|Book Review: Managing the Unknown|
Managing the Unknown: A New Approach to Managing High Uncertainty and Risk in Projects
Written by: Christoph H. Loch, Arnoud DeMeyer, and Michael T. Pich. Hoboken, NJ: John Wiley & Sons, 2006. 292 + xii pages.
Managing the Unknown is an important book, and it was a revelation for me. It takes a fresh look at project risk management (PRM), which is a vital skill in developing a new product, but it goes beyond conventional risk management in critical ways.
Many books, including one coauthored by me (Smith and Merritt, 2002), cover conventional project risk management, which is a process of identifying the risks in a project, sorting them, and planning how to address the most serious ones. This conventional process ignores so-called unknown unknown risks—commonly known as unk unks—because, by definition, they cannot be identified in advance. However, unk unks are critical to the more innovative types of development projects. In fact, they define the way you manage such projects.
The authors draw an important distinction between risky projects—for which conventional project risk management applies—and novel projects. “The fundamental logic of the PRM mind-set does not address novel projects” because “the basic backbone of PRM is that there is a real project plan …”; for novel projects, “the plan is but a starting point, an illusion, a simple sketch. The basic backbone of PRM, the project plan, does not really exist” (p. 3).
Managing the Unknown, written by three professors from INSEAD, focuses on unk unks and provides a comprehensive package for managing them. The book opens with two illuminating chapters. The first is a case study of a conventional project. Although it had its share of problems, it was completed satisfactorily by using conventional PRM. The second is a case study of a new-to-the-world technology scale-up project with many unk unks. When management applied conventional risk management, which was all they knew, they failed. As one participant put it, “You face an entire forest, and each tree is a potential problem. In principle, you can focus on any one tree beforehand and study it, but you can't look at them all. You have to choose a few and focus on them, but then it may turn out that the relevant trees are totally different” (p. 46). The unk unks—those unidentified relevant trees—slip entirely outside the purview of conventional PRM.
Drawing on the contrast between these projects, the authors discuss various types of uncertainty, showing which ones are amenable to conventional methods and which require a fundamentally new approach. Besides the inability to know about the risks, project complexity contributes to this decision. Complexity depends on the number of components of a system and the number of interactions among these components.
The core of the book, chapters 4–7, lays out the methodology for dealing with unk unks. Unk unks cannot, by definition, be identified, but the areas where they lie—where knowledge about the project is lacking—can be circumscribed. Thus, homing in on them is a gradual iterative process of discovering the parts of the project in which knowledge is weakest. Once they are hemmed in, two methods can be employed for overcoming them. Described in chapters 5 and 6, these are learning and selectionism, respectively. “Learning in projects is the flexible adjustment of the project approach to the changing environment as it occurs” (p. 103). It is a repetitive process of asking, “What do we know, what do we need to know, and what might we not know that we do not know” (p. 120)?
The authors define selectionism as running multiple trials in parallel. It is most appropriately used when the terrain is so uncertain that a single trial is unlikely to home in on an inappropriate solution. Chapter 6 addresses the key issues of how many trials to run, when to terminate them, and how to terminate them without demotivating the people who were working on them. Finally, chapter 7 addresses combining learning and selectionism, which is what occurs in actual projects. The blend of these two approaches depends on project complexity and cost structure. For instance, one could use a Darwinian approach of parallel trials throughout the project, or one could be more exploratory, starting with several trials and progressively pruning them as they reach cul-de-sacs.
Notice that selectionism is also called parallel development by product developers, and it has fallen out of favor as development budgets have been squeezed. Likewise, learning can be costly, as it can retrace past steps and employ experiments that would not be needed if the route were clear. The methods in this book are not for those who wish to minimize development expense but instead for those willing to spend a bit more for a truly novel product and the flexible methodology it demands.
The large last part of the book, chapters 8–12, covers the crucial issues of making these tools work in real organizations. Chapter 8 addresses the mindset needed, which is quite different than the one normally encountered in the corporate world. The authors illustrate this by describing a version of mindfulness drawn from Weick and Sutcliffe (2001), which encompasses five components:
Preoccupation with failure: reporting and investigating any deviation from plan since it may be a precursor of serious problems ahead
Reluctance to simplify: because unk unks fall outside of our experience, simplifying and focusing may blind us to them; the mindful manager lives with ambiguity and uncertainty
Sensitivity to operations: not accepting standard procedures and outcomes at face value but looking behind them for patterns or hunches that might indicate unk unks
Commitment to resilience: recognizing that the unexpected will happen and being prepared to change plans quickly and bounce back from disaster
Deference to expertise: pushing decisions down to levels where the experience to deal with them is greatest, response can be quick, and a diversity of opinion exists
They also shed light on the value of experience in novel projects, which is similar to MacCormack's (2001) findings. What counts is hands-on experience with unk unks rather than years of experience in the field. The chapter 2 case study failed in part because the manager, who had plenty of experience with conventional projects, persisted with an approach that was completely ineffective for a novel project.
Chapter 8 also emphasizes the great value of a project vision, which acts as a compass for navigating over unknown terrain. When stable project plans are fiction, such a compass is invaluable for maintaining a constant direction even as the territory shifts.
Chapter 9 covers management systems for novel projects, including planning approaches, monitoring systems, and evaluation and incentive systems. Chapter 10 expands into projects carried out with partners, which are common in high-uncertainty projects because of desires to share the risk and to obtain the best expertise available. A critical point here is the value of contracts, because, like plans, contracts are of limited value when unk unks will upset them repeatedly. Instead, fairness, integrity, and trust become more important. Chapter 11 addresses the role of stakeholders in the project. An illuminating case study describes an effort to develop a combined automobile and airplane. The project failed, not due to technical difficulties but because the team did not share progress with a marketing manager and certain members of senior management, and the corporate immune system eventually killed the project. Chapter 12 lays out challenges for senior management, primarily to set up an effective system for their few novel projects that differs completely in philosophy from the conventional approach that should be used for the majority of their projects.
The strengths of this book are its comprehensive approach to the subject and the case studies that appear in nearly every chapter to illustrate the difficulties and principles nicely. However, some parts are not easy to follow. There is no glossary, and the usage of key terms seems to vary. Coordination between the text and the figures is poor in several instances.
Unfortunately, this book is quite expensive, which will limit its availability for those on development project teams and those managing them who should be reading it (an unk unk in the authors' publishing project?). With the decrease we are seeing in truly novel new products (Cooper, 2005), we can only hope that the material in this book receives more attention and that the publisher issues a reasonably priced paperback soon.