynamic interrelations between control (controlling) processes and strategic (strategizing) processes (Chapman 2005, Jørgensen and Messner 2009). For researchers who fit into this movement (Hinings, 1997; Pettigrew, 1997), qualitative methodologies and longitudinal case studies seem particularly appropriate. In this perspective, we adopt a methodology in which researchers interact with their field, within the framework of the transformative intervention research. While this type of approach has long been relatively rare in management control research (Kasanen and al., 1993), it seems that this has evolved today with the development of action research such as «innovation action research» (Kaplan, 1998) or «constructive research approach» (Kasanen and al., 1993, Labro and Tuomela, 2003). Our research is particularly relevant to innovation action research (IAR) which consists, according to Kaplan (1998), of developing a methodology for action research involving the participation of researchers in managerial innovations through their field experimentation, which should lead to the development of new theories and practices. Kaplan (1998) then refers to his work with Norton on the BSC. In the IAR, researchers develop and refine the theory of a new managerial practice that is expected to be broadly applicable to a wide variety of organizations (Labro and Tuomela, 2003). Through our study, we wish to develop our understanding of a pre- existing management control tool—the BSC—to study its role in organizational learning processes. Doing so, our research is part of a project of knowledge little explored by Kaplan and Norton. It is a step in a longitudinal research programme carried out in two industrial enterprises over a period of about 12 years for one company and 8 years for the other. The research program aimed to provide answers to three sets of complementary questions: the role of the BSC in the design of the strategy; the control levers mobilized in the design and use of the BSC and, more specifically, the modalities and effects of an interactive BSC approach; the role of the BSC in organizational learning processes. Several publications have been produced on the first two themes. This article more particularly develops the third.

1.2.2 – Longitudinal study of two cases of industrial companies

27We conducted a longitudinal study on two cases of companies in which we initially conducted an intervention research. These two companies were selected for three reasons: firstly, their industrial activity, which allowed identifying the physical processes of value creation. Secondly, these two companies are of intermediate size which allowed us to grasp the strategic and operational levels. Thirdly, these production units were chosen because of the strong involvement of their management teams in the research project. Very interested in the integrated and multidimensional nature of the BSC, the leaders of the two companies were willing to commit the necessary means to the implementation of the device. For reasons of confidentiality, the two cases have been renamed AMIDON and MECATRONIC. AMIDON, which employs 200 people, is a profit centre of an international group in the carton packaging sector; MECATRONIC is a SME independent of the sector of precision mechanics. It has 130 employees.

28Both companies had similar aspects in many respects. In both companies, a participative management style was adopted with the intention of involving a large group of employees in the management of the organization. The CEOs of both companies also seemed to be driven by an entrepreneurial vision and eager to develop strategic thinking around a collective project. All the conditions seemed to be there to allow us to observe how the design of a BSC can play a role in forming a strategy. In both cases, the experiment responded to the need for a new joint project able to (re) mobilize a collective.

29The research method took the form of an active participation of researchers in the development of the BSC in close cooperation with the management teams extended to a set of collaborators at intermediate and operational levels. At every stage, interviews were used to collect stakeholder representations and regular observations were made each year.

30However, once the BSC reached an operational stage, the two companies did not wish to continue the intervention research experiment. At AMIDON, following its absorption by a competing group, the Chief Executive Officer left the company and his replacement did not wish to extend the experience, especially because the device was not part of the management tools of the group. At MECATRONIC, the company has suffered a severe crisis and a large part of the direction has been renewed. In both cases, although the research intervention experiment was terminated after the design phase of the BSC, however, the companies accepted that the researchers continue their research work as observers. If it was observed that the BSC had not passed the stage of its routine implementation in the form of a steering system, this observation established in the two companies, reinforced us in the idea that it was interesting to deepen our analyses and to continue our observations in order to study the learnings that could remain after the abandonment of the BSC.

31The AMIDON study was conducted over a 12-year period and the MECATRONIC study was conducted over a period of 8 years. The investigation method is part of a longitudinal approach structured in two phases: a phase of immersion and intervention; a phase of observation and distancing.

