Project Management Discussion 2

BenchmarkingofProjectManagementOfficeEstablishment.pdf

Home >Business & Finance homework help >Management homework help >Project Management Discussion 2

Benchmarking of Project Management Office Establishment: Extracting Best Practices

Bjørn Andersen1; Bjørnar Henriksen2; and Wenche Aarseth3

Abstract: This paper deals with best practices in establishing, developing, and implementing project management offices �PMOs�. First, a brief overview of the theoretical background for PMOs is presented. The research approach is described, along with an overview of the benchmarking partners used. In the main part of this paper, various aspects of a PMO’s life cycle are discussed based on observations from the benchmarking partners. Through the benchmarking study, we have discovered that although the PMO design differs greatly, certain key characteristics, responsibilities, and tasks are very similar. Successful PMOs take on responsibility for different project-related functions and core tasks related to development of shared methodology and processes for handling of projects, training and competence development within project management, proposing of new projects, and quality assurance of projects. The success of the PMO is related to ensuring the necessary authority of the PMO, real organizational authority as well as academic and social credibility, top management support, and that the PMO covers true needs in the organization.

DOI: 10.1061/�ASCE�0742-597X�2007�23:2�97�

CE Database subject headings: Project management; Bench marks; Best management practice; Change management; Life cycles.

Introduction

Many organizations, especially above a certain size and with an extensive degree of project work, have taken the step to establish centralized project management offices �PMOs� to take on responsibility for project-related functions and coordinate project- related activities. There are large variations in terms of organiza- tional location of such PMOs and the responsibilities/tasks they hold.

This gap in perceptions of PMOs and their impact led several companies we continuously work with to ask the question “what seems to be best practice in this area?” These companies were all in the process of establishing a PMO or redefining/formalizing existing project support functions in a PMO, and thus saw the need for some kind of roadmap for designing and implementing a project management office. As a result, a comparative bench- marking study was undertaken, using a sample of companies who had accumulated experiences in this field as data sources. The purpose was to identify any common factors, positive and nega- tive, that seemed to dictate the success rate of a PMO.

Theoretical Background

Historical Background and Development

Project offices have for quite some time been used as a means for administrating large projects, based on the need for an overall, coherent approach. Project offices were established to coordinate portfolios of projects, and these offices facilitated experience transfer and benchmarking among the projects. In addition, such project offices often functioned as a “project monitor” that re- ported directly to senior management about problems or devia- tions from plans. They might even intervene and adjust projects going astray �Winters 2000�. Even before such offices came to be called “project management offices,” they were in some pioneer- ing organizations termed program management offices, for ex- ample in NASA. In the early days of such support offices, the program management offices resembled closely what we today term project management offices, but later on, they have been given somewhat different meanings, as is explained later.

In terms of impact, studies by the Gartner Group indicate that organizations with a well-functioning project management office will experience half the cost and time overruns as those without one �Gartner Group 2000�. The Standish Group found that espe- cially the information technology �IT� industry has been less than proficient in managing its projects �Crawford 2001�. Considering how widespread IT projects have become, this is an ominous trend. One particular study showed that 46% of all IT projects ran over budget and schedule, with 28% of them construed as fail- ures. Other organizations, e.g., Robbins–Gioia, Inc., have found similar statistics; 90% of all IT projects underestimated the size and complexity. Almost half of them, 44%, had cost overruns in the magnitude of 10–40%, whereas only 16% consistently stayed true to the project plan.

In response to this, many organizations turn to solutions like project management offices. One question still remains; can it be proven that a PMO is the right solution to these problems? The trend seems to point in the direction of a higher project success

1Professor, Dept. of Production and Quality Engineering, Norwegian Univ. of Science and Technology, NO-7491 Trondheim, Norway.

2Researcher, Productivity and Project Management, SINTEF Technology and Society, S.P. Andersensv. 5, NO-7465 Trondheim, Norway.

3Researcher, Productivity and Project Management, SINTEF Technology and Society, S.P. Andersensv. 5, NO-7465 Trondheim, Norway.

