Open Strategy-Making at the Wikimedia Foundation: A Dialogic Perspective

Abstract

Dialogue is essential to open strategy processes, yet these processes have not been researched from a dialogic perspective. We therefore ask the question: What is the role of dialogue in open strategy processes? Our study of the development of Wikimedia’s 5-year strategy plan through an open strategy process reveals the endemic nature of tensions occasioned by the intersection of dialogue as an emergent, nonhierarchical practice, and strategy, as a practice that requires direction, focus, and alignment. Furthermore, our study suggests that context matters to both dialogic and open strategy processes and challenges universalist features of dialogue. Specifically, dialogic organization development and large group interventions perspectives can offer useful lenses to incorporate contextual features. Finally, our implications for practice suggest that particular organizing principles of open strategy processes can foster a healthy dialogic process in terms of enabling ongoing, diverse, and substantive interactions and inputs by stakeholders.

Keywords

open strategy dialogue dialogic organization development Wikimedia

Imagine a world in which every single person on the planet is given free access to the sum of all human knowledge. That’s what we’re doing.

—Jimmy Wales¹

Introduction

Open strategy without dialogue would be a contradiction in terms since key features of an open strategy process, such as transparency and inclusion (Whittington, Cailluet, & Yakis-Douglas, 2011), necessitate dialogic interaction. In this article, we explore how Wikimedia, an organization with a long-standing commitment to openness, participation, and transparency to its community of contributors, has developed a 5-year strategy through an open strategy process. We outline the dialogic perspective (Gergen, McNamee, & Barrett, 2001; Isaacs, 1993; Jacobs & Heracleous, 2005), an apt but missing perspective from the literature on open strategy, and employ it to analyze Wikimedia’s strategy process.

We define dialogue as communicative interaction between two or more parties that serves to accomplish particular outcomes as well as to shape agents’ ways of thinking and acting through effectively understanding and sharing each other’s perspectives. In this context, we employ textual data, including data from dialogic interactions, to identify features of Wikimedia’s open strategy process as well as key dialogic attributes at different stages of the process. Given the early stage of research in open strategy processes (Whittington et al., 2011), and the lack of a dialogic perspective in this field, we pose the exploratory research question: What is the role of dialogue in open strategy processes?

In addition to conducting the first study of open strategy processes from a dialogic perspective, and providing implications for the practice of open strategy, we make two key conceptual contributions. First, we propose that open strategy processes can be viewed as large group interventions (LGIs; Bunker & Alban, 1992, 2006), and we show that various types of tensions are inherent in this processes. These tensions arise from the intersection of dialogue as a nonhierarchical, emergent, ongoing practice between equals on the one hand (Gergen et al., 2001), and the domain of strategy from a traditional, planning perspective demands direction, focus, and alignment on the other hand (Andrews, 1971; Chaffee, 1985). In drawing from organization development (OD) literature to view open strategy processes as LGIs, we observe that open strategy and LGIs have significant similarities in their key features and challenges. These include the endemic dilemmas that arise in LGIs relating to actors’ voice, appropriate degree of structuring of the process and the need to take account of diverse perspectives to arrive at a commonly shared direction for the future. These tensions noted in LGIs are not absent from open strategy processes, where they take a special form given the directive, focused, convergent qualities of traditional conceptions of strategy, and the nonhierarchical, expansive, and often divergent nature of dialogic approaches. These two domains exhibit different and often incompatible expectations and features. We argue, therefore, that tensions are endemic in open strategy processes and that they have to be understood and negotiated if such processes are to accomplish their goals.

Our second conceptual contribution is that we challenge the universalist qualities of both dialogic and open strategy approaches and suggest that further theoretical development of these fields will depend on customizing insights to particular empirical contexts. We also suggest that dialogic OD (Bushe & Marshak, 2009; Marshak & Bushe, 2009) has an inherently contextual nature that can help advance our understanding of dialogic processes in different contexts.

Dialogic literature, more broadly, often posits “ideals of dialogic co-production” (Beech, MacIntosh, & MacLean, 2010, p. 1352), conditions of dialogue such as ongoing information exchange by parties regarded as equals, a genuine interest in others’ views, self-reflexivity, joint exploration of challenges, building convergence, and opening new avenues of being and doing (Beech et al., 2010; Gergen et al., 2001; Gergen, Gergen, & Barrett, 2004). The implicit assumption is that such features are applicable in different contexts. However, we find that dialogue was guided and occurred differently, and had a different role, in each phase of the open strategy process at Wikimedia. The literature on open strategy is still in its infancy, but essential dimensions such as transparency and inclusion have been identified. So far, there is little understanding of how such dimensions may play out in different contexts. We argue, therefore, that we need to conceive of dialogue as well as open strategy as contingent, context-dependent processes; particularly, when outputs such as robust strategic plans are sought as outcomes.

Open Strategy-Making and the Dialogic Perspective

Open Strategy-Making as a Large Group Intervention

The concept of “open strategy” has emerged as a response to empirical developments in business models that challenge fundamental assumptions of traditional strategy-making, such as the assumed need to own and control value-creating resources (Chesbrough & Appleyard, 2007). In contrast, organizations such as YouTube, Linux, and Wikimedia employ community-driven value creation, where value capture can be either by the company (e.g., YouTube) or by the ecosystem (e.g., Linux and Wikimedia; Chesbrough & Appleyard, 2007). Open innovation, the process of involving external parties in innovation processes or leveraging innovations occurring outside the organizational boundaries (e.g., Huston & Sakkab, 2006; West & Bogers, 2013) has been a conceptual and empirical precursor of open strategy. Traditional perspectives of innovation and strategy as occurring solely within firm boundaries have been challenged by developments in technology (Matzler, Fuller, Koch, Hautz, & Hutter, 2014) and new business models (Chesbrough, 2010), which enable and welcome new types of interaction and collaboration in value creation processes.

Involving a broader spectrum of internal as well as external stakeholders in strategy-making has been enabled and accelerated by information and communication technologies, such as the Internet and social media (Stieger et al., 2012), or in particular cases more specialist technologies such as Wiki software (Dobusch & Mueller-Seitz, 2012). Such technologies are embodied in material artefacts that shape strategy-making in significant ways, what Whittington (2015, p. S13) refers to as the “massification of strategy.” It has been argued that where strategy is crowd-sourced, crowds contributing to a common endeavor should be diverse so that their inputs are also diverse, should be independent (rather than directed or controlled) in arriving at those inputs, and finally accurate aggregation of inputs should take place (Stieger, Matzler, Chatterjee, & Ladstaetter-Fussengegger, 2012).

Matzler et al. (2014) propose a typology of open strategy based on two axes: whether the process aims to facilitate strategy generation or implementation and whether it engages internal or external stakeholders; leading to four quadrants of open strategy processes. As we argue, this is a useful typology but some instances of open strategy that involve broad scope of both strategy generation and implementation, and both internal and external stakeholders are difficult to categorize in this way.

Open strategy processes that involve a multitude of stakeholders can be seen as LGIs (Bunker & Alban, 1992) in which they attempt to meaningfully engage these stakeholders to collaboratively identify a direction for the future. LGIs aim to bring the whole system into the room to address complex problems that demand multistakeholder inputs, as Weisbord and Janoff (2005) describe in their work with IKEA to redesign the company’s product manufacturing, production, and distribution. The aim is not only to engage actors with a stake in a situation but also to promote ownership and create a commitment to act to implement the jointly developed plans (Coghlan, 1998).

Despite the large number of approaches to designing LGIs (Bunker & Alban, 1992), there are commonalities. As Worley, Mohrman, and Nevitt (2011, p. 405) note,

LGIs tend to have a common set of features, including a focus on the future, broad participation, an open-systems perspective, an intense period of planning prior to the meeting, information sharing, and a combination of divergent/creative as well as covergent/agreement processes.

Challenges in LGI processes include the dilemma of voice (how to enable all actors to have a voice when a multitude of actors are involved and a limited amount of time available), the dilemma of structure (to avoid too much or too little structure, both of which can derail LGIs), and the egocentric dilemma (how to balance and integrate individual and group perspectives; Bunker & Alban, 2006).

Worley et al.’s (2011) description of central features of LGIs can also describe central features of open strategizing, particularly the broad participation, future focus and planning prior to the process. An open systems perspective may not be explicit in open strategy processes or LGIs, but it is nevertheless consistent with such processes. The challenges relating to voice, structure, and perspective noted by Bunker and Alban (2006) are also present in open strategy initiatives.

