Despite decades of practice and research, projects continue to underperform (Hayes, 2004; Johnson, 1995; Johnson, Boucher, Connors, & Robinson, 2001; Williams, 2004). The underperformance of projects represents a significant but substantially avoidable loss of economic value. One factor that has been consistently ranked high among factors leading to project failure is the lack of executive support (Hayes, 2004; Johnson, 1995; Johnson et al., 2001; Schmidt, Lyytinen, Keil, & Cule, 2001).
Management support in the form of project sponsorship has consistently been cited as critical to achieving project objectives (Graham & Englund, 2004; Love & Brant-Love, 2000; Schmidt et al., 2001; Ross & Weill, 2002). Yet, there is little validated knowledge on the situations in which an organization should use project sponsorship to protect project outcomes. Senior management’s time and attention is limited and it is, therefore, impossible for them to participate in all aspects of operation (March & Simon, 1958).
Drawing on the management control theory (Simons, 1987, 1990, 1991), this paper argues that the level of perceived management priority and project risk leads to increased use of project sponsors, which, in turn, impacts on project outcomes. The propositions are tested with survey data from project managers and senior managers with project management oversight in Australian companies. The findings suggest that organizations could improve their project delivery capabilities by appointing sponsors to projects with high priorities and by developing a project-centered culture that focuses on results.
In the following sections, the literature is reviewed and hypotheses are developed. The research design and analysis are then described, results are presented, and their implications discussed. The findings have important implications for both research and practice development.
The Precedence and Effects of Project Sponsorship
A project is a complex endeavor to deliver a set of business objectives within constraints in a unique organizational setting. Reviewing the extant literature, there appears to be four imperatives for project sponsors. First, a project’s success in delivering the business objectives depends on the commitment and involvement of stakeholders beyond the control of the project manager. Support from senior management to secure commitments from stakeholders is often seen as necessary to realize desired benefits (Graham & Englund, 2004).
Second, project managers are not members of the senior management team and, therefore, are unlikely to be conversant with the latest strategic intentions of the organization. So, guidance and oversight from senior management on the conduct of the project is needed to ensure the project is on track to realize strategic benefits (Ross & Weill, 2002). Third, projects compete with other priorities for resources and commitments. Typically, a project manager does not have a stable power base and works outside the normal line reporting structure. Resources for projects have to be negotiated and bargained (Pinto, 2000). Therefore, it is important to have a senior manager to “provide air cover for the troops” (Graham & Englund, 2004; Sauer, Liu, & Johnston, 2001). Finally, as a temporary endeavor, a project may be treated as secondary to permanent and continuous operations. Strong and visible commitment from senior management is critical to motivate the project team (Grover, 1993; Jarvenpaa & Ives, 1991).
According to Simons (1994; 1995), there are four types of control systems that organizations can employ to manage their operations. These are diagnostic control systems, belief systems, boundary systems, and interactive control systems. This study focuses on the choice and implications of adopting interactive control systems, defined here as formal systems used by top managers to regularly and personally involve themselves in the decision activities of subordinates (Simons, 1987; 1991). With senior management involvement, the standard assumptions underlying project decisions can be challenged in the project team’s search for innovative solutions (Simons, 1994; 1995).
Typically, an interactive control system is invoked in situations of high strategic uncertainty, where strategic uncertainty refers to uncertainties that are fundamental to achieving business goals and could provide threats or opportunities as circumstances change (Daft & Macintosh, 1981; Simons, 1990; 1991; 1994; 1995). Implicit in the definition is that uncertainty and importance combine to create “strategic uncertainty.” It follows that effective senior managers focus their attention on issues that are both uncertain and important to business outcomes.
High uncertainty alone does not attract the attention of senior managers unless the events or factors causing the uncertainty are also seen as important to achieving critical organizational goals (Daft & Macintosh, 1981; Simons, 1990; 1991; 1994; 1995). Similarly, importance alone may not be sufficient to attract their attention. For example, Simons (1991) reported that top managers of low-cost, high-volume U.S. healthcare product businesses do not pay much attention to efficiency-related controls such as cost accounting systems. Instead, they focus their attention on the systems that produce and monitor information on the strategic uncertainties that threaten their vision of the future. In contrast, goal-setting and exception-based reporting are used to manage efficiency-related systems. In project management, the attention of senior managers can be measured by the perceived level of management priority. Therefore:
Hypothesis 1A: Project sponsorship is a function of the level of management priority of achieving project targets.