Note. Discussion open until September 1, 2007. Separate discussions must be submitted for individual papers. To extend the closing date by one month, a written request must be filed with the ASCE Managing Editor. The manuscript for this paper was submitted for review and pos- sible publication on February 7, 2006; approved on June 8, 2006. This paper is part of the Journal of Management in Engineering, Vol. 23, No. 2, April 1, 2007. ©ASCE, ISSN 0742-597X/2007/2-97–104/$25.00.

JOURNAL OF MANAGEMENT IN ENGINEERING © ASCE / APRIL 2007 / 97

rate. When the Standish Group investigated software projects, only 16% were successful in 1994 with regard to time, budget, and technical specifications. This rate had increased to 26% in 1998 �Crawford 2001�. In the book The Strategic Project Office, three main reasons given for this development are: 1. The project size tends to decrease, thus making the projects

less complex; 2. Better project management; and 3. More extensive use of standard project procedures, for ex-

ample as a consequence of the implementation of a PMO. A survey from the Gartner Group �2000� showed that:

• 40% of project-oriented larger enterprises have established a project management office;

• Organizations with effective PMOs will reduce overruns by 50% �cost, time, and resources�; and

• Project management offices represent best practice regarding project execution success.

PMOs at Different Levels

One of the aspects we studied in our research was practices and requirements with regard to PMOs at different organizational lev- els. From our definition of a PMO and from practices observed, it seems clear a PMO can be, but need not to be, a physical office. It can also be a virtual unit consisting of people with a special interest and expertise in project management, promoting good practices on behalf of the entire organization. This makes it easier to separate the PMO from a less productive discussion on its location in the organizational hierarchy and chart. However, there have been attempts at defining types of PMOs depending on the organizational level they are linked to �Crawford 2001�: • Level 1—Project control office or project office. This type of

PMO handles the management of large and complex indi- vidual projects, with a focus on control and monitoring of schedule, budget, and other more administrative aspects.

• Level 2—Unit project office, for example an IT project office. This type of PMO can also be used in managing individual projects, but the overall objective is to integrate all projects in a unit into one or more portfolios of projects.

• Level 3—Strategic project management office, for example a corporate project office. Level 3 office is located at the corpo- rate level, enabling the senior management to take part in the prioritization of projects, to support the goals of the organization.

Associated to Level 3 of project management office, we find the concept of program management office. The distinction might be dismissed as an unnecessary semantic exercise but we believe it merits attention as they often have different aims and roles in an organization. Russ Martinelli and Jim Waddell �Intel and Tek- tronix, respectively� described important distinctions �Kendrick 2005�: “Program and project management are related but distinct disciplines. It is important for everyone within the organization to fully understand the distinctions between the two, as well as the differing roles and responsibilities of program and project man- agers. In general, the greatest difference between program and project management is that program management focuses on achieving business results to create a competitive advantage while project management focuses on planning and executing the work required to deliver the end product.”

Programs often consist of several projects and a project man- agement office might be a part of the program management office.

Kendall and Rollins �2003� described important issues on the enterprise/program mission level by focusing on portfolio man- agement and the process to do that effectively.

In line with Crawford’s understanding of the different levels of PMOs, Rad and Levin �2002� suggested the following as suitable levels: • PMO for individual projects or a program of related projects; • PMO at the divisional level; and • PMO at the corporate level. Although there seems to be consensus about this issue, we will revert to it later when discussing our findings.

Establishing a PMO

There is of course no exact approach prescribed for establishing a PMO. The approach will depend on the size and structure of the organization, the purpose of the office, and so on. Some recom- mendations have been made, for example, Perry and Leatham �2001�, suggesting a three stage process: 1. Training project managers; 2. Launching the PMO; and 3. Deployment through active project consultation. There is little empirical data available on implementation pro- cesses for PMOs, but Rad and Levin �2002� provided some rules of thumb for implementation times: • Project level PMO, 3 months–1 year; • Division level PMO, 1–3 years; and • Corporate level PMO, 3–7 years. Beyond these sources, we have uncovered very little literature that bears on the issues we set out to study.

