Lagomizing, organic integration, and systems emergency wards
innovative practices in managing complex systems development projects
Jack Järkvik Management Ltd
Department of Management and Engineering, Linköping University
BI Norwegian School of Management, and Department of Management and Engineering,
In complex systems development, project management is a key factor for innovation, for bringing together system capabilities to actually working systems and taking them to the customer. The critical question then is: How can successful project management in this field be conceptualized, practiced, and understood? In the extant literature there is a plethora of suggested tools for advanced planning and scheduling, for system decomposition and modularization, for reducing interdependencies and avoiding errors. There is also a growing criticism of these “planning approaches,” suggesting various contingency and flexibility approaches, to reflect and adapt to complexity and change. This critique, however, tends to lack grounded suggestions for effective managerial practices and does not distinguish between general flexibility needs and specific project structures required to make complex systems development at all possible.
This paper centers on the development of large, complex systems with an empirical focus on the telecom industry. Key challenges here, it is argued, are to understand complexity, reduce complexity, and rapidly act on the consequences of complexity to ensure timely delivery of reliable and predictable systems to highly demanding customers. To cope with these challenges, a set of innovative practices has been developed within Ericsson, a world leader in mobile network systems. We focus on three such practices which together represent examples of a “neo-realistic” approach to project management: (1) “Lagomizing,” a top-down redefinition of project goals to reduce complexity and transform expectations; (2) “Organic integration,” an articulation and visualization of a shared understanding of system capabilities; and (3) “Systemakut,” the Systems Emergency Ward, a real-time, high-visibility agora for managing integration, handling errors, and making swift decisions and in public. The study is based on a research methodology involving knowledge co-production, where the team of authors represents both academic knowledge and practitioner experience of managing innovations in complex systems development projects.
Keywords: project management, integration, practice, knowledge co-production, neo-realism, systems development, complexity.
The Need for “A Practice Turn” in Studying Complex Systems Development
During the last decade, research in project management has developed considerably, introducing more elaborate analyses of, for instance, the evolution of project practice (Brady & Davies, 2004), the management of complexity and uncertainty in projects (Shenhar, 2001), and various contingency-inspired approaches (Dvir, Lipovetsky, Shenhar, & Tishler, 1998; De Meyer, Loch, & Pich, 2002, Pich, Loch & De Meyer, 2002). A number of these studies have criticized “mainstream” models of project management for an over-emphasis on the role of planning and scheduling and highlighted the need for developing models that take into account flexible and adaptive approaches (Shenhar & Dvir, 2007). These studies have singled out the importance of fitting project management to the situation and working out adaptive contingency formulae as critical for firm-level competitive advantage. This critique, however, tends to be overly general in character and lacks grounded suggestions for effective managerial practices. The Rethinking Project Management Network recently emphasized that “the current base of project management continues to attract criticism for its lack of relevance to practice” (Winter, Smith, Morris & Cicmil, 2006, p. 638). Scholars seem to disagree whether the fundamental problem lies at the practice side or within academia—if practitioners are ignoring the models and theories produced by researchers, or if academics ignore the complexity of practice and cognitive limitations of practitioners.
In this paper, we build on the experience of complex systems development to further analyze the limits of “the rational planning approach” but in more specific and concrete ways than is usual in the critique of planning models. We concentrate on two issues. First, we suggest a practice-oriented approach for handling the identified limitations. Thus the paper addresses the need of a “practice turn,” which has been advocated in several recent studies of project management (Blomquist, Gällstedt, Hällgren, Nilsson, & Söderholm, 2006; Söderlund, 2005a). We do this on the basis of the idea of knowledge co-production, where researchers collaborate with practitioners to articulate findings and try out solutions that conform to the dual criteria of rigor and relevance, rigorous in relation to academic research and relevant in relation to managerial practice (Nowotny, Scott, & Gibbons, 2001). Second, we explore the importance and practices of controlled flexibility. Contrary to much earlier research that has argued for the prime need of increasing flexibility, in the type of complex systems development projects analyzed in this paper, the managerial challenge is rather to build stability, to implement semistructures and standardized approaches (cf. Brown & Eisenhardt, 1997). However, previous research has only limitedly explored how these semistructures work in practice and how project managers use them. Thus, the paper highlights project management practices that build stability: for reducing and handling complexity and for controlling the consequences of complexity.
The paper is structured in the following way. In the next section, we discuss the close and important link between projects and innovation, and introduce the theme of innovative approaches to project management. We then present three major challenges of the management of complex development projects—challenges that we believe are particularly pressing in the telecom industry. After this, we outline the research methodology and steps taken to establish a mode for knowledge co-production where representatives from practice and research collaborate to generate findings that are both relevant and rigorous. The empirical part of the paper follows. Here we focus on the neo-realistic approach and three innovative practices of project management. The paper continues with a theoretical interpretation of the neo-realistic approach and the three selected practices. The paper ends with theoretical implications and main conclusions.
Linking Projects with Innovation
The intersection between innovation and projects is important due the intimate link between product and process innovation (see e.g. Van de Ven, Polley, Garud, & Venkataraman, 1999). For example, Hughes (1998) in his account of monumental projects documented the significance of projects for technological and product innovation as well as process innovation. In many ways, Hughes' analysis points out the necessity of studying project management and organizational practice in more detail, not only to improve our understanding of project management, but perhaps more notably, the knowledge about innovation in general. Researching the link between projects and innovation is important in a wide range of contexts. For instance, the launch of a new technology may require an innovative way of product development; and embedded technologies may require integrated project management designs—in other words, the ability to innovate project management becomes critical for technology innovation. A company's sustainable competitive advantage might in many contexts be associated with its capabilities in generating new ways of organizing its projects. Similar observations have been reported in studies of project capabilities and project competence (see e.g. Davies & Brady, 2000; Davies & Hobday, 2005; Söderlund, 2008). Therefore, in this paper we focus on what we call “innovative practices of project management”—a field that so far largely has been dominated by technique-oriented studies often with no clear empirical basis.
