Designing the digital organization

Organizations adapt to the needs of the market; to the technologies available for the design, production, and delivery of products and services; and in their means of organizing (Miles and Snow, 1978). The digitization of society is affecting customer needs, product and service properties and delivery mechanisms, and organization design (Langer, 2017). Changes across these areas are coalescing. For example, music has changed from a product to a service industry wherein consumer demand has evolved from purchasing packaged CDs to accessing customized, streamed playlists which are organized by aggregators such as iTunes, Tidal, Spotify, and Pandora, with distribution driven by artists instead of studios. “Products as services” is a business model that is growing in many arenas (Porter and Hempelmann, 2014).

Technological innovations can be incremental or disruptive. Incremental innovations are characterized by small improvements to existing products, services, and processes. Disruptive innovations, on the other hand, may create new markets and business models, and often may displace market leaders (Christensen, 1997; Christensen and Bower, 1996). In past decades, organizational responses to technological changes were mostly incremental and, in part, enabled by IT improvements that allowed greater scope and dimensionality of organizational control and coordination. Most of those adaptive responses were made within existing hierarchical forms of organizing (Altman et al. 2015). Digital technologies, however, often disrupt established ways of organizing and require adaptation through collaboration as well as self-organization around situation awareness (Endsley, 2000) and knowledge commons (Hess and Ostrom, 2006; Ostrom, 1990, 2010). Self-organization and collaboration, as an adaptive response, is faster and more effective than a hierarchical response.

The technological manifestations of disruption in organization design are clearly visible, as are workplace changes and changes in inter-organizational relationships. What is less visible are changes in the associated design paradigm, which enable organizations to obtain efficiency and effectiveness improvements by investing in digital technology. The new organization design principles are similar to those used in designing digital technologies themselves. Their roots are found in object-oriented systems design (Dahl and Nygaard, 1966) and in the architecture of the Internet (Krol, 1993). In organizational terms, these principles are embodied in actor-oriented architectures (Fjeldstad et al. 2012).

Digital technologies play a role in all aspects of operating, controlling, and coordinating the activities of organizations (Setia et al. 2013). Broadly speaking, they are used for automating and augmenting tasks, communicating internally among organization members and externally with customers and partners, and in collaborative decision-making among digital and human agents (Davenport and Kirby, 2015; Engelbart 1962; Huber, 1990; Licklider, 1960; Simon, 1973). At Tesla’s manufacturing facility in Fremont, California, technicians work alongside 185 robots made by the German firm Kuka Robotics to assemble the electric cars. By using artificial intelligence “reinforcement learning algorithms,” the robots are able to switch tools and perform certain tasks far better and faster than their human co-workers (Gershgorn, 2016). Surgeons at the Mayo Clinic use robots to augment a variety of surgical procedures in heart, head, and neck operations. The surgeons perform those operations by controlling surgical micro-instruments attached to robotic arms (Mayo Clinic, 2016).

Both intra- and inter-organizational transactions and communications have been performed digitally for a long time (Fedorowicz and Konsynski, 1992). Walmart exemplifies a highly digitized supply chain connecting its stores, distribution centers, and suppliers (Mata et al. 1995). Currently, social media such as Facebook and Twitter are used by companies to communicate with their customers and other stakeholders, and digital platforms such as Facebook at Work and Microsoft’s SharePoint allow for internal communication and for collaboration with partners.

Digital technologies are also used for learning, decision-making, and design. E-commerce companies such as Amazon, Google, Airbnb, and Uber study the data trails of consumer behavior to design markets for greater efficiency and build new markets (Lohr, 2016). Intelligent digital design tools are used in engineering and creative industries. Those tools typically offer 3D representation of the objects under design, and they allow designers to simulate the operations and performance of alternative design choices (Fujitsu, 2016). In semiconductor manufacturing, the designs are digitally transmitted to equipment that manufactures the product. With continuing development and wider adoption of 3D printing technologies, the design-to-manufacturing process will become fully digital across many more industries (Sasson and Johnson, 2016). MTR Corporation, which owns and operates the Hong Kong subway system, uses artificial intelligence to schedule maintenance tasks. In a typical week, more than 10,000 people carry out 2,600 work orders. The system was “trained” by using expertise extracted from human experts and then transformed into work rules. The main difference between normal software and MTR’s artificial intelligence is that it contains human knowledge that takes years to acquire through experience (Hodson, 2014). Some companies employ digital design tools in collaborating with their customers and partners. Lego provides toolkits on its website that enable entrepreneurs and customers to submit product ideas and start new Lego brick-based businesses (Heinerth et al. 2014). (See Table 1 for a summary of digital applications and leading digital firms and organizations.)