A phase of observation and distancing

33After completion of the construction phase of the BSC (2003 at AMIDON, 2006 at MECATRONIC), semi-structured interviews were conducted once a year in each of the two companies in order to study the practices adopted in the use of the BSC.

34It was particularly important to observe how it was used as a diagnostic or interactive control lever. Most importantly, we wanted to understand its role in strategic processes and learning mechanisms. So, at AMIDON, once a year between 2004 and 2007 we interviewed several actors: the General Manager, the Management Controller, the Sales Manager and the Production Manager. At MECATRONIC, between 2007 and 2009, we interviewed the General Manager and the Director of Sales and Purchasing (Business Manager). Then, between 2011 and 2014, we interviewed the Management Controller, the Quality Manager and a Workshop Supervisor at AMIDON, while at MECATRONIC the General Manager and his second in command, the Business Manager were interviewed. The choice of the various interviewees is explained by the fact that over the period, the acquisition of AMIDON by a competing group and the restructuring of MECATRONIC following significant economic difficulties, led to a strong turnover among the Managers in place. Most of the management teams were for the greater part renewed in both companies.

35In the appendix, the frame of the interview guides used during the two phases of the research (immersion and intervention, observation and distancing) is presented. We proceeded with a manual coding in two stages: an open coding (Strauss and Corbin, 1998) at first. This allowed us to highlight more key themes related to the different stakeholders’ view of the contributions and limitations of the BSC experience, the lessons they gained from it, the different stages of the process learning and knowledge, etc. Secondly, we carried out an axial coding (Strauss and Corbin, 1998), which enabled us to reveal elements of response with respect to our questions of research by intertwining the different themes highlighted. In addition to the various semi-structured interviews conducted with managers of both companies, it should also be noted that these two entities recruit trainees or apprentices with a Master’s Degree in Management Control each year. Researchers associated with the project, members of the Master’s teaching team, have access to continuous monitoring of the two organizations, on the occasion of learning visits, thesis supervision and defense. Tables 1 and 2 summarize the list of meetings and interviews conducted at AMIDON and MECATRONIC.

Table 1

List of meetings and interviews carried out at AMIDON

Research phases Content of discussions in chronological order Functions of questioned people Nature of interviews and meetings Time taken for interviews or discussions (in hours) 1st phase : Immersion and intervention (between 2001 and 2003) Presentation of the BSC tool and the company - Direction Committee Collective interviews 3 h - Direction Committee enlarged to include operational chiefs 4 h - Direction Committee 2 h Diagnosis of the current performance management system - Direction Committee Collective interviews 1 h - Commercial function 1 h 30 - Production function 1h 30 - Human Resources function 1 h 30 Discussion and formalisation of the Strategy - Direction Committee Collective interviews 2 meetings of 2 h each Conception of the strategy map - Direction Committee 3 h Construction of the BSC - Direction Committee 3 h 2nd phase : Observation and detachment (between 2004 and 2014) Tracking the use of the BSC - Membres of the Direction Committee Individual interviews 2 h per person and per year between 2004 et 2007 What remained of the BSC several years after - Management Controller, - Director of Production, - Person in charge of Quality - Operational Manager Individual interviews 2 h per person and per year in July 2011, 2012, 2013 and 2014

List of meetings and interviews carried out at AMIDON

Table 2

List of meetings and interviews carried out at MECATRONIC

Research phases Content of discussions in chronological order Functions of the questioned people Nature of the interviews or meetings Duration of interviews or meetings 1st phase: Immersion and intervention (between 2005 and 2006) Présentation of the company and the strategy - Managing Director and Business Manager Collective interviews 2 h Présentation and discussion concerning the BSC tool - Executive Committee expanded to operational managers 2 meetings of 6 h each Definition of the strategy - Executive Committee expanded to operational managers 3 h Conception of the strategy map - Executive Committee expanded to operational managers 2 meetings of 3 h each Reflexion on the performance indicators - Executive Committee expanded to operational managers 6 h Construction of the BSC - Executive Committee expanded to operational managers 3 h Feedback from experience 3 heads of line Semi-directive individual Interviews 3 h 2nd phase: Observation and distancing (between 2004 and 2014) Tracking the use of the BSC Directeur général et directeur commercial et des achats Collective interview and individual interviews 2 hours per year between 2007 et 2009 What remained of the BSC several years after New Managing Director (ex- Commercial and Purchase Director Individual interview 2 hours per year in July 2011, 2012, 2013 and 2014

