Interactional and procedural practices in managing coopetitive tensions

Purpose – The purpose of this paper is to explore interactional and procedural practices in managing tensions of coopetition (simultaneous collaboration and competition between ﬁ rms). Design/methodology/approach – Through an in-depth literature review of prior research within coopetition and strategy-as-practice ﬁ elds, and by using two illustrative empirical examples, the authors develop a framework for preventing and managing coopetitive tensions through combinations of procedural and interactional practices. Findings – The authors identify tensions related to strategizing, task and resource allocation, as well as knowledge sharing. Furthermore, they demonstrate potential ways of how these tensions can be prevented, resolved and managed. Research limitations/implications – The ﬁ ndings show that the analysis of tensions in coopetition would bene ﬁ t from a holistic, multilevel approach that recognizes practices that are interactional (i.e. face-to-face interactions) as well as procedural (i.e. organizational routines). Coopetitive tensions and their resolution are related to the use or neglect of both types of practices. Furthermore, interactional and procedural practices are mutually interdependent and can complement each other in tension management in various ways. Practical implications – The ﬁ ndings of this study shed light on the roles and activities of actual practitioners involved in coopetition, and shows how their work and practices in-use contribute to coopetition, related tensions and their resolution. Originality/value – By adopting the strategy-as-practice approach, this study generates valuable insights into the practices and tensions in coopetition, as well as illuminates the roles of the practitioners involved in managing coopetition relationships.


Introduction
This study explores interactional and procedural practices in managing coopetitive tensions. Coopetition, or the simultaneous occurrence of competition and collaboration between firms, has recently gained a lot of interest both in practice and theory. In current literature, two aspects of coopetition strategy surface continuously: first, the possibility to reap coopetition-specific benefits (de Resende et al., 2018;Gnyawali and Park, 2009;) and second, the existence of coopetition-specific tensions that need to be managed Fernandez et al., 2014;Tidström, 2014;Fernandez and Chiambaretto, 2016). Both aspects highlight the tension-laden nature of coopetition: such relationships are by nature paradoxical, consisting of two opposing logics of competition and collaboration in a single relationship (Raza-Ullah et al., 2014;Gnyawali et al., 2016;Ritala et al., 2017). In general, paradox is defined as "contradictory yet interrelated elements that exist simultaneously and persist over time" (Smith and Lewis, 2011, p. 382). The contradictory elements of paradoxas they persist over time and require attentionlead to tension between these elements (Smith and Lewis, 2011;Schad et al., 2016). Thus, given the paradoxical nature of coopetition relationships, such tensions are bound to arise (Wilhelm, 2011;Gnyawali et al., 2016).
Tensions in coopetition have been studied within the business network approach, where focus has been on roles (Bengtsson and Kock, 2000), power and dependence, as well as opportunism (Osarenkhoe, 2010). These tensions can be viewed as strategic tensions, which are important in order reap the strategic benefits of coopetition relationships. Such tensions often concern imbalances between activities and strategic goals of the companies (Ritala and Tidström, 2014;. Moreover, tensions related to division of work, or the allocation of tasks and resources are important to recognize. In coopetition, the task and resource allocation might be a source of conflict, as well as a resolution (Fernandez et al., 2014;Le Roy and Fernandez, 2015). A third type of well-documented tension recognized in coopetition literature relates to knowledge and the sharing and protecting of knowledge (Tsai, 2002;Solitander and Tidström, 2010;Ritala and Hurmelinna-Laukkanen, 2013;Fernandez and Chiambaretto, 2016). However, while the coopetition literature has recognized different types of tensions, we are still lacking an overarching perspective to how these inherently surfacing tensions can be managed.
Tension management studies in coopetition have focused on governance mechanisms (Vaaland and Håkansson, 2003), coopetition capabilities , conflict management styles (Tidström, 2014) and various types of organizing modes such as separation and integration logic (Fernandez and Chiambaretto, 2016). A majority of prior research in this context has focused on a firm-level mechanisms or relational level interaction. Furthermore, existing research has mostly viewed coopetition strategy and related tensions as given, i.e. as something the company has. There are two limitations to this approach: first, it often omits the examination of organizational actors and their social activities and, second, it views strategy primarily as a formal plan without considering how it is carried out in practice. In fact, Bengtsson and Kock (2014), Fernandez et al. (2014) and Tidström and Rajala (2016) have argued that there is a need for more research on coopetition from a multilevel perspective. Coopetition goes beyond mere interaction between firms, but it also involves employees engaged in handling the management and strategizing of coopetition (Dahl et al., 2016;Bengtsson and Kock, 2000;Walley, 2007). In this regard,  call for more research into the micro-foundations, i.e. the individuals and their activities in coopetition. Building on these suggestions, we argue that by adopting a strategy-as-practice perspective (Whittington, 2006;Jarzabkowski, 2005;Seidl and Whittington, 2014), we are able to form a richer understanding of coopetition strategy. The strategy-as-practice perspective has recently been used by coopetition scholars (Dahl et al., 2016;Tidström and Rajala, 2016). However, there is still a lack of coopetition studies focusing on tensions in coopetition from the perspective of both procedural and interactional levelsa perspective which would be useful for fully understanding the complexity of tension management in coopetition.
