سه نقش دفتر مدیریت پرتفولیو پروژه: تاثیر آنها بر اجرای مدیریت پرتفولیو و موفقیت

Project portfolio management offices (PPMOs) are a subset of project management offices (PMOs) that handle collections of multiple single projects and programmes, i.e. portfolios. PPMOs are centralised organisational units that cater to the demands of various stakeholders by performing specialised tasks. They are initiated by their organisation's leadership in response to increasing management challenges originating from project portfolios. Although there has been considerable research on PMOs in general, not only a clear understanding of multi-project PMOs' activity patterns set in specific contexts like project portfolio management, but also both existence and mode of multi-project PMOs' contribution to successful performance are still lacking. By quantitatively analysing PPMOs in 278 portfolios, we identify three different activity patterns, which are interpreted as distinctive roles. We show a significant positive effect of PPMOs' coordinating and controlling roles on performance in terms of project portfolio management quality, which is a predictor of portfolio success.

Managing multiple sets of projects simultaneously is a challenge organisations have to master today in order to implement their strategic objectives (Artto and Dietrich, 2004 and Dietrich and Lehtonen, 2005). Although the project management literature still focuses primarily on single projects, research in the last 5 years has increasingly acknowledged that multi-project issues have become critical for all organisations regardless of delivering projects to external or internal customers. Multi-project PMOs have emerged within these multi-project management environments as a major device to develop competence in project management, manage single project performance and coordinate multiple projects and actors. Project portfolios that include multiple unrelated single projects and/or various programmes are focus of this study. Recently, studies have furthered general knowledge on PMOs, including their description (Hobbs and Aubry, 2007) and their relationship to context and transition (Aubry et al., 2010a and Aubry et al., 2010b). Other studies have addressed PMOs' part in organisational change (Pellegrinelli and Garagna, 2009) or as a value-realising function (Hurt and Thomas, 2009). Hobbs and Aubry (2007) show that PMOs' organisational characteristics and mandates vary significantly, highlighting the existence of a wide and diverse range of PMOs. These authors explain, “The organizational reality surrounding PMOs is complex and varied … organizations establish a great variety of PMOs to deal with their reality” (p. 85). To provide a clearer picture of the mandates of multi-project PMOs, these findings suggest that PMOs should be differentiated based on comparable realities such as project portfolios. In this paper, we consider project portfolio management offices (PPMOs), which are multi-project PMOs dedicated to project portfolio management (PPM). Operative PMOs that only handle a single project or one programme are disregarded. Pellegrinelli and Garagna (2009) confirm that “multi-project PMOs are organizations' responses to their needs and environments — unique structural arrangements designed to fulfil a specific purpose” (p. 651). Continuing this notion of purpose specificity, we follow the task-oriented paradigm of organisational design and organisational theory, which states that tasks and sub-tasks should follow the requirements of the organisation ( Burton et al., 2011 and Mackenzie, 1986). Therefore, we posit that the task environment is critical for identifying and defining the appropriate tasks to be undertaken by PPMOs and for detailing the activities that enforce the objectives of these tasks. A major feature of the task environment of a PPMO is its organisation's first-tier senior management, who are typically the owners of all the firm's project portfolios. Fundamentally, the tasks of PPMOs may be derived from these key stakeholders' requirements and their need to delegate management obligations. From a task delivery perspective that considers all participating personnel who contribute and collaborate in managing project portfolios (Gemünden et al., 2008), the project portfolio manager is a prominent participant ( Blomquist and Müller, 2006 and Jonas, 2010). In this vein Jonas (2010) proposes multi-project PMOs to take up being project portfolio manager “as a central coordination unit that supports the senior management with its specialized knowledge about project portfolio practices” (p. 823). Jonas (2010) contributes the attributes of role clarity and role significance to provide a formal role definition for the project portfolio manager. The present paper builds on this conceptual work, which fundamentally grasps what project portfolio managers are and how others perceive them, to characterise the day-to-day practise of PPMOs as project portfolio managers. PPMOs' operational roles are based on the management demands of first-tier senior managers, which are linked to the typical PPM phases to determine the activity patterns that comprise PPMOs' involvement in PPM. We consider the activity patterns and actual contributions of PPMOs when performing assigned tasks (role taking), rather than their formally stated roles (role making), to outline the configuration of PPMOs in