Foreword
Today, innovations are key to survival in many companies. As more and more innovations are IT-enabled, corporate IT departments have to cope with a growing number of IT projects proposed by different organizational units. At the same time, funds for new IT investments are typically very scarce. Thus, procedures and mechanisms for selecting the most valuable projects in alignment with the corporate and the IT strategy are required. In this context, other factors like mandatory projects and the risk at the portfolio level need to be considered as well. For this purpose, the concept of IT project portfolio management has received growing attention in recent years.
In the current thesis, Thorsten Frey demonstrates how companies are struggling with the right balance between local autonomy and centralized control in the context of IT project portfolio management. This struggle is reflected in the governance arrangements employed. Thus, the primary research objectives addressed in this dissertation are to describe governance arrangements for IT project portfolio management as they are employed in practice and to identify antecedents for and performance impacts of distinct governance arrangements on the composition of IT project portfolios in different contexts. In order to address these research objectives, Thorsten Frey employs qualitative empirical methods as well as a quantitative analytical approach.
The first part of this book contains a comprehensive summary and analysis of theoretical backgrounds for the two covered research disciplines. With regard to IT governance research, relevant terms and concepts from organizational theory are introduced. While previous research on IT governance is only briefly summarized due to the pervasiveness of existing reviews, the state-of-the-art of IT project portfolio management research is analyzed in detail.
Based on an extensive review of scientific papers covering IT project portfolio management, Thorsten Frey identifies key concepts of IT project portfolio management research. He especially highlights the emergence of two different streams of research and the beginning merge of these two streams. Based on his structured analysis, he identifies opportunities for future research. Thereby, Thorsten Frey makes a significant contribution to the advancement of this important new research discipline. The second part of this book focusses on the design of governance arrangements for IT project portfolio management. For this purpose, outcomes of a qualitative empirical study are presented. Based on an extensive analysis of governance arrangements in ten case companies, Thorsten Frey identifies major concepts that are combined into a comprehensive framework. This framework explains relationships between different contingency factors and different governance designs for IT project portfolio management. Impacts of distinct governance arrangements on the composition of IT project portfolios are also analyzed. Based on insightful descriptions and citations, Thorsten Frey vividly demonstrates why different governance arrangements for IT project portfolio management exist in practice and how companies cope with the inherent conflict between local autonomy and centralized control.
The developed framework provides an important instrument for governance experts in order to evaluate the appropriateness of distinct kinds of governance arrangements in a given context.
In the third part of this book, coordination mechanisms for budget allocation and IT project portfolio management are investigated based on a mathematical analytical conception. This general conception allows for an analysis of the impact of various factors influencing the composition of IT project portfolios. In particular, the impact of different governance designs on synergy exploitation is formally analyzed in the current dissertation. This is of high relevance for IT project portfolio management research due to the pervasiveness of different kinds of inter dependencies between IT projects. The conception is also of high relevance for practitioners as it can be used in order to analyze the appropriateness of a particular governance arrangement in a given organizational context. Thereby, costly failures of reorganization projects can be avoided.
In the final part of the thesis the practical use of the findings and the theoretical frameworks described in the previous parts of the book are discussed. In this context, Thorsten Frey also outlines the steps IT governance experts should take in order to design appropriate governance arrangements for IT project portfolio management. Due to the provisioning of a comprehensive framework for governance arrangements for IT project portfolio management, the presentation of insightful examples from practice and the development of an adaptable analytical approach, this dissertation is of high value for researchers and practitioners alike. Therefore, I hope that this book will find a wide circulation in research and practice.
