Post-Project Reviews to Gain Effective Lessons Learned
Terry Williams, PhD, PMP
ISBN 13: 978-1-933890-24-1 ISBN 10: 1-933890-24-X
Published by: Project Management Institute, Inc. 14 Campus Boulevard Newtown Square, Pennsylvania 19073-3299 USA Phone: +1-610-356-4600 Fax: +1-610-356-4647 E-mail: firstname.lastname@example.org Internet: www.PMI.org
©2007 Project Management Institute, Inc. All rights reserved.
“PMI”, the PMI logo, “PMP”, the PMP logo, “PMBOK”, “PgMP”, “Project Management Journal”, “PM Network”, and the PMI Today logo are registered marks of Project Management Institute, Inc. The Quarter Globe Design is a trademark of the Project Management Institute, Inc. For a comprehensive list of PMI marks, contact the PMI Legal Department.
PMI Publications welcomes corrections and comments on its books. Please feel free to send comments on typographical, formatting, or other errors. Simply make a copy of the relevant page of the book, mark the error, and send it to: Book Editor, PMI Publications, 14 Campus Boulevard, Newtown Square, PA 19073-3299 USA.
To inquire about discounts for resale or educational purposes, please contact the PMI Book Service Center:
PMI Book Service Center P.O. Box 932683, Atlanta, GA 31193-2683 USA Phone: 1-866-276-4764 (within the U.S. or Canada) or +1-770-280-4129 (globally) Fax: +1-770-280-4113 E-mail: email@example.com
Printed in the United States of America. No part of this work may be reproduced or transmitted in any form or by any means, electronic, manual, photocopying, recording, or by any information storage and retrieval system, without prior written permission of the publisher.
The paper used in this book complies with the Permanent Paper Standard issued by the National Information Standards Organization (Z39.48—1984).
10 9 8 7 6 5 4
1 INTRODUCTION 1
1.1 Introduction 1
2 LITERATURE SURVEY 5
2.1 Introduction 5
2.2 Motivation 8
2.3 Concepts 13
2.4 The Current Situation 27
2.5 Creating Knowledge 30
2.6 Transferring Knowledge 42
2.7 Case Studies 56
2.8 Conclusions 57
3 SURVEY OF PRACTICE: DO ORGANIZATIONS LEARN FROM PROJECTS? 61
3.1 Introduction 61
3.2 Profi le of Respondents 62
3.3 What Are Organizations Doing? 64
3.4 How Successful Are these Processes? 71
3.5 Factors Contributing to the Perceived
Success of Lessons Learned 74
3.6 What Do You Think Is Best Practice? 81
3.7 Conclusions 86
4 INTERVIEWS: DO ORGANIZATIONS LEARN FROM PROJECTS? 89
4.1 Introduction 89
4.2 Organization C 90
4.3 Organization G 93
4.4 Organization P 94
4.5 Additional Case Studies 97
4.6 Organization W 98
4.7 Organization L 100
4.8 Organization S 102
5 CASE STUDY IN PRACTICE 105
5.1 Background 105
5.2 Interviews of Project Participants 105
5.3 Conclusion 107
6 CONCLUSIONS 109
APPENDIX 1: THE ZOOMERANG SURVEY 111
APPENDIX 2: THE ZOOMERANG RESULTS 123
APPENDIX 3 133
The need to manage projects well, and to learn from one project to the next, is of vital importance as our world becomes more and more project-based. Our management of complex projects is often seen as less effective than it might be, and consequently, we do need to learn from one project to the next. This is a well-known theory (Collison and Parcell 2001; Kerzner 2000)—but in practice, projects are often not reviewed at all for various reasons (Williams et al. 2001) If they are reviewed, the methods don’t account for the complexity or try to explain causality, so there is no understanding about what went wrong (or right) and why (MacMaster 2000). So with a few exceptions, it appears that project reviews are infrequently performed and useful lessons are not captured. If we did gain lessons, these need to be incorporated into the processes and policies of our organizations; a “lessons learned” process needs to be implemented and then refl ected upon to produce double-loop learning. The need to consider what is best practice in reviewing projects is evident.
At the end of 2004, while working at Strathclyde University, Glasgow, Scotland, UK, I responded to a request for research proposals from the Project Management Institute (PMI®), Newtown Square, PA, USA, with a proposal to look into post-project reviews, and PMI awarded me a research grant. In the middle of the contract, I moved to Southampton University, UK (also Concertante Consulting, London, UK) and the work was completed there.
The objectives of the research include identifying current practice as well as practice currently considered best practice for lessons learned in the project management fi eld, and comparing current practices with advances in understanding and modeling of project behavior to identify project lessons not being learned. This work sought to answer the following research questions which are of different types; therefore, different methodological approaches were required as is often the case with good management research (Easterby- Smith, Thorpe, and Lowe 1991).
2 Post-Project Reviews to Gain Effective Lessons Learned
1. Research Question 1 (RQ1) comprised two questions: “What is current practice?” and “What is currently considered best practice?” These are factual questions that are best explored using positivist techniques. A literature survey (Activity 1) and questionnaire (Activity 2) were used to investigate this.
2. Research Question 2 (RQ2) asked “Do these techniques actually achieve their purpose?” and is harder to answer with an unambiguous positivist answer. A literature survey (Activity 1) and questionnaire (Activity 2) along with interviews to gain subjective opinions (Activity 3), and some phenomenological analyses of an actual case study (Activity 4) were used to investigate whether these cases are indeed establishing the diffi cult-to-identify lessons.
3. Research Question 3 (RQ3) asked “Can we identify practical techniques to help draw out the diffi cult lessons from the projects?” is a practical question, and was approached by using the knowledge from Activities 1 through 3 and drawing out the experience of Activity 4, along the lines of Mode II research common in top- quality UK management research (Starkey and Madan 2001).
4. Research Question 4 (RQ4) asked “How can such lessons be incorporated into organizational practice?” Although a short research project, this is again a practical question, which will be approached by considering the literature (Activity 1), looking at actual practices (Activity 2), and considering both by interviews (Activity 3) and observations (Activity 4) as to how and whether lessons are taken into organizational procedures.
The program thus consisted of four main activities. The fi rst pair of activities established the current state of practice:
• Activity 1 carried out a survey of literature relevant to learning lessons in projects.
• Activity 2 surveyed the current practice of establishing and recording “lessons learned” throughout the project management community using an online questionnaire survey.
The second pair of activities, which formed the considerably smaller part of the study, looked at some particular examples:
• Activity 3 looked more in depth into current practice in six organizations, using semi-structured interviews.
• Activity 4 looked at a particular organization, and a particular project within that organization, to see how lessons are learned and how they can be learned.
I led the research work. A signifi cant part of the work for Activities 1 and 2 was carried out by Judith Stark (also of Strathclyde University, Glasgow, Scotland, UK). For Activity 1, there was some additional help with literature.
This research report on the work is divided into six chapters. After this introduction, Chapters 2 through 4 correspond to Activities 1–4; Chapter 2 covers the literature survey; Chapter 3 covers the online survey; Chapter 4 describes the interviews; and Chapter 5 reviews the case study. Chapter 6 provides some brief conclusions. Chapters 2–5 are designed to be self-standing and each can be read on their own.
This chapter describes the results of the survey of literature on the use of “lessons learned” from projects. The report was based on a Procite© database, which included not only references but a categorization by the keywords given in Table 2-1. The database included 280 records, of which 243 documents had been read fully and are discussed in the following paragraphs. For 28 references, only summaries, abstracts, or book reviews were studied or are included because they are referenced by one of the 243 documents; and another 9 appeared relevant but were not found suffi ciently appropriate for inclusion when the literature survey was written up and therefore do not appear here. Due to the tremendous number of documents, this survey should be considered as a summary rather than a full exploration of the literature.