A Dialogic Perspective and Open Strategy-Making

The very possibility of open strategy, however, presupposes effective dialogue among the parties involved. In particular, the key dimensions of open strategy, transparency, and inclusion (Whittington et al., 2011), require mutual and ongoing dialogic interaction between the organization and relevant stakeholders. Given that a dialogic perspective seems an appropriate lens for exploring the phenomenon of open strategy, it would be helpful to introduce it to the open strategy literature. In this article, we examine how Wikimedia employed dialogic processes to develop its strategy in an open manner, engaging over one thousand “Wikimedians,” agents at large who regularly contribute to its projects such as Wikipedia. Wiki technology is by its nature dialogical, as it enables both real time and asynchronous interactions among a large number of agents on particular points of interest.

At its simplest, a dialogic interaction involves two or more individuals exchanging information on an ongoing basis, with the aim of accomplishing a particular goal or building mutual understanding. As Mintzberg (1971) and subsequent studies have shown, the managerial work environment is largely verbal, discursive, and dialogical. Effective dialogue has been viewed as a useful way to accomplish a variety of organizational outcomes. These include strategic innovation (Jacobs & Heracleous, 2005), organization change (Hatch & Ehrlich, 2002; Skordoulis & Dawson, 2007), organizational learning (Oswick, Anthony, Keenoy, & Mangham, 2000; Schein, 1993), executive education (McCambridge, 2003), and strategy implementation (Pye, 1995).

Perspectives on dialogic processes differ however. Drawing on a typology of discourse studies (Heracleous & Barrett, 2001), we can distinguish between functional, interpretive, and critical views of dialogue. While this typology was developed to provide some structure to the broader, voluminous field of organizational discourse, we find that it can also usefully categorize different approaches and aims of dialogic studies. Functional dialogic processes focus on how dialogue can foster effective information sharing, recognizing other points of view, and facilitating particular organizational outcomes (e.g., Powley, Fry, Barrett, & Bright, 2004; Pye, 1995). From this perspective, “managing is about dialogue—listening and talking—and about ‘doing’—taking action” (Pye, 1995, p. 445). Organization change can be accomplished when the dominant discourses change; such discourses “emerge from a continuous, iterative and recursive process” (Grant & Marshak, 2011, p. 212).

Interpretive dialogic processes are concerned with how dialogue can open new vistas of seeing and being through an ongoing process of mutual exploration and reshaping of fundamental assumptions (Hatch & Ehrlich, 2002; Isaacs, 1993) within socially constructed realities (Gergen et al., 2004). Dialogue here is a “sustained collective inquiry into the processes, assumptions, and certainties that compose everyday experience” (Isaacs, 1993, p. 25).

Finally, critical dialogic processes aim to expose taken for granted power inequalities, inspire agents to take actions of resistance, and aid in agents’ emancipation from institutional constraints (Heath, 2007; Raelin, 2012). Dialogue is seen as “the genetic material for building a culture of democracy freeing people form institutional forces that limit their personal autonomy and leading to their acquisition of a collective consciousness” (Raelin, 2012, p. 819).

While analytically useful, these perspectives are not always as distinct as the typology suggests however. A key interpretive insight, that dialogue and more broadly discourse shapes ways of thinking and acting, is a shared assumption with functional and critical approaches, as can be seen, for example, in the argumentation of Grant and Marshak (2011). Furthermore, a connection between interpretive perspectives and functional concerns can often be found in the employment of a dialogical perspective. For example, Jacobs and Heracleous (2005) offering an interpretive perspective on how the functional goal of strategic innovation can be fostered, suggested that generative dialogue can facilitate strategic innovation through an ongoing interaction between diagnostic and generative moments, respectively, involving critical reviews of existing mental models and emergence of novel mental models.

The functional dialogic approach has affinities with what Bushe and Marshak (2009) label diagnostic OD. This is the traditional approach to OD, more objectivist in nature, where the emphasis is on gathering information to diagnose issues and plan and implement appropriate change processes, with the objective on changing behaviors. This is contrasted with dialogic OD, a more recent set of understandings and practices in OD, which is more interpretivist in nature. Organizations here are seen as meaning-making systems, reality as socially constructed, and the emphasis is on encouraging generating dialogue to foster change in mind-sets and facilitate self-organizing organization change.

Table 1 offers an outline of functional, interpretive, and critical perspectives on dialogue.

Table 1.

Dialogic Perspectives.

Dialogic perspective	View of dialogue	Relevant definitions	Selected references
Functional	Dialogue as communicative interaction and sharing of ideas, serving as a means of accomplishing outcomes such as organizational learning, strategy implementation, or organization change.	“Managing is about dialogue—listening and talking—and about ‘doing’—taking action” (Pye, 1995, p. 445). “Dialogue is a mode of communication that builds mutuality through the awareness of others, use of authentic or genuine discourse, and reliance on the unfolding interaction” (Putnam & Fairhurst, 2001, p. 116).	Powley et al. (2004); Putnam and Fairhurst (2001); Pye (1995); Schein (1993); Skordoulis and Dawson (2007)
Interpretive	Dialogue as shaping agents’ ways of thinking and acting through effectively understanding and engaging with each other’s perspectives; in the process giving birth to new ways of looking at and being in the world.	Dialogue is a “sustained collective inquiry into the processes, assumptions, and certainties that compose everyday experience” (Isaacs, 1993, p. 25) Dialogue is “a reflective conversational mode; a diagnostic moment where participants can review their existing mental models, and a generative moment where emergent mental models may be shaped” (Jacobs & Heracleous, 2005, p. 344).	Beech et al. (2010); Bushe and Marshak (2009); Hatch and Ehrlich (2002); Isaacs (1993); Jacobs and Heracleous (2005); Oswick et al. (2000)
Critical	Dialogue as a reflexive process of exploration that can expose taken for granted power inequalities, dominance, and institutional constraints, with the aim of fostering a critical consciousness and inspiring emancipatory or resistance-oriented actions.	“ . . . dialogue is seen as the genetic material for building a culture of democracy freeing people from institutional forces that limit their personal autonomy and leading to their acquisition of a collective consciousness” (Raelin, 2012, p. 819) “Defining dialogue provides the very closure of a subject matter that dialogue resists; however, three dialogic themes contribute a priori constructs . . . (a) dialogue as generative, (b) dialogue as grounded in diversity, and (c) dialogue as critical of power” (Heath, 2007, p. 149).	Heath (2007); Kersten (2000); Raelin (2012); Zoller (2000); Hammond, Anderson, and Cissna (2003)

Drawing from both the functional and interpretive views in Table 1, we define dialogue as communicative interaction between two of more parties that serves to accomplish certain outcomes, as well as to shape agents’ ways of thinking and acting through effectively understanding and sharing each other’s perspectives. Given the central role of dialogue in open strategy processes, the early stage of the open strategy field, and the paucity of research on open strategy from a dialogic perspective, our research question is as follows: What is the role of dialogue in open strategy processes?

Methodology

Given the emerging nature of the open strategy theme and the sparseness of available literature, we carried out an in-depth qualitative case study (Eisenhardt, 1989; Yin, 2014) on Wikimedia’s open strategy process. We saw the Wikimedia case study as an instance of phenomenon-based research, where little theory exists for a particular phenomenon of interest, and therefore, such study can provide empirical and theoretical insights (von Krogh, Rossi-Lamastra & Haefliger, 2012). Furthermore, we examine Wikimedia as an extreme or unique case (Yin, 2014) of open strategy, given the unprecedented extent of stakeholder involvement, and the scope and significance of this involvement in terms of the resultant strategic plan (Dobusch & Mueller-Seitz, 2012). This was therefore an example of theoretical rather than random sampling; examining a particular, selected case that will likely lead to useful insights on the topic of investigation given its relevance to this topic and its exemplary nature (Eisenhardt, 1989).

Given Wikimedia’s commitment to transparency, a broad variety of materials relating to its strategy-making process have been archived on the Internet. We drew on publicly available documents (including calls for participation, inputs by volunteers, blog posts, webinars, interactions among agents involved in the process, and reports). We also conducted two e-mail interviews of Wikimedia executives, Jimmy Wales (founder) and Philippe Beaudette (Director of Community Advocacy).

In addition, the research was informed by participant observation by one of the authors who initially acted as a project facilitator to the open strategy process, before being employed full time by Wikimedia. The other authors researched the strategy process via the voluminous public records of this process. The interaction of outsider and insider perspectives helped us gain a more rounded understanding of the process, control for interpretation biases that often accompany purely internal or purely external perspectives, and buttress the validity of our analysis and findings.

