It has long been recognized that good project management is not enough to ensure project success. For example, de Wit argued “good project management can contribute towards project success but is unlikely to be able to prevent project failure” (de Wit, 1988, p. 165). Similarly, a recent study of information systems (IS) executive perceptions found no significant difference in the project success experience of PMP® credential holders and uncertified project managers (Starkweather & Stevenson, 2011). This suggests that, at least for IS projects, good project management is a necessary but not sufficient condition for project success. Apart from the unfortunate imputation of this conclusion on the value of management, this leaves the question: What else can account for project success that is beyond the capabilities of project management?
The usual answer to this question in the literature is some function of the argument that projects are one-off, complex, high-risk, multidimensional socio-technical endeavors. In the case of IS projects, reference is also often made to the pervasive, cross-functional impact of information systems technology on most aspects of organizations. Underpinning these answers is the assumption that there is significant scope in projects for other, often unknown, factors than project management to negatively impact performance. However, while for many projects these descriptions might be true, they provide few clues to the underlying mechanisms that can result in variable project performance. Neither do they suggest how to manage the problem. The aim of this paper is to develop a capabilities-based explanation of project performance that can explain the dilemma of the insufficiency of project management to determine project outcomes as well as other anomalies that are found in practice such as high variability in project performance within a single organization from one project to the next. The paper focuses on the IS project domain but the fundamentals of the explanation are not IS-specific, so the mechanisms are also relevant to projects in other application domains (but generalization of the performance model is not addressed here).
Past approaches to answering the larger question of what determines project performance—particularly in software-related projects—have tended to take the functional form of a continuous effect and assume that project success is the expected or normal outcome of a project. Take, for example, the two most common explanations for project performance found in the literature: factor- and process-based approaches (Sauer, 1999). Factor-based approaches seek to identify antecedent variables (factors) that influence project performance. Essentially, they argued that project performance is determined by taking account of recognized critical success factors, and/or avoiding failure factors and managing risk factors. Lists of such factors are commonly known, for example: executive commitment; business/user involvement; clear objectives; stable requirements; minimized scope; project management expertise; change management; and suitable methodology and tools. Fundamentally, this approach argues that if you do the right things in terms of managing identified performance factors, the project will succeed. Furthermore, the more things that are done right, the greater the likelihood of success. That is, it argues that there is a continuous positive main effect of antecedent factors on project outcomes.
Similarly, process-based approaches examine the interactions between stakeholders and events to identify the specific nature of the relationships between factors and outcomes. In essence, they argue that project performance is determined by following best practice processes, methodologies, and techniques in each project situation. Fundamentally, they argued that if you do things in the right ways by following prescribed project processes, the project will succeed. Furthermore, the closer the fit between process and project requirement, the greater is the likelihood of success (again, the functional form of a continuous positive main effect).
Together, the factor and process approaches represent good project management. They are also popular because they have a strong practical appeal, providing specific guidance on what to do in practice to achieve project success. These approaches have dominated the field of project management throughout most of its history and have contributed significantly to the current state of the art and science of project management. However, a key problem with both of these explanations is that they contain no explicit drivers for underperformance. They assume that if you do the right things in the right ways then the project will succeed. Underperformance arises if insufficient attention is given to taking account of the key factors and applying suitable processes. In effect, underperformance or failure is the inverse of success. That is, it is not doing what is necessary (or enough of what is necessary) to succeed. If this occurs, and something is missed in one project, then stakeholders can learn from the experience and do better the next time around. By contrast, the alternative explanation developed in this paper takes the view that underperformance or failure is not just the inverse of success. Neither is it just a matter of learning from experience and doing things better next time. Rather, there are fundamental issues with the learning process that can introduce explicit drivers for failure that are independent of the drivers for success. These drivers make it possible for a project to underperform or fail despite good project management.
As implied previously, another key problem with the current dominant approaches is that they appear to be insufficient to guarantee project success. This dilemma is graphically illustrated in Figure 1. The figure plots survey data from The Standish Group on IS projects that succeeded outright against schedule, budget, and planned features; failed and were terminated before intended completion; and underperformed against schedule, budget, and/or features delivered (i.e., were “challenged”). The graph combines the data for failed and challenged projects and compares it to projects that succeeded outright.
Figure 1: IT Project Performance (data source: The Standish Group, Inc.).