The survey was wide-ranging and sought views beyond the project-management literature. Databases (including Google scholar) were used with relevant keywords. Key papers were selected and “forward chaining” was performed (i.e., searching citation indexes for papers that refer to these papers). Some particularly relevant journals were checked in their entirety for the previous 10 years, namely:
• Construction Management and Economics
• IEEE Transactions on Engineering Management
• International Journal of Project Management
• Management Learning (only from Vol 29 [March 1998] to present)
• Organization Science
• Project Management Journal
• The Learning Organization
In addition, some conferences were checked in their entirety, namely:
6 Post-Project Reviews to Gain Effective Lessons Learned
• PMI full conferences 2001–2003
• PMI Research Conferences 2000–2004
• IPMA conferences 1996–2004
• IRNOP (International Research Network on Organizing by Projects) conferences 2002-2004
• PMI Europe conferences 2000–2001
The Index of PMI Periodical Publications: 1996–2002, supplied by PMI, was also studied. The topic of lessons learned can be regarded as very narrow or very broad. Clearly, it
is important to gain the theoretical underpinnings of the how-to-learn lessons, and this means going beyond reports of implementations of project learning. However, the review of the entire organizational learning and knowledge management literatures was outside of the scope of this study; therefore, this survey contains only articles particularly relevant to project learning or fundamental articles or review articles of state-of-the-art theory. For those who want to read further about the literature on learning, DeFillippi (2001) gives an overview on different perspectives of learning from projects and where to fi nd more about them, covering: (a) action learning (Revans, Smith)—people learn by working on real problems; (b) action science (Argyris and Schon)—project participants refl ect on theories in use with the help of a facilitator; (c) action research (Lewin)—combines theory building with research on practical problems; (d) communities of practice (Wenger, Brown, and Duguid)—learning occurs naturally through communities, with the deepest learning occurring when people’s positions move within a community and at the intersections of multiple communities; and (e) refl ective practice (Raelin). Much of this will be touched upon as part of this book.
The problem of how to learn from projects has long been an issue with project-based organizations, and this book will discuss some of the reasons why. Almost forty years ago, Middleton (1967) identifi ed this as a problem with project-based organizations as opposed to functional ones: “Lessons learned on one project may not be communicated to other projects. One executive who was transferring from a project being phased out to a new project found the same mistakes being made that he had encountered on his former assignment three years earlier. He felt that the problem resulted from splitting normal functional responsibilities among project organizations and from not having enough qualifi ed, experienced employees to spread among all organizations.” Although our views of projects might be different today, Packendorff (1995), a leader in the Scandinavian school looking at projects as organizations rather than tools and seeking to go beyond normative statements, said “theories on learning in projects . . . are almost non-existent today.” This is despite a clear recognition, as Sense (2002) points out, that projects offer enormous potential for learning even though this is neglected in traditional project management practice.
Literature Survey 7
Learning clearly always happens to some extent by the nature of undertaking projects. Schofi eld and Wilson (1995), for example, looked at capital projects within the UK National Health Service and observed that learning did happen (in three ways: termed mechanical, cognitive, and behavioral) but they concluded “none of the team set about achieving organizational learning in a premeditated way.” In this book, we are looking towards the more systematic collection and distribution of lessons from projects. Ayas (1998) said that “learning within a project does not happen naturally; it is a complex process that needs to be managed. It requires deliberate attention, commitment, and continuous investment of resources” (cited in Sense 2003a).
This document is restricted to learning about project processes or how to manage and execute projects. It is not aimed towards learning technical knowledge from projects (such as in work along the lines of Nobeoka 1995)—it is about the project-management process rather than the content of projects.
Table 2-1 Keywords used in database
8 Post-Project Reviews to Gain Effective Lessons Learned
We will not be considering cases of particular projects that have gone wrong (e.g., Glass 1998) and trying to analyze them; rather, we will be looking at the literature to learn how organizations learn lessons from projects they have undertaken.
The structure of this document is as follows. Following this introduction, we shall look at:
• Motivation: why look at learning from projects?
• Basic concepts: what is knowledge? And what is learning?
• The current situation: what do project-management and other standards say about learning lessons? How prevalent actually are such practices?
• Creating knowledge: techniques, the use of narratives, dealing with systemicity, and factors that facilitate or hinder creating knowledge from projects.
• Transferring knowledge: organizational learning and knowledge management in projects, practices for distributing lessons, factors that facilitate or hinder distributing lessons from projects, and one particular idea, that of “communities of practice.”
• Case studies: a brief mention of some case studies.
• Conclusions: fi nally, some conclusions.
It might be thought that it goes without saying that learning from one project to another is a worthwhile aim, upon which it is worth expending effort. Abramovici (1999) for example tells us that “lessons learned” is a good thing to do, while outlining some basic practices. Similarly, Pinto (1999) states that information systems projects have a poor success record, describes evidence supporting this and reasons for it, then goes on to stress the need to pass lessons learned downstream through post-project review meetings. In the fi eld of complex product systems, Davies and Brady (2000) demonstrate the learning cycle of such projects, which requires a step: “lessons learned from the project and recommendations for improvements can be transferred to current or succeeding projects.” Kerzner (2000) places continuous learning and improvement as the highest level of project management maturity in an organization, and states that “without ‘discounted’ lessons learned, a company can quickly revert from maturity to immaturity in project management. Knowledge is lost and past mistakes are repeated.” Berke (2001) says that “Organizational learning and organizational knowledge make up the CIQ [corporate intelligence quotient] . . . Best practices and lessons learned are the building blocks of organizational learning and organizational knowledge.”
Why is learning lessons important? And is there evidence that it gives some use? This
Literature Survey 9
section provides a few sources that support the motivation to carry out lessons learned activities.
There are fundamental aspects of the nature of project-based organizations that require concentration on learning. Projects are, by nature, temporary organizations, and any learning that is accumulated in a project will largely dissipate at the end of the project unless attention is paid to the collection and dissemination of that knowledge. This has become increasingly recognized as the nature of projects as temporary organizations have been studied, particularly in the Scandinavian project management school. Ekstedt et al. (1999), for example, discuss the central importance of “knowledge formation” (by which they mean the combination of learning and embedding that learning): (a) they say that permanent organizations, which have a long-term perspective, generally have mechanisms that are built in for learning, but the new project-intensive organizational structures (i.e., “temporary organizations”) are action- and task-oriented and not geared for learning; and (b) individuals become more able and experienced, but there is often no mechanism or motivation for that learning to be shared within the company. They claim that “in temporary organizations . . . the most important thing in the design of a project is the combination of project members and the resources allotted to the project that is relying on the combinatorics of different stocks of knowledge”; however, “the outcomes of the knowledge processes that take place are diffi cult to feed back to the permanent organization.” Brady, Marshall, Prencipe, and Tell (2002) describe a number of barriers to learning from projects, in particular the absence of departmentally held “knowledge silos”; the uniqueness of projects, with long life cycles, so a long time interval elapses before lessons are retrieved; and their temporary nature, requiring new “human encounters” for each project. Similarly, Disterer (2002) looks at the need to manage the retention of knowledge because of the fragmentation of the organization into project teams, concluding: “only a few fi rms manage systematically to identify and transfer valuable knowledge from projects to following projects.” Bresnen, Goussevskaia, and Swan (2004) discuss the contradiction between short-term aims of projects and long-term aim of organizational learning, showing that knowledge management will depend on the degree of projectization of the fi rm (pure project fi rms will be different from matrix, and these will be different from functional organizations that only do occasional projects).
Project processes are generally temporary and unique, with non-routine features. Gann and Salter (2000) say that a particular challenge for project-based fi rms is to integrate their project and business processes, so the experiences of projects need to be integrated into the business processes, and knowledge acquired from projects needs to fl ow back to the core resources of the fi rm. Brady, Marshall, Prencipe, and Tell (2002) describe a further number of characteristics of projects, which contribute to the diffi culties of learning, including customization (either because previous solutions are obsolete or are being driven by customer demands), discontinuity (both temporal and organizational; project work is highly discontinuous, with a narrow focus and task oriented), complexity, interdependence
10 Post-Project Reviews to Gain Effective Lessons Learned
(systemic, so it is not always possible to rely on past experience to solve current problems), and uncertainty. However, that is not to say that all projects are completely different—and the “misguided belief” that they inhibit learning (Cooper, Lyneis, and Bryant 2002).