An organization is a goal-directed, boundary-maintaining activity system (Aldrich and Ruef, 2006). In traditional organizations, technological artifacts such as manufacturing equipment and computers are controlled by human operators. With the declining costs of global communication and information processing, hierarchy is being replaced by radically different ways of organizing (Fjeldstad et al., 2012), the digital elements of which include cloud computing, big data analytics, cognitive computing, and collaboration platforms. Artificial intelligence embedded in machinery and tools, as exemplified above, plays an ever-larger role in emerging digital organizations (for a survey see Kolbjørnsrud et al. 2016). As a result, employees collaborate with, rather than merely control, the technology in use, and organizational designs have to encompass both human and digital agents. Organizing digitally means collaboration with more entities and less reliance on hierarchy for control and coordination. It also entails empowering employees, partners, and customers who use digital tools for the co-creation and co-production of products and services as well as providing digital platforms for self-organized collaboration (Boudreau et al. 2011).

Actors – whether they are individuals, teams, or firms – must possess the capabilities and values to self-organize. They engage in self-management rather than wait to respond to directions received from the hierarchy. They also act with integrity, developing a reputation for consistent, competent behavior – social capital that can be used at other times and in other arenas to develop new relationships and business opportunities. The trust that is built up between actors saves the costs of designing and using elaborate control mechanisms that monitor and ensure proper actor behavior. In an actor-oriented organization, actors understand the overall structure and processes of the organization, and their decisions and actions are taken in pursuit of the organization’s common good. By focusing on the common good, actors can take advantage of shared values, norms of reciprocity, and trust in the self-governance process (Hess and Ostrom, 2006; Ostrom, 1990, 2010).

Commons refers to resources that are collectively owned and available to actors (members of the organization). One type of commons is a knowledge commons, a repository of knowledge that organization members can both contribute to and use. For example, Blade.org, a collaborative community of more than 70 firms in the computer server industry, posted all of the solutions developed by member firms on the organization’s website. Any Blade.org member firm could access the website and examine the solutions for ideas that might apply to its particular market or customer base (Snow et al. 2011). Smart Aarhus, the smart-city initiative of Aarhus, Denmark, has a database called Open Data Aarhus whose purpose is to make relevant data and information accessible to Aarhus citizens and organizations. These datasets can be used to develop new products, services, and digital applications. Any firm or individual citizen can access the more than 75 datasets in Open Data Aarhus and use the data for collective purposes (Snow et al. 2016).

Shared situation awareness is a commons that facilitates self-organization. Situation awareness refers to knowing what is going on in the organization (Endsley, 2000). Digitally shared situation awareness provides an up-to-date portrait of problems and opportunities in the organization’s environment as well as the current availability of resources to address those problems and opportunities. Through digitally shared situation awareness, valuable information is widely available to organization members in their decision-making. Originally developed as an operational tool for fighter pilots in World War I, situation awareness is used today in air traffic control, power plants, and advanced manufacturing systems (Endsley, 2000), and its use is growing in healthcare and other sectors. When actors share an up-to-date awareness of the organization’s situation, everybody in the organization can make the right decision or take the correct action without seeking direction or authorization from the hierarchy.

Infrastructures connect actors and allow access to the same information, knowledge, and other resources. Actors who have the knowledge, information, tools, and values needed to set goals, and who can assess the consequences of potential actions for the achievement of those goals, can self-organize. Self-organizing actors use protocols to guide their collaborative interactions. Protocols are “codes of conduct” used by organizational actors in their exchanges and collaboration activities. An important category of protocols deals with the division of labor, the mobilization and linking of actors for a particular project or task. Examples are protocols by which actors advertise problems or opportunities as well as their own capabilities and availability, and protocols by which actors search for potential collaborators.

In summary, actor-oriented organizational architecture is focused on the organization’s actors: the work that they do, and the principles and processes by which they relate to one another. Competent actors working for the common good of the organization can self-organize and self-manage with only minimal need for hierarchical control and coordination. Commons, infrastructures, and protocols are used to guide and support actor behavior, connecting organization members with one another and facilitating their work activities.

A competent actor is one who possesses the knowledge, skills, and values suited to an actor-oriented system. In building a digital organization, the effective composition and mobilization of a set of competent actors may require a combination of selection, training, mentoring, and replacement of personnel. An actor-oriented digital organization is especially conducive to use by millennials (people born after 1980) who have acquired knowledge and expertise from their Internet activities (Langer, 2017: Ch. 10), and it may be difficult to use by employees who lack social media skills and who have been ingrained with hierarchical approaches to organizing and managing (Espinoza and Ukleja, 2016; Saxena and Jain, 2012).