Research Approach

The research project had the following main objective: “Estab- lishing insight into good/best practices in establishing and oper- ating project management offices in larger organizations.”

By large, we did not specify a minimum size limit, but the research included organizations of at least 1,000 employees, very often also organized in different units and/or geographical loca- tions. Given this data material, we think the results are not valid for smaller organizations, i.e., with less than 500 employees. The chosen research approach combined literature review with an in- terview survey and a process benchmarking study: • Literature review: To the extent relevant literature or other

types of documentation was available, this was examined for any answers to this research question.

• Interview survey: A survey was conducted in the organizations that represented the “customers” of this project. The survey focused on understanding the needs of project managers and participants in terms of PMO functionality.

• Benchmarking: A process benchmarking study was undertaken to identify any common denominators in PMO practices. The benchmarking study was based on a stepwise model called the benchmarking wheel �Andersen 1995�. More specifically, the purpose was to analyze the PMO imple-

mentation processes in the benchmarked companies to understand whether generic recommendations could be identified. The approach is typically qualitative in nature, with a sample of benchmarking partners too small to allow any type of statistical analysis. Thus, the reliability of the results is limited by the benchmarking partners included in the sample. However, repre- senting a cross-section of different sectors and types of organiza-

98 / JOURNAL OF MANAGEMENT IN ENGINEERING © ASCE / APRIL 2007

tions, the results should be widely applicable. Many of the bench- marking partners, and indeed also the companies participating as discussion partners in the research project, represent large engi- neering companies. Engineering companies aiming to establish or change existing PMOs should use the results from our work as guidelines when planning their change processes.

Empirical Data

The empirical data collected in the study originated partly from the interview survey of the participating organizations, partly from the studies of the benchmarking partners.

Internal Interview Survey

The interview survey took place in the organizations participating first hand in the research, conducted by the research team together with a representative from the respective organization.

The most important findings from this study are presented in the following: • Armed forces: Have already implemented a project manage-

ment office with certain responsibilities. The main support expected from the PMO is strategic portfolio management, coordination among projects, being a resource base for the projects, having “mentors” to offer aid, monitoring the projects’ performance, harmonizing approaches and tools be- tween the projects, and competence development.

• Oil company: Does not have one centralized PMO but rather project organizations within each business unit. These perform project analyses for the business units, and handle harmoniza- tion of approached and tools. In the future, PMO customers expect clearer links to the project customer, coordination of project resources, shared methodologies, resource allocation support, training, “super project managers” for hire, and some kind of best practice sharing.

• Telecom operator: Several of the subcompanies have estab- lished PMOs, servicing projects as “stimulants” of good project practices throughout the project duration. They also handle training of project management personnel, while future expectations from the projects are supplying them with people, creating value for the projects, comparing projects and trans- ferring experiences, shared methodologies, and ultimately making the project managers perform better.

• Defense contractor: A PMO was established just before the survey was undertaken, with the PMO having only a consult- ing authority toward the projects. Expected, prioritized func- tions of the PMO are maintenance of, training in, and support for the project management process, coordination of project management approaches, support to projects in the start-up phase, auditing of projects, and establishing arenas for experi- ence exchange.

External Benchmarking

The search for benchmarking partners was directed at organiza- tions known for successful PMO operation. In the end, the fol- lowing were used in the study: • Lego, the Danish toy company, to which a site visit was con-

ducted with interviews with both the person responsible for developing the PMO and a project manager.

• Danish Oil & Natural Gas �DONG�, interview conducted with the person responsible for developing the PMO.

• Dyno Nobel, the Norwegian branch, interview conducted with the person responsible for developing the PMO.

• Fiat, experiences in PMO development exchanged over e-mail with the person responsible for developing the PMO.

• BHP Billiton, telephone interview conducted with the PMO manager.

