Project Management Institute

The role of the project management office in achieving project success

This paper reports findings from a seminal large-scale research study of project management offices (PMOs). The study investigated the impact of the presence of PMO functions and services on project performance.


In the last several decades there has been an increase in use and acceptability of project management (PM) and its related methodologies. This increase arises from recognized advantages of applying the project approach. An emerging theme is that effective PM increase project success rate, and hence, improves organizational performance. The use of a PMO has been seen as one way for improving this effectiveness, particularly by enabling the effective transfer of knowledge from earlier success/failures and by providing a range of support services both for projects and various management levels. However, the existing literature on this important subject is largely anecdotal.

Pinto (1986) reported an empirical study of critical success factors (CSFs) as predictors of project success. It included 10 CSFs including project mission, top management support, project schedule/plan, client consultation, technical tasks, communication, and trouble-shooting. Besides Pinto's study, there are several other CSF lists reported in the literature, such as Kerzner's report for PM excellence (1987). Most of these lists include factors related to the project manager and the organization to which the project belongs; none of them directly address the question of the impact of PMO functions and services on project success. A natural next step beyond any of these studies was to determine how organizations could systematically foster key success factors on an ongoing basis.

Objectives of Study

For the study, the PMO was defined as an organizational entity established to assisted project managers and teams throughout the organization in implementing PM knowledge, principles, tools and techniques. Six major PMO functions and services emerged from the literature review: developing and maintaining PM standards and methods, developing and maintaining project historical archives, providing project administrative support, providing human resources/staffing assistance, providing PM consulting and mentoring, and providing or arranging PM training. PMO presence index was formed from responses to survey questions in each of these six categories.

The specific goals of this study were (1) to investigate the impact of a PMO on reported project success; (2) to construct a set of indices of functions and services representing the presence of PMO practices in organizations; and (3) to extend the previous research conducted by Pinto (1986) on the impact of critical success factors (CSFs) on reported project success.

General information was gathered in addressing issues on (1) the growth of PMOs, (2) the circumstances associated with the establishment and use of PMOs, and (3) main environmental factors surrounding established PMOs.

Research Methodology

The study used the PMI® year 2000 membership list as the population for sample choice. One thousand members in North America were randomly selected from the pool of 35,880 members who had not chosen to exclude their names from mailing lists. Response rate was 23.4%. In addition, this research utilized a targeted group of organizations that were identified in advance as having some version of a PMO. There were 96 PMO representatives who participated, along with 52 project managers who were invited by a subset of the representatives. Three survey instruments were used under a web address hosted at the George Washington University. Identified PM and PMO field experts were consulted in creating the survey instruments.

Key Research Findings

The study found a distinct trend showing an increasing number of PMOs being established in the mid-1990s. The establishment of an overwhelming proportion of PMOs was approved at the top/upper management level, and most of the PMOs also reported to senior management. A very high percentage of full-time staffing was reported in both the random and targeted surveys. About 60% of PMOs operated at less than 2% of their organizations’ budget.

The study also obtained a rich collection on motivations for PMO establishment, PMO mission statements, and PMO policy documents. The most frequently reported motivations were: (1) improving all elements of PM—including performance outcomes, lessons learned, support for project managers and reducing the number of “troubled projects,” (2) achieving a common PM approach—including methodologies, standards, and accountability, (3) improving quality and customer satisfaction, and (4) incorporating PM with strategic goals and developing competitive advantage.

About 26% of the random respondents and 31% of the targeted respondents reported their organizations as being part of the software development, data processing, or computer/IT industries, while 53% and 62% of the reported projects, random and targeted respectively, were associated with these areas. This finding suggests that (1) in the world of projects and project management, the growing importance of “soft” projects in comparison with physical products is evident, and (2) as more of the world's output becomes increasingly complex and sophisticated, software, hardware, and process development become ever more intermingled.

Across many industry and governmental areas, the respondents from both the random and the targeted samples presented a very similar work pattern of on average, 20 years of work experience with about 12 years in PM.

A basic set of PMO functions and services identified from the literature review were, for the most part, ratified by the research findings. Regression analysis modeling revealed that from both random and targeted samples, PM Standards and Methods explained the most variation in project success (compared with other PMO functions and services). This result was fully compatible with and supportive of the PMO motivation and mission statement data. Moreover, the Fortune 500 Benchmarking Forum on PM (1997) reported a similar finding on the importance of a standard methodology.

As for the other five PMO functions and services (Project Historical Archival, Project Administrative Support, Human Resource/Staffing Assistance, PM Training, PM Consulting and Mentoring), they also explain some variations in reported project success, although there are a few anomalies that seem to require further investigation. For example, responses from the randomly selected PMI members showed very low positive correlation between the use of project archives and project success; on the other hand, data from the targeted respondents showed a high positive correlation for this factor.

In general, PMO functions and services are less appreciated by the random sample respondents than by the targeted sample respondents. Also, from the random sample, results do not show that reported project success is higher in organizations that have a PMO than those that do not have a PMO.

The regression analysis models indicate that the overall PMO presence index does have a positive linear influence on reported project success, in varying degrees from both the random and targeted survey groups.

Research results show that a second research hypothesis, that the PMO presence index would have a 2nd order influence on reported project success (a nonlinear regression, intended to see whether the regression changes its slope over the range of observations) was not supported. Reported project success does not diminish when the PMO presence index grows too strong. In other words, there is no indication that a PMO appeared to introduce bureaucratic impediments. The PMO appears to be supportive and facilitative as opposed to being seen as another level of directive management.

Research results suggest that a third research hypothesis, the PMO presence index has influence on reported project success beyond the influence of previously identified CSFs alone, was not supported. The PMO presence index did not qualify as another last-in variable for CSF multiple regression models both from random and targeted samples. Also, the CSFs identified by Pinto nearly 15 years ago have held up reasonably well, but the topic could benefit from a large-scale update.

For reviewing the entire paper (about 300 pages), you may contact UMI Dissertation Service at 1-800-521-0600 or 734-761-4700 in Ann Arbor, Michigan.


Pinto, Jeffrey. 1986. “Project Implementation: A Determination of Its Critical Success Factors, Moderators, and Their Relative Importance Across the Project Life Cycle,” dissertation, University of Pittsburgh.

Kerzner, Harold. 1987. In Search of Excellence in Project Management. Journal of Systems Management, 38 (2).

Toney, Frank, & Ray Powers. 1997. Best Practices of Project Management Group in Large Functional Organizations: Results of the Fortune 500 Project Management Benchmarking Study. Project Management Institute.

This material has been reproduced with the permission of the copyright owner. Unauthorized reproduction of this material is strictly prohibited. For permission to reproduce this material, please contact PMI or any listed author.

Proceedings of the Project Management Institute Annual Seminars & Symposium
October 3–10, 2002 • San Antonio, Texas, USA



Related Content