Darmstadt, January 2014 Prof. Dr. Peter Buxmann
Table of Contents
1 Introduction 1
1.1 Motivation 1
1.2 Research objectives 3
1.3 Outline 5
2 IT governance – Structures, processes, and relationships in IT decisionmaking 7
2.1 Historical developments in corporate IS functions 7
2.2 Centralization and decentralization 10
2.2.1 Definitions 11
2.2.2 Comparison of centralized, decentralized, and federal arrangements 17
2.3 From centralization and decentralization to IT governance 21
2.4 IT governance 24
2.4.1 Emergence of the IT governance concept 25
2.4.2 Definitions 25
2.4.3 The IT governance framework of Weill and Ross 27
2.4.4 Structural, procedural, and relational mechanisms 29
2.4.5 The difference between governance and management 30
2.5 Business/IT alignment 31
3 IT project portfolio management – Evaluating, selecting, and staffing IT projects 35
3.1 Research approach 35
3.1.1 Definition of the review scope 37
3.1.2 Conceptualization 39
3.1.3 Literature search 42
3.2 Literature analysis and synthesis 45
3.2.1 Definitions 46
3.2.2 Modern portfolio theory 54
3.2.3 Key characteristics of IT projects 56
3.2.4 Theoretical backgrounds of IT project portfolio management research 64
3.2.5 Classification of the identified contributions 66
3.2.6 Mathematical approaches and decision support systems 68
3.2.7 Empirical findings and concepts 75
3.2.8 Convergence of empirical and mathematical contributions 81
3.3 Research agenda and implications for future research 83
4 Governance arrangements for IT project portfolio management – A case study in ten companies 87
4.1 Purpose of the investigation 88
4.2 Research questions 88
4.3 Conceptual framework 89
4.4 Research approach 90
4.4.1 Research design 90
4.4.2 Data collection 92
4.4.3 Data analysis 95
4.5 Findings 96
4.5.1 Fields of activities 96
4.5.2 Governance arrangements for IT budget allocation 97
4.5.3 Governance arrangements for IT demand management 101
4.5.4 Governance arrangements for IT project portfolio selection 108
4.5.5 Governance arrangements for IT resource management 115
4.5.6 Contingency factors 122
4.5.7 Consequences 141
4.6 Integration of findings 151
4.7 Summary and limitations 152
5 The impact of different governance arrangements on IT project portfolio selection outcomes – A quantitative modeling approach and simulation studies 157
5.1 Motivation 157
5.2 Problem background 158
5.2.1 Centralized, decentralized, and federal arrangements for IT project portfolio selection 158
5.2.2 IT project interdependencies and synergies 160
5.3 Related contributions 163
5.4 Conception and model development 166
5.4.1 Conception 166
5.4.2 Assumptions 169
5.4.3 Model parameters 170
5.4.4 Model formulation 173
5.4.5 Illustration 175
5.5 A computational study based on a 2k-factorial simulation design 183
5.5.1 Simulation design 183
5.5.2 Results 188
5.5.3 Discussion and limitations 195
5.6 A visual comparison approach based on a risk/return perspective 198
5.6.1 Data input and visualization 199
5.6.2 Problem solving process 201
5.6.3 Presentation of results 204
5.6.4 Comparison of centralized and decentralized perspectives 206
5.6.5 Discussion and limitations 208
5.7 Final discussion 210
6 Practical implications 213
7 Summary and outlook 217
Appendices 223
Appendix A – Contributions identified based on the structured literature search 223
Appendix B – Concept matrix for mathematical approaches 224
Appendix C – Success factor descriptions 228
Appendix D – Propositions for future research 229
Appendix E – Interview guideline 232
Appendix F – Case descriptions 234
Appendix G – Impacts of contingency factors on governance arrangements 244
Appendix H – Theoretical integration of the identified contingency factors 246
Appendix I – Theoretical integration of the identified outcome categories 247
Appendix J – Simulation statistics 248
Bibliography 251
Abstract
In recent years, the importance of project-based work has strongly increased in many organizations. As a large fraction of corporate expenditures falls into the category of IT investments, organizations particularly have to manage a growing number of IT projects.
While the requirements for efficient project implementation have long been discussed in the project management literature, the need for effective project evaluation, project selection, and resource allocation has moved into the focus of research only recently. These aspects fall into the domain of IT project portfolio management (IT PPM). Different conceptions for IT project portfolio management are currently discussed in the scientific literature as well as in practical contributions. In this context, it is often argued that in particular the governance, i.e. the appropriate organizational embedding of IT project portfolio management, is of eminent importance. The current dissertation is devoted to this topic. To this end, existing concepts from the domain of IT governance research are employed in order to examine the antecedents and impacts of the use of different governance arrangements for IT project portfolio management. Methodically, the present work is based on a structured literature review, case study research and a mathematical modeling approach.