To address the aforementioned gaps, our study examines how procedural and interactional practices can be used for preventing and managing tensions in coopetition. We aim to identify tensions related to strategizing, task and resource allocation, and knowledge sharing, and how these can be prevented and managed. Through an in-depth literature review of prior research within coopetition and strategy-as-practice, and by using illustrative empirical examples, we develop a framework for preventing and managing coopetitive tensions through combinations of procedural and interactional practices. Our study provides both theoretical and managerial contributions, as we provide a practice perspective to how tensions can be prevented and/or managed.

Coopetition from a strategy-as-practice perspective
Recently, an increasing amount of studies have focused on individuals and their activities in coopetition Dahl et al., 2016;Kock, 2016a, 2016b;Pattinson et al., 2018;Tidström and Rajala, 2016). Thus, coopetition practitioners consequently confront a multi-dimensional, tension-laden strategic situation where they act as individuals and also as the representatives of their own organization (Bengtsson and Kock, 2000). Both individualand organizational-level activities or practices are recognized within the strategy-as-practice-approach.
Strategy-as-practice views strategy as something a firm does rather than as something it has (Whittington, 2006;Jarzabkowski, 2005). Focus is on the practices of strategy as well as on the practitioners of strategy; the human actors who are engaged in the making, shaping and executing of strategies (Whittington, 2002(Whittington, , 2006Johnson, Melin, and Whittington, 2003). A practice-based view of strategy and strategizing focuses particularly on the actions and interactions of strategy practitionersthe routines, discourses, concepts and technologiesthrough which strategy is conducted (Jarzabkowski et al., 2007;Jarzabkowski and Whittington, 2008). Practices are perceived as the shared routines of behaviors composed of the traditions, norms and procedures for thinking, acting and using "things" (Whittington, 2006, p. 619, quotes in original). In other words, practices are "the things people do to, for, or with each other insofar as they follow recognizable patterns" (Van Leeuwen, 2005, p. 286). Jarzabkowski (2005) suggests that practices can be categorized as procedural and interactional practices.
Procedural practices consist of organizational routines, such as strategic plans and budgets, and their associated meetings and committees. Procedural practices provide the externally legitimate, traceable realizations of institutionalized rules for activities, such as strategy making or coopetition (Jarzabkowski, 2005). These formal and structural practices have legitimacy and long duration because they are embedded in the social order of the organization. Procedural practices comprise the institutional backbone for coopetition activities. These organizational level practices are related to managing, organizing, decision-making and to communication between the companies. As organizational practices take a long time to develop and are difficult to change (Gherardi, 2009;Feldman, 2000), companies are often inclined to rely on their existing practices rather than to develop new practices. On an organizational level, procedural organizing and managing practices consist of systems and structures that define how resources and information are allocated and shared (Ambrosini et al., 2007). These include, for example, the overall management systems, quality systems and control, job and role descriptions, training programs, problem-solving and decisionmaking practices, IT systems and their use, and other organizational practices related to information distribution and knowledge creation, such as information coding, distribution and storing practices, meetings, workshops, committees, boards and task forces, among others. Additionally, locations for the mutual coopetition activities and the people involved in them play an important role (Van Leeuwen, 2005).
Interactional practices involve direct, purposive, face-to-face interactions, which enable the practitioners to reinforce their own interpretations of activities as well as to negotiate these interpretations with others (Jarzabkowski, 2005). All practices have an interactive dimension in some sense, and in this paper, we define interactive practices as interpersonal communicative encounters, both mediated by technologies and face-to-face, enacted by and between the companies' key people. By interactional coopetition practices we refer to the communicative practices between actors in the coopetition. These practices may entail a wide range of interactions in the form of written communications, such as letters, e-mails and text messages, as well as direct interactional encounters in the form of phone calls, one-to-one discussions, less-formal/ad hoc meetings and gatherings, dinners and other social events as well as the so-called corridor talk (Samra-Fredericks, 2003). Faceto-face interaction, in particular, incorporates the broadest range of possible resources for communication (Suchman, 2007;Jarzabkowski, 2005).
Obviously, the distinction between procedural and interactional practices is not clear-cut, as most practices entail social interaction. However, distinctive to procedural practices is that they often are institutional in nature and can be sustained by various members of the organization. Interactional practices, in turn, are dynamic communications in evolving situations in specific relationships and between specific people, and less prescribed in formality and content than procedural practices. Next, procedural and interactional practices will be elaborated in light of coopetitive tensions.

Procedural and interactional practices in managing coopetitive tensions
Strategizing Strategizing, or strategic goal setting and framing, is practices that lie at the heart of value creation in coopetition relationships. Strategy and strategizing are particularly important, as coopetition involves a specific feature of a variable-sum game. This means that some of the goals are mutually shared, while at the same time, there are also potentially conflicting private goals (Padula and Dagnino, 2007;Kim et al., 2013). This paradoxical situation involves strategic tensions, as goals and strategic objectives of competitors are seldom fully aligned in coopetition (Tidström and Åhman, 2006;Tidström, 2009;; see also Das and Teng, 2000).