We focus on the way project management is executed in a particularly demanding area, namely that of complex systems development in the telecom industry. Such a focus, we submit, will not only contribute to an improved understanding of projects and project management, but also, ultimately, to the general knowledge of managing complex innovation. Nightingale (2000) has illustrated the problems of managing complex systems development, the costs of “redesign feedback loops” and the value of avoiding such costly redesigns. In line with much conventional thinking he also stressed the role of making sure that the “design matches its specifications” and that the “design specifications are correct” (Nightingale, 2000, p. 927). In this paper we will show the limitations of this and similar “planning approaches” and by describing three innovative managerial practices demonstrate the feasibility of alternative methods. In several ways, the approach discussed here resembles ideas presented in Eisenhardt & Tabrizi (1995) of how to manage highly uncertain development projects: the importance of avoiding excessive computerized planning, the value of rapid development cycles to create shared understanding, and the role of strong project management. This paper, however, is more clearly focused on the particular problems of complex systems development, and more detailed in discussing the relevant managerial practices to deal with these problems. In particular, we stress the importance of reducing complexity by transforming expectations, of articulating specific visual images of interdependencies as a basis for systems integration; and of creating highly visible public arenas for handling the unknown amount of errors in these types of development projects.
According to Miller and Lessard (2007, p. 2), “…projects are dynamic, iterative, and often chaotic systems, and project management architectures must reflect this.” Our view is very different: project management in complex technology areas should not reflect, but aim at controlling complexity: understand complexity, reduce complexity, and create mechanisms to make it possible to rapidly act on the consequences of complexity. Thus our approach is different both from the planning approach and its theoretical opposition, the flexibility and adaptation model.
Innovative Approaches to Project Management
Reviewing the many articles within the field of project management reveals one striking thing: There is no lack of ideas on how to improve practice. This observation is true for both the academic literature and for practitioner-oriented books. Several novel approaches such as SCRUM and Agile Project Management have been suggested (Highsmith, 2004). In several ways these methods have been driven by technological development and opportunities within the software industry—of rapid prototyping and compilation of beta versions. From a research-oriented perspective, however, relatively little of these new practices are based on well-founded empirical analyses. In our view, innovative approaches to project management constitute a field of inquiry that requires constant research as firms and managers elaborate on new ways of organizing, of trying out new concepts, and of transferring ideas from one context to another to meet new challenges.
The practice-turn subscribed to in this paper has gained support in the recent debate on project management research. For instance, scholars within the Rethinking Project Management Network summarized their main findings in a special issue of the International Journal of Project Management (Maylor, 2006) and argued for the need of studying the “actuality of project-based working and management” (Cicmil, Williams, Thomas, and Hodgson (2006: 675). Winter, Smith, Morris, and Cicmil (2006), in the same issue, emphasized the importance of breaking away from the “hard systems model” with its focus on planning and controlling to develop new theories for practice: “new images, concepts, frameworks and approaches to help practitioners actually deal with…complexity in the midst of practice” (Winter et al., 2006, p. 643). In project management textbooks, the hard systems model still has a prominent role in the recommended practice for everyday project management. A major problem, however, with the textbook theory of project management is its lack of realism and relevance in fast-changing businesses. The need for understanding the real management challenges have also been reported by organization and management scholars studying related industries (Brown & Eisenhardt, 1997).
Doing research on innovative project management in innovative ways constitutes a dual challenge. Our way of responding to this is (1) to focus on leading-edge practices within one particularly important player in the global telecom industry—an industry characterized by furious competition, complex systems innovation, and high requirements on project management capabilities; and (2) to do the research and writing together in a team consisting of two academic researchers (Berggren and Söderlund) and one senior project manager (Järkvik) with decades of experience as a management innovator and development leader in Ericsson's efforts to try out new forms of managing business-critical projects. Our ambition is to show how this type of co-production of knowledge can contribute to the understanding of innovations in project management, both as a theoretical and practical field. As mentioned earlier, we believe that such studies can also enhance the general understanding and knowledge of how firms develop project capabilities (Davies & Brady, 2000).
Complex Systems Development: Three Challenges
A number of empirical reports have documented the managerial challenges of complex systems development (Adler, 1999). The sheer number of involved technologies, functionalities, and knowledge bases poses difficult challenges for management and the integration capability of the firm. Furthermore, changes in customer requirements and the difficulties of handling them in complex multi-functional development organizations add to the integration challenge. Highly complex systems, customer demands on rapid delivery, and the need for interaction and integration of a multitude of knowledge bases make the telecom industry particularly interesting for a study of innovative project management practices. Managing systems development in this industry could be seen as a test bed for trying out new ways of organizing projects in much the same way as the defense and aerospace industries were in the 1960s (Morris, 1994; Sapolsky, 1972), and the automotive industry was in the 1980s (Womack, Jones, & Roos, 1990). The complex image of systems development in the telecom industry could be divided into three major challenges that project management must respond to:
- First, rapid technical development and intense competition mean that contract specifications and customers' initial technical expectations seldom square with what is possible to do within the delivery schedule. This could easily result in costly delivery delays or, perhaps even worse, a timely delivery of a defect system, which is later thrown out by the customer. Moreover, research has also shown that time is absolutely crucial and that speed is a critical competitive factor to generate new business and to be able to stay at the technological edge (see e.g. Eisenhardt & Tabrizi, 1995).