List of meetings and interviews carried out at MECATRONIC

2.1.1 – The design of the BSC at AMIDON

37The organizational context facilitated the implementation of the BSC project within the AMIDON plant. The commitment of the site management was paramount to the BSC design project. From the start of the project it was open to all forms of innovation and led to an inter-service dialogue. For the site director, this management approach should represent «an investment for the future», while specifying: «the return on investment not being necessarily measurable in the short term». The management found it necessary to communicate to facilitate the implementation of the project but also to make it survive. At the start of the project, the plant had an environment conducive to the integration of projects and this infatuation has undoubtedly played a large role in the emergence of a collective learning approach. All the people interviewed were very sensitive to the arguments generally accepted in favour of the BSC: the BSC is not only interested in financial indicators. It is also interested in more qualitative indicators such as customer or staff satisfaction. The BSC is therefore based on a global approach to business, a multidimensional performance; the BSC is part of a long-term value of the company by including indicators focusing on intangible aspects such as innovation and learning.

38The members of the BSC project started with one observation: on the one hand, all the dashboards used by the plant before the BSC project combined only result indicators that favoured the financial aspect; on the other hand, these indicators of delayed performance only allowed us to apprehend the effects of actions already undertaken by taking stock of past performance. But the leaders were getting the results they were subjected to.

39With the BSC project, management wanted to gain a more holistic view of performance and not be content with financial data. Several meetings organized by working groups, involving the management team and several process managers, made it possible to identify key success factors to access the strategic vision. Their identification was completed with the design of the strategy map. The work was based in particular on a mapping of the processes established in the framework of certification. Each process resulted in a collective definition of the missions, conditions of success, action variables, indicators and finally a process dashboard. The working group then asked about the upstream factors that influenced outcome indicators. This led to the creation of a strategy map based on the four axes advocated by Kaplan and Norton. Its design then made it possible to highlight the key elements of management of the AMIDON site and to create a dynamic of continuous improvement of the process of value creation. Figures 3 and 4 describe the strategy map and the indicators selected in the AMIDON BSC.

Fig 3

The strategy map of the BSC at AMIDON

The strategy map of the BSC at AMIDON

Fig 4

AMIDON’s BSC

AMIDON’s BSC

40The BSC approach, which has since been geared towards a long-term promotion of the company, has made it possible to clarify the strategy which was then turned towards customer satisfaction and to translate it into concrete actions that everyone can understand. Thus, the local objectives of managers were coordinated with the strategic objectives defined by management. All processes have been questioned in a similar way. Finally, existing dashboards have been improved and other control systems have been introduced.

41We must emphasize here the importance of the interactive dimension of the project which is above all a collective work, “a human adventure” according to the director of the site:

42

“The project demands a desire for transparency and cooperation; It improves internal communication, creates a dialogue between previously compartmentalized services and reduces the phenomenon of information retention. There is indeed strong territorial logic in French companies and the Gallargues factory is no exception to this rule: lack of cooperation between the different services, retention of information to continue to preserve some power. The BSC breaks down these barriers and serves as an instrument of mediation between the actors: the interservices meetings give the possibility to each department to expose the constraints that it faces. The debates give them an opportunity to work together toward a common goal: the performance of the site.”

43The discussions, however, highlighted contradictions in terms of steering. The interactive dimension of the approach has led to initiatives that would not have been perceived without this approach and to create value added for the client.

44

“I am asked to buy the cheapest pallets possible to optimize the cost, but the shipping department and customers ask for quality pallets to be able to move them easily.”

45The BSC project was a real success during its creation phase but it quickly ran out of steam and finally was abandoned a few years later. This experience, however, marked the management of the organization and allowed its control tools to be improved.

2.1.2 – The design of the BSC at MECATRONIC

46At MECATRONIC, the BSC project was part of a change of management perspective.

“I wanted to use the BSC as a tool for change.”