We first prepared a narrative of the process indicating the context, the sequence of key events, and the main actors involved, which also informed our interpretation and analysis of the archived materials (Mauthner, Parry, & Backett-Milburn, 1998). We proceeded to analyze the data through a coding process, beginning with open coding via iterative reading of the various documents through which initial themes emerged. We proceeded to axial coding where we aimed to understand the relationship among these themes and to group them into categories. Finally, we carried out selective coding where we incorporated information relevant to the categories that emerged. The process followed the tenets of grounded theory (Glaser & Strauss, 1967), in particular, the approach proposed by Gioia, Corley, and Hamilton (2012).

Gioia et al. (2012) make the assumptions that organizational realities are socially constructed, agents are knowledgeable about these realities and can be articulate about them, and patterns should emerge from the data rather than be structured by concepts and theories imposed a priori by the researchers. While we were sensitized by conceptual issues relating to open strategy and dialogue, following Gioia et al. (2012), we suspended judgment about this literature and instead gave primacy to the data. The second-order concepts therefore emerged inductively based on first-order communications as captured in the various data sources.

Through this analytical process, we gradually reached an appreciation of key aspects of Wikimedia’s culture as well as its open strategy process, as represented in eight categories: transparency, participation, collaboration, coordination, process design, context creation, community empowerment, and decision-making control. The first three categories (transparency, participation, and collaboration) represented broader Wikimedia values that imbued the open strategy process. The next five categories represented particular aspects of how the open strategy process was operationalized. Figure 1 shows the data structure that emerged.

Figure 1.

Data structure.

Table 2 gives illustrations of the empirical data that led to each category. We selected the themes of transparency, participation, and decision-making control in order to show the potential tensions that can be occasioned when an open dialogic process intersects with strategy, where decisions between competing objectives need to be made.

Table 2.

Representative Quotes for Selected Central Themes.

Theme	Representative quotes
Transparency	“As much as possible, work should be done in public, and be visible to all” (Wikimedia Foundation, 2011b).^a “Task forces will be working in a transparent way unless there is a critical reason to keep something confidential. . . . We would strongly encourage those self-organized groups to also work transparently on this Wiki, and ask that they report out their findings for the final, synthesis, stage” (Wikimedia Strategic Planning, 2011r).^b “The core tenet of Wikipedia (and all Wikimedia communities) is radical transparency and collaboration” (Beaudette, personal communication, January 30, 2012). “When the task forces begin to meet, they will do their work transparently and on this wiki, and any member of the community may join fully in their work” (Wikimedia Strategic Planning, 2011r).^b “The fact that the project is being run in such a way as to maximize participation and transparency strikes me as a strong indication that there is sufficient harmony between the staff and the volunteer community to move forward productively” (User Peteforsyth; Wikimedia Strategic Planning, 2011r).^b
Participation	“Our principles for facilitating this process: Empower our community and stakeholders to participate . . . ” (Wikimedia Strategic Planning 2011c).^c “The more people who contribute to this process, the more meaningful it will be. We strongly encourage people to participate, and we will be thoughtful about ways to get more people engaged” (Wikimedia Strategic Planning, 2011x).^d “We want a representative segment of the movement’s stakeholders to participate in our process” (Wikimedia Strategic Planning, 2011a).^e “The mechanisms used to solicit input should be designed to be as open as reasonably possible . . . ” (User Dafer 45; Wikimedia Foundation 2011b).^a “ . . . we ask that the strategic planning process be designed to include input from a wide range of sources, including Wikimedia volunteers and supporters representing a diversity of geographies and projects” (Wikimedia Foundation, 2011b).^a
Decision-making control	“Although we strive for an open decision-making environment, there may be times when decisions must be made by the convenors” (Wikimedia Strategic Planning, 2011e).^f “We recognize that we will not develop a consensus strategy that pleases everyone. We will need to make difficult decisions that may prove unpopular” (Wikimedia Foundation, 2011b).^a “ . . . in some cases we actually said: OK, we are going to put together a committee, we will have this process open so everyone can contribute. But if there is some kind of conflict or there is a decision to be made and consensus is not being reached around it we will empower this group of people to basically make a decision” (Kim, 2011).^g “You can’t have that principle [NPOV (Neutral Point of View)] in strategic planning. . . . . Strategic planning is about making choices . . . we knew that was going to be different from an open source project. . . . We knew it was even going to be different from Wikipedia. Because we could not use all of the same principles at strategic planning and so on that Wikipedia uses” (Kim, 2011).^g “ . . . a democratic process was never a goal for this project. Rather, it was a community facing/community influenced process. For that to work, you don’t need participative democracy” (Wikimedia Strategic Planning, 2011d).^h

Wikimedia Foundation (2011b). [Foundation-l] Strategic plan resolution (http://lists.wikimedia.org/pipermail/foundation-l/2009-April/051565.html). ^bWikimedia Strategic Planning (2011r). Village pump/Archive3 (http://strategy.wikimedia.org/wiki/Village_pump/Archive3).^cWikimedia Strategic Planning (2011c). Purpose and principles (http://strategy.wikimedia.org/wiki/Purpose_and_principles). ^dWikimedia Strategic Planning (2011b). Call for action (https://strategy.wikimedia.org/wiki/Process. ). ^eWikimedia Strategic Planning (2011a), Process/Evaluation/Community engagement/en (http://strategy.wikimedia.org/wiki/Evaluation/Community_engagement). ^fWikimedia Strategic Planning (2011e). Community guidelines (http://strategy.wikimedia.org/wiki/Community_guidelines). ^gKim, E. E. (2011, March 22). Followup to Strategic Planning for Networks Webinar by Eugene Eric Kim (http://leadershiplearning.org/blog/natalia-castaneda/2011-03-22/strategic-planning-networks-webinar). ^hWikimedia Strategic Planning (2011d). Village pump/Archive1 (http://strategy.wikimedia.org/wiki/Village_pump/archive1)

Throughout the analytical process, we were conscious of the relevance and role of dialogic interaction among the parties involved. We combined a coding analysis of textual data with exploration of the key features of the dialogic process, which enabled us to explore the role of dialogue during each phase of Wikimedia’s open strategy process (Table 3 and Figure 2), as well as with respect to key parameters of the process (Table 5), as we outline in subsequent sections. This also helped us clarify the role of dialogue in our discussion of implications for practice (Table 6).

Table 3.

The Four Phases, Main Actions, and Dialogic Process of Wikimedia’s Strategic Planning.

Phase	Actions	Dialogical process
Level-setting	Creation of online platform for strategic conversations Creation of knowledge base on Wikimedia to inform planning process Outreach plan (call for participation) Proposals for strategy projects invited, then debated by community Three emergent strategic priorities identified and 14 task forces formed	Creating the technical infrastructure for facilitating dialogue, inviting parties to engage, and allowing broad dialogic themes to emerge. Dialogic funnel starts off widely and gradually narrows in subsequent stages.
Deep dives	Definition of mandate of 14 task forces Selection of suitable contributors Allocation of decision-making authority Development of detailed strategy proposals	Setting dialogical boundaries, narrowing down number of parties to dialogue, aiming for initial strategy outputs.
Synthesis	Formation of Strategy Task Force by 20 engaged contributors Synthesis of prior discussions with focus on feasibility Development of first draft of Wikimedia’s 5-year strategic plan with five strategic priorities	Intensification of dialogic process among a smaller number of parties, application of specific criteria, creation of detailed strategy outputs.
Call to action	Review and refinement of the five strategic priorities, rationales, and measures Wikimedia’s “theory of change” published Call to action to implement proposals Publication of final strategy document	Community review process applied to dialogic outputs and final outputs made public. Dialogic funnel expands again.

Figure 2.

Dialogical process of open strategy at Wikimedia Foundation.

The narrative of the process and the emergent categories were shared with the participant observer member of the authorship team who offered clarifications and inputs to the overall findings, as well as specific factual information that added to our understanding. This member also offered insights based on their involvement in the open strategy process early on in the data-gathering phase. This research therefore represents the fruit of an academic–practitioner collaboration, which aimed to understand the features of an effective open strategy process and the role of dialogue in this process, themes that are important to both communities.

Wikimedia as Collective Creation: Community and Collaboration

Wikimedia Foundation was founded in June 2003 as the umbrella, nonprofit organization of hugely influential Wikipedia and its sister projects, with the aim of providing a governance structure to support their growth and development. Headquartered in San Francisco, the foundation had 191 full-time employees in 2013,² with its projects supported by 80,000 active, engaged editors around the world and by two million donations to a total of US$44.6 million in 82 different currencies in 2012-13.³ Wikimedia is led by a Board of Trustees consisting of 10 members, assisted by an advisory board consisting of an international network of 22 experts.