Procedural strategizing practices are essential in the continuous process of negotiating and bargaining over how the pie is created, and how it is divided between the companies (Dyer et al., 2008). As coopetition involves collaboration between competitors, the strategizing process suffers from the threat of opportunism involved in the relationships, which may hinder the value-creation efforts (Hamel, 1991;Park and Russo, 1996;Quintana-García and Benavides-Velasco, 2004). In coopetition, cooperation can impede a firm's operations by enabling the competitor first to obtain sight of and then to imitate the firm's core competences (Lado et al., 1997). The opportunism threateven without being actualizeddisturbs the strategic process of coopetition in closing some opportunities that could otherwise be available. Jointly agreed strategic objectives, often stated in the form of written and verbal contracts, may help to manage such tensions. Other formal and procedural strategizing practices include, for example, strategic plans, budgets, reviews, analyses, forecasts, performance targets and the like. These kinds of formal practices are quite critical in coopetition relationships because they provide the externally legitimate, traceable realizations of the institutional rules of strategy making (Jarzabkowski, 2005, p. 51). Formal practices are explicitly articulated and documented, and in this sense open for access and evaluation by both parties. Their (ostensive) transparency and accessibility further build trust and common ground for mutually beneficial coopetition. Therefore, they are also important in the proactive management and alleviation of tensions in coopetitive relationships.
Interactional coopetition practices that relate to strategizing take place at the level of managers and decision makers involved in coopetition relationships. Managing the everyday complexities in combining the logics of competition and collaboration is a key skill for individuals operating in a coopetition context (Raza-Ullah et al., 2014;Fernandez et al., 2014;. In particular, individuals involved in coopetition relationships have to simultaneously make strategic decisions related to the achievement of their own firm's objectives as well as the objectives of the coopetition relationship in general. This phenomenonwhich is a major source of individual-level stress and tensionis known as the "dual allegiance" (Husted and Michailova, 2010;Husted et al., 2013).
In coopetition relationships, strategic activities are related firstly to setting and agreeing on (framing) mutual objectives and goals for the cooperation as well as to how to share the risks and revenues of joint operations. Second, participants need also to acknowledge which issues and activities are framed out of the coopetition relationship. Through interactive strategizing, participants can focus each other's attention in specific ways, for example by framing particular issues as "opportunities" or "risks", while maybe placing less emphasis on some other issues. Interactive strategizing may, thus, be a means of generating normative control over others through the development of dominant sets of meanings (Jarzabkowski, 2005, p. 57). Meanings are, however, not durable and need to be constantly reinforced and renegotiated. Consequently, the strategists' communicative competences and interactional skills are critical to how various strategic issues and options are viewed and perceived, and which choices are seen feasible (Samra-Fredericks, 2003).
Tensions in coopetition relationships may occur when frames become frozen (Lems et al., 2013). This may happen if the parties are not willing to re-negotiate their interpretations and understandings, and at the same time, reject the arguments and positions of the other party/parties. On an individual level, employees involved in coopetition relationships can realign the elements of coopetition strategy objectives by communicating and negotiating ways of working that combine the views of both parties. An alternative route to realignment is interaction, and activities being implemented based on the best opportunities for value creation. From a practice perspective, this has been described as "changing hats" between the strategic objectives being pursued .

Resource and task allocation
Resource and task allocationi.e. how the procedural practices are actually used to organize the activities in coopetitionis important to fully use the potential benefits of a coopetition strategy. There is some case-based evidence showing, e.g. how coopetitors allocate their resources to share risks and costs (Tidström and Åhman, 2006;Gnyawali and Park, 2011) and how they divide between different tasks in the value-creating process (Ritala and Tidström, 2014). There are also studies showing that competitors from a value-chain perspective tend to usually cooperate in activities far away from the customer and compete in activities closer to the customer (Bengtsson and Kock, 2000;Walley, 2007).
Procedural resource and task allocation practices include, for example, regular planning and decision-making activities and their related analyses and reviews. A potential tension related to the allocation of resources and tasks is an imbalance in the division of benefits and revenues between the companies. One company may perceive that it has had to offer too much in comparison with the other company/companies involved in the coopetition relationship.
In terms of resource and task allocation, there are several often-cited procedural tension management practices. One commonly mentioned solution is the separation vs integration principle (Dowling et al., 1996;Bengtsson and Kock, 2000;Fernandez et al., 2014). This means that depending on the situation, competitive and collaborative tensions may be either integrated into the same function or separated to different functions where they are handled by different people. When such tension exists at the highest levels, creating unnecessary tensions and stagnation, the separation logic should be followed. On the other hand, when competition and collaboration are more mutually reinforcing and a source of "positive tension", such integration logic can be a good solution (Lado et al., 1997). For example, Hutter et al. (2011) found that the combination of cooperative and competitive behavior turned out to be beneficial when it came to the creation of highquality and breakthrough ideas.
Furthermore, the role of third parties as a procedural tension alleviation practice has been emphasized in many coopetition studies Tidström and Hagberg-Andersson, 2012;Fernandez et al., 2014). In addition, relational governance is also needed for creating practices that ensure mutual trust. Indeed, existing case evidence demonstrates how mutual trust is helpful in opening up communication and collaboration potential in coopetition Tidström, 2014).