As part of the structured literature review, the existing state of research in the field of IT project portfolio management is identified and discussed. In this context, it is highlighted that the discipline is based on two different research strands. The first research strand is concerned with quantitative mathematical models that provide decision support for project portfolio selection and resource allocation. The second strand includes empirical and conceptual work relating to the organizational integration and design of (IT) project portfolio management arrangements. Based on the literature review, recent developments in the two research strands are discussed and remaining research gaps are presented.
In a case study investigation with ten participating companies, deeper insights into the design of governance arrangements for IT project portfolio management in practice are provided. In this context, different fields of activities and different contingency factors influencing the governance of IT project portfolio management are investigated. It is also demonstrated how different governance mechanisms in different fields of activity complement each other. Moreover, the consequences arising from the use of different governance mechanisms are discussed.
Based on the findings of the case study analysis, a mathematical modeling approach is derived in order to study the impact of different governance arrangements on outcomes of IT project portfolio management. The approach is based on a generic coordination mechanism identified during the case study research. The approach is illustrated by simulation experiments. In particular, the influence of different types of project interdependencies on the outcomes of IT project portfolio selection in different decision-making constellations is examined. For example, it is illustrated to which degree decentralized decision-making can lead to a lack of synergy exploitation, redundancies, and budget overruns. The results of this thesis contribute to a better understanding of the importance of adequate governance arrangements for IT project portfolio management. In particular, the factors that influence the design of these arrangements are explicitly described. At the same time, the generic mathematical approach presented in this dissertation enables an examination of the preferability of different decision-making arrangements in different contexts. Based on this approach, hypotheses for future research can be derived.
1 Introduction
1.1 Motivation
In recent years, information technology (IT) has become indispensable in most large companies, and for some industries, like the financial industry, IT is the main production factor today. At the same time, many companies are heavily investing into new IT systems and capabilities. In order to cope with technological and strategic changes but also to drive innovation, constantly new IT project proposals accrue within these organizations. In this context, there is a trend to manage more and more organizational activities as projects. While constant renewal and innovation are of high importance for the prospering of an organization, some project proposals are not of benefit to the company and, therefore, should be rejected. Moreover, IT projects typically underlie significant risks, wherefore even some of the project proposals promising high returns should not be approved. Most importantly, IT resources such as project managers or programmers are typically very scarce. Consequently, there are often more potential IT projects than can be staffed with the available resources. As IT projects compete for the same scarce resources and are often subject to additional interdependencies, they should be assessed and managed in comparison and not in isolation.
This situation briefly describes the background for IT project portfolio management, the main subject of the current dissertation. Simply speaking, in contrast to project management which is concerned with “doing projects right”, project portfolio management is in particular concerned with “doing the right projects”. In order to effectively evaluate, select, and manage IT projects as a portfolio, appropriate governance arrangements are required.Establishing IT governance arrangements involves specifying decision rights and accountabilities by implementing structural, procedural, and relational mechanisms. There is no IT governance arrangement that fits to all companies. Consequently, when implementing new governance arrangements in a given organization, several contingency factors need to be taken into account and different groups of stakeholders need to be considered. This in particular applies to the IT project portfolio management context, where many different kinds of stakeholders are involved. Importantly, IT projects commonly are not only a matter of the IT department. As information systems are the backbone of many large companies, they affect stakeholders in various business units and departments. Consequently, requirements for changes to these systems or demands for new information systems are typically triggered by business stakeholders from different parts of the organization. In this context, previous literature on IT project portfolio management has emphasized the need for obtaining a centralized view on the corporate-wide IT project portfolio. However, this does not necessarily reflect the way IT project portfolio management arrangements are implemented in practice. In contrast, in many companies rather decentralized governance arrangements for IT project portfolio management are installed, where IT projects are not approved and managed at a corporate-wide level but at a divisional or departmental level. Moreover, federal arrangements can often be encountered, where different kinds of IT projects are handled at different hierarchy levels, depending on the project characteristics.