Three points are noteworthy about Figure 1. First, the scientific validity of the survey results on which the graph is based, particularly for the first two surveys, has been challenged in the research literature by claims that it overstates the problem (see, for example, Jørgensen & Moløkken-Østvold, 2006; Eveleens & Verhoef, 2010). Overall, however, the data reported appear to accord anecdotally with the views held by many executives and managers that IS projects are high-risk ventures. Furthermore, a more recent study suggests that “tech” projects have an even worse performance record than that reflected in the graph (Flyvbjerg & Budzier, 2011). Second, ignoring the first two years (in recognition of the literature-based challenge), a visualized trend line suggests a steady improvement in project performance over the period of the surveys. Failed and challenged projects have gradually declined, commensurate with an increase in projects that have succeeded outright. This implies that good project management, as it exists today, may be having a positive effect on project outcomes. Third, the bad news is that there is still a substantial gap between the two trend lines, and intersection seems to be a long way off. Even worse, these data indicate that project underperformance or failure is a more likely outcome than outright project success. That is, IS projects are more likely to fail or underperform than succeed. This is contrary to the assumption of the factor and process approaches to project performance, which assume that success is the normal or expected outcome of an IS project. In sum, despite the apparent improvement trend, this evidence supports the views expressed above that good project management is insufficient to determine successful project outcomes.
Before proceeding, one other explanation of the problem of project performance that is starting to emerge is to attribute outcome variance to black swans (see, for example, Flyvbjerg & Budzier, 2011). Proposed by Taleb (2010), a black swan is an improbable, unexpected event that has massive consequences, either positive or negative. The notion evolved from Taleb’s earlier work on luck and randomness, and the human tendency to expect that if 1,000 white swans were seen to fly by, the next swan would also be white (Taleb, 2008). Black swans disrupt our expectations. However, after the event, they can be rationalized in a way that could have made them expected. Applied to IT projects, it is argued that black swans can account for the unexpected disruptions that arise causing a $US 5 million project that is believed, ex ante, to be low risk, to blow out to a $US200 million loss (Flyvbjerg & Budzier, 2011). In hindsight, many of the red flags should have been evident.
In the explanation developed in the next section, disruption is also an important element in the model but it arises through a much more concrete mechanism than randomness. The approach draws on literature-based evidence that learning from experience (“learning by doing”) is not a simple continuous process that continually accumulates the organizational capabilities needed to deliver projects successfully, time after time. Rather, discontinuities and other barrier conditions can arise that reduce, block, or negate learning from experience, reducing the capabilities available to the organization to apply to projects. These disruptions can constantly refresh a liability in projects to under perform or fail.
Capability-related research in project management is not new. Substantial prior work has been done on identifying individual capabilities that are important in achieving project success. Arguably, capability-related approaches comprise the third, next most dominant explanation of project performance found in the literature after factors and processes.1
This prior research is distinguished from the current paper in two main ways. First, research on specific individual/team/organizational capabilities fits within the existing factor-based explanation paradigm. It focuses on identifying capability factors that influence project outcomes. Second, the prior research focuses on the causal link between capabilities and project performance, and the generative mechanisms whereby capabilities are built. That is, it also assumes a continuous positive main effect between capability factors and project success as the normal project outcome. Typically, it has not directly focused on degenerative mechanisms that can reduce or negate capabilities or account for failure outcomes.
The paper is conceptual. The research method used is theory development from organizational theory and the phenomenon of project underperformance, despite apparent good project management in practice. The unit of analysis is the organization and the capabilities it has to apply to the projects it undertakes. The proposed model draws from capability-based theory, organizational learning theory, and concepts from the organizational ecology literature. The distinctive contribution is an alternative capability-based project performance model that can explain how the circumstances reflected in Figure 1 might be possible. The theory is illustrated by application to a qualitative longitudinal case study, providing initial validation. In applying the theory to practice, the paper also discusses how the problem might be managed to improve project outcomes.
You can attend our Online Agile Training facilitated by our trainers who have more than 15+ yrs. of training and industry experience.
MSys Training is the markets-leading learning services company. Our customized training solutions are efficiently tailored to meet organization and individual goals. With various training formats, technologies, and approaches, we recognize the need for custom solutions that fit your company’s systems. MSys Training is highly recognized for its global expertise on trainings to co-create significant business value.