practice. Formal role statements often raise expectations that PPMOs cannot fulfil with their limited power and resources. Furthermore, these roles are formulated purposefully as generic statements because a more precise formulation would not reach consensus due to the high potential for conflict among the expectations of the stakeholder groups of a PMO. By examining activity patterns, which are translations of the required tasks into everyday work, while each pattern makes up a role, we clarify the action of PPMOs when managing project portfolios. Thus, we refrain from listing every individual task performed and use a higher level of abstraction to understand roles as dimensions of social behaviour ( Morrison, 1994 and Webster and Wong, 2008). Research on roles in the governance of project management and PMOs is not new, but this research has been extremely sparse and varied. First, Turner and Keegan (2001) proposed two roles, the “broker” and the “steward”, when discussing governance mechanisms in project-based organisations, confirming two separate activity patterns in interface and resource management, respectively. Second, Blomquist and Müller (2006) identified the role of middle managers in programme and portfolio management. Third, Hobbs and Aubry (2007) provided a fundamental understanding of the roles of PMOs through a framework that grouped five sets of tasks, potentially forming five general PMO roles. Recently, interest in the roles of PMOs has intensified, with Hobbs and Aubry (2011) and Müller et al. (2011) proposing typologies of PMOs. In general, however, the understanding of PMOs' roles and the impact of these roles on value contribution and creation remains unclear. Thus, the acceptance, existence and legitimacy of PMOs remain at stake (Pellegrinelli and Garagna, 2009). Acknowledging that PMOs have a low life expectancy of four years, on average (Hobbs and Aubry, 2007), Pellegrinelli and Garagna (2009) illustrate the effects of changes in an organisation on PMOs: “The PMO can be the battle ground between empowerments and control, between people and processes, and between political factions” (p. 652). This observation suggests that the closure of PMOs may be a case of collateral damage rather than a natural death due to a lack of purpose or unnecessary activities. Thus, with the existence of PMOs in question, Pellegrinelli and Garagna (2009) point out the need for PMOs to battle for altered organisational needs and their stakeholders' changing preferences by acting as “the fulcrum between forces for centralisation — the tendency for decision and policy making, executive powers and resources allocation to reside in a dedicated (line of) business unit or corporate function — and decentralisation — the tendency for decision and policy making, executive powers and resources allocation to be devolved throughout the organisation to individuals or operating unit” (p. 652). Thus, in an attempt to outlive fads and fashion manoeuvres and to justify and sustain the existence of multi-project PMOs, Pellegrinelli and Garagna (2009) recommend that multi-project PMOs be re-shaped as an organisational construct, allowing PMOs to become agents, rather than reactionaries. In this paper, we respond to this call and conceptualise a PPMO that is in charge and operationally manages project portfolios to produce a clear value proposition. Hence, we identify the performance contribution of individual PPMOs' roles. We aim to extend the findings of a quantitative study in which the engagement of project portfolio managers, who are often the heads of PPMOs, has been shown to have a significant positive impact on the execution quality and success of PPM ( Jonas et al., 2011a). Ultimately, these results should establish the fundamental legitimacy of PPMOs. In showing the positive performance impact of PPMOs, we substantially extend the existing research. Previous attempts to provide evidence for a performance increase as a result of PMOs' involvement in project management or in overall organisational performance offered little empirical validation. Quantitative research on the impact of PMOs on single project management has failed to show a relationship between PMOs' involvement and improved performance (Dai and Wells, 2004 and Kwak and Dai, 2000). In a qualitative study of 65 organisations, Thomas and Mullaly (2008) showed the near impossibility of calculating the direct impact of single projects managed by PMOs for return on investment. In two other qualitative studies, Aubry and Hobbs (2011) and Aubry et al. (2011) suggested that the performance of a PMO should be assessed by its contribution to organisational performance. These studies proposed a multi-dimensional framework that acknowledged the coexistence of competing values in any organisation. From this perspective, assessing a PMO's performance is likely to be an organisational dialogic process involving criteria and indicators. More recently, Hobbs and Aubry (2010) used the concept of a PMO's embeddedness in its context to explain 48% of the variance in project performance, whereas the structural characteristics and functions of the same PMOs explained only 28% of the variance. These are encouraging findings, but they are based on exploratory data analysis, and the performance measures are broadly defined. Further research is needed to explain and validate these findings. We therefore posit the following: despite increasing research efforts, solid empirical evidence for the positive impact of multi-project PMOs on performance is still lacking. In light of so many unsatisfactory attempts to answer the fundamental question of PMOs' contribution to performance, it is not surprising that consultants and academics increasingly focus on this topic (Aubry et al., 2010a). Of primary interest is to extent previous inconclusive research on the value of PMOs (Dai and Wells, 2004 and Hurt and Thomas, 2009) and to answer the critical open question of PMOs' performance contribution, which, in principle, would justify the existence of a multi-project PMO. The popularity of this research area despite its failure to produce fundamental evidence has also generated criticism. Some academics dismiss the concept of PMOs as a temporary fashion (e.g., Crawford, 2010). It is argued that the academic community participates in constructing and nurturing an empty shell and that isomorphism may be at play, considering the high level of information flow (DiMaggio and Powell, 1983). Against all odds, this study analyses PPMOs as a subset of PMOs that manage project portfolios. Departing from organisational theory, we aim to describe PPMOs' activity patterns as a way to cater to organisational and stakeholder needs, independent of individual preferences. Thus, we strictly focus on management requirements in connection with the project portfolio setting, and we disregard PMOs that are designed to manage single projects or programmes. We aim to present the roles derived from PPMOs' typical activity patterns by statistically integrating similar tasks to reduce complexity. Hence, our first research question is as follows: What roles does a project portfolio management office perform? To adequately address the question of how PPMOs contribute to performance, relevant performance metrics must be applied. We use the execution quality of PPM (i.e., project portfolio management quality, or PPM quality) to incorporate the project portfolio setting. Furthermore, any relationship must be empirically shown. We quantitatively analyse how role fulfilment impacts on PPMOs' performance. Hence, our second research question is as follows: What impact do PPMOs' roles have on performance in terms of project portfolio management quality? This study makes significant contributions. First, we empirically derive and characterise three distinct roles of PPMOs. We clarify the generic PPM tasks performed by PPMOs and we corroborate initial evidence on general groups of PMOs' tasks or roles. Second, we show the significant positive impact on performance of two roles assumed by PPMOs, the roles of coordinator and controller. Third, we provide evidence for the positive direct effect produced by PPMOs' supporting role on the success of project portfolios. This evidence points to the underlying principle of our model: the mediating effect of PPM quality on portfolio success.

These findings contribute to the academic literature on multi-project PMOs in general and the roles and performance of PPMOs in specific. PPMOs' impact on performance and success is especially relevant from a theoretical standpoint. First, three distinct roles of PPMOs were theoretically identified, and the activity patterns characterising their social behaviour were outlined. This analysis reduced and integrated the vast number of tasks currently proposed for PMOs to a comprehensive set at a multi-project management level. Thus, the gestalt of the PPMO as a coordinating, controlling and supporting unit at the portfolio management level was established. By clarifying PPMOs' behaviour, we furthered understanding of PMOs' distinct roles at a multi-project management level compared to a single project management level. Second, two of these roles – the coordinating and controlling roles - were shown to impact on PPM quality, which predicts project portfolio success. Hence, this paper introduced the first quantitative empirical evidence showing that PMOs are instrumental in attaining project success, as exemplified in the three roles of PPMOs' and their positive impact on PPM quality and portfolio success. Third, the coordinating role was shown to have a twofold impact, positively influencing both cooperation and resource allocation in the portfolio management process. Thus, we may assume a graded impact scale of the three roles of PPMOs. The controlling role only impacts on information quality, and the supporting role only shows a direct effect on the average single project success. Thus, both of these roles have a lesser impact range than the coordinating role. 9.1. Implications for management Practitioners may benefit from these findings. The implications of this study are mainly tied to the understanding and differentiation of the various roles assumed by a PPMO, which condition its power and threshold of action in a multi-project environment. Some suggestions are provided on the potential capacity of PPMOs. First, when establishing a new PPMO, the three roles of coordinator, controller and supporter provide insight into how a PPMO can participate in a PPM. From a behavioural point of view, in addition to implementing the fit of patterns of activities, PPMOs help to instil a certain attitude (or root a distinct culture). Relying on these roles facilitates the shaping and designing of each PPMO's activity pattern, intensity, size and