- Second, there is the basic problem of interdependencies, of understanding interdependencies, and getting project members to share a common view of relevant interdependencies and how they will evolve. Some of this complexity can be taken out by modularization efforts (Sanchez & Mahoney, 1996), but despite all the extensive attempts to separate and isolate sub-components, research within the telecom industry in particular has shown “the existence of a high number of elements, interdependencies, and interfaces that cannot and do not need to be decomposed to a level of independent units” (Yakob & Tell, 2007, 390). In many cases, the problem of tight interdependencies and complexity appear at the project level and need to be handled by its management team.
- Third, and related to the second challenge, is the problem of the so-called “fog of errors,” where the number of errors is unknown and the first errors have to be identified and corrected to make it possible to identify and correct the next ones. Handling these errors successfully is like sailing in a deep mist in an archipelago of thousands of islands, islets, shallows, grounds, rocks, and banks without the help of any radar or GPS. It is urgent to identify and correct the errors acceptably without introducing any new errors, but also to do it swiftly—otherwise, the project will lose steerage-way and start drifting more or less uncontrollably. Previous research in the telecom industry shows the importance of project management in understanding and managing this problematic (Lindkvist, Söderlund, & Tell, 1998, Yakob & Tell, 2008; Lilliesköld & Taxén, 2006): when the first round of errors has been detected and corrected without inducing new errors, this only means that the next errors can be detected and corrected, and so on. In complex systems development, speed is therefore important for two additional reasons: the extent of serious errors is unknown, and the number of correction steps needed to come “out of the fog” is also unknown.
These three challenges imply that, in this industry, traditional project success criteria, defined as meeting time, cost, and (original) technical specifications, are almost impossible to satisfy in a meaningful way. The challenges also imply that many of the conventional ideas of project management need to be readdressed. As suggested by Lindkvist et al (1998), project management in this setting must rely on a “coupling logic” that allows for concurrent work and tight integration between inter-functional units. In this paper we analyze three innovative management practices responding to the three aforementioned challenges. Taken together they constitute a “neo-realistic approach” of project management. The bottom line is that project success is possible, if the goal is revisited and the methods revised in order not to mirror but to consistently reduce and control complexity.
Research Methodology and Empirical Foundations
Empirically, this paper is based on studies and experience of complex systems development within Ericsson between 1992 and 2005. It is written by a team of one practicing manager and two academics and draws on several different sources. Our cooperation started in 1995 when one of the academics conducted a multi-year study of complex project management at Ericsson. During the same time, the practicing manager wrote a booklet for practitioners explaining the fundamentals of managing complex projects, which had a strong impact on the practice of managing projects within Ericsson; several thousands of copies were distributed and used as course material among project managers. Since 1999 we have together been involved in project management courses in other companies, teaching and testing the same basic concepts. Thus the paper is an attempt to engage in a process of knowledge co-production, which has been seen as vital by the Mode 2- theorists of modern knowledge production, where robust knowledge needs to be generated in the “context of application” (Nowotny et al., 2001, p. 1). In the paper we refer to the practical experiences of one of the authors (Järkvik) as “According to the experience of one of the authors,” or in short “According to our experience.”
To broaden the empirical base, we also draw on parallel studies of projects within Ericsson to look for patterns and improve the conceptualization of our observations and experience, carried out by one of the two academics in the research team (Söderlund), or by other researchers (e.g. Adler, 1999; Lilliesköld, 2006; Lilliesköld & Taxén, 2006; Yakob & Tell, 2007). The latter studies will be referred to explicitly in the text.
The research process consists of five separate but overlapping parts:
- Research involving five case studies of complex development projects at Ericsson. Reported in separate research reports; written by one of the researchers in the team.
- Conceptualization of own practice developed by one of the authors; booklet on key issues in managing complex development projects at Ericsson, presentations to practicing project managers and sharing of experiences.
- Comparisons with other studies of projects within the Ericsson Group, with a focus on projects that have adopted innovative models of project management.
- Intensified cooperation, selection of the most relevant issues from a practice as well as a theory perspective; workshops with other researchers and writing of different drafts from various perspectives.
- Documenting the findings, searching for rigor and relevance; joint conference presentations; completing the final conceptualizations.
This paper can be viewed as a case-based study in a special sense: statistical generalizations are obviously neither possible nor relevant. A key aspect is whether the ideas presented here work or not. Based on our analysis and empirical research by others of projects within Ericsson, there is strong support that they do. As for generalization, the most important aspect is whether the paper inspires innovative practices in other settings and contributes to a more in-depth understanding of the challenges in a highly demanding project management area.
The Features of a Neo-Realistic Approach
Our analysis concentrates on innovations within three areas of project management: (1) success criteria and goal reformulation; (2) project planning and system configuration; and (3) error handling, detection, and reaction. The selection of these areas is based on their inter-reliance when organizing and managing complex projects. The selection is further based on the conviction that our observations within these three areas of project management, although based on a limited empirical material, have some constructive theoretical implications by providing a specific set of critique of conventional wisdom within the project management field, and thereby might stimulate other “neo-realistic” management approaches. The selection of these three areas builds on a straightforward way of analyzing research findings and reflected experience. We documented the innovative approach developed and adopted within Ericsson in a set of categories. In each category we focus on one core idea for the management of complex systems development. Below is a first presentation of the aim and points that we are addressing.