The level of project risks in achieving project targets represents major uncertainties facing the project. It follows:
Hypothesis 1B: Project sponsorship is a function of the level of perceived risks of achieving project targets.
Organizations today are faced with strong competition and demanding clients. Frequently, at the beginning of a project, the clients have conflicting and incomplete objectives embedded in their requirements. Business needs must be translated into requirements and specifications. Based on those requirements and specifications, the project team delivers the products and is responsible for delivering the project to the client, on time, to budget, and of quality and of functionality. This task is made particularly difficult by changes in business and functional requirements. Requirement changes or scope creep have consistently been identified as a major cause of project failures (Boehm, 1991; Johnson, 1995; McConnell, 1996; Schmidt et al., 2001).
Similarly, other studies show that the quality of the requirements-analysis phase impacts on later phases (McConnell, 1996; Zmud, 1980). Errors not identified in the early stages of a software project are expensive to fix later (Boehm, 1991; McConnell, 1996). A large number of methods has been developed to improve requirements analysis, such as user involvement and participation, prototyping, and incremental delivery. Generally, these methodologies have not met users’ objectives (Iivari, Hirschheim, & Klein, 2000; McConnell, 1996). New methods are being proposed and tried (Iivari et al., 2000) and warrant close attention from senior management (Simons, 1990; 1991; 1994; 1995).
Appointing a project sponsor is a mechanism to exercise interactive control over projects (Briner, Geddes, Hastings, 1990; Frame, 1994). Project sponsors are not responsible for the execution of projects. Rather, their critical responsibility is to ensure the overall success of projects (Briner et al., 1990; Graham & Englund, 2004; Ross & Weill, 2002). It follows that project sponsors should take interactive control of issues of strategic importance, such as achieving project objectives. The project manager retains responsibility for the day-to-day management of the project.
There are two levels at which project performance is measured. One focuses on the performance of a specific project. The other is an organization’s capability to deliver projects according to the client’s expectations, defined here as an organization’s project delivery capability (PDC).
The recent process movement, including, for example, the capability maturity model (CMM) approach to managing software projects, reflects the shifting of attention to PDC from the previous dominant research focus on individual project performance. PDC differs from the term project performance in that PDC considers an organization’s consistency when delivering various projects over an extended period of time. In contrast, project performance typically refers to the one-off performance in delivering a project (Liu, Yetton, & Saucer, 2003; Liu & Yetton, 2004).
Researchers have divided project performance into two subdimensions: product performance and process performance (Nidumolu, 1996; Wallace, Keil, & Rai, 2004; Yetton, Martin, Sharma, & Johnston, 2000). Product performance refers to the success of the product developed, while process performance refers to the successful delivery of the product within project constraints (Wallace et al., 2004). Formally:
Hypothesis 2A: Project sponsorship has a positive effect on PDC product performance.
Hypothesis 2B: Project sponsorship has a stronger effect on PDC product performance when management priority is high than when it is low.
The impact on process performance is secondary, remaining the primary responsibility of the project manager (Yetton et al., 2000). The primary concern of a project sponsor is to protect the product performance. The sponsor’s attention to and influence on process performance is limited. Formally:
Hypothesis 3: Project sponsorship has a weak or nonsignificant effect on PDC process performance.
A high level of management priority gives the project legitimacy, elevates team morale, and ensures sufficient resources for undertaking the project. Formally:
Hypothesis 4: Management priority has a significant effect on both PDC product performance and PDC-Process Performance.
In contrast, a high level of project risks negatively impacts on PDC product performance and PDC process performance. Formally:
Hypothesis 5: Project risk has a significant negative effect on both PDC product performance and PDC process performance.
Research Design
The hypotheses are tested using survey data collected from registered project managers of the Australian Institute of Project Management (AIPM). While project management practices are likely to vary across business units within the same organization, within the business unit across projects, the difference in project management practices is likely to be small. Therefore, the business unit is adopted as the unit of analysis. Project managers were asked to respond to the questionnaire in relation to the business unit they had been most associated with. The questionnaire was sent to 746 program/project managers and 95 responses were received, resulting in a response rate of 13.0%.