• SunTrust Bank, telephone interview conducted with the person responsible for developing the PMO for IT projects �who had previously developed similar PMOs in Coca-Cola and the in- surance company Aflac�.

The main findings from the benchmarking partners are presented in Table 1.

Analysis

The following questions formed the basis for analysis and discussion: • Is it possible to identify a pattern of organizational-dependent

factors, its challenges, and its projects that dictate the type of project management office that is suitable and how it should be implemented?

• What typical responsibilities does a PMO have, and which are normally not included in the functionality of a PMO?

• What objectives are pursued in establishing a PMO, and what stakeholders and what needs does it attempt to satisfy?

• Will a PMO display a dynamic development over time, i.e., have some kind of a life cycle?

• How are PMOs typically organized, what size do they have, and how should they be staffed?

• How can one ensure that a PMO has the necessary authority in the organization to have a real impact on the projects, and what role does top management support play in this?

• To what extent is the success of a PMO influenced by the existence of an individual with a true passion for the PMO, and how to ensure the transition from one person’s pet project to regular operations?

• What competence requirements must be fulfilled by a PMO and its employees?

The analysis section of the paper is thus structured accordingly.

Classifying PMOs Based on Organization Type and Other Aspects

An, albeit unscientific, observation many in the research team made during the interviews was that the project management of- fices seemed to be “colored” by the person who had been the obvious driving force behind its establishment. If this person was inclined toward IT solutions, the PMO focused on selecting and implementing project management systems, while an HR person would direct the attention of the office toward competence devel- opment. This observation certainly belongs to the heading of sponsor influence, but it also led us to ponder the question about how different a PMO is from organization to organization.

After some discussion, we still concluded that they are basi- cally quite similar. Their status and use can depend somewhat on the sponsor, also their size, organizational level, and how they are staffed. However, the key characteristic, the responsibilities and tasks they have, are often surprisingly similar, if not identical. This led to the formulation of our definition of a project manage- ment office: “A systematic coordination and unified handling of key project-related tasks, as an enterprise-wide responsibility.”

From this definition, the actual physical manifestation of the “office” is of less importance as long as it assumes a systematic

JOURNAL OF MANAGEMENT IN ENGINEERING © ASCE / APRIL 2007 / 99

Table 1. Findings from Benchmarking Partners

Benchmarking partner Rationale for PMO implementation Tasks handled by the PMO PMO success factor and evaluation

Lego Lego is in a market where about 70% of the revenues each year are generated by new products of that year, giving rise to an extreme need for speedy product development. The average product development time has been reduced from 36 months to about 24, with the target being 12 months. This was also really the triggering factor in the establishment of Lego’s PMO, an office that even “owns” a group of close to 20 project managers.

• Defining and improving business processes across the organization • Providing project managers to projects • Supporting planning and support systems • Training project managers

Co-location of the project teams is a very important success factor, and the project manager interviewed considered the PMO a large success. It was seen as a unit representing the project managers toward the rest of the organization, as opposed to being a tool to “teach the project managers to behave” as is the case in many other organizations.

DONG The PMO is located inside the Exploration and Production �E&P� division, and was established as a response to a need for more standardized project execution.

• Offers the projects support and reports from the project to the senior management • Methodology standardization • Portfolio reporting • Coordinates the resource portfolio • Trains project managers • Conducts peer reviews and project evaluations • Aids project start-up processes

A success factor has been the inclusion of senior project management competence in the PMO to give it credibility, while pitfalls seen are that the PMO becomes too static and bureaucratic or introduces too advanced systems.

Dyno Nobel Europe

The PMO is a staff function with only one employee, established following a reengineering process in the company.

• Project management methodology • No decision authority • Training of project managers

Being able to redo the process, the financial depart- ment would have been more closely involved, simply because financial aspects are an integral part of projects.

BHP Billiton The PMO consists of 16 persons and was established following a series of very poorly managed projects. The PMO is not a centralized project organization, but rather a support unit distributed across the different locations.