Interactional coopetition practices are related to the practical resource and task allocation between individuals working in the coopetition relationship interface. At best, tasks and resources are allocated in a value-creating way, whereby the competitive position of the collaborating firms is used as a facilitator, rather than impediment. However, the competitive background might create tensions if there is too much overlap in skills. This, in turn, may create conflicting interpretations over what the most suitable solutions would be. If specialists are involved in organizing the tasks, the solutions suggested by the competing firm's specialists may seem inferior, and thus not useful. On the other hand, they may be perceived superior and therefore threatening to the initial specialists. Jarzabkowski et al. (2013) note that on a micro-level, individuals struggle with conflicting demands within their own roles, and also with the conflicting demands of the roles of others with whom they interact. This situation leads to contradictory actions (Smith and Lewis, 2011). Asymmetries in capabilities and resources among the companies can also lead to unleveled opportunities to negotiate, and consequently also lead to divergent opportunities of learning (Jordan and Lowe, 2004).
These sources of potential tensions emphasize the importance of ways to collectively decide how problems in the coopetition relationship are solved. One important way is to increase the understanding of the persons that interact. Creating such "high-bandwidth" relationships may help to negotiate over the collective task-skill-set that is most beneficial to coopetition practice. On the other hand, sometimes competitive and collaborative tensions are so high, that the individuals in competitive domains should be separated from those operating in the collaborative domain (Fernandez et al., 2014). According to Bengtsson and Kock (2000), the benefit of separating cooperation and competition at the individual level relates to individuals' tendency in properly coping with just one of these relational logics at a time.
Knowledge sharing A recurring theme in coopetition studies concerns the difficulties in balancing between sharing knowledge externally and retaining relevant knowledge and resources in-house (Von Hippel, 1987;Ritala and Hurmelinna-Laukkanen, 2013;Fernandez et al., 2014;Fernandez and Chiambaretto, 2016). As the knowledge useful for collaboration can often also be used in competition (Bengtsson and Kock, 2000;Oxley and Sampson, 2004), knowledgesharing issues are a major source of coopetition tensions.
Another possible source of problems is employing excessive reservations and protectiveness regarding knowledge sharing, which may be harmful for the relationship (Hamel, 1991;Grandinetti, 2017). This gives rise to procedural decisions on what knowledge will not be shared, what knowledge is protected and with what kind of mechanisms (Ritala and Hurmelinna-Laukkanen, 2013), and how different issues are communicated between competitors (Dahl, 2014). Other important issues to be agreed upon are when knowledge is shared, who is involved, what is to be done with the knowledge (Levy et al., 2003). To enable this, concrete choices are needed on the types of employees working on collaboration interface, and which types of knowledge-sharing activities they are supposed to engage in. Further, concrete practices related to this include the utilization of different knowledge protection mechanisms that enable safe knowledge exchange between companies. These include not only formal mechanisms for utilization such as intellectual property rights, trademarks and contracts but also more informal mechanisms (Ritala and Hurmelinna-Laukkanen, 2013).
Overall, the formalization of the procedural coopetition practices serves the purpose of defining, articulating, and making visible the principles and sets of rules that are used in the coopetition relationship. However, as earlier research of practices has demonstrated, there can be a wide variation in how formal practices are carried out in practice (Feldman, 2000;Jarzabkowski, 2005). For example, a meeting can be arranged in a fixed location with all participants present, or it can be arranged as a teleconference, or even as an e-mail meeting. Ultimately, however, the greatest difference in coopetition relationships is made by the people employing these practices (Ambrosini et al., 2007).
The sharing of substance knowledge in coopetitive business relationships is a key practice in coopetition. The frequency of communication is an important factor in intra-team work, also in coopetition relationships (Ambrosini et al., 2007). Thus, in coopetition relationships both the frequency and fluency of knowledge sharing are emphasized. Knowledge sharing in coopetition requires judgment and care, so that the sharing of knowledge ensures the value creation potential. At the same time, it is paramount to keep proprietary knowledge within organizational boundaries. This is particularly critical in coopetition because knowledge transfer and acquisition are typically fast and effective due to the shared cognitive frame related to the competitors' being in the same technological and market environment . Selecting how and with whom to share knowledge requires experience and seniority from the parties. Due to its critical nature, knowledge sharing in coopetition is a constant source of interactional tensions. Harmful knowledge leakage often takes place at the employee level (Baughn et al., 1997;Hannah, 2005) and may cause negative effects for the firm, including loss of revenue, reputational damage, lower productivity and costs related to breached confidentiality agreements (Ahmad et al., 2014).
An important interactional tension management practice is, therefore, the maintenance of open communication on what issues are and are not put on the table in the coopetition relationship, as well as how proprietary knowledge is protected. According to Zineldin (1998), in a successful business relationship, individuals communicate and cooperate in an atmosphere of frank debate and trust. Indeed, several cases in a coopetition context show that individuals appreciate transparent communication about the firm's goals and agendas in their communication between rival firms Fernandez and Chiambaretto, 2016).
Interplay between procedural and interactional practices Coopetition activities and relationships are managed and enacted through both procedural and interactional practices regarding strategizing, resource and task allocation and knowledge sharing. Formal procedural practices and interactional practices are intertwined in various ways. They construct and contribute to each other both by creating opportunities and by restricting each other.