Individual community members are the driving force behind the creation and functioning of Wikimedia Chapters, which are independent organizations aiming to support and promote Wikimedia projects within specified geographical regions. Chapters are governed by their own boards of directors or trustees and remain decentralized (P. Beaudette, personal communication, January 30, 2012). Wikimedia’s unique governance structure is characterized by community self-governance on the one hand through local chapters and volunteers, as well as the institutional umbrella of the Wikimedia Foundation on the other (Morell, 2011). Neither the founder nor the board and other Wikimedia staff commonly intervene in content and local governance decisions. The Wikimedia board and staff bypass the community only in situations that demand swift action due to potential legal implications, for example (Konieczny, 2010); or to address continuous conflict and editing wars.

Registered contributors and users of Wikimedia sites wield various forms of technical power and social authority. They often join formal and informal subgroups dedicated to ideological, functional, and content-related themes based on their areas of expertise or interest. Trusted Wikimedians can apply for one of the many levels of volunteer stewardship, receive administrator rights, serve as “bureaucrats” on local projects or “stewards” in global roles with almost full access to wikis, user rights and groups (Forte, Larco, & Bruckman, 2009). Communication and coordination within the global community is accomplished through a variety of online platforms: Wikimedia Meta-Wiki focused wikis such as Wikimedia Outreach or Wikimedia Strategic Planning, local Wikimedia Chapter wikis, the Wikimedia Foundation Blog, Wikimedia Signpost, and various mailing lists.

Over time, various guidelines, policies, and formal processes have arisen as a response to specific situations and pressures exercised by stakeholders (Forte et al., 2009). Written and edited by community members like any other Wikipedia content, they can be seen as dynamic and evolving collective-choice agreements.⁴ Many Wikimedia projects aim for decision making based on “consensus over credentials,” a fundamental principle that has been labelled antielitism. “Neutral point of view,” “verifiability,” and “no original research” have emerged as Wikipedia’s nonnegotiable, core content policies aimed at determining and implementing quality standards and control. “Verifiability” and “no original research” require that content published on Wikipedia needs to be based on reliable published sources; and according to the “neutral point of view” principle, articles must represent “all significant views fairly, proportionately, and without bias.”⁵

Contributors to Wikipedia are advised to apply the relevant etiquette, refrain from personal attacks, aim for consensus, avoid edit wars as well as act on and assume good faith. A process of dispute resolution has emerged involving intereditor negotiation, formal mediation and, in case of user conduct disputes, if all other resolution mechanisms have failed, the enforcement of binding solutions based on arbitration.⁶

Wikimedia’s Open Strategy Project

Developing Strategy the Wikimedia Way: Transparency, Participation, and Collaboration

Wikimedia’s strategic planning efforts began to take concrete shape after the first Board of Trustees was formed in 2004. Public brainstorming pages were created and users were encouraged to suggest strategic priorities to the board. Participation, however, was slim at the time.⁷ By 2007, Wikimedia’s board undertook several additional strategic planning efforts: community wide, board, staff, and project committees were formed, the foundation’s mission and vision statements revisited, a basic strengths, weaknesses, opportunities, and threats analysis conducted, and several resolutions passed.

At its April 2009 board meeting, the board decided to take a fresh approach to strategy-making, drawing on the principles of building a transparent, collective vision, open collaboration, and stakeholder involvement.⁸ Both the executive director at the time, Sue Gardner⁹ as well as Michael Snow, the chair of the Board of Trustees, made public statements supporting a collective, transparent, participative, and collaborative approach to strategy-making. Snow noted that,

we recognize that we will not develop a consensus strategy that pleases everyone. We will need to make difficult decisions that may prove unpopular. But we believe that people who want to have a voice in the process, should be heard.¹⁰

These early statements recognize the value that can emerge from dialogical processes that engage several stakeholders; they also are pragmatic, in highlighting that strategy involves difficult decisions and is likely to be controversial, even if it is developed in an open manner. This realization indicates the need for a governance process and thematic as well as participation boundaries placed around dialogical processes of open strategy formation.

The collaborative, participative approach ran into Wikimedia’s veins.¹¹ Philippe Beaudette, a facilitator of the strategic planning project and subsequently Director of Community Advocacy for Wikimedia, noted in 2009 that

If we attempted to go off into a board room somewhere and design a strategy, it would most likely have been universally decried as not being “the Wikimedia way” [ . . . ] Experience has shown that when things are “imposed” upon the Wikimedia community they usually are not adopted wholeheartedly. (P. Beaudette, personal communication, July 27, 2011)

The themes of transparency, participation, and collaboration that pervaded Wikimedia more broadly and their constituent elements are shown in Figure 1.

Designing the Process: Openness and Flexibility Versus Structuring From the Center

Wikimedia strategy project facilitators realized that the process had to balance the tension between the opposing needs for flexibility and openness that was part of Wikimedia’s DNA, with clarity and a structure able to inform and drive strategic decisions:

We also need to make sure that we deliver results. There can and should be an on-going conversation about the details of this process, but we can’t wait until we all agree on everything before moving forward. We have to be both open and agile, meeting our individual needs, being thoughtful and deliberative, and at the same time, moving to action.¹²

Beaudette clarified that the process was collaborative but not necessarily democratic: “a democratic process was never a goal for this project. Rather, it was a community facing/community influenced process.”¹³

Wikimedia’s open strategy process begun with a set of community principles that were introduced at the outset rather than being allowed to gradually develop as collective-choice agreements, as would have been typical for other projects of the Wikimedia movement. These principles encouraged users to edit pages and to exchange ideas, to anticipate particular decisions by convenors when needed despite the openness of the process, and to act in good faith in contributing to the process.¹⁴

Thirty-seven dedicated Wikimedians offered to become hosts (convenors) with the primary goal of creating and maintaining a user-friendly and productive wiki environment. Their work involved organizing and merging proposals, facilitating participation, welcoming new contributors, encouraging conversations, and resolving disputes constructively. Coordination between hosts took place through a regularly updated to-do list. Utilizing Wikimedia’s IRC channel infrastructure, they conducted regular chats where real-time discussions took place and queries from volunteers were answered. Following Wikimedia’s quest for openness and transparency, summaries and entire IRC logs were posted online on the strategy wiki.¹⁵ As a complement to the work of hosts, governance of the process was conducted through experienced users equipped with “sysop” (system operator or administrator) rights, the appointment of six “bureaucrats” and engagement of the Wikimedia Foundation’s steward team.¹⁶ This process illustrates how self-organizing teams can achieve effective coordination, as shown in Figure 1. Also, the themes of process flexibility and clarity (comprising process design) in Figure 1. These and other themes shown in Figure 1 continue to manifest in the description that follows.

Despite the open, collaborative nature of the process that was received positively by many Wikimedians, the top-down initiation and shaping of the project (shown by the second-order themes of context creation and decision-making control in Figure 1) via structures such as task forces challenged the established culture of community-led projects. Several members made critical comments about these aspects, followed by responses explaining why such a guided approach was necessary given the objectives of the process.¹⁷ Such exchanges illustrate the tension between the need for central guidance and coordination of the process on the one hand and norms of open community development on the other. Table 3 outlines the four phases of the process and the main actions taken in each as well as the dialogical aspects of each phase of the process.

Phase 1: “Level-Setting”¹⁸—Preparing the Infrastructure and Inviting Parties to Dialogue

A strategic planning platform called “strategy wiki” was created to facilitate the process of collecting, analyzing, and synthesizing relevant information at a central location. Wikimedia’s project team established a shared knowledge base, “Wikimedia-pedia,” which aimed to illuminate the larger context of the planning process and to support the identification of knowledge gaps. Wikimedia-pedia contained a collection of facts, existing research, and analysis relevant to the Wikimedia movement that could be accessed by everyone.¹⁹ In addition, it contained interviews of advisory board members, members of the Board of Trustees, external experts, and Wikimedia Foundation employees outlining what they believed to be appropriate directions for Wikimedia.²⁰

The project team developed a template with a standardized proposal format and invited volunteers to submit proposals on what they believed were strategic issues that should be addressed. This is an illustration of the “community empowerment” theme shown in Figure 1. Once published, a traditional wiki process was applied to polish, translate, debate, group, and merge the proposals. A box at the bottom of each proposal page enabled visitors at the site to rate it in terms of its priority, impact, feasibility, and desirability from very low to very high. A page listing all proposals was introduced and the suggestions were categorized according to their aims. The page further provided an overview of the most active proposals, the proposals that received the highest ranking, and those which most contributors volunteered to implement.²¹

Even though the sheer size of the movement provided a large talent pool for Wikimedia’s strategic planning efforts characterized by cultural and geographical diversity, recruiting volunteers from the community proved to be a challenge.²² The project team heavily engaged in relationship building with individual volunteers as well as Wikimedia Chapter organizations, online and on a face-to-face basis. An “outreach plan” aimed at reaching as many of Wikimedia’s stakeholders as possible early on in the process.²³ The launch of a broad Call for Participation on September 21, 2009 comprised the final milestone of Wikimedia’s outreach plan. An appeal letter written by Wikipedia founder Jimmy Wales and Michael Snow was translated into 69 languages, highlighting the main ways in which volunteers could support the yearlong planning process. During the first year, in addition to online discussions, 31 face-to-face meetings took place in 19 different countries, from which minutes were shared online.²⁴

In terms of dialogical process, Phase 1 involved the creation of the technical infrastructure for facilitating dialogue along the requirements of the open strategy process, invitation of the parties to engage, and the emergence of broad dialogic themes. The dialogic funnel starts off widely at this stage and gradually narrows in subsequent stages.