The first area centers on goal-oriented behavior and expectations—a topic that has received much attention within the project management community, including studies of the role of goals for understanding why projects exist and how they are organized (Engwall, 2002; Lindkvist & Söderlund, 2002). The corresponding innovative practice, lagomizing, addresses the problem of realistic success criteria and how the goal and objective of the development effort could be reformulated. The result is a critique of much of the conventional literature on project management where meeting budget, time constraints and specifications are considered to be at the core.
The second area deals with integration—a topic of equal importance for understanding the role and practice of project management, and related to a stream of research on technology integration (Iansiti, 1998) and knowledge integration in projects (Söderlund, 2002; Enberg, Lindkvist, & Tell, 2006). To turn complex structures into neatly delimited sub-systems, researchers normally suggest modularization and decomposition. As argued by Lilliesköld and Taxén (2006), the product structure is perhaps the most common way of mapping interdependencies in development projects. Increasingly, these decomposition efforts are supported by various methods for visualization (see for example Forsberg, Mooz, & Cotterman, 2007). However, these tools seldom capture the relations between system functions. A different kind of visualization is needed that builds on a process of knowledge articulation to build a shared understanding of system capabilities and interdependencies.
The third area is about innovative ways of creating visible and fast-paced forums of project management: the everyday project-decision agora, where real-time status is analyzed and rapid decisions are publicly made. Such meeting places resemble some aspects of the concept of “Ba” discussed by Nonaka and Konno (1998), as critical for integrating knowledge. But in the context of complex systems development, this forum takes on a new sense of urgency and visibility. At Ericsson this forum has been labeled the “Systemakut” (a possible translation might be “Systems Emergency Ward”). According to Ericsson experience, this type of forum is fundamental for timely delivery of telecom systems given the time-consuming activities of identifying and correcting errors, which often appear in multiple layers and areas.
Lagomizing Project Management: Top-Down Reduction of Project Scope and Complexity
Mainstream project management often portrays delivery on time and according to specifications to be the keys for successful project management. However, according to our observations and experience, it is normally not possible to meet both delivery time and specified customer requirements. The real challenge is to be able to deliver anything on the right day, with properties and functionality of which the project has a reliable knowledge. The starting point to deal with this dilemma is captured by the idiosyncratic concept of “lagomizing” (from the Swedish word ‘lagom' which may perhaps be translated as ‘just right’). Essentially this means a top-down driven reduction of specifications with no customer involvement, against a superficial understanding of the dogma of the quality movement of “exceeding the customer's expectations.” Accordingly, the task at hand for a neo-realistic project manager is to drive down designers' interpretations of the specifications from technical perfection to the simplest possible level. This means that project management takes on the task of delimiting scope, imposing non-negotiable constraints, and reducing complexity. Obviously this requires a particular type of heavyweight project manager, a “shusa” (Clark & Fujimoto, 1991). The “lagomized project management” observed in Ericsson reinterprets and reformulates success criteria and organization used in systems development, in opposition to the conventional model of “reflecting complexity.”
In the Ericsson case, the concept of lagomizing rests on many years of experience of systems development projects and the general tendency of underestimating the effect of complexity on development efforts. Fundamentally, lagomized project management is a way to deal with this underestimation. Moreover, it is a “heavyweight” approach to deal with the problems of scope creep and “drifting ambitions” characterizing complex systems development in many high-tech sectors. According to our experience, the involved engineers have a tendency to add ambitions either as an attempt to do things they believe will be asked for later on or just as a way to deal with vague specifications where one possible way to deal with vagueness is to “over-interpret” specifications. There is also the very natural tendency to do more than the plan envisages unless one is hindered to do so. What is at stake here is closely linked to the engineering ethos of what a good engineer is and how he or she behaves (cf. Vincenti, 1990).
The lagomized approach to project management adds customer value in the very specific sense of delivering on time with a tested and tried functionality, which is not necessarily, if ever, the same as the one found in the original specifications. The lagomized approach also acknowledges the fact that engineers frequently have limited communication skills. Simple answers can be received from just asking a colleague or a project manager, however, but often this does not happen. Similarly, specifications are assumed to be clear and transparent, although they seldom are. There is always room for interpretation and subjectivity, creating opportunities for highly trained engineers to do more than what is necessary. This issue is made more troublesome by the fact that specifications are the result of several layers of handovers. As has been documented by hard-won experience and series of studies in product development, handovers do not always add quality and they most certainly never add simplicity. Thus, lagomizing as a top-down effort to control and limit project scope and stop design efforts, which cannot be thoroughly tested is one key aspect in reducing and controlling complexity.
Organic Integration: Nothing is Right from the Beginning and Traditional Project Plans are Never Sufficient
Mainstream project management research has devoted much attention to planning techniques and breakdown tools, or in other words, efforts to reduce the need for developing a common understanding. This is an efficient way of dealing with complexity and knowledge integration in cases of known and stable technologies. However, according to our experience, in complex systems development in rapidly changing industries this is not enough; there is also a need for collective efforts to articulate and visualize a view of system interdependencies by key actors within the project. A powerful tool to build this shared view is the functional anatomy/system anatomy, which shows how the core functions of a system—its capabilities—interact and which functions build on other functions. This anatomy is then used for organic integration where basic functionality are assembled and tested first and then more advanced functionality is integrated in rapid iterations. It should be kept in mind that functionality here is understood as the capability of the system to do something, which should not be confused with “function,” for example “alarm function.” Figure 1 presents a simplified version of a system anatomy and the basic process to test functionality in a step-wise fashion.