Procedural practices, such as systems, protocols, rules, norms and routines form the backbone for organizing, managing and developing activities within coopetition networks. They represent the formal, legitimate and structural aspects of coopetition. Procedural practices are central in defining the goals and aims of the mutual field of cooperation as well as agreeing upon the principles according to which issues and areas of operation are included or excluded in coopetition. A feasible set of functional procedural practices brings order and structure to coopetition relationships and activities and contributes to value-creating activities in various ways. However, poorly designed and used procedural practices may be harmful to the relationship. Procedural practices need to be agreed upon in cooperation to ensure that they cover essential areas of activities and that they allow for both the sharing of critical knowledge and for creating new mutually useful knowledge.
In coopetition, companies may use various means and strategies of interaction to manage the relationship and its activities. In face-to-face interactions the skillful use of interactional strategies, such as aligning goals, negotiating, giving concrete examples and highlighting mutual benefits are emphasized. Abilities to construct particular issues as strategic facts, opportunities and capabilities and frame particular actions as desirable are valuable communication capabilities (Samra-Fredericks, 2003;Jarzabkowski and Seidl, 2008). Therefore, given the sensitive strategic nature of coopetition relationships, interactional capabilities and the fluency of interaction are emphasized.
Interactional practices are intertwined with procedural practices, but particularly close personal relationship may dominate the relationship, and even override the more formal organization. This is more likely in relationships where the companies are acquainted or friendly also outside their work context. This familiarity can also pose a threat to the coopetition, as noted by Solitander and Tidström (2010). Companies usually have strategic plans, technologies and processes that need to be withheld from partners in cooperative relationships. However, even if there is a company-level decision to withhold certain knowledge, this knowledge may, nevertheless, leak on other levels in the organization; valuable knowledge may spread to competitors through the personnel (Solitander and Tidström, 2010). Based on this, it is possible to say that interactional practices may follow a logic based more on personal relations than formal norms and rules.
A procedural practice for managing coopetitive tension is to use various types of contractual and institutional governance. In this case, legal frameworks are established to clarify the objectives, responsibilities and positions of the coopetition partners, and institutional arrangements are made to enable sound execution of coopetition relationships, e.g. via a thirdparty mediation or separation of competitive and collaborative activities. However, Fernandez et al. (2014, p. 224) argue that "a legal framework offers little help as the relationships are evolving over time and have multiple dimensions." This implies that, for instance, separation of collaboration and competition (as suggested by Bengtsson and Kock, 2000) can be difficult from a practice perspective, for example, in situations where individuals are related to each other through various kinds of activities. Therefore, we argue that while procedural arrangements may prove helpful in managing coopetition, they cannot completely rule out tensions arising at the individual and interpersonal level. Therefore, tension management should always take into account the interactional tension as well. The focus of the study is illustrated in Figure 1.
In this section, literature on coopetitive tensions and interactional and procedural practices in tensions management has been discussed and integrated. However, there is a gap in existing research especially related to combinations of interactional and procedural practices in relation to different kinds of coopetitive tensions. Next, this gap is addressed through an analysis of two illustrative case examples, as well as further reflection of strategy-as-practice perspective to coopetition.

Empirical illustrations
In this section, we will present two empirical examples of procedural and interactional practices related to tensions in coopetition. These examples are drawn from two previously conducted qualitative case studies on coopetitive business relationships and should be treated rather as illustrations of the developed theoretical insights, rather than full-fledged case studies. In general, empirical illustrations based on qualitative case studies can be considered as appropriate for this study, as we focus on a scarcely studied research topic (Eisenhardt, 1989). Moreover, the focus of our study is related to multiple levels of analysis (Yin, 1984). Finally, qualitative research approach is also recommended when studying strategy-aspractice (Jarzabkowski and Spee, 2009).
The empirical studies were completed in 2009 and the material consists of interviews with the managing directors of one of the companies involved in the coopetitive dyad. The cases were purposefully chosen and considered suitable as empirical illustrations, as they represent two different angles of procedural and interactional practices related to coopetitive tensions. Case 1 concerns SMEs and Case 2 involves coopetition between a small and a large company. Moreover, Case 1 concerns tools for the public education sector, while Case 2 involves vehicles, boats and machines for the private sector. Both examples concern coopetitive business relationships that have evolved from being merely cooperative to becoming competitive.
Case 1 discusses Company A and one of its partners, Company B. Company B was established as a sub-contractor for Company A, but eventually became a competitor to Company A. The business is within the education sector and the companies manufacture tools, machines and interior design elements for educational sectors' technical departments. The business also involves public procurement, which means that the tenders are public. It is fairly easy for all interested actors to get access to important product knowledge. Moreover, the industry of the sector is small and is continuously shrinking.
Case 2 involves a small company (Company H) manufacturing boats, vehicles and machines for use on land and on water. The company was established in the 1990s, and it uses about 30 persons. Company H started cooperating with a large company (Company Y) manufacturing similar products on other markets. The aim of the cooperation for Company H was to reach new global markets through the utilization of Company Y's sales network. Company Y, in turn, would get a complementary product to their portfolio and serve the needs of more customers.
Case illustration 1procedural and interactional practices for tensions management Case 1 provides an illustration of a coopetition relationship, where insufficient management of the relationship and the unbalanced use of appropriate practices leads to tensions and problems in the relationship.