Furthermore, projects not only cross organizational functions, they are trans- disciplinary, so even beyond the project management literature, Gibbons et al. (1994) have explored changes in the manner in which knowledge is produced. They discuss the change in “mode of knowledge production” by contrasting the traditional mode in which discipline-based (academic) institutions defi ne a problem and use these institutions to develop knowledge and maintain quality control, with a new mode in which knowledge is produced in the context of application, knowledge is transdisciplinary, and knowledge production has to be socially accountable. So we need new mechanisms to capture and disseminate knowledge beyond the traditional disciplinary and functional structures.
Cooke-Davies (2002) describes a major ongoing empirical study, and in this paper identifi ed 12 key success factors in project-oriented organizations; one of these is “an effective means of learning from experience.” Menke (1997) describes a study of 79 R&D companies and gives 10 “best practices’” for R&D decisions which confer R&D competitive advantage. Once again, “learn from post-project audits” is one of the 10 best practices, although it is worth noting that it comes in as the lowest when measured by frequency of use (in this survey, in only 24% of companies). Kotnour’s (2000) infl uential survey of 43 project managers within the Project Management Institute, based on subjective measures, indicates that project management performance is associated with project knowledge; project management knowledge is supported by project learning activities; the level of activity of producing lessons learned is related to inter-project learning; and that “learning support” (for example, collecting data and having appropriate corporate culture about learning lessons) is needed for all learning activities.
The literature provides a number of reasons for, or perceived benefi ts from, managing the learning from projects. These include
• Project managers learn how to manage experientially, and it is important to refl ect and gain these lessons. “Research shows that the majority (85%) of project personnel have gained their knowledge, both explicit and tacit, through experiential learning” (Turner, Keegan, and Crawford 2000).
• Learning lessons from projects can feed into the project assessment, risk analysis, or initial planning of the next project. Neale and Holmes (1990) describe a survey of fi nance directors in 1,000 companies regarding post-auditing procedures for capital projects, so the emphasis was on improving decisions about whether to invest in a project, for example better proposals, better evaluation, and better fi nancial control: 38% of the survey identifi ed “encouraging more realistic project assessments” as the main aim of post-project auditing and 30% identifi ed
Literature Survey 11
“encouraging greater realism in project appraisal” as a main benefi t; they conclude that post-auditing can “radically improve the quality of investment decision- making.” Williams (2005) describes an extensive case study in which a number of lessons learned were used as feedback to the formal risk management pre-project process of a large corporation.
• In general, lessons are used to feed into improving project-management processes— the primary reason for carrying out post-audits, according to Azzone and Maccarrone’s (2001) survey of 124 Italian fi rms (albeit with only 34 responses).
• Similarly, lessons are used to improve management decision-making, which is identifi ed as the second reason in Azzone and Maccarrone’s (2001) survey, and as a main benefi t by 26% of the respondents in Neale and Holmes’ (1990) survey.
• Projects are part of a cycle and lessons learned can be tested and experimented with during the next generation of the cycle. Kotnour (1999), for example, presents learning from projects as a plan-do-study-act cycle (adapted from quality management), occurring both intra- and inter-project, where the “act” is making use of the lessons learned in future projects. He views learning cycles as an inherent part of the project and series of projects, not just a separate activity at the end (an empirical study of 43 project managers showed this is how they view learning), and indeed “every step in the project management process, if viewed from the learning perspective, can serve as the basis for producing and sharing knowledge for the project team.”
• Lessons learned procedures are important to disseminate knowledge within the project team, beyond the team to other projects, and even to other organizations. Busby’s (1999a) study of post-project review meetings highlighted one main strength as disseminating knowledge within the project team and promoting remedies; dissemination to other projects was also mentioned. Going beyond the project team, Gulliver (1987) describes the infl uential post project appraisal (PPA) unit within British Petroleum, with two key elements being (a) its independence and (b) the fact that PPA is company-wide so projects can be reviewed and lessons passed on to project teams elsewhere in the world. Holt, Love, and Li (2000) describe how learning underlies inter-company alliances: being able to learn collectively.
• Lessons learned are useful for benchmarking. Garnett and Pickrell (2000) describe some action research developing a methodology for benchmarking in the construction industry, noting that much of the benefi t was derived from generating and sharing ideas in the interactive activity (a social constructivist view) rather than fact fi nding about hard measures (positivist). (Note that the idea of project benchmarking [Ottmann 2000], for example, learning from others’ projects is a different topic.)
12 Post-Project Reviews to Gain Effective Lessons Learned
• Post-project audits to capture lessons learned can also have the side-benefi t for senior management of being able to check on the performance of their personnel (Azzone and Maccarrone’s  third reason) or on their project managers’ expertise (the main aim for 22% in Neale and Holmes’  survey).
• Kumar and Terpstra (2004) note the key role that a post-mortem can play at the stage-gates of the new product development process, as lessons learned in one (usually less costly) phase can feed through to the next phase.
However, Learning lessons and disseminating the knowledge gained from them is not simple. Barnes and Wearne (1993) in their look into the following 25 years of project management said “A problem which may well continue is that of presenting the lessons of the completed projects . . . in a form which is brief enough to attract busy people . . . yet is specifi c enough to persuade them.” Two particular issues in the exercise need particular attention to make the lessons useful:
• The need to gain depth in the lessons rather than obvious or simple lessons. Busby’s (1999a) survey identifi ed one limitation of the reviews he studied as shallow diagnosis, he claimed because of (a) a preference for causal rather than diagnostic reasoning (diagnostic learning is harder to do and involves more blame, which is socially awkward but leads to deeper diagnosis), (b) not enough “why,” (c) a norm of constructive criticism inhibits criticism with no immediate solutions. This highlights the main issue—the need to look into the systemic reasons for project outcomes—which we will look at further in Section 2.5.3; the issue is highlighted in the case study in Williams (2005).
• Secondly, there is the need to gain generalizable lessons rather than lessons specific to that one project (see Toft’s  distinction between organization- specific learning and isomorphic learning [universally applicable lessons gained from analysis of factors]). Lack of generalization was another main limitation that Busby (1999a) saw in the reviews he studied. One good example of reviews being used for general experimentation is Cooper, Lyneis, and Bryant (2002), who present their work as a dynamics-based “learning system” for cross-project learning, used by managers to test ideas and see impacts, and to record best practice.
Procedures to learn and disseminate lessons from projects need to be organized. Davies and Brady (2000) claimed in their study of suppliers of Complex Product Systems that “learning tends to be on an ad-hoc basis, with few systematic efforts to spread the initial learning throughout the organization.” Ayas (1996) says that “learning within a project does not happen naturally; it is a complex process that needs to be managed. It requires deliberate attention, commitment, and continuous investment of resources . . .
Literature Survey 13
Learning . . . has to be managed together with the project and must be integrated into project management as standard practice.” These procedures cannot stand alone: Cavaleri and Fearon (2000) considered that organizational learning is unmanageable if seen as an adjunct to other processes and has to be integrated into core processes; they provide a framework for integrating organizational learning with project management.
Learning is so fundamental to projects that many writers espouse projects particularly as learning mechanisms. Ayas and Zeniuk (2001) discuss promoting projects as learning vehicles and developing communities of practice. Similarly, Sense (2003b) discusses the importance of managing learning in project teams and describes project teams as embryonic communities of practice as they provide an opportunity to learn provided that their focus is shifted to include learning. Arthur, DeFillippi, and Jones (2001) also look at project-based learning, classifying project success by both dimensions of project performance and learning.