Figure 1. An example of an anatomy of a radio base station
The practice of organic integration builds on another insight: in non-repeatable endeavors such as complex development projects, components or subcomponents are normally not right the first time. The management challenge is therefore to develop arenas where problems will be understood and discovered quickly. The practice of organic integration is a way to start detecting errors and misunderstanding as soon as possible. The opposite method is to isolate subsystems and try to make them independent and in the process put all the burden of proof on the interface specifications. In rapidly changing and complex systems this has turned out to be very difficult, if not impossible.
According to Lilliesköld and Taxén (2006), the anatomy approach has had a profound impact on the development practice within the Ericsson Group. Since its introduction in a few critical projects in the early 1990s, the approach has been used in more than 250 projects. The anatomy provides a basis for coordination and integration in complex development projects, different from much theorizing on how to deal with interdependencies such as design structure matrices and work breakdown structures. Thus on the basis of an understanding of the system anatomy, an organic integration plan can be developed. The focus in this plan is on testing functionalities in a way that makes it possible to always safeguard the basic operational functionality. The concepts of system anatomy and organic integration rest upon the insight that a priori separation and decomposition are desirable, but often impossible to accomplish. Hence the focus in the anatomy approach is on visualizing the non-decomposable “functional interdependencies” in such a way as to create a shared understanding within the project. This is critical for detecting, diagnosing and eliminating errors in the system under development (or for justifying errors: “this is not a defect, it is a functional property”). The organic integration plan is not a plan for scheduling the actual design work. Instead, the organic integration plan visualizes the process and the logical steps for starting and testing the functionalities the system is supposed to be capable of. This type of plan must be supported by increment plans detailing the deliveries to be made to the integration plan.
The Systems Emergency Ward: When Errors Are Layered They Can Easily Overwhelm the Project
The view of handling errors as a challenge of navigating in tricky and misty waters, and the high risk that correcting one error introduces a new one, call for new ways of rapid, concentrated, and visible action. One possible method for accomplishing this is what Ericsson terms the “Systemakut” (“Systems Emergency Ward”) where the project management team and relevant specialists meet on a daily basis to discuss and decide on all defects reported from the development project. Operating the Systems Emergency Ward is a heavy burden on management and cannot be delegated: it is high visibility management par preference. According to our experience, this is also one of its virtues. When management presence subsides, project members take immediate notice, not after weeks or months of non-functioning operations. The Systems Emergency Ward is therefore a way to signal high management attention.
The Systems Emergency Ward plays a critical role in coping with complexity and for integrating knowledge and activities within the project. During the meetings the primary focus is on the effects on the system at the present stage, if deliverables are in order and if receivers have understood what has been delivered to them. These meetings have a hands-on character, dealing with problems item by item and making fast decisions not to lose pace in the project. As stated earlier, many changes in a complex project tend to introduce yet more errors and therefore require an increased amount of attention. The challenge is to keep these changes at a minimum and only implement corrections that actually improve the capabilities of the system. The introduction of the Systems Emergency Ward has considerably improved error handling capacity within Ericsson and turned projects into much more public endeavors than normally have been the case.
In general, the ideas of organic integration and error detection and correction as a daily and public exercise lead to the following insight: In complex development projects in the telecom industry, new requirements are regularly added during the development period, and the delivery time is often instable. Nevertheless, projects are required to deliver on short notice and still have complete knowledge regarding the usefulness, functionality, and properties of the system delivered. Thus, for several reasons, the system under development needs to be frequently and realistically tested to verify that added functionality operates according to intentions, that added functionality does not destroy previous functionality, and that added functionality is continuously improving. The only way to accomplish this is to build “complete” systems regularly and frequently and in the process build a shared knowledge among participants about the system, its properties and its problems.
Theoretical Interpretations of a Neo-Realistic Approach to Project Management
These practices all represent important measures to deal with complexity in a fast-changing environment and of handling the fundamental problem of interdependencies and emergent behavior. In the following section, we will discuss the theoretical interpretations and implications, starting with lagomizing, moving over to organic integration and ending with the Systems Emergency Ward. Table 1 below provides an overview of the three practices and associated challenges and principles.
Table 1. Comparing Innovative Practices of Project Management
|“Lagomizing”||“Organic Integration”||“Systemakut” (the Systems Emergency Ward)|
Specifications of complex systems tend to include an abundance of features never used or hard to implement
Existence of a high number of elements, interdependencies, and interfaces that cannot be decomposed
Number of errors are unknown and correction of errors lead to new errors
Adhere to specifications; meet time, cost, and budget constraints
Schedule and control projects by detailed WBS and network plans
Decompose system, to avoid errors and test each compoenent sufficiently
Neo-realistic managerial insight
Scope and complexity needs to be reduced top-down by project management
Nothing is right from the beginning, and traditional project plans are never sufficient
Control the interpretation of errors detected and correct errors on a daily basis
Reduce complexity: Re-interpret specifications and keep doing this throughout the project.
Handle complexity: Test the different functions in the order they will be used when the system is started, configured and used in real life
Control consequences of complexity: Test each new build from the bottom up in real equipment
In complex projects, there is inherent over-optimism and underestimation of the consequences of complexity
In complex projects, visual structuring of interdependen-cies are critical to create shared mental maps
In complex projects, decision-making needs to be enacted on stage and exercised daily as a core capability
Inherent in project undertakings is the basic problem of over-ambitious goals and drifting scope (Kreiner, 1995). According to Lovallo and Kahneman (2003), most projects are plagued with over-optimism and this is the basic reason why projects fail according to standard criteria such as time, cost, and quality. This problem is particularly serious in complex systems development and can result in a complete loss of management control. Research has shown that large-systems are often “operating failures that neither do not function as intended or are not used at all” (Nightingale, 2000, p. 914). But as stated above the solution is not, as suggested by Nightingale and others, to make sure that the “design matches its specifications,” but to reinterpret these specifications and transform customer expectations.