Fig. 5

The strategy map of MECATRONIC

The strategy map of MECATRONIC

2.1.3 – A design phase rich in learning

47In both companies, the collective reflection of the actors to build the strategy, formalize it with a strategy map and finally define the BSC, seems to have played a positive role in organizational learning. In particular, the BSC enabled actors from different sectors to interact around the same theme. Actors from different functions and different skill levels and representations gathered around the design of the BSC. Three of the four foundations of organizational learning (Huber, 1991) were found: knowledge acquisition, information distribution, and information interpretation.

2.2.1 – The finding of abandonment of the BSC in both companies

70In both cases, the BSC experience did not go beyond the stage of designing strategy maps and defining performance indicators. In concrete terms, during the semi-structured interviews conducted in companies during the years following the BSC experiment, the researchers were unable to observe any diagnostic or interactive use of the BSC in the form of scorecards or dashboards. It is now important to consider the reasons and modalities of this abandonment of the BSC in both case studies.

2.2.2 – Lessons on favourable factors and barriers to learning

95It is appropriate at this stage to focus on the lessons that can be learned from this abandonment of the BSC with respect to the favourable factors and barriers to learning. Among the factors that played a positive role in both companies we may note in particular the very strong involvement of managers in an interactive dialogue with their employees around strategic issues. This fits more generally into the implementation of a participative management. It confirms Huber’s analyses (1991, p. 103), which emphasize the decisive role of social interactions in the interpretation process in particular. These observations are consistent with the recommendations of Kaplan and Norton which are based on the work of Simons (1995) and consider that the use of BSC as an interactive control lever is likely to favor double-loop learning. Our observations in the AMIDON and MECATRONIC companies are however different from this analysis showing that interactions are crucial at the design stage of the BSC. In other words, this would suggest that interactive control can occur during the design phase of a BSC and not only during its phase of use as a steering instrument. But, above all, the two cases studied enlighten us regarding the obstacles to learning born from the design of the BSC. Many authors (Argyris and Schön, 1978; Huber 1991; Bollecker 2002; Seo 2003) have identified several obstacles. Bollecker (2002) points in particular to the influence of human obstacles related to limits in cognitive abilities or the existence of defensive routines. These have been particularly highlighted since their first publications by Argyris and Schön (1978). Seo (2003) takes up this element, which he classifies as one of the emotional barriers to learning. To this he adds political barriers as well as barriers related to the imperatives of management control and this, in particular, in the context of capitalism and globalized economies. The abandonment of the BSC in both case studies alerts us to the risks that are detrimental to organizational learning. At AMIDON, it is the change of direction of the management control system towards financial control that is at the origin of the abandonment of the BSC: faced with shareholder pressures, the group strengthened its reporting and, in the unit studied, it replaced the BSC. The turnover of a party of executives, starting with the Director of the establishment, could also be a barrier to learning, especially in memorizing knowledge (Huber, 1991, p. 105). At MECATRONIC, the financial crisis experienced by the company and the departure of its main executives are at the origin of the abandonment of the BSC. New priorities were then defined in the direction of dashboards which, while taking over the architecture of the BSC, have a more operational vocation.

96Finally, despite these obstacles, if in both cases the BSC was abandoned, we note that it has inspired the various management control tools that replaced it. In both companies, it seems that a capitalization of learning, through the memorization of knowledge acquired during the design of the BSC, then occurred.

2.2.3 – A memorization of the knowledge acquired during the design of the BSC

97In both organizations, despite the departure of several executives and collaborators present during the BSC experience, it seems that there was a phase of memorization of the knowledge acquired during the design of the BSC.

98As Huber (1991, p. 105) points out, the storage of knowledge can involve the memorisation and archiving of information on the one hand and the computerization of organizational memory on the other. In both cases studied, the memorisation of information as a result of the design experience of the BSC was mainly reflected in the practices and representations of the actors. For Huber (1991) indeed, much of the organizational knowledge regarding how to do things is stored as standard operating procedures, routines and scripts. Managers routinely acquire and mentally store «soft» information. This is precisely what our observations tend to show at AMIDON and MECATRONIC.