Indeed, Bredillet (2004a) claims projects are the key learning arena in organizations. Brady and Davies (2004) show how knowledge generated by learning from projects can lead to far-reaching changes in the strategic focus of an organization. Project-led (bottom- up) learning happens in phases: an exploratory “vanguard project” phase, project-to- project (through lessons learned), and project-to-organization, which is fed back to senior management and used to formulate new strategy. As fi rms advance through the phases, there is a transition from exploration to exploitation, and businesses can plan to move quickly to exploitation (top-down learning).
There are four general concepts that need to be briefl y considered before we can look at how people learn from, and increase their knowledge within, projects:
• For an individual, what is knowledge?
• For an individual, what is learning?
• For an organization, what is organizational learning?
• For an organization, what is knowledge management?
2.3.1 Knowledge The question of what knowledge is has exercised many authors since the time of Aristotle; clearly “knowledge” is not the same as “information” (McDermott 1999). It is not within the scope of this book to cover all of these arguments.
In this matter, however, the work of three authors clearly stands out. The fi rst is Polanyi
14 Post-Project Reviews to Gain Effective Lessons Learned
(1962), who established the idea of knowledge being internal and personal, or “tacit” and thus not necessarily easy to codify (as opposed to “explicit” knowledge, which can be expressed and shared in highly specifi ed formats). Subsequent work on how people learn and what personal knowledge means is based on his seminal work. However, individuals who learn and keep knowledge to themselves is not suffi cient to help organizations learn and develop. As the requirement to help develop “learning organizations” grew, the work of two other seminal authors has been particularly infl uential. Senge’s work (Senge 1990 and Senge et al. 1994) described the “learning organization,” and this work will be referred to later. The other particularly infl uential author (and the second whose work on “knowledge” has been particularly infl uential) was Nonaka (particularly Nonaka and Takeuchi 1995 but also Nonaka 1991), who described how Japanese companies working in innovation created “knowledge-creating” companies, and whose work requires a careful explication of what is “knowledge.” This work includes a model of knowledge creation within the organization based on the interrelationships between tacit and explicit knowledge.
When looking at knowledge as “tacit,” “tacit knowledge” is diffi cult to defi ne and operationalize. Cook and Brown (1999) provide a distinction between explicit and tacit knowledge as follows: explicit knowledge can be spelled out or formalized, and tacit knowledge is that associated with skills or “know-how.” Ambrosini and Bowman (2001) look at the defi nition of tacit knowledge and within the context of the resource-based view of the fi rm, redefi ne it as tacit skills; this enables the authors to propose ways in which tacit knowledge can be operationalized, using a methodology incorporating causal mapping, “Self-Q,” semi- structured interviews, and the use of metaphors. Johnson, Lorenz, and Lundvall (2002) say that learning (of “tacit” knowledge) is not just about codifying knowledge; when one is deciding whether to codify one needs to take into account: (a) the amount that will be lost in the transformation process and (b) whether codifi cation is an improvement or not. They recognize that codifi cation can stimulate learning when used to refi ne models and create shared vocabulary and when used to support the process of “refl ection, explication, and documentation of practices.” They then make distinctions between four different kinds of knowledge: know-what, know-why, know-how, and know-who.
Isabella’s (1990) work looks at the development of knowledge within an organization— particularly relevant to the development of knowledge within projects—as a process of sense-making. Isabella looks at how managers construe events around them and make sense of events in their organization. This will become an important theme later in this section, and also as we explore the ideas in Sections 2.3.2, 2.5.2, and 2.6.1. A key element of knowledge for this study is how groups of managers learn rather than individual managers. Knowledge creation is a social process as well as an individual process, sharing tacit knowledge (see von Krogh 1998, based on some ideas of Nonaka and Takeuchi ). Thus our understanding of what knowledge is has to look at the subtle interplay between tacit and explicit, and between individual and group knowledge. Cook and Brown (1999) consider that traditional understanding of the nature of knowledge is based on the “epistemology of
Literature Survey 15
possession” as it treats knowledge as something people possess, but that this epistemology cannot account for the knowing found in individual and group practice—so that knowing as action calls for an “epistemology of practice.” Cook and Brown claim that the literature tends to treat knowledge as being essentially of one kind and that the epistemology assumed in the literature tends to favor the individual over the group and explicit knowledge instead of tacit knowledge. However, they believe that: “Organizations are better understood if explicit, tacit, individual, and group knowledge are treated as four distinct and coequal forms of knowledge (each doing work the others cannot), and if knowledge and knowing are seen as mutually enabling (and not competing)” [Paper Abstract], and go into detail on these four distinct and coequal forms of knowledge. They also see a distinction between knowledge that is “part of practice” and that knowledge which is “possessed in the head.” They refer to the latter as knowledge and the former as knowing. The focus of the article is more on knowing, and the authors discuss the interplay between the two terms as a way in which new knowledge and new ways of knowing are formed. The authors draw on Dewey’s concept of productive inquiry in examining the way in which knowledge can be visualized as “a tool at the service of knowing.” To help in this exposition, the authors apply their perspective [of seeing distinct forms of knowledge and of viewing knowledge as a tool of knowing] to three cases, which they say “help make clearer some of the actionable and theoretically signifi cant implications of this perspective.”
Schulz (2001) also looks at the relationship between the production of knowledge and the fl ow [vertical or horizontal] of knowledge within organizations. This topic is felt to be important as “. . . each process conceivably depends on the other.” The author fi nds that the production of knowledge in organizational subunits of a fi rm affects the outfl ow of knowledge to other units of the same fi rm: horizontal fl ows are to peer units and vertical fl ows are to supervisory units. The author found that “exposure to internal and external sources of newness” and the “uniqueness of experiences” intensify vertical outfl ows of knowledge, but do not affect horizontal outfl ows. On the other hand, “reciprocating” and “substitution” affect horizontal outfl ows but affect vertical outfl ows much less (the paper includes descriptions of these variables). “It thus appears that collecting new knowledge increases vertical outfl ows and combining old knowledge intensifi es horizontal outfl ows . . . different kinds of knowledge fl ow in different directions: New knowledge fl ows mainly vertically, and incremental knowledge fl ows mainly horizontally” (p. 674).
Later, we shall return to the way in which knowledge fl ows within an organization, but it is worth noting that Lesser, Fontaine, and Slusher (2000) explain how organizations are using communities in order to enhance the creation, sharing, and application of knowledge. They highlight people, places, and things as being the three basic components of communities; while IT is also important, the main emphasis is on the social aspects involved in the processes of creating and sharing knowledge.
A key element in understanding the nature of knowledge creation (that is, the manner in which knowledge is made) is the manner in which knowledge is justifi ed. von Krogh and
16 Post-Project Reviews to Gain Effective Lessons Learned
Grand (1999) “. . . conceive justifi cation as the permanent corporate and management activity of relating issues and tasks to a generally accepted corporate knowledge base . . . [justifi cation] means understanding the mechanisms which decide whether new insights, concepts and ideas are rejected, returned for further elaboration, or fi nally appropriated as relevant” (p. 16–17). They use the concept of “dominant logic” to try to understand justifi cation.
This discussion pertained to the nature of knowledge in general. Why are these aspects particularly relevant for understanding knowledge within projects? This is because of the following:
• Projects are complex systems, which means that the way in which we organize our thinking about those systems is also complex (Tsoukas and Hatch 2001). They point out that the nature of this complexity affects how we generate knowledge, in particular making narrative types of thinking more appropriate than logico- scientifi c (i.e., propositional) thinking.
• Real projects are often very much more concerned with sense-making rather than carrying out a full-formed plan, which affects how we can learn from projects (Ivory et al. 2004). Projects frequently take place in a context of confusion, and controversy can help to create new knowledge (Fernie, Green, Weller, and Newcombe 2003).