Instrument Design and Validation
Dependent Variables
PDC is an organization’s capability to satisfy expectations on cost, time and quality, functionality, and achieving business objectives, as measured by each respondent’s perceptions of overall performance, relative performance with competitors, and satisfaction of clients (see Appendix 1 for instruments). Following Wallace et al. (2004) and Nidumolu (1996), PDC is divided into two dimensions—product performance and process performance. Table 1 reports acceptable reliability indices for PDC on the two dimensions, respectively.
| PDC Product Performance |
0.89 (6 items) |
| PDC Process Performance |
0.89 (9 items) |
Table 1: Cronbach alphas for the dependent variables
| Instruments | PDC Process Performance |
PDC Product Performance |
| QA2_1A | .626 | |
| TI2_1B | .693 | |
| COS2_1C | .738 | |
| QA2_2A | .684 | |
| TI2_2B | .564 | |
| COS2_2C | .656 | |
| QA2_3A | .709 | |
| TI2_3B | .773 | |
| COS2_3C | .723 | |
| SPE2_3D | .523 | .631 |
| BIZO2_3E | .653 | |
| SPE2_1D | .486 | .636 |
| BIZO2_1E | .800 | |
| SPE2_2D | .810 | |
| BIZO2_2E | .902 |
Extraction Method: Principal Component Analysis
Rotation Method: Varimax with Kaiser Normalization
Table 2: Exploratory factor analysis for the dependent variables
Exploratory factor analysis results are reported in Table 2. Out of the 14 questions measuring PDC, there are two cross-loadings of above 0.4 from the questions measuring PDC product performance on PDC process performance demonstrating good construct validity.
Independent Variables
Project sponsorship is measured by the percentage of IT projects that have been assigned a project sponsor. Out of the 95 responses, 44 (46%) assign project sponsors to all their IT projects, while four (4%) do not assign project sponsors to any IT projects. On average, business units assign project sponsors to 71% of their projects.
Management priority is measured by the respondents’ perception of senior management’s priority in achieving specific project targets for a typical project (see Appendix 1). Project risk is measured by the level of risk in achieving specific project targets for a typical project undertaken by the business unit. Table 3 reports acceptable reliability coefficients for the two constructs.
| Management Priority | 0.71 (5 items) |
| Project Risk | 0.86 (5 items) |
Table 3: Cronbach alphas for the independent variables
Analysis
Hypotheses 1A, 1B, 2A, 3, 4, and 5 are tested by regressing the dependent variables on the corresponding independent variables using Equation 1:
Where Y denotes dependent variables and X denotes the independent variables. A significant β1 indicates the significant effect of the independent variable on the corresponding dependent variable.
The moderating effect as hypothesized in H2B is tested using Equation 2. A significant β3 supports the hypothesis on moderating effect (Venkatraman, 1989). Following Cohen, Cohen, West, & Aiken (2003), the variables are standardized to minimize the effect of multi-colinearity.
Results
Hypothesis 1A is supported—project sponsorship is significantly influenced by perceived level of management priority. Hypothesis 1B is not supported—project sponsorship is not significantly associated with project risk. Table 4 reports that the standardized regression coefficients are:
Hypothesis 2A is supported - Project sponsorship has significant positive effect on PDC product performance. Table 4 reports that the standardized regression coefficients for regressing project sponsorship on PDC product performance is = 0.22 (p<=0.05). Hypothesis 3 is also supported—project sponsorship has a nonsignificant effect on PDC process performance. Table 4 shows that
=0.19 (n.s.) and the adjusted R square is 0.03.
| Project Sponsorship | PDC Product Performance | PDC Process Performance | ||||
| P | P | P | ||||
| Management Priority | 0.23 | 0.05 | 0.43 | 0.00 | 0.52 | 0.00 |
| Project Sponsorship | N/A | N/A | 0.22 | 0.05 | 0.19 | n.s. |
| Project risk | -0.11 | n.s. | -0.28 | 0.01 | -0.36 | 0.00 |
Table 4: The main effects
Hypothesis 4 is supported—Management priority has a significant effect on both PDC product performance and PDC process performance. Table 4 reports that the standardized regression coefficients are: PDC-Product Performance = 0.43 (p<=0.00) and
Process Performance = 0.52 (p<=0.00).
Hypothesis 5 is also supported: Project risk negatively impacts on both PDC product performance and PDC process performance. Table 4 reports that the standardized regression coefficients are: PDC Product Performance = -0.28 (p<=0.01) and
Process Performance = -0.36 (p<=0.00).
| PDC Product Performance (Adj. R2 = 0.23) | PDC Process Performance (Adj. R2 = 0.28) | |||
| P | P | |||
| Management Priority | 0.51 | 0.00 | 0.58 | 0.00 |
| Project Sponsorship | 0.15 | n. s. | 0.08 | n.s. |
| Management Priority * Project Sponsorship | 0.25 | 0.05 | 0.17 | n.s. |
Table 5: The moderating effects of management priority
Hypothesis 2B is supported: Project sponsorship has a stronger effect on PDC product performance when management priority is high than when it is low. Table 5 reports that the standardized regression coefficient for the interaction term is as follows: = 0.25 (p<=0.05).