99At AMIDON, even though the BSC was abandoned, the BSC experience has left a lasting impression. The latter has, as indicated by its CFO, «left a trace» to the extent that;

100

“It has created a new working method (…) it has enriched the other management tools by a multi-business approach leading to contradictory debates with a pooling of the knowledge of each to decide on appropriate actions”.

101This is confirmed by the Production Manager:

102

“The corporate culture remains impregnated by this collective approach (…) The cost reduction imposed today by the group that controls AMIDON gives rise to the constitution of working groups that reflect and decide at the same time on the objectives and actions to be taken”.

103According to him, this participative management logic is «directly inspired by the BSC approach». In addition, according to the CFO, «this teamwork systematically takes into account the performance axes of the BSC in the construction of new steering dashboards» which allows the actors to maintain the multidimensional vision of the BSC without being constrained «by its formatting, deemed too standard and therefore too rigid».

104However, the increasing mobilization of the actors in the activities related to the reporting tends today to reduce the periods of collective exchange. For example, according to a supervisor, the progress groups were removed and «everyone is working in his corner». The Quality Manager states that she «has less and less room for maneuver and that she «spends her time reporting». This recurrent reporting practice is likely to create new learning and challenge those emanating from the BSC.

105At MECATRONIC, if the BSC did not materialize in a management and monitoring chart of the indicators, the strategy map now constitutes a frame of reference for the CEO, who uses it as an instrument of dialogue with the shareholders and with his main collaborators:

106

“The strategy map helped me a lot to open up directions and make decisions but we did not keep the formalism of the BSC (…) I always communicate around the map—it allows me to set up blockages (…) The Philosophy is still there but formalism is no longer our goal (…) Today, the strategy map helps me a lot in the answers I need to give to employees. I share it with most of them (…) We discuss the map with the main shareholder—it allows us to discuss the business plan”.

107In addition, the four axes of the BSC structure the monthly management dashboards:

108

“Today we have monthly management dashboards in which we use almost all four axes of the BSC: Finance, Sales, Quality, Human Resources”.

109Another element also seems to have to be considered as learning coming into being during the BSC experiment: the development of an interactive approach to management control, insofar as the new top management encourages a participative dialogue, especially around the management dashboards:

110

“Once a month, we make a point at a Board of Directors meeting in which the shareholder participates. I would like to do the same thing, a notch below, at the level of the line leaders”

3.2.1 – Paradox, equilibration and learning

115The case of the AMIDON company is particularly illustrative of the paradoxes that appear during the implementation of a BSC. To meet all functional expectations and those of stakeholders to meet the needs in the same way, measures have tended to be multiplied at the risk of undermining the identification of primordial values and drowning in an ocean of sometimes incomplete information. This phenomenon has resulted in a pile-up of untapped figures and information. Following the same logic, it is sometimes difficult to simultaneously manage all the selected indicators to evaluate the performance. Some goals can conflict: a policy to reduce manufacturing costs can influence product quality indicators that will in turn influence customer satisfaction; the development of quality can oppose a zero stock policy to optimize cost management. The entity may therefore be unable to maximize simultaneously all items of the dashboard. Moreover, the actors of the two companies have often expressed their reservations about a too prescriptive vision of the BSC, especially with respect to the operational dashboards. We realize that in the face of the upheavals of the organizational context, the actors encounter difficulties in adapting the BSC and making it evolve. One of the failings of the BSC could be to neglect the specificity of the company. The BSC is composed most of the time of the same axes regardless of the activity of the company, its size and its strategy, as if the dimension of the BSC became a universal instrument. We see, through the doubts expressed by the actors on the ground, that one of the major stakes of the BSC and, no doubt, its conditions of success, lies in its ability to provide an idiosyncratic representation of the organization. In other words, its design is more «tailored» than «ready to wear». This particularly suggests the conclusions of a special issue of the French Journal of Management published in 2011: in reality, the exercise of the design of a BSC should proceed more from the construction of a LEGO, than from the assembly of a PUZZLE (Naro and Travaillé, 2011).