Phase 2: Setting Dialogical Boundaries, Recruiting, and Performing Deep Dives

In October 2009, the Wikimedia Foundation appointed a Task Force Selection Committee comprising 15 members, whose core responsibilities were to define the mandates and to select suitable volunteers among over 1,923 applicants. These individuals would be accountable for reaching the aimed for outcomes by the 14 task forces related to the three emerging strategic priorities of sustainability, development, and accessibility as depicted in Table 4.²⁵

Table 4.

Strategic Priorities and Task Forces.

Sustainability	Development	Accessibility
Advocacy	Reader conversion	Offline
Financial sustainability	Expanding content	Local language projects
Alliances and partnerships	Wikipedia quality	Arabic
Community health	Technology	China
Movement roles		India

Table 5.

Comparative Parameters of Wikimedia’s Strategy Process and Dialogical Features.

Parameters	Traditional strategy process	Open source peer production	Wikimedia’s strategy process	Dialogical tensions
Leadership of process	Top-down, transactional	Relational, emergent	Relational, emergent, with boundaries set centrally	Facilitated open dialogue, but within “container,” aiming for particular types of outcomes
Communication flows/openness	Top-down, usually confidential process	Peer-to-peer, transparent	Peer-to-peer, top-down, transparent	Horizontal communications characterized by open dialogue, vertical communications guided by requirements of strategy
Structuring of process	Hierarchy	Network	Hybrid, both hierarchy and network	Network engenders open dialogue; hierarchy creates instrumental guidelines
Outcomes of planning process	Deliberate	Emergent	Hybrid, both deliberate and emergent	Strategy plan created, based on broad ranging dialogical engagement and delicate negotiation of tensions

Table 6.

Key Success Factors for Open Strategizing.

Organizing factors	Description	Effects on dialogical process
Recruit proactively	Identify relevant groups and invite their involvement. A higher number of engaged individuals mean more extensive and productive exchanges	Broadens inputs and expands perspectives engaged in process
Incentivize with symbolic rewards	Collaborative strategizing relies on voluntary participation, where social cues, relationship building and symbolic rewards are key. Monetary incentives, on the other hand, may foster competition rather than collaboration	Enhances participation to process in a way consistent with the values of the community
Actively moderate the process	Process design and coordination, and engaged relationship management can amplify positive network effects associated with open collaboration while minimizing the threat of unfavorable behavior	Ensures dialogic process occurs within parameters suitable to the intended outcome
Enable distributed leadership	Create micro-opportunities for volunteers to participate in designing and leading aspects of the process, which can lower the burden on organizational resources while serving as additional motivator for contributors	Enhances frequency and depth of dialogic process, creates opportunities for air time and individual inputs
Participate, listen and respond	Open collaboration does not negate management’s role in the process which remains crucial. Participation, active listening, and engagement with contributors goes a long way toward positive outcomes	Maintains energy and trajectory of process by active engagement, shows appreciation to contributors
Customize	There is no one-size-fits-all approach; customize the process to your own stakeholder context, organization culture, and network	Ensures dialogic process is consistent with features and demands of context

The committee categorized the applications by individuals who offered to participate in a task force, most of which were received from India, followed by the United States, the Russian Federation, the United Kingdom, and the People’s Republic of China. The criteria were availability, country, fluency, and specialty of applicants. Indeed, 30% of task force applicants were offering to invest 10 hours or more into the process on a weekly basis.²⁶

Ultimately, each task force was composed of 5 to 10 members, one of whom was in charge of the group’s coordinative and administrative tasks; the working language was English. Even though the work of task forces occurred transparently on strategy wiki (apart from the financial planning task force which met by conference call) and theoretically everyone was allowed to participate, task force members were considered “responsible and accountable [ . . . ] for seeing that the work is done and the deliverables are in on time.”²⁷ The task forces were advised to aim for decision making based on consensus among all participants. Nevertheless, individual group members with final decision-making powers had been identified at the beginning of the process to address situations in which reaching consensus was not feasible.²⁸ While receiving support from Wikimedia staff facilitators, the community members themselves were responsible for success or failure of the task forces.²⁹

Nevertheless and despite tremendous relationship building efforts by the project team,³⁰ the deep-dives phase was initially less successful than had been hoped for, with only 9 of the 14 task forces delivering recommendations and only 4 of those being of the quality that had been anticipated.³¹ Further engagement efforts then led to the required outputs. The foundation’s deliberation on the causes for the initial issues led to diverse explanations ranging from design failures of specific task forces,³² asking too much from volunteers,³³ resource limitations of the project team (P. Beaudette, personal communication, July 27, 2011), or driving away engaged volunteers by misframing the question.³⁴

In terms of dialogic processes, this phase involved the setting of dialogical boundaries, a narrowing down of the number of parties to the dialogue, and aimed for developing initial strategy outputs. The initial challenges with producing outputs of sufficient quality point toward the tension between the guided nature of the process that included task forces and particular deliverables under deadlines on the one hand, versus the more unstructured interaction process without any central shaping that characterized Wikimedia.

Phase 3: “Synthesis” of Dialogic Outputs

The publication of the task force recommendations marked the beginning of the third, “synthesis” phase of the strategic planning project that took place between January 2010 and April 2010. Community members were encouraged to translate the recommendations in as many languages as possible and engage into further deliberation. In February 2010, a group of 20 contributors that had demonstrated particularly deep involvement in the strategic-planning process formed a Strategy Task Force that was to serve as the central hub for further discussions. In order to maintain the openness of the process, everybody was free to join the task force through the end of February 2010. Membership, however, presupposed active participation and familiarity with a large amount of relevant background material. In four 3-week cycles and with the help of a guidelines page defining the characteristics of movement-wide goals,³⁵ the Strategy Task Force synthesized previous discussions, amounting to over 900 proposals and recommendations, into roughly 1,500 content pages, with a particular focus on the feasibility of their implementation. By the end of April 2010, this work had resulted in a first, rough draft for Wikimedia’s 5-year strategic plan.³⁶

In dialogic terms, this phase involved the intensification of the dialogic process among an even smaller of parties, application of particular criteria to the ideas such as feasibility, and the delivery of detailed strategic outputs. Arguably, at this stage dialogic interaction was even more structured by the requirements of the strategy process, with the dialogic funnel becoming tighter and converging on particular themes and specific contributors.

Phase 4: “Business Planning/Call to Action”—The Dialogic Funnel Expands but Under Defined Parameters

During the fourth and final part of Wikimedia’s strategic planning process the community was once more invited to review and refine the five emergent strategic priorities of quality content, innovation, increasing participation, growing readership, and stabilizing infrastructure. Also, the community was invited to debate the rationales underlying the priorities, key indicators, possible targets, and other measures.

In the last month of the project, Michael Snow and Jimmy Wales launched an official “Call to Action” to enthuse volunteers to implement the directions identified throughout the process:

As we bring this process to a close, our biggest challenge is to put these ideas into action. Here on Wikimedia’s strategic planning wiki, you’ll find a list of action opportunities organized around the priorities they support. We’d like to invite you to volunteer for and take ownership of these action opportunities.³⁷

As the strategic planning process came to an end, Wikimedia launched a dedicated celebration page and encouraged contributors to share their experiences and thoughts about the process, award barnstars to committed volunteers and post a Virtual Champagne Toast. Volunteers that had been deeply engaged in the project and representatives of the Wikimedia Foundation commented on a variety of topics.

The final, official document of the yearlong strategic planning process, titled “Wikimedia Strategic Plan: A collaborative vision for the movement through 2015” was distributed and published on strategy wiki in February 2011.³⁸ In correspondence after the completion of the strategic planning process, Jimmy Wales (personal communication, July 27, 2011) emphasized how “all [ . . . ] investment at the Foundation is now driven by the strategic plan” and noted that the project left him with the same insight that he got from Wikipedia itself many years ago: “people can cooperate and can produce great value with harmony as long as some fundamental values of mutual respect are used to set the ground rules.”