Company B was established to manufacture workbenches for Company A, which sold the benches on to its customers. Company B had also worked for several years as a subcontract manufacturer of shelves for Company A. Knowledge sharing between Company A and B involved the whole concept of manufacturing shelves, including, for example, the technical description of the tools in the shelves. Collaboration between the parties was based on informal agreements and on active dialogue between the key persons. The relationship was thus managed through interactional rather than procedural practices, and there was a mutual agreement on coopetition strategy, task allocation and knowledge sharing.
As the business grew, Company B was not able to manufacture the increased quantity of shelves required by Company A. These capacity problems led Company A to seek to terminate the delivery of shelves from Company B. Incompatible views on who had developed the product and had rights on certain product features of the shelves led to disagreement and tensions between the companies. At this time, Company A wanted to protect its knowledge and competitive advantage and opted to manage the tensions and settle the disagreement by drafting a formal written contract presented to Company B; however, Company B declined to sign the contract. Over time, Company A found another supplier and Company B stopped manufacturing the shelves. The loss of business caused financial issues for Company B, which sought to improve turnover by selling workbenches directly to the same customers as Company A. This, in turn, led to the emergence of tensions between the two companies over task and resource allocation. The managing director of Company B contacted Company A asking for a permission to start selling workbenches directly to customers. As times were tough within the tools business for the education sector, Company A accepted that proposal, and since then, Company B has continued to do so. Despite its initial compliance with Company B's request, Company A has continuously communicated its disappointment, and stressed that this is not Figure 1 The focus of the study Case 1 illustrates a partnership developing from subcontracting to coopetition, as well as the emergence of tensions that relate to both procedural and interactional practices. At first, mutual trust and collaboration were established on the basis of personal relationships and informal verbal agreements between key persons in Company A and Company B. Potential tensions were dealt with continuous dialogue. Interactional coopetition practices worked well as long as the business relationship was steady and unproblematic. However, changes in the business triggered a set of tensions related to knowledgesharing and strategy. Company A unsuccessfully attempted to manage those tensions with a new set of coopetition practices governed by formal written agreements.
An even more critical change in the relationship, strategy and task allocation between the parties occurred when Company B was able to informally re-negotiate the relationship with Company A, and thus change its status from a sub-contractor to competitor/coopetitor. By using interactional practices, in this case, one-to-one discussions between managing directors, Company B established a new business area for itself and created a highly tense situation between the parties.
Case 1 is illustrated in Figure 2 and elaborated upon below. Case 1 highlights multiple noteworthy issues regarding the management of tensions in coopetition relationships. First, a balanced set of both procedural and interactional coopetition practices would help to prevent and manage tensions more effectively than relying on only one type of practice; in this case, close interaction between the parties. Interactional practices may seem sufficient in the beginning of the relationship, but they provide little protection when problems arise in business, and therefore, interactional practices should be complemented with formal procedural practices, and particularly in sensitive areas such as immaterial rights and knowledge. Second, if the relationship is strongly based on just one form of coopetition practice, in this case, interactional practices, it is difficultand perhaps even impossibleto introduce other types of practices to manage the relationship, particularly in crisis situations. Third, verbal agreements on major changes in the relationship should be formalized and secured by written agreements to safeguard the interests of both parties.

Case illustration 2
Company H and Company Y had a written contract regulating their collaboration. The benefits of the collaboration included Company Y selling the products of Company H to complement its own product range, and Company H acquiring a sales channel to a new market. The contract prohibited Company H from selling its products to other companies. In practice, Company Y preferred to sell its own products and would only offer the products of Company H if customers declined to buy Company Y's products. The relationship between the two companies (H and Y) was unbalanced, as Company Y was able to engage in opportunistic behavior favoring its own products in the sales channels. Tensions regarding the division of tasks built up and representatives of the companies initially discussed the situation. However, Company Y was a larger company with better resources, and was, for example, able to use lawyers in the contract negotiations. As the informant put it, Company H was stuck in the relationship with Company Y.
Through their cooperation, Company Y was able to acquire information and knowledge as well as to learn from Company H. This information concerned both technical aspects of the products and working practices, which were later applied by Company Y to improve its own products. Moreover, Company Y had Company H send it some machines, ostensibly to evaluate and to demonstrate the product to customers. The tactic allowed Company Y to test and examine the product to learn the details of its composition. Simultaneously, Company Y had started developing its own new product similar to the product made by Company H, and thereby the relationship between the firms was also competitive. The fact that there were no practicesinteractional or proceduralto control and manage the dissemination, use and ownership of valuable information and knowledge, contributed to a situation where Company Y, as the larger, more powerful and more resourceful company, was able to exploit the collaboration and the acquired knowledge.
The strategic aims of the parties in the coopetitive relationship were also misaligned. Company H sought to grow Figure 2 Practices for managing coopetitive tensions in Case 1 and develop through collaboration, whereas Company Y's strategy was to prevent Company H's growth and to start competing with it. Accordingly, there were certainly strategic tensions between the parties. However, these aspects and risks in coopetition were not addressed in the contracts between the parties. Obviously, the companies did not have wellfunctioning interactional practices either. A strong line of communication between key persons from both companies could have been used to address the gaps in the contract and the terms of collaboration. In the absence of procedural practices for proper risk management, Company H was disadvantaged. The main findings of Case 2 are illustrated in Figure 3 and elaborated on below.