While decentralized arrangements reduce complexity and leave a certain degree of autonomy to local units, these arrangements are susceptible to inefficiencies caused by redundancies and the negligence of synergy potentials. Centralized arrangements, in contrast, may foster the exploitation of synergy potentials and the implementation of strategic initiatives, but are often incompatible with the organizational culture and other contingency factors. Federal arrangements can provide a compromise but often lead to an even more fragmented view on the portfolio and are particularly prone to ineffectiveness. Consequently, some fundamental tradeoffs need to be regarded when choosing a particular governance arrangement. This dissertation sets out to explore this tradeoff in detail in the context of IT project portfolio management. For this purpose, contingency factors fostering or inhibiting different governance arrangements are explored and consequences of the use of different arrangements are investigated.
From a theoretical point of view, this dissertation contributes by transferring concepts from the domain of IT governance research to the IT project portfolio management context and by integrating existing theoretical foundations in the IT project portfolio management discipline. Major theoretical contributions resulting from this research are a contingency model and a formal model of decision-making in organizations. This research is also of practical relevance as IT project portfolio management is particularly susceptible to political interventions and many companies struggle to install appropriate governance arrangements. In this context, IT governance experts and middle managers in charge of IT project portfolio management are often torn between conflicting requirements. In order to support these experts and to prevent failures of costly change initiatives, a thorough assessment of antecedents and consequences of the use of alternative governance arrangements is of vital importance. For this purpose, an approach for assessing the impact of different governance arrangements for IT project portfolio management, dependent on the given organizational environment, is outlined in this dissertation.
1.2 Research objectives
The main objective of this thesis is to analyze the antecedents for and the impact of the use of different governance arrangements for IT project portfolio management. The following general research questions have motivated and guided the work presented in the subsequent chapters:
1. What is the current state of research in the domain of IT project portfolio management?
2. Which governance arrangements are employed for IT project portfolio management in practice?
3. Which contingency factors influence the design of governance arrangements for IT project portfolio management?
4. Which advantages and disadvantages apply to different governance arrangements for IT project portfolio management?
5. How can different governance arrangements be modeled and compared in the particular context of IT project portfolio selection?
The answering of the first research question provides the grounds for the following investigations. Project portfolio management in general and IT project portfolio management in particular is a relatively new field of research.19 Still, significant progress has been made in recent years. Consequently, it is an important endeavor to identify and integrate the current literature, to demarcate the field of research, and to identify opportunities for future studies.
The second research question has been posed in order to gain deeper insights into the implementation of IT project portfolio management arrangements in practice. Although a significant body of literature on IT project portfolio management already exists, much of the previous work has been of conceptual nature. More recently, a growing number of empirical contributions have been published, but comprehensive insights into governance arrangements for IT project portfolio management are still considerably scarce. Empirical research in this area has typically focused on specific aspects such as roles and responsibilities of middle managers or management control mechanisms. Moreover, most existing empirical studies are concerned with project portfolios in general and not IT project portfolios in particular, wherefore the particularities of the specific governance context are not taken into account. Consequently, the second research question is intended to motivate a comprehensive study of governance arrangements in the specific context of IT project portfolio management. The third research question takes account of the large impact of contingency factors in the IT governance context. Typically, different variants of IT governance arrangements can be encountered in practice, dependent on a number of organizational and environmental factors. In contrast to the second research question, which primarily motivates a description of the governance arrangements employed in practice, the third research question aims at an investigation of the impact of the environments in which these governance arrangements are embedded. This investigation is of high relevance as it helps to explain why different arrangements are used in practice and which arrangements are appropriate in a given context.
The fourth research question motivates an investigation of the consequences of the use of different governance arrangements in the IT project portfolio management context. While the appropriateness of a particular governance arrangement in a given context depends on different contingency factors, it is still of interest to investigate the general advantages and disadvantages of different arrangements in order to anticipate outcomes on the IT project portfolio level.
Finally, the fifth research question is inspired by previous research on decision-making in organizations.By formally modeling and comparing different governance arrangements for IT project portfolio management, relationships and effects identified in previous research can be retraced and explained. Moreover, propositions for future empirical research can be derived. If adapted to the specific organizational context, such an approach also provides a means to support IT governance experts in evaluating the appropriateness of different potential governance designs.