Discussion: Open Strategy-Making Through a Dialogic Perspective

We started this study by asking “what is the role of dialogue in open strategy processes?” In Wikimedia’s case, the broad scope and genuine commitment to engaging stakeholders in dialogue in an open strategy process on the one hand, interacting with the need to align initiatives and make decisions centrally about what strategies were desirable as well as feasible on the other hand, led to tensions. We believe that these tensions would be endemic to open strategy processes more broadly. This is because they juxtapose a dialogic process, with assumptions of equality among parties and emergence of joint visions and understandings, with the domain of strategy, a traditionally directive, structured endeavor. As our first contribution, therefore, we discuss the endemic nature of dialogic tensions in open strategy processes, and the need to understand and negotiate them constructively in order to accomplish desirable outcomes.

As our second contribution, we highlight the need for context sensitivity in dialogic theory as well as open strategy. “Ideal” features of dialogue are often noted in the literature, with the implicit assumption that these features are universal and applicable in all cases. We found instead that dialogue was guided and occurred differently, and had a different role, in each phase of the open strategy process. Key dimensions of open strategy such as transparency and inclusion (Whittington et al., 2011) may play out differently in different contexts, depending for example on the particular values of the organization and stakeholder expectations. Rather than conceiving of dialogue and open strategy as having universal features that would be applicable in any context, we need to start thinking of these as contingent, context-dependent processes; particularly when outputs such as robust strategic plans are sought as outcomes.

Our third contribution relates to implications for practice. The Wikimedia case poses valuable lessons for productively facilitating open strategy processes. We discuss six such lessons that arose from our case, and their effects on the dialogic process.

The Endemic Nature of Dialogic Tensions in Open Strategy Processes

The dialogic process of open strategy-making at Wikimedia Foundation was imbued with a variety of tensions. Flexibility and openness on the one hand versus structure and control on the other; broad participation of stakeholders versus selection of particular contributors to lead parts of the process and integrate ideas; aiming for a collective creation, but within clear, directed parameters. These tensions were occasioned by conditions of dialogue where interlocutors are seen as “equals within a conversational space” in ongoing interaction (Gergen et al., 2001, p. 705) on the one hand, intersecting with strategy as a structured, hierarchical practice with clear expected outputs (Chaffee, 1985) on the other hand.

We argue that managing such tensions is almost an existential condition of open strategy. On the one hand, the “ideal” conditions for dialogue involve information exchange by equals, self-reflexivity, and a joint ongoing exploration aiming to appreciate others’ points of view, build convergence, and to open new vistas of being and doing (Beech et al., 2010; Gergen et al., 2001; Gergen et al., 2004). On the other hand, strategy is an endeavor that involves tough choices between competing alternatives, resource commitment and alignment, defined timescales, and above all decision makers who have the authority and power to choose. The ideal conditions of dialogue and directive conceptions of strategy are different and potentially conflicting; yet open strategy processes bring them together in a precarious balancing act.

Wikimedia’s strategizing process employed both elements of open-based peer production as evidenced in offerings, such as Linux and WordPress, combined with a more centrally guided process. Despite the open nature of the process, central guidance was occasioned by the nature of strategy as a discipline that entails clear choices among competing alternatives and allocation of resources to realize those choices within a process of implementation and monitoring, none of which can be naturally or easily achieved by open-source communities.

Such tensions are not unique to open strategy processes but are characteristic of LGIs (Bunker & Alban, 2006) more broadly. As aforementioned, LGIs involve challenges of voice, structure, and perspective. With respect to voice, Wikimedia case demonstrates that through the use of technology, the voice of a multitude of stakeholders can be facilitated since there is much more leeway to contribute in comparison with an LGI that takes place within a limited amount of time such as a few days, with the stakeholders literally in the same room (e.g., Weisbord & Janoff, 2005).

With respect to structure, through a delicate balancing act of openness and participation but within directed parameters of process and required outcomes, Wikimedia negotiated the path between too little and too much structure. The online debates among Wikimedians on what many perceived as unprecedented direction by the center may have acted as steam valves to externalize such feelings but also gave Wikimedia the opportunity to explain why such direction was needed, given the need to develop a viable, implementable strategic plan.

Finally, with respect to perspective, and again via the affordances of technology in enabling continuous communication at a distance, a multitude of stakeholders were able to exchange their views and arrive at a commonly developed plan for Wikimedia’s future. By adopting the Wiki software, an already established technology that stakeholders were comfortable with, Wikimedia mitigated several of the risk factors mentioned by Denyer, Parry, and Flowers (2011) such as limited participation, cynicism, and political use of the technology.

The Wikimedia case shows that such use of technology could contribute to addressing some of the inherent dilemmas of LGIs in terms of enabling the balancing of such tensions. Technology as an enabler of the second-order theme of collaboration, and wikis as vehicles for the second-order theme of transparency, appear directly in our data (Figure 1). Technology also appears indirectly in our data in terms of the existence of self-organizing teams where members are dispersed around the world, where technology is the only way to accomplish such self-organization (a dimension of the second-order theme of coordination) and infrastructure provision by Wikimedia (a dimension of the second-order theme of context creation). Our findings suggest that LGI research could benefit from a more explicit focus on the role of technology in such processes; in particular, how technology can enable the balancing of inherent tensions in LGI processes.

“Ideal” dialogic conditions would be more likely found in open-based peer production where there is a lower degree of centralization rather than in traditional strategy processes. Dialogic tensions in open strategy processes are endemic, occasioned by the intersection of two practices (dialogue and strategy) that are not always mutually consistent in terms of their key features and expectations. Table 5 outlines key features of the Wikimedia process and relates these to the various dialogical tensions that were present.

We argued that tensions are an ontological feature of open strategy, given the juxtaposition between values of openness and participation, with the need for direction and choice demanded by the institution of strategy. In Wikimedia’s case, these tensions were constructively managed to lead to the desired outcomes. Further research could explore in more detail what kinds of tensions exist in different types of organizations, how tensions could be managed constructively, and what could potentially lead tensions to become destructive. Concepts such as organizational ambidexterity or paradoxes (Papachroni, Heracleous, & Paroutis, 2015) may be fruitful in offering avenues for exploring such questions. Furthermore, Gergen et al.’s (2001, p. 682) concept of “transformative dialogue” refers to “any form of interchange that succeeds in transforming a relationship between those committed to otherwise separate and antagonistic realities . . . to one where common and solidifying realities are under construction.” This concept may be particularly helpful to understanding and addressing tensions in contexts where stakeholders differ markedly in their views.

Context Sensitivity, Dialogic OD, and Emergence

Wikimedia’s open strategy process involved facilitated dialogue within the “container” to use Isaacs’s (1993) term, both horizontally and vertically, and aimed for particular outcomes (an actionable strategic plan fit for purpose, based on broad inputs) in accordance with the functional approach to dialogue (Heracleous & Barrett, 2001). Our analysis shows that the dialogic process of open strategy was shaped by the objectives of dialogue, which were to produce a strategy in an open manner. The literature suggests that effective dialogue is a meeting of equals (Gergen et al., 2001) involving “deep listening” and reframing (Marshak, 2004) and interspersed with both diagnostic and generative moments (Jacobs & Heracleous, 2005). When a directive, focused practice such as strategy intersects with dialogue; however, the dialogic process is placed in the context of, and shaped by the requirements of that practice, as illustrated by Figure 2.

Matzler et al. (2014) typologize Wikimedia’s process as aiming to facilitate strategy implementation (rather than strategy generation), and as engaging external (rather than internal) stakeholders. While indeed external engagement was crucial, Wikimedia’s employees were also actively engaged in contributing to and shaping the process. Furthermore, the process aimed both to generate strategy as well as to engage stakeholders in its implementation. As our analysis shows, Wikimedia’s open strategy process was broad-ranging in scope and hard to categorize in terms of a two-by-two framework.

Furthermore, Wikimedia’s long-standing values of transparency, collaboration, and participation shaped stakeholders’ expectations of how the open strategy process should be carried out. Open strategy processes in other contexts may be shaped by different values and stakeholder expectations. An implication for dialogic theory as well as for open strategy is the need for higher context sensitivity in order to appreciate to what extent and how particular dialogical features manifest in practice; and how such features may be observed or challenged when the circumstances are not necessarily conducive to their functioning. Such context sensitivity may potentially be accomplished through all three dialogic paradigms based on Heracleous and Barrett’s (2001) typology. However, the interpretive approach may hold the most promise in this respect, given its commitment to uncovering actors’ first-order perspectives and building on them to develop theory.