• Projects are temporary organizations, and there are issues around the lack of time for developing trust in such organizations, which means that knowledge generation and sharing is different from that in permanent organizational structures (Koskinen, Pihlanto, and Vanharanta 2003).
• The whole idea of project management imposes an ontology and a specifi c way of thinking within a company; this immediately frames ways of thinking when reviewing projects, and can pose diffi culties in critically thinking through what really happened (Hodgson 2002).
• We have to learn not just about technical aspects but how our social structures have behaved. For example, Vaughan’s (1996) analysis of the Challenger launch decision concluded that the Challenger disaster was not “a technical failure due to managerial wrongdoing and production pressures” (as cited for instance in the U.S. Presidential Commission Reports) but “a mistake embedded in the banality of competition, an unprecedented, uncertain technology, incrementalism, patterns of information, routinization, organizational and interorganizational structures, and a complex culture.” The argument is that “mistakes and disasters are socially organized and produced by social structures and culture” (quotes taken from Mitev 1998).
Literature Survey 17
2.3.2 Learning If this is what knowledge is, what does learning mean within the context of projects? Clearly these are related entities: “Learning and knowledge are intertwined in an iterative, mutually reinforcing process. While learning (the process) produces new knowledge, knowledge impacts future learning” (Vera and Crossan 2003 cited in Scarbrough et al. 2004).
Mumford (1994) says that we need to enable managers to recognize opportunities for learning and learn effectively from experience. He describes four approaches to learning: (a) intuitive (no explicit process, not aware of learning happening), (b) incidental (refl ection following some jolt, e.g., a mishap), (c) retrospective (learning by routinely reviewing, not just mishaps), and (d) prospective (opportunities to learn are identifi ed in advance). Managers are not shown how to refl ect on what has happened or to plan to learn. Some managers are not aware of the benefi ts of doing so, and this limits the quality of learning they can achieve.
However, clearly refl ection plays a key part in this process. Smith (2001) says that “Most of the time we have experiences from which we never learn” and describes the framework and tools for refl ective learning in an organization. Winter and Thomas (2004) believe that project management is less about applying specifi c techniques and more about the powers of managing, which implies that a project manager’s education or professional development should focus on developing practitioners’ critical awareness and refl ective practice. Scarbrough, Swan, and Preston (1999) provide a good review of recent literature on the process of learning by refl ection.
Zollo and Winter (2002) describe three types of learning behaviors: tacit accumulation of experience (semiautomatic), knowledge articulation, and knowledge codifi cation (deliberate). They indicate that dynamic capability (“a learned and stable pattern of collective activity through which the organization systematically generates and modifi es its operating routines in pursuit of improved effectiveness”) arises from the interaction of the three learning behaviors.
As discussed in the previous section, refl ection and learning within projects takes place within a team. Learning is not only about acquiring information, but also socialization and requiring appropriate social contexts (Gherardi, Nicolini, and Odella ). Raelin (2001) discusses learning through refl ection with others and provides the what, why, and how to do it, claiming this is particularly applicable to learning from projects.
Two names with very particular lines of approach should be noted. The fi rst is Ralph Stacy, whose work on complex responsive processes in organizations shows the importance of socially constructed knowledge creation such as narratives and community of practice. Stacy (2001), for example, outlines Weick’s ideas about sense-making and mental models and incorporates them into the structure of complex responsive processes. This work will be important as we develop ideas about knowledge and understanding arising from the complexity of organizations and projects. The second is von Glasersfeld (1995), who can be identifi ed as the key writer in radical constructivism and who stated its basic principles
18 Post-Project Reviews to Gain Effective Lessons Learned
as being that knowledge is not passively received either through the senses or by way of communication but is actively built up by the cognizing subject; and that the function of cognition is adaptive and serves the subject’s organization of the experiential world, not the discovery of an objective, ontological reality: “Causality, then, is part of the design that reason imposes on experience to make it understandable” (p. 42); this work will be important as we consider “knowledge” and “knowing” with a suitably critical approach (whatever our own individual ontological or epistemological stance).
Johanssen, Olaisen, and Olsen’s (1999) paper clearly explains some of the principles related to learning. For example, the relationship between information and action in the concept of learning: “For most people learning will most likely be synonymous with information acquisition (Machlup).” However, we do not learn by acquiring information. We neither learned how to swim, nor to read or cycle, by acquiring information on swimming, reading, or cycling (Polanyi). There is, in other words, no learning separated from action (Nonaka and Takeuchi). Action is not enough to learn—time has to be built in for refl ection relative to the action as well as what is learned (Bandura, Piaget, Rolf). “A major part of learning is then carried out by means of the processes: planning, action, refl ection.”
2.3.3 Organizational Learning The move from the individual to the organization is not simple. Simon (1991) notes that: “All learning takes place inside individual human heads; an organization learns in only two ways: (a) by the learning of its members, or (b) by ingesting new members who have knowledge the organization didn’t previously have. But . . . what an individual learns in an organization is very much dependent on what is already known to (or believed by) other members of the organization and what kinds of information are present in the organizational environment.”
Kim (1993) stresses the importance of making the distinction between the organization and the individual, explicit in the analysis of organizational learning, and develops a model showing the links between individual and organizational learning by means of “shared mental models,” what he describes as “the thought constructs that affect how people and organizations operate in the world.” Cross and Baird (2000) talk about the need to embed learning within the wider company; they take a proactive stance, targeting learning when it needs to be learned.
As we move from the individual to the organization, we should note the inter-relation between these two: DeFillippi and Arthur (2002) show how workplace learning (knowledge creation and sharing) happens in four contexts: individual, company, community, and industry; and while these are usually studied in isolation, they argue that they are strongly inter-related so it is just as important to understand the connections between them. “Each learning context is a contested domain in which the context and actors operating within the context exert reciprocal infl uence on each other. It is out of such contested,
Literature Survey 19
reciprocal interactions that meaning is constructed and learning arises.” Furthermore, for many, the process of refl ection and knowledge creation is a social process. Purser, Pasmore, and Tenkasi (1992), for example, looked at deliberation in R&D teams and showed that learning for complex project development was enhanced by deliberations. They noted that small informal forums were conducive to knowledge-sharing and active inquiry—by exposing people to the big picture of the overall product system, using a participative decision-making approach. Newell and Huang (2005) use Nonaka’s work to look at knowledge creation in multidisciplinary teams, and note the importance of the collaborative nature of teams. Fong (2002) similarly assesses Nonaka and Takeuchi’s (1995) knowledge creation model in looking at knowledge creation in multidisciplinary project teams. Fong proposes a model of the processes of knowledge creation within such teams made up of fi ve processes: boundary crossing, knowledge sharing, generation, integration, and collective project learning. Ramaprasad and Prakash (2003) point out that one needs to take local knowledge and integrate it with generic knowledge, and their methods of learning are based on ideas “emerging” from the project, based on concepts such as constructionism and critical thinking. Crossan, Lane, and White (1999) put this into a neat framework as they give what they call “the four I’s of organizational learning”: intuiting, interpreting, integrating, and institutionalizing, referring to the individual, individual, group, and organization level, respectively.
Spender (1996) tries to pull together the ideas of organizational knowledge, organizational learning, and memory. “. . . The fragmentation of the [organizational learning] literature is the result of the two methodological manoeuvres institutionalized into the contemporary analysis of organizational knowledge. The fi rst separates the notions of knowledge, learning, and memory, presuming each can be treated independently. But we see the three concepts are interdependent parts of a single system of ideas about organizations and their knowledge processes . . . This triangle of interdependency and interdefi nition is the foundation on which the rest of the organizational system must be built.” Spender then presents different types of organizational knowledge and how they interact.
The motivation for companies to become learning organizations clearly gained considerable momentum with the infl uential work of Senge with his “fi fth discipline” work (Senge 1990 and Senge et al. 1994). This work recognizes the complexity and systemicity inherent in organizational activity and sets the theme for our discussion of “systemicity” in Chapter 5 (see in particular Kofman and Senge 1993).