Discussion
In this section, the results are summarized, validity threats to the findings are reviewed, and the implications for theory and practice are discussed.
Findings
Consistent with the predictions derived from management control theory (Daft & Macintosh, 1981; Simons, 1987; 1990; 1991; 1994; 1995), Table 4 reports that management priority has a significant impact on project sponsorship. In contrast, project risk has no significant impact on project sponsorship. Project sponsorship is found to have a significant positive effect on PDC product performance but not on PDC process performance.
Further, the study finds that management priority has significant main effects on both PDC product performance and PDC process performance. Similarly, project risk is found to have significant negative effects on both PDC product performance and PDC process performance.
Extending extant theory, Table 5 shows that there is an interactive effect between management priority and project sponsorship. The two variables jointly explain 23% of the variance in PDC product performance.
Implications for Theory
The specific finding of the positive effect of project sponsorship on PDC product performance but not on PDC process performance, provides further evidence for the contingent nature of management support. Specifically, the results question the view that top management support is required across all dimensions of project performance.
Recall, Sabherwal, and King (1992) found that top management participates only in the projects they perceive as important. Here, the results show that management priority interacts with project sponsorship to impact on PDC-product performance and its effect on the service delivered to the client.
This study refines findings from several studies (Yetton et al., 2000; Sabherwal & King, 1992), concluding that project sponsors have a direct effect on the PDC product performance but not on PDC process performance. When product performance is the key project performance criterion, high-risk projects are associated with high levels of user participation (Barki, Rivard, & Talbot, 2001;Yetton et al., 2000). In those situations, the involvement of a project sponsor supports client (user) participation. The sponsor engages senior managers from the client organization as actively involved users (Yetton et al., 2000). The sponsor is better positioned than the project manager to engage and negotiate with the client over variations to the project deliverables.
The finding that management priority has a positive main effect on both PDC product performance and PDC process performance suggests that management priority impacts on PDC even without the presence of project sponsorship. The lack of evidence supporting the link between project risk and project sponsorship suggests that project level risks need to be managed with other interventions.
Implications for Practice
The findings reported in Table 4 suggest that, when a project faces a high level of uncertainty in project outcomes with significant business implications for both the organization and its clients, assigning a senior manager as the sponsor to the project improves the chances of delivering the project as required. The project sponsor focuses attention on protecting the product performance, while leaving the process issues as the responsibilities of the project manager. The project sponsor’s primary concern is to deliver the functionalities that meet the client’s business needs. Project sponsors do that by exerting leadership, allocating the necessary resources to deliver quality and providing “political” protection for the project management team.
More importantly, the findings show that, above all, management priority placed on protecting project outcomes is paramount regardless of whether project sponsors have been appointed. Management priority influences the effective use of project sponsorship and they jointly impact on project outcomes. The results suggest that elevating project management priority could result in significant improvement in project outcomes.
There is an opportunity to reduce the level of strategic uncertainty. In that case, senior management can be relieved of the need for direct intervention to protect PDC product performance. Instead, they would be able to focus their attention on other strategic opportunities to add value. The current developments in methodologies and process improvement are essentially efforts to reduce task uncertainty and improve PDC, which will eventually lead to the reduction of strategic uncertainty.
One promising approach lies in improving product performance by introducing best practice, and benchmarking with other project teams and organizations. In the long term, the focus should be on reducing task uncertainties to capture improved business benefits. The coordination of resources, roles, and responsibilities can improve task programmability and, therefore, reduce organizational uncertainty. Similarly, methodologies and tools for the conduct of project tasks can contribute to reducing strategic uncertainty by reducing technical uncertainty.
Summary
This paper shows that management priority and the interaction between management priority and project sponsorship protect the strategic project outcome of an organization’s capability to deliver projects, influencing PDC product performance but not PDC process performance. This contributes to the emerging literature on the contingent nature of project performance. The implications for both theory and practice are to reinforce the need to further refine and to extend the models of the contextual contingencies influencing project business deliverables. The findings provide insight on how to focus senior management attention and actions to protect the strategic outcome of projects.
To generalize the findings to other industries, further studies are needed to identify strategic project outcomes that need the protection of executive sponsors and to test the effect of project sponsorship on project outcomes in different contexts.