Taking the consideration of context sensitivity in dialogic theory further, Bushe and Marshak (2009) proposed the concept of “dialogic” OD to distinguish an emerging practice of OD centered on interpretive, social construction assumptions in self-organizing systems from the more traditional, rationalist OD practice in the context of planned change. Dialogic OD “emphasizes discourse, emergence and generativity” in local environments (Bushe & Marshak, 2014). As such, dialogic OD is inherently contextual and also consistent with the interpretive approach to dialogue (Heracleous & Barrett, 2001).

Bushe and Marshak (2009) suggested that differentiating dialogic from diagnostic OD would help us “avoid unknowing mixing and matching of diagnostic and dialogic practices that in combination may be inappropriate, out of alignment, or even counterproductive” (p. 363). Oswick (2009) however critiques this conceptual division between the two forms of OD as potentially counterproductive and suggests that these forms could be fruitfully combined in practice, beginning with dialogic OD to open up options, and moving to diagnostic OD to plan for change.

Our analysis has shown that Wikimedia’s process included elements of both diagnostic and dialogic OD but not in the sequence Oswick (2009) envisages. Initial diagnosis and data gathering was consistent with traditional diagnostic OD. The attempt to cocreate a future for Wikimedia through broad dialogue, creating and sharing a vision and common narratives, is consistent with dialogic OD in terms of its generative potential and emergence (e.g., Gergen et al., 2001; Gergen et al., 2004). Indeed, a sequence of first diagnostic and then dialogic OD is seen by Marshak and Bushe (2009, p. 382) as “less plausible.” Therefore, in addition to illustrating the value of a dialogic perspective in open strategy processes, we also explore how variants of OD may manifest in practice as Oswick (2009) recommended. Taking account of the context, it is likely that the sequence employed at Wikimedia was related to the need to converge toward an actionable strategy rather than just seek stakeholder input, where the dialogic objectives influenced the mix and sequence between diagnostic and dialogic OD.

The presence of emergence within a “container” of a guided process at Wikimedia is consistent with the literature on emergent strategy (Chia, 2014; Mintzberg & Waters, 1985), exhibiting characteristics of what Mintzberg and Waters (1985) referred to as “ideological” and “process” types of emergent strategy. In particular, the multitude of Wikimedians involved in the process share common values of openness, transparency, and collaboration; but these were only able to produce a strategy when balanced with structuring and process guidelines, bringing some structure to the diverse perspectives and inputs. Chia (2014) warns against a “direct confrontational approach to achieving desired outcomes” (p. 15) since it can lead to resistance and undesirable side effects. He recommends that we recognize the “inherent potentiality” (p. 19) already present in social reality that can allow change to happen. In Wikimedia’s case, this inherent potentiality was the active, engaged, participative multitude of Wikimedians, who were keen to contribute to developing the strategic plan through the orchestration provided by the adopted process.

Further research could take account of the fact that empirical studies in open strategy are still scarce. The present study of open strategy is the sole one to date from a dialogic perspective. It would be useful to know more for example about the features of open strategy and dialogic exchanges in organizations that have more traditional, hierarchical structures and values compared with Wikimedia. Given its inherent context sensitivity, a dialogic OD perspective (Bushe & Marshak, 2009) as well as more broadly dialogical processes that are explicitly recognized as contextually embedded (Gergen et al., 2004) could fruitfully be employed to study open strategy processes.

Furthermore, we do not yet know the degree to which open strategy is emergent or directed, and what shapes the balance of emergence versus direction. Are there certain elements that can be deliberate, for example, and others that can be emergent, and what shapes this mix? Do configurations of emergence and direction change in different contexts and how? Open strategy research is in its infancy and such questions merit further investigation.

Implications for Practice

Wikimedia’s experience holds lessons for organizations that have a large number of stakeholders with genuine interest in the organization’s future as well as the means and motivation to be involved. These might include virtual networks where digital communication and networking are a way of life; nonprofit organizations or public agencies providing public services who wish to gain or sustain legitimacy and involvement of stakeholders; purpose-driven networks or NGOs that bridge organizational and governmental boundaries such as health, education, or transportation; or communities, urban or metropolitan areas, regions, and states who wish to develop a common vision for the future by involving as many people as possible in the process.

In addition to the above, however, corporations in competitive markets could employ such a process to inform politically contentious or complex strategic decisions involving adaptive challenges, which would benefit from stakeholder involvement and from engaging collective wisdom. Corporations that display flat network structures, knowledge-sharing cultures, a general outward orientation, and a relational approach to leadership would be most suited to such an open strategizing process.

Organizations adopting an open approach to strategizing would need to find an appropriate balance between endemic tensions such as openness and control, or broad participation as well as selection. This could be achieved, as with Wikimedia, by setting the boundaries and objectives of the process, an overall framework to guide the process, involving a broad mix of contributors until synthesis, alignment, and decision making between competing options are needed, and retaining final decision-making rights. Such a process does require senior management to change their strategy paradigm toward an adaptive leadership (Heifetz & Laurie, 1997) one. From this perspective, leaders acknowledge that they do not have all the answers to wicked, unstructured problems, and invite the collective intelligence of a broader set of relevant actors to reach broadly informed directions.

An open strategizing approach also requires effective, motivated participants who are willing to invest time and energy in the process. In the absence of monetary rewards, a value-based foundation and a genuine interest in the future of the organization would be needed for sustained involvement. Views will vary significantly among contributors based on their personal beliefs, experiences, and stakeholder position. Thus, disagreements and conflicts are likely to occur and a process as well as common values would be needed to resolve them. A framework would be needed to ensure that contributions agree with the objectives of the organization, and would be feasible and appropriate. The shaping role of the center is crucial, leading to some hybrid elements of both networks and hierarchies.

Table 6 outlines some key factors that emerged from Wikimedia’s experience, and the corresponding dialogical effect of each of these factors.

The organizing factors that arose from our study such as incentives, participation, and distributed leadership are in fact productive features that can be employed more broadly in organization change processes (Kotter, 1995). The dialogic features that these organizing factors link to, aim to maintain the health and vitality of the dialogic process in terms of the ongoing breadth and depth of diverse contributions to open strategy or more broadly organization change processes. While the tensions that can be occasioned by the juxtaposition of dialogical processes employed in open strategy with the more directive and often nonparticipative practices of traditional strategizing are endemic and still need to be negotiated, such organizing factors can help keep tensions in check and make the process more productive.

Despite the challenges, the approach pioneered by Wikimedia holds promise as a way of engaging diverse groups of relevant actors in an organization’s strategizing process. Diverse viewpoints can enrich the discussions and implementation can be facilitated after broad engagement. An effective dialogic process of open strategy-making can enable organizations to engage collective intelligence (Heifetz & Laurie, 1997) to identify, refine, and address fundamental strategic challenges.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Andrews

K. R.

(1971). The concept of corporate strategy. Homewood, IL: Irwin.

Beech

MacIntosh

MacLean

(2010). Dialogues between academics and practitioners: The role of generative dialogic encounters. Organization Studies, 31, 1341-1367.

Bunker

Alban

(1992). Editors’ introduction: The large group intervention: A new social innovation? Journal of Applied Behavioral Science, 28, 473-479.

Bunker

Alban

(2006). Large group interventions and dynamics. In Gallos

J. V.

(Ed.), Organization development (pp. 309-321). San Francisco, CA: Jossey-Bass.

Bushe

Marshak

R. J.

(2009). Revisioning organization development: Diagnostic and dialogic premises and patterns of practice. Journal of Applied Behavioral Science, 45, 348-368.

Bushe

Marshak

R. J.

(2014). Dialogic organization development. In Jones

B. B.

Brazzel

(Eds.), The NTL handbook of organizational development and change (2nd ed., pp. 193-211). San Francisco, CA: Wiley.

Chaffee

E. E.

(1985). Three models of strategy. Academy of Management Review, 10, 89-98.

Chia

(2014). Reflections: In praise of silent transformation: Allowing change through “letting happen.” Journal of Change Management, 14, 8-27.

Chesbrough

(2010). Business model innovation: Opportunities and barriers. Long Range Planning, 43, 354-363.

10.

Chesbrough

H. W.

Appleyard

M. M.

(2007). Open innovation and strategy. California Management Review, 50(1), 57-76.

11.

Coghlan

(1998). The process of change through interlevel dynamics in a large-group intervention for a religious organization. Journal of Applied Behavioral Science, 34, 105-119.

12.

Denyer

Parry

Flowers

(2011). “Social,” “open” and “participative”? Exploring personal experiences and organizational effects of Enterprise 2.0 use. Long Range Planning, 44, 375-396.

13.

Dobusch

Mueller-Seitz

(2012). Strategy as a practice of thousands: The case of Wikimedia. Academy of Management Proceedings, 1, 43. doi:10.5465/AMBPP.2012.43

14.