Some authors describe the power of learning to transform and reinvent the organization: Coutu (2002) (quoted in Morris and Loch 2004b) interviewed by Edgar Schein said “Despite all the time, money and enthusiasm that executives pour into corporate change programs, the stark reality is that few companies ever succeed in genuinely reinventing themselves. That’s because the people working at those companies more often than not fail at transformational learning—they rarely get to the point where they are eagerly challenging deeply held assumptions about a company’s strategies and processes and, in
20 Post-Project Reviews to Gain Effective Lessons Learned
response, thinking and acting in fundamentally altered ways. Rather most people just end up doing the same old things in superfi cially tweaked ways—practices that fall far short of the transformational learning, learning that most experts agree is the key to competing in the twenty fi rst century.”
Garvin (1993) defi nes and describes the “gritty details” of a learning organization. He discusses fi ve building blocks to becoming a learning organization: (a) systematic problem solving, (b) experimentation, (c) learning from past experience, (d) learning from others, for example, benchmarking (“learning organizations . . . cultivate the art of open, attentive listening”), and (e) transferring knowledge, for example, reports, tours, and staff transfers. For building block (c), he discusses IBM’s 360 computer series, one of the most profi table ever made, which was based on a previously failed program, but he says this was by chance rather than policy and compares this with Boeing, who carried out a lessons learned after the 737 and 747 programs, and then transferred the lessons and some of the people who carried out the exercise onto the very successful 757 and 767 programs.
We are not going to summarize here all of the vast literature on organizational learning. However, key reviews of the vast literature on organizational learning can be identifi ed. Huber (1991) gives an early and authoritative summary of the literature about organizational learning to 1991, dividing it into four aspects: knowledge acquisition, information distribution, information interpreting, organizational memory, listed in decreasing magnitude of maturity. Dodgson (1993) provides a similar bibliography of organizational learning generally, and notes the range of disciplines that contribute to the fi eld, with organization theory plus some psychological interpretations to comprehend the process, problems, and learning. He also discusses the approaches related to economics/management/innovations, which look at the motives and source of learning, concluding also that “together, the literatures reviewed contribute to the understanding of the complexity of factors that encourage and restrict learning.” Nair’s (2001) literature review, charting developments in the understanding of organizational learning since the 1950’s, concludes that organizational learning systems still did not have a sound theoretical base (he proposed developing a classifi cation of organizational learning systems by their complexity as a way to learn more about them). Edmondson and Moingeon (1998) also review the literature and categorize it into four groups based on the unit of analysis and research objectives: (a) organizations as residues of past learning, (b) organizations as collections of individuals who can learn, (c) organizational improvement gained through intelligent activity of individual members, and (d) organizational improvement gained through developing individuals’ mental models (note that this review is focused on the individual rather than the team).
Two recent authors analyze the literature with regard to the development of the fi eld over the recent past. Bapuji and Crossan (2004) review literature on empirical organizational learning research published between 1990 and 2002, and some of their results are noted later. Easterby-Smith, Thorpe, and Lowe (2000) try to map the development of the fi eld
by reviewing the set of papers (102 in total) submitted to a particular 1999 organizational learning conference; they follow the genesis, progression, and decline of several debates in the field, some of which include:
Debate about units of analysis: if organizational or inter-organizational learning is the sum of what individuals learn, or if there is something more (including the role of the group);
Debate on distinguishing between changes in cognition and changes in behavior;
Debate around single- and double-loop learning;
Debate between organizational learning and the learning organization;
Debate about the nature and location of organizational learning (through interpersonal interactions rather than within individuals or organizational systems);
Debate on how to investigate organizational learning;
Tension between the ideas of organizational learning and knowledge management;
Shifting in focus towards a closer scrutiny of workplace activities and work practices;
Need to reconcile learning with diversity; particularly due to globalization;
Focus on power, politics, and trust “three fundamental dimensions of learning.”
One authoritative text drawing the work together is given in the Handbook of Organisational Learning and Knowledge (Dierkes et al. 2001), giving 42 chapters of work ranging from insights on organizational learning from the major social science disciplines through to putting the knowledge into practice.
This body of work has contributed to a number of influential “how to” books. For example, DiBella and Nevis (1998) wrote How Organizations Learn, a strategic package that can be used to examine and enhance the “learning capacity” of any organization. The book describes the circumstances in which organizational learning takes place, with 10 practices or conditions which promote learning within all kinds of organizations (including experimental mind-set, climate of openness, systems perspective, etc.). It defines seven learning styles and a measurement tool for gauging an organization’s performance along a continuum for each of the seven learning orientations and 10 facilitating factors. (Early versions of this work are mentioned in Nevis, DiBella, and Gould ). Schwandt and Marquardt (2000) wrote Organizational Learning: From World-Class Theories to Global Best Practices, which presents the authors’ model for organizational learning, Organizational Learning Systems Model (OLSM), which is founded on social action theory (specifically
22 Post-Project Reviews to Gain Effective Lessons Learned
the general theory of social action posed by Talcott Parson). In addition to setting the theoretic framework for the OLSM and describing it in detail, the book also includes “action-based recommendations for organizations,” recommending seven steps (said to have been gleaned from best practices in the authors’ experience) towards organizational learning (using the OLSM as a frame of reference).
Two issues are clearly important in understanding organizational learning: culture and organizational structure. Culture plays a key role in how organizations learn and whether or not they learn. This applies both to national culture and internal corporate culture. Carmona and Grönlund (1998), for example, looked at two situations, in Sweden and Spain, whereby learning was achieved by problem-solving at the operations level. They suggest that the subsequent deterioration of learning in one of the situations presented was infl uenced by both the structure of the organization and national culture. High power distance and high uncertainty avoidance in Spain meant that middle managers had less responsibility and less decision-making power, so they were less able to implement changes than in Sweden. Kidd (1998) also analyzed organizational learning in some Japanese-Italian companies using Nonaka’s SECI model, and found a clash of culture and context, so learning in the donor company (Japan) could not easily be translated to the receiving company (Italy). Kidd suggested that optimal learning happens in cases where there is (a) training by the donor company, (b) opportunities for peer group discussion, (c) empowerment at the local level (not just compliance), (d) appreciation of tacit knowledge held by the receiving company, and (e) good exchange of data. Kidd noted that elitism and rigid processes or systems were both hindrances. In terms of corporate culture, Lipshitz, Popper, and Friedman (2002) claimed that learning depended on structure, culture, psychology, policy, and context. Five cultural values that promote productive learning are: (a) transparency (the willingness to expose one’s thoughts and actions to others in order to receive feedback), (b) integrity (the willingness to seek and provide info regardless of its implications), (c) issue-orientation (focusing on the relevance of information to the issue under consideration regardless of the social standing or rank of source or recipient), (d) inquiry (persisting in investigation until full understanding is achieved), and (e) accountability (willingness to assume responsibility for learning and implementation of lessons learned). Reger and von Wichert-Nick (1997) state that structure and culture of an organization are important factors in determining how effectively it can learn, which requires a culture that supports teamwork, a culture that supports experimentation and a culture that is open to risks. Wreme and Sorrenti (1997) describe cases where process consultants used systems-thinking tools with organizations to help them to learn, highlighting the importance of a safe environment for learning rather than a controlling organizational culture.
The works previously cited note the importance of culture and organizational structure. Lipshitz, Popper, and Friedman (2002) state that “For learning to become organizational, there must be roles, functions, and procedures that enable organizational members to systematically collect, analyze, store, disseminate, and use information relevant to their
Literature Survey 23
own and other members’ performance.” They use the term “organizational learning mechanisms” for “observable organizational subsystems in which organization members interact for the purpose of learning.” Reger and von Wichert-Nick (1997) argue that organizational learning needs hierarchy-free communication and fl ow of information, a primary structure that is hierarchical and a secondary structure that is “supra-hierarchical and coordination-oriented.” Carmona and Grönlund (1998) suggest that strict budgeting in an organization can constrain implementation of learning, teams need to be recognized by the rest of the organization, and the benefi ts of a reward system (both intrinsic [e.g., making people feel their jobs are important] and extrinsic [e.g., fi nancial]) are important.