In some publications within the research community, project management is seen as a problem of avoiding deviations (e.g. Hällgren & Maaninen-Olsson, 2005). According to our experience, however, deviations are always to be expected. The key challenge is to develop approaches where deviations, for example from original specifications, are controlled and managed. Consciously managing deviations of this type might be seen as an illustration of what project management is about. The following proposition summarizes this approach.
Proposition I: Project management in complex systems development is not about meeting expectations, rather to transform expectations and establish goals that make a controlled system delivery on time possible.
The second innovative practice presented in this paper revolves around the idea of organic integration. As documented in many previous studies (e.g. Yakob & Tell, 2007), modularization can play a key role in managing complex systems, to separate tasks and to be able to reduce complexity. We also acknowledge the importance of modularization and a priori interface management. But these separation and decomposition mechanisms are not sufficient to cope with the irreducible system dependencies and emergent properties characterizing the complex systems development projects discussed here. Previous studies have come to similar conclusions. As mentioned, Lindkvist et al (1998) highlights the need for a coupling logic of project management to be able to deal with the dual challenge of systemic complexity and error detection. The authors argued that project management in such contexts must concentrate on integration by error detection where a trial-and-error process plays a fundamental role. They also demonstrated the significance of public arenas in such development projects to be able to detect errors in organizational interfaces and in product features.
Integration and coordination are classical issues in product development and organization theory. They have been researched in a number of seminal publications since the 1960s (e.g. Lawrence & Lorsch, 1967), and continue to receive much attention from scholars around the world. In the project management community, much emphasis has been put on how to resolve interdependencies by the use of work breakdown structures and various work plans. In recent literature, however, changing plans are seen as a vital ingredient in successful project management As argued by Dvir & Lechler (2004, p. 2). “original project plans and project goals will have to be changed to address the dynamics caused by uncertainty and to maximize project success.” Similarly, they suggest that the higher the uncertainty, the later the point of design freeze. This corresponds with the idea of “project flexibility” found in much project management literature (cf. Olsson, 2006). However, our analysis of organic integration and the role of anatomies in projects illustrate rather the contrary—that stable integration plans are critical for project success. These are fundamental for handling interdependencies and for managing the error detection and correction process. In other words, system anatomies are important semi-structures that provide enough structure so that project members will create sense making, facilitate organic integration and be confident to act in complex situations (cf. Brown & Eisenhardt, 1997).
According to the perspective presented in this paper, defining system capabilities is a first step to create a required level of stability in a complex project. The concept of system capabilities is important here for two reasons. First, it sheds light on the most critical integration process in complex systems development (that of system capability integration); second, it focuses on a specific type of integrative activity: rapid and repeated testing cycles. The increment plans should be synchronized with the integration plan, so ideally no new functionality should be delivered which is not tested at once. The virtue of this approach is that flaws and faults are immediately exposed and not embedded deep down in the system for later discovery and massive rework. Further, no design work is wasted on activities, which in the error-detection cycles may be found to be unnecessary or impossible to handle. This theoretical interpretation leads us to the following proposition:
Proposition II: Project management in complex systems development centers on developing means and arenas for shared understanding of system properties and interdependencies as a basis for integration planning and testing.
The third innovative practice deals with the Systems Emergency Ward. Previous studies have documented the positive relationship between communication-intensive means, including high profile project management, and product development performance, but without exploring the actual integration mechanisms in any details (Clark & Fujimoto, 1991). At the same time, previous research has highlighted problems in bringing about real knowledge sharing in project settings. For instance, Hoopes and Postrel (1999) stress the importance of shared knowledge between functional departments. Their study focuses on the reasons underlying the so-called “glitches” in product development projects; i.e. problems caused or allowed by a lack of integration between specialists. The reason why a glitch occurs in a project can either be a lack of relevant forums and similar integrative mechanisms or the sheer complexity or stickiness of the knowledge involved (Hoopes, 2001). The general approach to avoid glitches or deal with glitches once they occur is about finding ways of organizing and integrating specialist knowledge. The Systems Emergency Ward is a concrete example of an integrative forum for dealing with knowledge integration in practice. In the complex systems development discussed here, there is also the challenge of detecting and handling unforeseen system interaction effects and emergent properties. The Systems Emergency Ward could be seen as a forum both to detect and handle glitches, and to detect and react on this type of problems and errors. The daily emergency wards also trigger frequent iterations, which establish an understanding about the system. This type of combined testing and iterations plays a critical role for generating “factual and concrete information” so important for maintaining short development lead times (Eisenhardt & Tabrizi, 1995).
Another important feature of the Systems Emergency Ward is the combination of skills and experience within a development organization. In the forum a unique set of highly qualified people gather to sort out the most pressing and difficult problems. This forum then also makes sure that design changes and error handling of a strategic character are controlled for. As previously discussed, such agoras constitute important arenas for communication but equally significant for hands-on problem solving and decision-making. As stated, powerful project managers are important, however, the practice of the Systems Emergency Ward in addition illustrates the collective aspects of leadership and decision-making in complex development settings. Such groups of highly skilled people may have the capacity to hold the development process together despite an often-overwhelming complexity.
Proposition III: Project management in complex systems development requires innovative and highly visible forums with the capacity to integrate diverse knowledge, to handle system interaction effects, and for making project-wide decisions.