• The project governance process • Reviewing projects • Training in project management • Benchmarking • Project management software • Analyzing the company’s project

Today, the large projects are well under control, and the focus ahead is on medium-sized projects.

SunTrustBank Established a PMO, to move away from a situation where the divisions worked in “silos” and prioritized projects motivated by the priorities of the divisions, not the company as a whole. The PMO mainly handles the bank’s IT projects, and is a separate unit reporting to a member of the top management team �giving it weight and authority�.

• Developing methodology and tools for project management • Implementing quality assurance • Portfolio management • For each project, a scorecard is developed and monitored continuously

The PMO has produced benefits like more effective projects, that the right projects are prioritized, and less project failures.

1 0 0

/ JO

U R

N A

L O

F M

A N

A G

E M

E N

T IN

E N

G IN

E E

R IN

G ©

A S

C E

/ A

P R

IL 2 0 0 7

responsibility for a set of important tasks related to the execution of projects. Considering this set of tasks, there are only minor differences among the observed PMOs. The next logical question then was: Are there aspects of the organizations, e.g., challenges they are facing or their projects that dictated the design of the However, no such pattern could be found in our data. This topic has not been covered in existing literature either, supporting the interpretation that this is not an important topic.

PMO Responsibilities and Tasks

From the benchmarking data, we found that the following repre- sent the core of tasks handled by the typical project management office: • Establish, continue to develop, and manage shared methodol-

ogy and processes for handling of projects in the organization. Very often, a project model lies at the bottom of this, often a phase model with decision gates. On top of this, there are usually process descriptions, routines, tools, etc.

• Training and competence development within project manage- ment, either through offering courses by the PMO itself or providing external training. In some cases, the office also man- ages a “competence ladder” or some other form of certification where project managers gradually and systematically develop more advanced knowledge.

• Offering support to projects, in the form of, for example, methodology and tools, participation in meetings, and present- ing recommendations and consulting. Normally, this is a free service for the projects. In some cases, it is mandatory for the projects to use these services, in others it is voluntary. Most PMOs are also clear about not wanting to act as operative resources in the projects, but rather to maintain a certain dis- tance and be “consultants.”

• Contribute to the governance processes of the projects, i.e., the proposing of new projects, ensuring the quality of the decision basis, and selecting projects. Common for all the PMOs ob- served is that their role is limited to offering recommendations, not having formal decision power.

• Quality assurance of projects, at different stages and in differ- ent ways, from evaluating decision basis documents in the selection phase to midterm/after-the-fact evaluations, to facilitating peer reviews. Again, the office normally has no authority to instruct or stop the projects, only to make recommendations.

• Offering support to the project owner, i.e., support to the man- agement of the organization and those in charge of the project portfolio, as opposed to the single project. This can include different types of quality assurance, consulting about project development, competence development, and so on. Various literature lists several additional tasks handled by

PMOs, and the benchmarking study also uncovered other respon- sibilities. However, these were specialized functions handled by only one or two PMOs and did not represent key functions.

To the extent that the previous list has evolved as a kind of consensus set of tasks for PMOs, it seems sensible for an organi- zation planning to implement one to include these responsibilities. They have emerged as a common denominator of a number of successful PMOs in reputable organizations, and thus represent much experience about what are suitable tasks for a PMO.

Objectives Pursued When Establishing a PMO

From existing literature and the observed organizations, there are a number of different triggering factors of PMO implementation processes. Some examples of objectives pursued are:

• Large differences in how projects are run, thus a PMO is es- tablished to ensure a more uniform project execution based on best practice.

• Problems with cost and time overruns in projects, thus aims to improve this by ensuring a central competence unit within project management in the organization.

• Lack of qualified project managers, as a consequence imple- menting a PMO to develop such competence.