Bapuji and Crossan (2004) in their analysis of the empirical literature since 1990, identify facilitators to organizational learning, as well as culture and structure. They highlight strategy, environment, organizational stage, and resource position (although the contribution of the last of these is unclear). It is worth noting that Örtenblad (2002), rather than looking at structure, culture, etc., as facilitating factors to organizational learning, synthesizes the literature (including DiBella, Easterby-Smith and Araujo, Finger and Bürgin Brand, and Argyris) and sees four different meanings of the term learning organizations: those with organizational learning, those where people learn at work, those organizations with a learning climate, and those with a learning structure.
There are also a number of other issues that need to be considered when looking at how organizations learn. These include:
• It is important that organizations learn productively: Lipschitz, Popper, and Friedman (2002) say that “productive organizational learning is a process that is (a) conscious and systematic, (b) yields valid information and (c) results in actions intended to produce new perceptions, goals, and/or behavioral strategies.”
• It is important to learn from a balance of both success and failures. For example, Denrell (2003) shows the bias towards success when observing management practices, because the ones that fail aren’t there to be observed.
• The role of IT support is important. Venugopal and Baets (1995), for example, discuss how IT tools can support organizational learning; they describe different learning processes (learning through cases, participative strategy formation, sharing individual knowledge, and exploratory knowledge) and the IT tools which can support them (that is, in particular databases, group decision support systems, cognitive mapping, and artifi cial neural networks).
• The increasing use of contingent work (such as contractors), especially among high tech fi rms, offering reduced cost, greater fl exibility, and technical expertise, can have a signifi cant effect on a fi rm’s ability to create and accumulate knowledge. “Contingent work can bring public knowledge into the fi rm, such as industry best practices; moreover, it can have a catalytic effect on the knowledge-creation process, helping to create new
24 Post-Project Reviews to Gain Effective Lessons Learned
private [fi rm specifi c] knowledge. However, it may also act as a conduit through which private knowledge leaks into the public domain.” (Matusik and Hill 1998).
2.3.4 Knowledge Management As an organization learns, there is a need to manage proactively the knowledge within the organization. The discipline of knowledge management has been developing over a number of years. Starbuck (1992) said that “creating, applying and preserving [knowledge] intertwine and complement each other. At least over long periods, merely storing knowledge does not preserve it. For old knowledge to have meaning, people must relate it to their current problems and activities.” Its importance for the development of organizations is well-recognized: for example, Coombs and Hull (1998) note the role of knowledge management in promoting innovation and describe how different knowledge management practices can infl uence the extent to which an organization is able to generate variety and create radically new knowledge rather than being limited by existing knowledge, shared routines, shared mental models, etc. While knowledge management is often given a fairly narrow defi nition, O’Dell and Jackson Grayson (1998) describe it simply as “a conscious strategy of getting the right knowledge to the right people at the right time and helping people share and put information into action in ways that strive to improve organizational performance” (cited in Levene and Gale 2000). It is now a widespread and accepted part of corporate life—for example, in the construction industry (a particularly projectized industry), Robinson, Carrillo, Anumba, and Al-Ghassani (2004) report on the state of knowledge management in 2004 in the construction industry and describe a framework for structuring and implementing a knowledge management strategy, noting the importance of linking it to business strategy and performance.
There are different types of knowledge management processes for different types of situations. Baumard’s (1999) Tacit Knowledge in Organizations, at a basic level, quotes Nonaka’s presentation giving four different ways of transferring knowledge: (a) tacit to tacit is socialization, (b) tacit to explicit is articulation, (c) explicit to explicit is combination, and (d) explicit to tacit is internalization; this book also provides a complex table of different tacit and explicit knowledge embodiments in organizations.
In a more complex level of categorization, Blackler, Crump, and McDonald (1998) divide knowledge into the following categories:
• embodied (action-oriented)
• embedded (in systemic routines)
• embrained (abstract knowledge)
• encultured and/or
• encoded (conveyed by signs and symbols).
Literature Survey 25
These categories use a number of types of knowledge processes, such as:
• provisional and refl exive (actively and creatively constructed rather than eternal verities)
• mediated by linguistic and technological infrastructure (people operate within interpretive or discourse communities)
• situated and pragmatic
• contested and political
• emotional as well as rational.
Although a full treatment of the huge literature on knowledge management in general is outside of the scope of this book, a few relevant summaries of the fi eld will be referenced. von Krogh, Roos, and Kleine (1998) give an overview of the state of the art in 1998. In terms of actual practice, Ruggles (1998) describes some knowledge management activities, based on a study of over 400 organizations and which include barriers to successful implementation. Prichard, Hull, Chumer, and Willmott (2000) give a compilation of articles that provide critical perspectives on work and organization. Specifi cally, broad perspectives on the developmental aspects of knowledge management, particularly exploring (inter al) the impact of knowledge management on the working life of professionals, the effect of knowledge management upon the internal mechanisms within organizations, issues relating to decision-making, and the extraction of wealth from knowledge. And of particular note, Scarbrough, Swan, and Preston (1999) give a full review of literature on knowledge management and learning organizations from 1993 through 1998. They note that literature regarding learning organizations has been declining in the latter years while that on knowledge management was increasing, largely in the IS/IT literature, much of it in practitioner-oriented journals: “The dominant discourse of KM [Knowledge Management] (to capture, codify, use and exploit the knowledge and experience of employees by developing better tools and methods and by developing a willingness and ability to use those methods) is fundamentally different to that of the LOs [Learning Organizations] (to harness the learning capability of the fi rm and individuals within it through people development, empowerment, leadership and culture change).”
A number of authors have warned against the frequency with which knowledge management work restricts its attention to knowledge that can be codifi ed. Johnson, Lorenz, and Lundvall (2002) make the common distinction between know-what, know-why, know-how, and know-who knowledge and note that learning isn’t simply about codifying knowledge. When deciding whether to codify knowledge, one needs to take into account both the amount that will be lost in the transformation process and whether codifi cation is
26 Post-Project Reviews to Gain Effective Lessons Learned
an improvement or not (although they recognize that codifi cation can stimulate learning when used to refi ne models and create shared vocabulary and to support the process of refl ection, explication, and documentation of practices). Connell, Klein, and Powell (2003) cover the literature on knowledge management and point to shortcomings with both the personalization view of knowledge management (knowledge is considered inseparable from the person or group that holds it) and the codifi cation view (knowledge is seen as a commodity which can be isolated and codifi ed); they make the point that one needs to consider knowledge as a property of the system, and knowledge needs to be embedded within the context.
The concentration on codifi cation is perhaps due to the frequent view of knowledge management (KM) as part of IT. Scarbrough, Swan, and Preston (1999) quote Cole- Gomolski (1997), who states “the idea behind knowledge management is to stockpile workers’ knowledge and make it accessible to others via a searchable application” to illustrate the view that KM consists simply of making a database. Venugopal and Baets (1995) do take a wide view of how IT tools can support organizational learning (as was seen in 1995). They describe different learning processes (for example, cases, participative strategy formation, sharing individual knowledge, and exploratory knowledge) and the IT tools that can support them (database, group decision support system, cognitive mapping, and artifi cial neural networks). McDermott (1999) has already been quoted previously as warning that IT cannot deliver knowledge. In more detail, Scarbrough, Swan, and Preston (1999) talk about the problem of knowledge management being only interested in IT solutions: “This obsession with tools and techniques falls foul of at least four basic problems. First, it assumes that all knowledge is codifi able, which clearly it is not . . . Secondly, it overemphasizes the utility new information technologies have for delivering organization performance improvements . . . Third, it assumes that, even if perfect systems existed, people are willing to make them work, i.e. to contribute their knowledge to the systems, to share their knowledge and to use the knowledge from the systems, which clearly they are not, or at least not readily. Fourth, the codifi cation and objectifi cation of all tacit knowledge into formal systems . . . generates its own psychopathy – i.e. the fl uid, organic, informal and intuitive practices that are essential in allowing the fl exible fi rm to cope with uncertainty will rigidify.”