Case 2 points out to several important issues regarding preventing and managing coopetitive tensions. First, it generally is in the best interest of the coopetition partners to use appropriate procedural practices, such as contracts, non-disclosure agreements, and other documentation to manage both the benefits and the risks associated with coopetition. Agreements and contracts signal commitment to collaboration and ensure that investments in time, resources and money are worthwhile. Indeed, contracts and other documentation would have helped both Company H and Y to set more concrete and mutually beneficial goals for the collaboration and prevented many of the disagreements between the parties.
Second, there was a lack of important interactional practices such as inter-individual communication. The companies had not established proper channels and practices of inter-personal communication, and therefore all interaction took place in an irregular, ad hoc manner. In addition, the companies mostly communicated with each other regarding negative issues, such as attempts to terminate the agreement or re-negotiate the terms of the contract. In the absence of proper interactional channels and practices, the coopetition parties did not actually collaborate towards joint value creation, but rather pursued only on their own private goals within the partnership. Here, proper lines of communication with key people involved as well as frequent and versatile interaction practices between coopetition partners could have strengthened the cooperation, facilitated commitment to collaboration and prevented much of the tensions.

Discussion
The findings of this study are illustrated in Figure 4, and they are discussed below.
In line with prior studies on coopetitive tensions, the findings of this study also indicate that coopetitive tensions are related to resource and task allocation, knowledge sharing and strategizing. However, our study provides and integrative framework of how these coopetitive tensions might be beneficially managed by combinations of interactional and procedural practices.
Coopetitive tension related to customer interface With regard to resource and task allocation, our findings indicate that this tension is highlighted in the customer interface, or sales to similar customers. On a procedural level, there are studies stressing that separation and integration can be applied for the management of coopetitive tensions related to resource and task allocation (Dowling et al., 1996;Bengtsson and Kock, 2000;Fernandez et al., 2014). The findings of our study show that these kinds of coopetitive tensions can be managed beneficially by combining procedural and interactional practices. In particular, we suggest that tensions related to activities in the interfaces with third parties such as customers should be alleviated through formal contracts that are set already at the beginning of a coopetitive business relationship. The contract should determine the companies' obligations and rights affecting the relationships with third parties. If the companies' roles and responsibilities have been set in a formal contract, on an interactional level, there can be frequent and open communication between individuals concerning task and resource allocation. Open communication can facilitate trust, which again can increase knowledge sharing between the companies Tidström, 2014).
Coopetitive tension related to core competences In line with prior research, the findings of this study show that knowledge sharing creates tensions in coopetition. The findings support those of Jordan and Lowe (2004), stating that asymmetries in capabilities and resources among companies can also lead to different kinds of opportunities to negotiate, and consequently lead to differentiated opportunities for learning. However, our study also offers a complementary perspective on knowledge leakage in coopetition, as it we found instances where it related to core competences such as productspecific knowledge and established ways of working. According to prior studies (Baughn et al., 1997;Hannah, 2005), harmful knowledge sharing usually take place on employee level, whereas our findings highlight that it can also occur through institutionalized, firm-level processes. These, in turn, affect and regulate how individuals behave. Those findings suggest it is important to have formal written contracts and procedures in place regulating what to share and what not to share, as this has a connection to individual level behavior. On an interactional level, there may then be dialogue between individuals within frames of the contracts and rights. More broadly, formal intellectual property rights are essential in that they provide a contractual regime for knowledge protection in coopetition (Ritala and Hurmelinna-Laukkanen, 2013).

Coopetitive tension related to strategy and strategizing
The findings of our study also show that coopetitive tensions related to corporate strategy and strategizing are common. While the value between partners is often divided in different ways, one party's over-acquisition of benefits can be seen as opportunistic activity (Osarenkhoe, 2010), and this is a key source of coopetition tension. The opportunism threateven if not actualizeddisturbs the strategic process of coopetition in closing some opportunities that could otherwise be available. As with any business relationship, strategy-related tensions in coopetition relationship can obviously be managed by various kinds of formal agreements. According to Jarzabkowski (2005, p. 51), formal practices are important because they "provide the externally legitimate, traceable realizations of the institutional rules of strategy making." However, procedural practices or formal agreements are not enough; interactional practices for communicating goals are of upmost importance. Our findings are in line with coopetition studies Fernandez and Chiambaretto, 2016) that suggest that individuals appreciate transparent communication about the firm's goals and the agendas influencing communication with rival firms.

Interaction of interactional and procedural practices for managing coopetitive tensions
In addition to managing tensions in the three aforementioned categories, procedural and interactional practices have interdependent features that warrant further discussion. Agreeing on strategic objectives mutually and reviewing them regularly provides structure, predictability and transparency to the coopetition relationship, and helps to prevent ambiguity and conflicting views, and the potential tensions associated with them. Additional elements that should be mutually agreed upon, properly documented, and regularly reviewed are financial assessments and targets for the required investments, as well as the anticipated results and profits of cooperation. A robust set of procedural practices can, at its best, function as a backbone for the coopetition relationship by framing the mutual goals, objectives, obligations and expected profits. In that case, procedural practices can be used to anticipate and prevent many of the tensions inherent to coopetition relationships, as well as to manage tensions in new situations.