116In addition, the BSC excludes other stakeholders, and, in particular, the environment in the dual sense of ecology, competition and technological change, from society (Atkinson and al., 1997). In their multidimensional vision, the creators of the BSC do not seem to include the external environment of the company as if this external environment had no impact on the performance of the company. However, precisely these elements of the external environment and, in particular, the taking into account a broader set of stakeholders to different and even conflicting representations and interests are sources of paradoxes. How in these conditions can a balanced model of performance be achieved? As suggested by Denis (2003), is it not precisely towards a dynamic model of equilibrium, going beyond the static vision of a «balanced» model and embracing the paradoxes of performance, which should be orientated? Denis (2003) thus proposes an ago-antagonistic representation of management control which proves extremely stimulating in that it offers very rich research perspectives on a renewed vision of the BSC. But even more so by placing the tensions and paradoxes of performance at the heart of the debates and dialogue that can be established between actors and stakeholders around the design and use of a BSC, this opens the way for us to deepen the issues of interactive control. It may be in this approach of a «paradoxical scorecard» (Naro and Travaillé, 2015), that the conditions of a permanent questioning of the early stages of the strategy can be exercised and, in doing so, of organizational learning. However, the observation of corporate practices in a context of financialization and advanced liberalism (Miller and Rose, 2008) tends to reveal the fragilities of the BSC and its learning at the risk of financial control

3.2.2 – Learning from the BSC at the risk of financial control

117Management control and, in general, accounting systems come from a social and institutional practice (Hopwood and Miller, 1994). In this, they are part of a épistémè in that they constitute a representation and a media of values and controversies of a society at some point in its history. The years 1990-2000 in which multidimensional models of performance appeared, such as the BSC, also correspond to a time when financial control geared towards the creation of shareholder value developed: on the one hand, integrated models of measurement control and performance management that are not only financial, taking into account the latest advances in strategic thinking, such as resources, knowledge, competencies and dynamic capabilities as intangible assets; on the other, performance models favouring the measure of value creation for the shareholder, notably with the rise of indicators such as the Economic Value Added (EVA) and strengthening the role given to financial reporting. The case of AMIDON illustrates this evolution. Following a change in the ownership structure of the profit centre, reporting takes precedence to the detriment of the BSC. Our study definitely shows that the design of the BSC was conceived from learnings that will be found in the tools set up by the new management, as also the interactive approach that had initiated the timing of the design of the BSC will tend to continue. But for how much longer? In its turn the practice of daily reporting is likely to generate new learning. This example illustrates the paradoxes and controversies that surround the control of management today. An article by Jensen (2010), one of the founding fathers of the agency’s theory, testifies by very explicitly criticizing the BSC, concerning the content of the debates. For Jensen (2010) the only score that must mobilize the leaders of a company is that of its long-term market value. Any other approach, including the stakeholder theory, in which he inscribes the approach of the BSC, cannot only mislead the leaders of this sole purpose, but also provide them with a motive and justification to appropriate the free cash flow of the company. The criticism is undoubtedly excessive, all the more so since the causal pattern of the strategy map of a BSC, at least in its canonical form, converges towards a shareholder axis which, precisely, can measure the long-term market value of the firm. As a prospective dashboard, as its translation in French indicates, the BSC is indeed oriented towards a performance in the long term, by taking into account key performance indicators or leading indicators. The generally accepted logic of causality is that the performance or the creation of value on the «shareholder» perspective, will be obtained thanks to a performance corresponding to the creation of value for the customer (customer perspective). Excellence in an «internal processes» perspective will then create value for customers and shareholders. Such long-term performance will depend on the company’s ability to innovate and maintain its aptitude for change, apprehended in a «learning-innovation» perspective. Thus, far from being reduced to a simple «human resources» perspective, as some examples of the BSC suggest, this last perspective accurately reflects the resources and business skills and also, its dynamic capabilities. It is undoubtedly in this fourth dimension that the most promising research perspectives lie. These latter refer us to the question of the intellectual capital of the company and, in doing so, taking intangibles into account in management control models. More importantly, they integrate organizational learning at the heart of the BSC performance model. Presented as such, the BSC presents itself as being at odds with a model of financial control focused on short-term financial performance that may be detrimental to organizational learning.