structure and suggests how the PPMO should relate to other PMOs and stakeholders in the PPM context. What is at stake in relying on these roles is the objective that a PPMO should appropriately correspond to the organisational demand. Thus, depending on the intended effect of the PPMO, a particular role may be emphasised to facilitate a specific PPM quality. However, from the perspective of project managers, creating these types of PPMOs can be seen as removing substantial degrees of autonomy, creativity and ownership. As project managers typically value self-determination and loathe formal processes and documentation (Pellegrinelli and Garagna, 2009), they are likely to resist imposed authority by PPMOs that, prima facie, fall short to state value to project managers. Project managers' lack of trust in PPMOs may spark disagreement. In this case, the organisations' leaders may delegate minimally and not very truthfully. In a worst-case scenario, the PPMO will become the common enemy of both groups, leading to collective action to dismantle the PPMO. All in all, a new PPMO without a convincing value proposition in terms of its supporting, controlling and coordinating roles in PPM, including positive performance contributions, may find itself challenged by its own community. Second, when assessing an existing PPMO, these roles provide a self-assessment tool to diagnose and chart the existing activity patterns. PPMOs' leaders can thus understand the PPMOs' current involvement in PPM. However, roles also allow for easy mapping of the desired future situation. Thus, new value may easily be generated by re-defining the PMO's purpose and activities (Pellegrinelli and Garagna, 2009). Presenting roles as activity patterns that mirror social behaviour rather than as lists of tasks should be especially beneficial for practitioners in designing, assessing and reshaping their respective PPMOs by offering abstract descriptions. To aid in configuring PPMOs, we summarise the results in a comprehensive framework, contributing a detailed description of the three roles of PPMOs as coordinators, controllers and supporters and their effects on performance (see Fig. 2). A positive consequence of understanding the roles of PPMOs and their effect is to prevent overlapping responsibilities among actors, which can pose conflicts of competencies. This role clarity helps to better position the PPMO as a PPM actor, and the PPMO unit can be assumed to perform more effectively when it is clear what needs to be done (Hall, 2008 and Tubre and Collins, 2000). Third, with the trend of multiple PMOs in organisations that perform specialised tasks or distinct roles, the design and integration of each PMO within such a multi-project environment becomes crucial. The three roles help to manage the interconnectedness of multiple PPMOs. In a typical scenario, if one PMO only fulfils one out of the three proposed roles and another PMO fulfils the other two roles, both parties need to be aware of their foci to avoid conflicts of competence (Müller et al., 2011). Finally, by differentiating three roles, this study reminds practitioners of the need for the embeddedness of PPMOs. By considering embeddedness the transparency of PPMOs' mission and action is augmented (Hobbs and Aubry, 2010). 9.2. Limitations and future research Some limitations need to be addressed. Two often-cited activities of PPMOs, “human resource management tasks” and “customer interface”, failed to load in our analysis. These items were dropped, and are not included in the model. Therefore, it is unclear who performs human resource-related tasks, such as recruiting and selecting personnel for PPM, as well as motivational tasks or the execution of a performance incentive system. Moreover, PPMOs' interface activities with internal and external stakeholders, such as customers, are unclear. The sample for this analysis is comprised of six national subsamples. The German and Swiss samples were generated from fewer items (7 items) compared to the Austrian, Canadian, Finnish and South Korean samples (15 items). This variation in items means that missing values had to be complemented for four items, two in the coordinating role and one in the controlling and supporting roles, respectively, in part of the database. Future research should follow up on the effort to make sense of the proposed PMO tasks, such as “customer interface” and “human resources”. These tasks echo typical activities adopted in practice by PPMOs but have not been successfully included in analyses thus far (compare, e.g., Hobbs and Aubry, 2007). It may be necessary to attempt a re-conceptualisation of these items to accurately measure them. Moreover, the supporting role of PPMOs requires more attention because of its critical relationship to success, at least from a theoretical standpoint. The impact on PPM quality has not been significant, but a direct effect exists on average single project success. These mixed findings may serve as a starting point for future research. The theme of this study may be extended to further research into PMOs' various and varied roles depending on their location in the hierarchy and within the multi-project environment. Future research should analyse the process beyond project management maturity (Pinto et al., 2010), which relies heavily on the notion of progression and transition and fails to establish a link to performance.