Concluding Discussion and Implications
On the basis of a combination of academic research and reflected real-life experience we have argued for the need of a “neo-realistic practice turn” in project management research to study the transformation of goals and expectations as a key management practice; the importance of visualizing systemic interdependencies as a basis for integration management; and the need for highly public project agoras to deal with multiple layers of errors and for rapid detection-action-reporting cycles. We have focused on the experiences from Ericsson, a company that has become widely known for its capability of managing and organizing complex systems development, but the basic approach has also been tested and adopted by project-intensive firms in other industries.
Project management as discussed in this paper is a theoretically informed, hands-on practice involving technical knowledge as well as capabilities for building forums for daily decisions and systems integration. Goal setting in such a process is much more about transforming expectations and goals to fit the system delivery requirements than to interpret specifications literally. In such a process, creating conditions for system reliability is more important than meeting original expectations. This underlines the need of analyzing goals as such in a project management context.
Moreover, our study stresses the importance of the neo-realistic insight that errors cannot be avoided in complex systems development, and thus the role of project management in creating effective arenas for detecting and handling errors. The traditional view that the task of project management is to organize to avoid errors and work out detailed delivery plans is in stark contrast to the image of project management practice reported here. Moreover, our study has documented the value of project management as a mechanism for knowledge integration. This requires images and artifacts, supported by public forums to arrange for such knowledge integration.
The practices analyzed here are all based on a reflective, experience-based approach to organizing projects, which we have labeled neo-realistic project management: “lagomized project management,” “organic integration,” and “Systems Emergency Wards.” These management innovations add to the understanding of the role played by project management in complex systems development. For instance, the Systems Emergency Ward is a symbol for the rapid organizational cycles characterizing the neo-realistic innovations in project management. This is important for error correction but also a virtue in its own right. A functioning organization and reliable procedures can never be taken for granted. Instead, such procedures have to be built and built again by practicing project managers, by their everyday decision-making and integration activities in the project. Because time is short, this requires daily exercises, not in advanced simulators, but in doing the real thing by delivering real increments to the real system. Identifying and articulating effective means to deal with the seemingly chaotic and emergent character of complex systems development are important steps in increasing the innovative potential of neo-realistic project management.
This paper made its point of departure in a dual criticism, both of the hard systems model so prominent in the textbooks in project management and in the planning-oriented research studies, and of the flexibility and adaptation approaches suggested in recent literature, which proposes that project management must reflect the dynamic, iterative, and often chaotic character of actual projects. In contrast to this we suggested an approach to understand, reduce and control for complexity using a set of innovative practices, which might be seen as key elements of the art of the project management of complex systems development. This has implications for how to look upon the evolving nature of capabilities and competence development in project-intensive firms. According to several studies (e.g. Davies & Brady, 2000; Söderlund, 2005b), the continuous development of project capabilities and project competence management within firms has not been analyzed in any detail, or, as pointed out in other studies, project management competence has been outsourced to outside specialists, thus making continuous learning and innovation difficult (Berggren, Söderlund, & Andersson, 2001). Our study documents the importance of identifying and developing inside views of operating dynamics in complex project management and conceptualizing new approaches. The study reported here is primarily an articulation and reflection on particular firm practices. More research is needed, both studies using the practice-oriented, knowledge co-production approach suggested here, and broad-scale studies with the possibility to generate and generalize insights across industries and firms.
Adler, N. (1999): Managing complex product development: three approaches, Published PhD Thesis, Stockholm School of Economics.
Berggren C., Söderlund, S. & Andersson, C. (2001). Clients, Contractors, and Consultants, The consequences of organizational fragmentation in contemporary project environment. Project Management Journal, 32(3), 39–48.
Blomquist, T., Gällstedt, M., Hällgren, M., Nilsson, A., & Söderholm, A. (2006). Project-as-practice: Making project research matter. Paper presented at the IRNOP Conference, Xian, China.
Brown, S. & Eisenhardt, K. (1997): The art of continuous change: linking complexity theory and time-paced evolution in relentlessly shifting organizations, Administrative Science Quarterly, 42, 1–34.
Cicmil, S., Williams, T., Thomas, J., & Hodgson, D. (2006). Rethinking project management research: Researching the actuality of projects. International Journal of Project Management, 24(8), 675–686.
Brady, T. & A. Davies (2004): Building project capabilities: from exploratory to exploitative learning, Organization Studies, 25(9), 1601–1621.
Clark, K. & T. Fujimoto (1991): Product development performance: strategy, organization and management in the world auto industry, Boston: Harvard Business School.
Davies, A. & T. Brady (2000): Organisational capabilities and learning in complex product systems: towards repeatable solutions, Research Policy, i, 931–953.
Davies, A., & Hobday, M. (2005). The business of projects. Cambridge: Cambridge University Press.
De Meyer, A. Loch, C., & Pich, M. (2002). Management project uncertainty: From variation to chaos. Sloan Management Review, Winter, 60–67.
Dvir, D. & T. Lechler (2004): Plans are nothing, changing plans is everything: the impact of changes on project success, Research Policy, 22, 1–15.
Dvir, D., S. Lipovetsky, A. Shenhar, & A. Tishler (1998): “In search of project classification: a non-universal approach to project success factors,” Research Policy, 27, 915–935.
Eisenhardt, K. & B. Tabrizi (1995): Accelerating adaptive processes: product innovation in the global computer industry, Administrative Science Quarterly, 40, 84–110.