In 2015, millenials became the largest generation in the U.S. workforce, and by 2025 they will constitute 75% of that workforce (Meister and Willyerd, 2010). A forecasted skill set for the digital-age workforce is shown in Table 4. As shown, the digital organization will require its members to have a demanding set of both hard and soft skills. Hard skills include computational thinking and trans-disciplinarity. Soft skills include social intelligence, cross-cultural competency, and the ability to collaborate. To capitalize on these skills, the digital organization must provide flexible workspaces and policies that motivate millennials and enable them to be productive. Such workspaces will include appropriate collaboration tools and be designed according to sound psychological theories and principles. It is well established that people are extrinsically motivated by reward systems, evaluations, and the opinions others have of them (Herzberg, 1966). Self-determination theory (Ryan and Deci, 2017) holds that just as frequently people are motivated from within – by their interests, curiosity, concern for others, and abiding values. Following self-determination theory, jobs must be designed so that they meet the core psychological needs of competence, relatedness, and autonomy. Such designs foster the most volitional and high-quality forms of motivation and positive work outcomes, including enhanced performance, persistence, and creativity. Conversely, the degree to which any of these basic psychological needs is unsupported or thwarted in an organizational context will have a detrimental impact on people’s well-being and performance in that setting.

The individual capabilities of actors must be turned into collective capabilities in order for the organization to operate at its desired scale and speed. The development of organizational capabilities occurs through a managed learning process in which individuals, technology, and organizational culture evolve together (Langer, 2017). At Valve Corporation, a digital distribution platform that operates with few hierarchical mechanisms, organizational capabilities are developed along actor-oriented lines (Felin and Powell, 2016). First, Valve recruits individuals who, in their estimation, have the capacity to create value in a marketplace of ideas. Second, those actors are allowed to self-select the projects on which they want to work. Third, a new project can be started if at least three peers agree that it is worthwhile (the “rule of three”). Fourth, project teams are empowered to sense, shape, and seize their own market opportunities. It is expected that teams will engage with external stakeholders via open innovation methods such as crowdsourcing, user communities, innovation contests, and so on.

Historically, organizations have been populated by humans using tools and equipment to accomplish their tasks. Increasingly, human actors work collaboratively with digital agents such as robots, adding a digital actor to the organization. Newer robots with abilities in social interaction are able to learn from their human counterparts through cooperation and tutelage (Green et al. 2008). Effective collaboration requires human and digital agents to share goals and situation awareness as well as the capability to communicate directly with one another.

Designing commons for a digital organization will be specific to each organization and its needs, but two commons in particular deserve attention: situation awareness and knowledge. To be effective, actors need a shared awareness of the resources and activities in their environment. For example, in the self-dialysis clinic at the Ryhov Hospital in Sweden, all dialysis patients share a common electronic calendar that allows them to schedule their own treatment sessions. In addition, the equipment of the center is designed in a way that allows patients to perform their own treatment. The roles of the actors in this example are different from those of a hierarchically organized treatment center. Here patients self-organize their treatment while doctors and nurses use shared information about the patient’s condition to intervene only if necessary (IHI Annual Report, 2012). Functionality that supports shared situation awareness is emerging across a wide variety of software platforms. Such platforms are extensively used by military organizations and advanced technology firms for collaborative problem solving and decision-making (Kolfschoten and Briggs, 2015; Nunamaker et al. 2009). Several recent software platforms support project management, and office productivity tools increasingly support real-time collaboration around documents and spreadsheets. Galbraith (2010) provides an account of how Procter & Gamble uses a collaborative spreadsheet program to speed up the reconciliation of corporate plans and budgets.

A knowledge commons is information and data that is shared by the actors who make up the organization. It is typical to distinguish between declarative and procedural knowledge (Kogut and Zander, 1992). Declarative knowledge is factual whereas procedural knowledge is know-how. For example, the Open Source Drug Discovery community maintains declarative knowledge in a repository of all scientific discoveries made by its members (Kolbjørnsrud, 2017). This allows all contributors to have a real-time awareness of the problem state (Newell and Simon, 1972) – how far the members of the community have come toward a solution and the knowledge gaps that still need to be closed. Many consulting firms, such as Accenture, maintain procedural knowledge of the best practices used by their consultant teams (Langer and Yorks, 2013). The overall purpose of a procedural commons is to provide a set of shared resources that helps an organization learn and adapt.

In a given digital organization, all actors may not use the entire commons. Some software platforms allow the commons to be subdivided such that actors can collaborate around selected subsets. Subdivision of the commons permits actors to use only those resources relevant to their task and situation.

Guiding interactions among actors and accessing commons require protocols that reduce ambiguity and increase the effectiveness and efficiency of interaction. There are self-organizing processes associated with each protocol that actors follow in order to achieve control and coordination. In a digital organization, actors use their collaborative tools to signal both the tasks to be done and the availability of resources to perform them. Although protocols are specific to the type of collaboration, there are commonalities. Military and emergency response platforms, for example, have protocols and processes for publishing updates to the situation awareness database and subscribing to particular types of information, such as new events occurring in a geographic area (DeMarco, 2016; Liang and Gao, 2010). Similarly, platforms for collaborative project management (Chen et al. 2003) have protocols for inviting collaborators, sharing information about the whole project or particular tasks, and alerting collaborators to changes affecting their portion of the project (c.f. www.​smartsheet.​com). The infrastructure that allows the actors to connect with one another consists of communication networks and computer servers.