• Lack of holistic practices with regard to project selection and synergies among them. A PMO can then be established to appoint responsibility for such an overall view on the project portfolio. On an overall level, the objective of a PMO will always be to

realize the given definition of a project management office, i.e., to ensure a systematic handling of key project management related tasks. One question we asked ourselves was whether the chal- lenges motivating the decision to implement a PMO decide how the PMO is designed. We have found this to be true to a certain extent, combined with the influence of the interests of the sponsor of the PMO. However, no matter what objectives are initially defined, we have observed that a gradually emerging goal is to develop a PMO that is really in demand, that answers to a need in the organization and which will be missed should it be closed.

Many of the interviewees pointed out the danger that a PMO could end up being one more bureaucratic element incurring more disadvantages than support for the projects and their managers. Developing a PMO that is truly needed counters this tendency, and this can be achieved by performing a thorough stakeholder analysis before designing the office.

Life Cycle of a PMO

One common trait in most of the PMOs observed is that they go through a dynamic development over time. This is in accordance with so-called maturity models that exist within the fields of qual- ity management, software development, and indeed project man- agement. The basic idea is that you must learn to crawl before you can learn to walk, i.e., starting with simpler tasks and expand- ing these to more advanced ones as experience is gained and competence developed, as shown in Fig. 1. This holds for true for many areas, also for a PMO.

For a project management office, we have seen such a dy- namic development in terms of: • Responsibilities of the PMO; • Size and staffing; and • Organizational location and level of authority. A classic life cycle goes from initiation via development to regu- lar operations and perhaps finally to closing, and is probably also relevant for a PMO. Partly inspired by some of the observed

Fig. 1. Phases of a PMO

JOURNAL OF MANAGEMENT IN ENGINEERING © ASCE / APRIL 2007 / 101

PMOs, a more specific model consists of three phases: 1. Development of common approached and tools for project

management; 2. Introduction of governance processes and quality assurance;

and 3. Implementing true project portfolio management. As the PMO progresses through these phases, the maturity in- creases both in the project office and in the organization as a whole, and the focus of the PMO changes. Typically, the “object” in focus evolves as follows: 1. Support to individual projects; 2. Support to project owners of individual projects; and 3. Support to the accumulated project portfolio and senior

management. These phases are also in line with other studies of PMOs and their evolution, for example Rad and Levin �2002�.

PMO Organization

At the start of this research project, we assumed that the question of organization of the PMOs would be a key topic, influencing the level of success of the offices. Throughout the work, we have indeed seen several approaches to PMO organization, but without being able to correlate this with the level of usefulness for the organization, and we have also revised our own perception of a PMO, from a physical office to a virtual function. Thus, the top- ic’s importance has been downgraded, but we have seen some interesting solutions, and it seems that the most preferred organi- zational alternatives are: • An integrated staff function, either on a high hierarchical level

and reporting to senior management, on a lower level of the organization, or both, i.e., having several offices with differen- tiated areas of responsibilities.

• A staff function, not integrated into one unit, but rather com- posed of resources located throughout the organization.

• A looser, distributed network of persons with a special interest and competence in project management who collectively handle the functions of a more traditional integrated PMO. Most existing literature in the field describes the first of these

as the typical PMO organization, while some hint at the distrib- uted staff function. We have not come across sources describing the network type of PMO organization. It is, however, more dif- ficult to uncover a pattern in terms of which types of organiza- tions and in which situations opt for one or the other approach. It seems rather as if the choice is more dependent on where in the organizations changes or project management practice standard- ization are desired.

Ensuring the Necessary Authority of the PMO

No matter what tasks are placed with the PMO, one of the main purposes is that it should contribute to changing the project man- agement practices of the organization. To achieve this, such an office and its resources are dependent on both real organizational authority as well as academic and social credibility. Our findings indicate that the extent to which this is achieved does not so much depend on the organizational solutions, but is rather defined through: • The respect the PMO enjoys in the organization through the

competence and seniority the office and its resources possess. To establish a PMO staffed with fresh graduates will likely not be very effective toward seasoned project managers.

• The type of “attitude” the PMO displays. If its focus is on

controlling the projects more than offering services and sup- port, the perceived benefits produced could be scarce and the PMO could lose its authority. Mapping the future users of a PMO and their needs and expectations is thus essential.