Previous literature on coopetition has often stressed the value of contracts in managing tensions ; see also Möhring and Finch, 2015). However, there are also studies questioning the role and value of contracts (Fernandez et al., 2014). The current study found examples of contracts both enabling and restricting coopetitive tensions. Based on our examples, we suggest that contracts have a potential for restricting coopetitive tensions if they are applied consistently from the initial phases of the relationship, and if they openly state the specific goals of the partners to the contractual regime.
Contracts and non-disclosure agreements are generally used to regulate the obligations of the parties regarding intellectual property rights, such as patents and other types of proprietary information, and also the knowledge generated during the coopetition relationship. Contracts and agreements can also be used to exclude certain types of actions in the relationship. However, contracts need to be complemented with other types of procedural practices. The strategic objectives of the collaboration must be agreed upon and documented and distributed among the partners. In addition, partners need to document the activities and practices they will use to accomplish the mutually agreed objectives. A regular review of Figure 4 Summary of findings: key practices for managing coopetitive tensions the coopetition strategy both in terms of objectives and actual activities is necessary to ensure that the interests and aspirations of both parties remain aligned.
Procedural coopetition practices work best when accompanied by frequent and open communication between the parties. Interactional coopetition practices may also be formalized in the sense that the coopetition partners agree on key persons, communication channels and the frequency of communication, on how these interactions are documented, and how and to whom such information is distributed.

Implications and future research directions
The study at hand provides useful implications for practitioners and managers working in coopetitive relationships. Our findings show that the management of tensions in coopetition would benefit from a holistic, multilevel approach that recognizes both how tensions are related and how they can be beneficially managed through various practices. Traditionally, coopetition strategy has been treated on a corporate level with a focus on strategy as a fixed plan rather than as a bundle of practices used in the various stages of strategy making. We assert that examining coopetition from a practice-based perspective can shed light on the details of how coopetitive relationships and their related tensions can be managed. At the same time, the practice-based approach illuminates the roles and activities of actual people involved in coopetition, and shows how their work and practices in-use contribute to coopetition. Third, it is important for managers to consider the ways in which contracts and other types of procedural practices can best be used to manage tensions. Companies engaged in coopetition should diligently document both the obligations of the parties and the objectives set for the coopetition from the start, as doing so can deter opportunistic activity. Fourth, one-to-one communication as an interactional practice is not sufficient to prevent opportunistic behavior in coopetition relationships. Instead, all critical and strategic aspects of the coopetition need to be formally documented to harness potential opportunism in the relationship. It is important to set the strategy and terms of coopetition at the beginning of the relationship and to expressly communicate what should be shared and what should remain confidential. These terms should be transparent on both the procedural and interactional levels.
An interesting opportunity for future research would be to investigate and compare the findings of this study with purely cooperative or competitive business relationships, to determine the extent to which tension-management practices differ in various types of relationships. Furthermore, given the illustrative nature of our case study examples, future research could adopt a more in-depth case study approach to find out more of the micro-dynamics of how tensions emerge and how they are resolved in coopetition, particularly among coopetition practitioners. Indeed, it has been shown that managers adopt different individual-level resolutions to deal with paradoxical tensions (Jarzabkowski et al., 2013;Karhu and Ritala, 2018), and coopetition as a context provides an excellent arena for empirical inquiry into individual-level analysis of paradox management. Another relevant field for further investigation is to explore practices for tensions management from a network perspective, by including actors outside the focal business relationships (Chou and Zolkiewski, 2017). Overall, it is clear that research on managing coopetitive tensions by focusing on practicesthat is, the activities, doings and sayingsremains undeveloped, which emphasizes the importance of encouraging further research in this field.

Conclusion
Our study outlines a strategy-as-practice logic for examining coopetition tensions and their management. In doing so, our main contribution is in connecting coopetition practices in two levels directly to the tensions that arise from the coopetition relationship, and on practices that can be employed to manage and alleviate the related tensions, as well as the interplay between these dimensions. Our findings add to the available insights into the practices of coopetition at different levels of analysis, including the individuals involved (a research gap flagged by e.g. Bengtsson and Kock, 2014;Dahl et al., 2016), and it also complements studies that examine the management of tensions in coopetition (Fernandez et al., 2014;Tidström, 2014;Ritala et al., 2017). Prior research on managing coopetitive tensions has mostly focused on one level of analysis, such as separation and integration on the company or individual level (Fernandez et al., 2014), the use of contracts on a relational level  or the need for open communication between individuals (Fernandez and Chiambaretto, 2016). Our study is the first to suggest an integrative framework for alleviating and managing coopetitive tension via procedural and interactional practices, while taking account their mutual interdependences, synergies and limitations. We argue that adopting the strategy-as-practice approach generates valuable new insights into the practices and the related tensions in coopetition, and illuminates the roles of the actual people involved in managing coopetition relationships in a novel and unprecedented way. While our short case illustrations provide a view on the potential of the approach, we advocate for more in-depth examinations of interactional and procedural practices that will help to understand how tensions arise in coopetition and how they can be effectively managed.