Enberg, C., L. Lindkvist & F. Tell (2006): Exploring the dynamics of knowledge integration: acting and interacting in project teams, Management Learning, 37(2), 143–165.
Engwall, M. (2002): The futile dream of the perfect goal, in In K. Sahlin-Andersson & A. Söderholm (Eds.), Beyond project management, Malmö: Liber Abstrakt.
Forsberg, K, Mooz, H, & Cotterman, H. 2005. Visualizing Project Management: Models and Frameworks for Mastering Complex Systems. J. Wiley & Sons: New Jersey.
Hällgren, M., & Maaninen-Olsson, E. (2005). Deviations, uncertainty and ambiguity in a project-intensive organization. Project Management Journal, 36(2).
Highsmith, J. (2004). Agile project management. New York: Addison Wesley.
Hoopes, D. (2001): Where are the glitches in product development?, R&D management, 31(4), 381–389.
Hoopes, D., & Postrel, S. (1999). Shared knowledge, “glitches,” and product development performance. Strategic Management Journal, 20, 837–865.
Hughes, T. (1998). Rescuing Prometheus. New York: Vintage Books.
Iansiti, M. (1998): Technology integration, Boston: HBS Press.
Kreiner, K. (1995). In search of relevance: Project management in drifting environments. Scandinavian Journal of Management, 11(4), 335–346.
Lawrence, P. R. & Lorsch, J. W. (1967). Differentiation and integration in complex organizations. Administrative Science Quarterly, Vol. 13 (1), 1–47.
Lilliesköld, J. (2006). Managing complex industrial projects: A comparison between holistic models. PhD Thesis, Stockholm: Royal Institute of Technology.
Lilliesköld, J., & Taxén, L. (2006). Operationalizing coordination of mega projects: A work-practice perspective. Paper presented at the IRNOP Conference, Xian, China.
Lindkvist, L., & Söderlund, J. (2002). What goes on in projects: On goal-orientation and learning. In K. Sahlin-Andersson & A. Söderholm (Eds.), Beyond project management, Malmö: Liber Abstrakt.
Lindkvist, L., Söderlund, J., & Tell, F. (1998). Managing product development projects: On the significance of fountains and deadlines. Organization Studies, 19(6), 931–951.
Lovallo, D., & Kahneman, D. (2003, July). Delusions of success: How optimism undermines executives' decisions. Harvard Business Review, 56–63.
Lundin, R. A., & Söderholm, A. (1995). A theory of the temporary organization. Scandinavian Journal of Management, 11(4), 437–455.
Maylor, H. (2006): Special issue on rethinking project management (EPSRC Network 2004–2006), International Journal of Project Management, 24, 635–637.
Merrow, E. (1988): Understanding the outcome of megaprojects, Santa Monica: RAND.
Miller, R., & Lessard, D. (2007), Evolving strategy: Risk management and the shaping of large engineering projects. Working Paper 4639–07, MIT Sloan School of Management.
Morris, P. W. G. (1994): The management of projects, London: Thomas Telford.
Nightingale, P. (2000). The product-process-organisation relationship in complex development projects. Research Policy, 29, 913–930.
Nonaka, I. & N. Konno (1998): The concept of “Ba”: Building a foundation for knowledge creation, California Management Review, 40(3), 40–54.
Nowotny, H., Scott, P. & Gibbons, M. (2001): Re–Thinking Science: knowledge and the public in an age of uncertainty, Oxford: Polity Press.
Olsson, N. O. E. (2006): Management of flexibility in projects, International Journal of Project Management, 24, 66–74.
Packendorff, J. (1995). Inquiring into the temporary organization: New directions for project management research. Scandinavian Journal of Management, 11(4), 319–334.
Pich, M. T., C. H. Loch & A. De Meyer (2002): On uncertainty, ambiguity, and complexity in project management, Management Science, 48(8), 1008–1023.
Sanchez, R. & J. Mahoney (1996): Modularity, flexibility, and knowledge management in product and organization design, Strategic Management Journal, 17, 63–76.
Sapolsky, H. (1972). The Polaris system development: Bureaucratic and programmatic success in government. Cambridge: Harvard University Press.
Shenhar, A. (2001). One size does not fit all projects: Exploring classical contingency domains. Management Science, 47(3), 394–414.
Shenhar, A., & Dvir, D. (2007). Reinventing project management. Boston: Harvard Business School Press.
Söderlund, J. (2002). Managing complex development projects: Arenas, knowledge processes and time. R&D Management, 32(5), 419–430.
Söderlund, J. (2005a). What project management really is about: Two alternative perspectives on the role and practice of project management. International Journal of Technology Management, 32(3/4), 371–387.
Söderlund, J. (2005b): Developing project competence: empirical regularities in competitive project operations, International Journal of Innovation Management. 9(4), 451–480.
Söderlund, J. (2008): Competence dynamics and learning processes in project-based firms: shifting, adapting and leveraging, International Journal of Innovation Management. 12(1), 41–67.
Van de Ven, A., D. E. Polley, R. Garud & S. Venkataraman (1999): The innovation journey, New York: Oxford University Press.
Vincenti, W. (1990). What engineers know and how they know it, Baltimore: Johns Hopkins University Press.
Winter, M., Smith, C., Morris, P., & Cicmil, S. (2006). Directions for future research in project management: The main findings of a UK government-funded research network. International Journal of Project Management, 24, 638–649.
Womack, J., D. Jones & D. Roos (1990): The machine that changed the world, New York: Rawson.
Yakob, R. & F. Tell (2007): Managing near decomposability in complex platform development projects, International Journal of Intelligence and Planning, 3(4), 387–407.