• The support enjoyed by a PMO from senior management, both in the establishment phase and later in its life span. This of course depends on senior management’s seeing a need for the PMO. Accordingly, the PMO must prove its worth by creating effects and documenting these, perhaps by undertaking bal- anced performance measurement of the project portfolio.

• The official mandate given to the PMO and the processes it practices.

Most of these are in line with existing literature, but few authors have discussed the need for the PMO to represent an answer to true needs in the organization. All of the benchmarking partners mentioned this as a key issue.

Importance of a Sponsor

An observed common trait of many of the PMOs in our data material is the existence of a sponsor either initiating their imple- mentation, putting pressure on senior management to make such a move, or has been hired from the outside to undertake the establishment. One obvious question has therefore been the sig- nificance of such a sponsor in ensuring a successful PMO devel- opment. This is of course very difficult to assess objectively, but our interviews seem to confirm that these sponsors have been extremely important in crafting the PMOs’ vision and managing their implementation and pushing them forward in the face of difficulties.

Competence Requirements for a PMO and Its Resources

To which extent a PMO manages to carve out a role for itself, both in the establishment and operational phase, does of course also depend on the competence it possesses. We have previously mentioned seniority as one requirement. Beyond this, we have uncovered the following important competencies required: • Obviously, solid insight into the craft of project management; • Holistic understanding of the organization and its projects; • Communication skills; • Business understanding; • A sense of innovation; and • Competence within other areas besides the core of project

management—a blend of personnel should cover all the com- petence areas of the PMI body of knowledge.

Analysis Summary—Recommendations and Success Criteria of a PMO

Table 2 summarizes is based on the questions this chapter aimed to shed light on, and presents the main answers from the analysis. At the conclusion of this analysis, we have attempted to summa- rize the most important factors that we have identified throughout this study as predicting the level of success of a PMO. These are based on the analysis presented above and take the form of factors to emphasize or avoid: • Ensure top management support; • Cover true needs in the organization to ensure that the users

see the benefits of the services of the PMO, based on conduct- ing a stakeholder analysis prior to designing the PMO;

• Employees in the PMO should be service-minded but also avoid becoming secretaries for the projects;

• Let the services of the PMO be free of charge for the projects; • Do not automatically turn to an organizational form of a cen-

tralized staff unit, but design the office based on the objectives and needs of it;

• Allow time for the progression of the PMO through the life cycle, starting slowly with some core tasks;

• Create some distance to and independence from the projects—be a support for, not a resource in;

• Man the PMO with senior project managers and other people with broad skills and project experience;

• Do not develop the PMO into a bureaucratic control unit; • Focus on improved project management practices; and • If possible, find a sponsor to support or run the implementation

process, but do not become too dependent on this one individual.

Conclusions

The sample of organizations, both for the internal interview survey and the benchmarking study, has been relatively small. Furthermore, all of these organizations are large entities, albeit representing a highly diverse set of business sectors. A valid ques- tion is thus to what extent the results generated are valid for other types of organizations than those represented in this sample?

This question of extending research findings beyond the sample used is always difficult to answer. In this case, we are confident the recommendations will be valid for organizations above a certain size. It is difficult to define a size limit, but per- haps in the area of 500–1,000 employees. For organizations smaller than this, we believe the challenges will be quite different, the need for centralized coordination less pronounced, and the logic of organization and communication different. There are of course also smaller organizations in charge of large projects, projects that could involve hundreds of people from contractors and other partners in the project. However, also in such cases, there is not the same type of need for a PMO to coordinate the project management practices across the organization. We thus choose to view our findings as relevant for larger organizations and assume they are of less relevance for smaller ones.

The recommendations presented come in the shape of “good advice.” How to proceed making use of these is another question. Devising a general stepwise approach is impossible, but we be- lieve the following are some generic issues to keep in mind: • Undertake a stakeholder analysis internally in the organization

to understand the needs and expectations toward a PMO and its responsibilities. Make sure to include in this analysis both the management of the organization, the project managers, and project participants.