Stein and Zwass (1995) indicate that while an organizational memory system can both support and impede higher level learning, an over reliance on it can lessen the degree to which people will question underlying assumptions and procedures and engage in exploratory learning.
Literature Survey 27
2.4 The Current Situation
2.4.1 Standards and Maturity Models Which standards are used for drawing lessons from projects? This is such a key element of project management that one might expect that there should be clearly defi ned guidelines and ways of measuring an organization’s effectiveness at learning project lessons.
Morris, Patel, and Wearne (2000) in their research used to revise the British project management “Body of Knowledge” (BoK) looked into what should actually be included in a BoK, and 80% of project practitioners agreed that “post-project evaluation review” should be included. In the expanded version of the British BoK, the Pathways book, Wateridge authors a chapter on project-implementation reviews (which considers “the reasons behind the variances between the plan and actual spend and effort” and “what would have been done differently with the benefi t of hindsight”), project health-checks (which look at the stakeholders’ view of the direction of the project and whether the project as a whole is moving in the direction to succeed), and project audits (particularly looking at fraud), although this chapter only gives general guidance (Wateridge 2002).
The Project Management Institute’s A Guide to the Project Management Body of Knowledge (PMBOK® Guide – Third Edition), an ANSI standard and the basis of PMI’s professional qualifi cation, gives the simple statement that “The causes of variances, the reasoning behind the corrective action chosen, and other types of lessons learned should be documented so that they become part of the historical database . . .” (Section 126.96.36.199). Strang (2003) notes the fact that lessons learned are included as inputs or outputs to almost all of the nine knowledge areas of the PMBOK® Guide, and in all fi ve process groups (and details many of the relevant subsections). Strang looks at theory about organizational learning in projects and looks at practice but concludes that the theory is rarely applied in practice and discusses the diffi culties. Lessons learned do come into the Project Management Institute’s OPM3 Organizational Project Management Maturity Model (Project Management Institute 2004), but with little emphasis or much guidance beyond the PMBOK® Guide.
PRINCE2 (Offi ce of Government Commerce 2002) also has a process for recording lessons learned and reporting on them. Lessons are captured in the lessons learned log; at the end of the project these are collated into a lessons learned report.
It is not just the project management BoKs that recognize this need: Turner et al. (2000) describe the use of procedures to represent captured knowledge and experience and stress the vital role of project reviews and say that “End of project reviews play a vital part in capturing experience within organizations. PRINCE2 and ISO 10006 suggest a review be conducted at the end of every project, and company standard procedures updated to refl ect this.”
So the standards generally require project reviews to be carried out, although with little guidance as to how this should operate. But is “learning lessons” featured in the more sophisticated “maturity models” that seek to defi ne how mature an organization is
28 Post-Project Reviews to Gain Effective Lessons Learned
in project management? Cooke-Davies and Arzymanow (2002) describe a study looking at how practices in project management differ between industries, carried out for a community of practice of project managers in the pharmaceutical industry; they identifi ed different levels of maturity in different industries but do not discuss differences in the way organizations learn. Nor did recent work on Organizational Project Management Maturity (e.g., Kalantjakos 2001); nor really the risk management maturity model standard (Risk Management Research and Development Program Collaboration 2002), in which there is no signifi cant discussion of post-project analysis and no closing the learning loop (“Level 3” has some data collection, and “Level 4” has a direct mention of “learning by experience”). But recent maturity standards work reported in Schlichter (2001) required the following:
• A process to capture and disseminate lessons learned
• “Lessons learned” stored in an accessible location
• Evidence of capture/dissemination and the reuse of information in subsequent projects.
Further, he looked for learning from projects that is translated into the corporate approach, and learning that takes place in all of its dimensions (i.e., individual, team, organizational).
Another work building further on this is from von Zedtwitz (2002), who gives a capability model only for post-project reviews. Using the standard fi ve-stage capability model, he defi nes stages:
• Initial: ad-hoc methods, reaction-driven reviews
• Repeatable: with sound review practices
• Defi ned: a standardized process with sound and consistent review criteria, “usually” with a small unit responsible for training
• Managed: review goals with “quantifi ed and measurable” quality criteria
• Optimized: post-project reviews established organization-wide with proactive review of processes.
2.4.2 Prevalence Given that the standards do require project reviews to be carried out, the literature is somewhat divided regarding its prevalence in practice.
Some papers state as an accepted fact that learning is rare. Gulliver (1987) says “in talking with business people from large British and multi-national corporations, I have found that few companies examine their completed projects in any depth.” Harris agrees,
Literature Survey 29
saying that “learning from past mistakes, or even building upon past successes, continues to be the exception rather than the rule.” Disterer (2002) says that “Only a few fi rms manage systematically to identify and transfer valuable knowledge from projects to following projects.” Carrillo et al. (2004) state that in construction management—a particularly relevant fi eld as it is project-based—“. . . teams frequently disband upon project completion without conducting post-project reviews and disseminating the lessons learned.” Keegan and Turner’s (2001) research suggested that “learning was unsatisfactory.”
Some papers on the other hand state as accepted fact that generally lessons learned activities and learning from projects generally occur in practice. Newell (2004) describes post-mortems as “ubiquitous.” Scarbrough, Swan, and Preston (1999) comment that “as fi rms increasingly become more innovative and project-based, many are recognizing the need to capture the learning from individual projects, and make it available throughout the organization. Consultancies, professional service fi rms, aerospace companies, etc., are in the vanguard of developing systems to codify and communicate such knowledge.”
More useful, of course, is actual statistics on the prevalence of project learning activity (this current PMI project will of course provide some large-sample statistics in a later document). None of the results give defi nitive or conclusive conclusions. Many describe post-project review as fairly common:
• Besner and Hobbs (2004) describe a survey of the usage of project management (PM) tools and techniques across different contexts and project types, in which “Lessons learned/post-mortem” is one of the most frequently used on all project types; however, the use of “database of lessons learned” differed signifi cantly between low and high PM maturity organizations.
• Kotnour’s (2000) infl uential paper was based on a survey of 43 project managers who were attending a PMI chapter meeting (so could be expected to be interested generally in project management learning)—of these, 31 said they completed lessons while 12 said they did not.
• Carrillo et al. (2004) (with 53 responses in the construction fi eld) found that 26% of companies did not have (or plan to have) a knowledge management strategy (although only 9.5% in the case of companies with more than 1,500 employees)
• Fong (2005) describes a survey of knowledge management among quantity surveyors in Hong Kong and the UK: in a table of “practices used,” “lessons learned from previous projects” came in as the third most common practice in the UK and fi fth in Hong Kong (although it wasn’t clear whether this referred to formalized procedures or not); however, this was clearly not seen as suffi cient in terms of what they wanted their companies to develop, 43% wanted them to develop knowledge- sharing skills, 39% wanted them to develop knowledge-capturing skills, and 18% knowledge-creating skills.
30 Post-Project Reviews to Gain Effective Lessons Learned
Others, on the other hand, found post-project reviews to be less frequent:
• An early piece of research here is described in Neale and Holmes (1990), who carried out a survey of fi nance directors from companies (with 410 replies) about their post- auditing procedures for capital projects (so the emphasis is on improving decisions about whether to invest in a project, e.g., better proposals, better evaluation, better fi nancial control): 48% used post-auditing techniques, with a bias towards the largest companies (large being a relative term as all the companies were taken from the Times 1000) and manufacturing companies.