Eisenhardt

K. M.

(1989). Building theories from case study research. Academy of Management Review, 14, 532-550.

15.

Forte

Larco

Bruckman

(2009). Decentralization in Wikipedia governance. Journal of Management Information Systems, 26, 49-72.

16.

Gergen

J. K.

Gergen

M. M.

Barrett

F. J.

(2004). Dialogue: Life and death of the organization. In Grant

Hardy

Oswick

Putnam

(Eds.), The SAGE handbook of organizational discourse (pp. 39-59). London, England: Sage.

17.

Gergen

J. K.

McNamee

Barrett

F. J.

(2001). Toward transformative dialogue. International Journal of Public Administration, 24, 679-707.

18.

Gioia

D. A.

Corley

K. G.

Hamilton

A. L.

(2012). Seeking qualitative rigor in inductive research: Notes on the Gioia methodology. Organizational Research Methods, 16, 15-31.

19.

Glaser

B. G.

Strauss

(1967). The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine.

20.

Grant

Marshak

R. J.

(2011). Toward a discourse-centered understanding of discourse and change. Journal of Applied Behavioral Science, 47, 204-235.

21.

Hammond

S. C.

Anderson

Cissna

K. N.

(2003). The problematics of dialogue and power. Communication Yearbook, 27, 125-158.

22.

Hatch

M. J.

Ehrlich

(2002). The dialogic organization. Transformative Power of Dialogue, 12, 107-131.

23.

Heath

R. G.

(2007). Rethinking community collaboration through a dialogic lens. Management Communication Quarterly, 21, 145-171.

24.

Heifetz

R.A.

Laurie

D. L.

(1997). The work of leadership. Harvard Business Review, 75(1), 124-134.

25.

Heracleous

Barrett

(2001). Organizational change as discourse: Communicative actions and deep structures in the context of IT implementation. Academy of Management Journal, 44, 755-778.

26.

Huston

Sakkab

(2006). Connect and develop: Inside Procter & Gamble’s new model for innovation. Harvard Business Review, 84(3), 58-66.

27.

Isaacs

W. N.

(1993). Taking flight: Dialogue, collective thinking, and organizational learning. Organizational Dynamics, 22(2), 24-39.

28.

Jacobs

Heracleous

(2005). Answers for questions to come: Reflective dialogue as an enabler of strategic innovation. Journal of Organizational Change Management, 18, 338-352.

29.

Kersten

(2000). Diversity management: Dialogue, dialectics and diversion. Journal of Organizational Change Management, 13, 235-248.

30.

Konieczny

(2010). Adhocratic governance in the Internet age: A case of Wikipedia. Journal of Information Technology & Politics, 7, 263-283.

31.

Kotter

J. P.

(1995). Leading change: Why transformation efforts fail. Harvard Business Review, 73(2), 59-67.

32.

Marshak

R. J.

(2004). Generative conversations: How to use deep listening and transforming talk in coaching and consulting. OD Practitioner, 36(3), 25-29.

33.

Marshak

R. J.

Bushe

G. R.

(2009). Further reflections on diagnostic and dialogic forms of organization development. Journal of Applied Behavioral Science, 45, 378-383.

34.

Matzler

Fuller

Koch

Hautz

Hutter

(2014). Open strategy: A new strategy paradigm? In Matzler

Pechlaner

Renzl

(Eds.), Strategie und Leadership (pp. 37-55). Wiesbaden, Germany: Springer.

35.

Mauthner

N. S.

Parry

Backett-Milburn

(1998). The data are out there, or are they? Implications for archiving and revisiting qualitative data. Sociology, 32, 733-745.

36.

McCambridge

(2003). 12 Angry men: A study in dialogue. Journal of Management Education, 27, 384-401.

37.

Mintzberg

(1971). Managerial work: Analysis from observation. Management Science, 18, 97-109.

38.

Mintzberg

Waters

J. A.

(1985). Of strategies, deliberate and emergent. Strategic Management Journal, 6, 252-272.

39.

Morell

M. F.

(2011). The Wikimedia foundation and the governance of Wikipedia’s infrastructure: Historical trajectories and its hybrid character. In Lovin

Tkacz

(Eds.), Critical point of view. A Wikipedia reader (pp. 325-341). Amsterdam, Netherlands: Institute of Network Cultures.

40.

Oswick

(2009). Revisioning or re-versioning? A commentary on diagnostic and dialogic forms of organization development. Journal of Applied Behavioral Science, 45, 369-374.

41.

Oswick

Anthony

Keenoy

Mangham

I. L.

(2000). A dialogic analysis of organizational learning. Journal of Management Studies, 37, 887-901.

42.

Papachroni

Heracleous

Paroutis

(2015). Organizational ambidexterity through the lens of paradox theory: Extending the research agenda. Journal of Applied Behavioral Science, 51, 71-93.

43.

Powley

E. H.

Fry

R. E.

Barrett

F. J.

Bright

D. S.

(2004). Dialogic democracy meets command and control: Transformation through the appreciative inquiry summit. Academy of Management Executive, 18(3), 67-80.

44.

Putnam

L. L.

Fairhurst

G. T.

(2001). Discourse analysis in organizations: Issues and concerns. In Jablin

F. M.

Putnam

L. L.

(Eds.), The new handbook of organizational communication (pp. 78-136). Thousand Oaks, CA: Sage.

45.

Pye

(1995). Strategy through dialogue and doing: A game of “Mornington Crescent”? Management Learning, 26, 445-462.

46.

Raelin

J. A.

(2012). The manager as facilitator of dialogue. Organization, 20, 818-839.

47.

Schein

E. H.

(1993). On dialogue, culture and organizational learning. Organizational Dynamics, 22(2), 40-51.

48.

Skordoulis

Dawson

(2007). Reflective decisions: The use of Socratic dialogue in managing organizational change. Management Decision, 45, 991-1007.

49.

Stieger

Matzler

Chatterjee

Ladstaetter-Fussengegger

(2012). Democratizing strategy: How crowdsourcing can be used for strategy dialogues. California Management Review, 54(4), 44-68.

50.

Von Krogh

Rossi-Lamastra

Haefliger

. (2012). Phenomenon-based research in management and organisation science: When is it rigorous and does it matter? Long Range Planning, 45, 277-298.

51.

Weisbord

Janoff

(2005). Faster, shorter, cheaper may be simple: It’s never easy. Journal of Applied Behavioral Science, 41, 70-82.

52.

West

Bogers

(2013). Leveraging external sources of innovation: A review of research on open innovation. Journal of Product Innovation Management, 31, 814-831.

53.

Whittington

(2015). The massification of strategy. British Journal of Management, 26, S13-S16.

54.

Whittington

Cailluet

Yakis-Douglas

(2011). Opening strategy: Evolution of a precarious profession. British Journal of Management, 22, 531-544.

55.

Worley

C. G.

Mohrman

S. A.

Nevitt

J. A.

(2011). Large group interventions: An empirical field study of their composition, process, and outcomes. Journal of Applied Behavioral Science, 47, 404-431.

56.

Yin

R. K.

(2014). Case study research: Design and methods. Thousand Oaks, CA: Sage.

57.

Zoller

H. M.

(2000). A place you haven’t visited before: Creating the conditions for community dialogue. Southern Communication Journal, 65, 191-207.

Open Strategy-Making at the Wikimedia Foundation: A Dialogic Perspective

Abstract

Keywords

Introduction

Open Strategy-Making and the Dialogic Perspective

Open Strategy-Making as a Large Group Intervention

A Dialogic Perspective and Open Strategy-Making

Methodology

Wikimedia as Collective Creation: Community and Collaboration

Wikimedia’s Open Strategy Project

Developing Strategy the Wikimedia Way: Transparency, Participation, and Collaboration

Designing the Process: Openness and Flexibility Versus Structuring From the Center

Phase 1: “Level-Setting” 18 —Preparing the Infrastructure and Inviting Parties to Dialogue

Phase 2: Setting Dialogical Boundaries, Recruiting, and Performing Deep Dives

Phase 3: “Synthesis” of Dialogic Outputs

Phase 4: “Business Planning/Call to Action”—The Dialogic Funnel Expands but Under Defined Parameters

Discussion: Open Strategy-Making Through a Dialogic Perspective

The Endemic Nature of Dialogic Tensions in Open Strategy Processes

Context Sensitivity, Dialogic OD, and Emergence

Implications for Practice

Footnotes

Declaration of Conflicting Interests

Funding

Notes

References

Phase 1: “Level-Setting”¹⁸—Preparing the Infrastructure and Inviting Parties to Dialogue