• Based on the stakeholder analysis and a general knowledge of the challenges of the organization, especially with regard to its projects, define a clear objective for the PMO establishment

Table 2. Summary of Analysis Findings

Analysis question Tasks handled by the PMO

PMO design dependent on organization and project type • Size, staff type, and organizational location can vary considerably • The main function of the PMOs are still very similar, independent of organization type or organizational characteristics

PMO responsibilities and tasks • Common core tasks are project management methodology, training, project support, governance processes, and quality assurance

Objectives pursued when establishing a PMO • Ensure a systematic handling of key project management related tasks • Shared concern that the PMO must not become an extra bureaucratic element in the organization

The life cycle of a PMO • Most PMOs evolve through phases of gradually taking on more advanced responsibilities • The phases typically progress from support to individual projects to support to project owners to support to project portfolio owners and senior management

PMO organization • Typical ways to organize a PMO are as an integrated staff function, as a distributed staff function, or as looser network of qualified persons throughout the organization

Ensuring PMO authority • Cannot be simply “designed” into the PMO • Is a function of competence and seniority of the PMO staff, the PMO’s attitude toward acting as a controlling element as opposed to a support function, the senior management support to the PMO, and its official mandate

Importance of a sponsor • A common trait of the benchmarkingpartners is the existence of a powerful sponsor

Competence requirements • Seniority of the PMO staff seems very important • Other competences include solid project management skills, holistic organizational understanding, communication skills, business understanding, and sense of innovation

and make sure this is understood throughout the organization. This way, unrealistic expectations are avoided.

• If possible, try to identify a sponsor that can assume responsi- bility for driving the process ahead and be the person to represent “the face” of the office toward the rest of the orga- nization. Preferably, this person should come from within the organization, but can also be recruited or hired from externally.

• Compare the needs for PMO tasks uncovered in the stake- holder analysis with the list of typical responsibilities of such an office. Functions that are rarely handled by a PMO should be kept outside the office’s coverage, as there probably are good reasons why few PMOs encompass these.

• Plan a gradual development of the PMO, in line with the life cycle outlined in this paper, to ensure that the office does not take on too many or too complex tasks before it is ready for the challenge.

• Do not be too concerned with “boxology” of organizational and physical location when planning the PMO. Many respon- sibilities of the PMO could benefit from a different design than the traditional integrated, physically co-located group of people.

• Staff the PMO with the number of people and competence that seems required, and choose the people with care. Make sure the PMO represents senior project management experience and enjoys respect throughout the organization.

• Launch the PMO formally, making use of both promotional events and perhaps “missionary work” in the organization. Make sure the PMO becomes well known among its potential customers.

References

Andersen, B. �1995�. “The results of benchmarking and a benchmarking process model.” Ph.D. dissertation, Norwegian Institute of Technol- ogy, Trondheim, Norway.

Crawford, K. �2001�. The strategic project office, a guide to organiza- tional performance, Center for Business Practices, New York.

Gartner Group. �2000�. “The project office: Teams, processes, and tools.” �www.techrepublic.com� �Dec. 2005�.

Kendall, G. I., and Rollins, S. C. �2003�. Advanced portfolio management and the PMO: Multiplying ROI at warp speed, J. Ross Publishing, Fort Lauderdale, Fla.

Kendrick, J. �2005�. “Project and program management offices.” �http://www.p2c2group.com/apr05nws� �Dec. 2005�.

Perry, S. S., and Leatham, L. �2001�. “The case for a full-function project office.” �www.bizforum.org/whitepapers/kanbay001.htm� �Dec. 2005�.

Rad, P. F., and Levin, G. �2002�. The advanced project management office: A comprehensive look at function and implementation, St. Lucie, Boca Raton, Fla.

Winters, F. �2000�. “Ensuring success: PMOs bring it together.” �www.gantthead.com� �Dec. 2005�.