Now that IT spending is on the rise, more IT projects are beginning to find their way onto the front burner. So the time is right to take a close look at IT project best practices.
In order to help CIOs plan future projects and increase their probability of success, veteran project manager Al Taylor teamed with CIO Canada to develop a survey examining the factors that contribute to project success and failure. The survey resides on the Internet under the domain name www.itprojectmetrics.com, where it has appeared for over a year. At press time it had been filled out by about 100 respondents.
The survey asks participants to enter the results of completed projects, including whether or not the project was deemed a success. Participants were asked to rate their project on a scale of 1 to 10. These ratings allowed us to quantify success and failure factors.
A separate survey by Athabasca University indicates that fewer than 50% of IT managers agree with the statement “We typically have good IT project management practices”. It is obvious that the current state of PM processes and procedures represents an opportunity in many organizations.
This article shares the collective wisdom of the large group of stakeholders who contributed to our project-management survey. The article also provides useful metrics that can be used to influence project stakeholders and external parties who control project budgets.
The following are several best practices that we have extracted from our survey database:
Sponsor involvement and title
The data shows a direct relationship between the level of sponsor involvement and the probability of success. CIOs need to ensure that project sponsors are engaged and motivated. Project managers should be able to distinguish an interested sponsor from an uninterested one, and should be encouraged to respond to the latter scenario.
Intuition suggests that the higher up in the organization the sponsor, the greater the probability of project success. The data does not support this; projects sponsored by Managers and Directors appear to be just as likely to be deemed successful as those delivered by more senior executives. This is good news for CIOs; there appears to be no need to defer or cancel important projects due to a lack of senior executives available or willing to serve as sponsor.
The project charter and PM contribution
Projects need a charter. In our survey, projects without a charter have a significantly lower average project rating. There appears to be significant value in having the PM position staffed early in the project. It is reasonable to assume that the PM who contributes to the definition of the charter understands the business objectives and can use that knowledge to motivate the project team. If you bring the PM on to the project after the initiation phase, or have to replace the PM during the project, be aware that this introduces risk.
The project manager
PM skills appear to be portable; almost half of all projects were managed by a vendor or a contractor. These external resources appear to be doing a good job; their average project ratings are in fact slightly higher than the average rating where projects were managed by internal service providers. This is good news for organizations where the demand for PM skills exceeds supply. Rather than cancel or defer projects because of a lack of internal PM resources, CIOs can use external resources with confidence.
Communications and decision-making
The survey demonstrates that strong project communication and decision-making models lead to success. We see a direct relationship between the effectiveness of these models and average project results. These models represent opportunities for reuse – CIOs should ensure that there are standard or template models in place that can be used as the base models for every project. Project managers can fine tune the base models as required. Of course it is not enough to have these models in place; they must be applied effectively. To measure the effectiveness of these models, watch for the following: confusion around project status, and items remaining on the issue log for extended periods of time.
Our data confirms the value of speedy delivery and the phased approach. We see a direct relationship between project length and risk; specifically, the shorter the project the greater the probability of success. CIOs need to be aware of the risks associated with long-running projects and should challenge project plans that propose long investment periods without concrete deliverables. Where this type of project cannot be avoided, risk mitigation is required.
Project manager designation
In today’s recruiting marketplace we see significant demand for project managers with the PMP designation; this demand appears to be warranted. The average project rating of projects delivered by PMs with the designation is slightly higher than those delivered by PMs with other designations and significantly higher than those delivered by PMs without designations. (Note: much of our survey data was provided by PMs with the PMP designation, which may introduce a bias into the data).
Project management office (PMO)
Many organizations have established a project management office. The PMO was involved in 62% of all projects. We see a direct relationship between the effectiveness of the PMO and average project rating. We note that the PMO was deemed ineffective 25% of the time, so it is clear that some organizations need to work to improve the structure and role of the PMO. CIOs who have not established a PMO should consider doing so. Those who already have a PMO in place should monitor its effectiveness to ensure optimum results.
The survey supports the notion that the effective use of project-management process and procedures leads to project success. We noted a significant positive affect of PM Tools, PM Methodologies, and Systems Development Methodologies on successful projects. As well, Project Implementation Reviews (PIRs) had a significant effect on approximately one-half of all successful projects. This provides a clear message that there is value in looking at the results of completed projects and applying those learnings to future projects.
Key factors influencing success and failure
Survey respondents were asked to identify key factors that had a significant effect on the outcome of the project. Responses were unprompted, so a wide variety of factors were identified. The accompanying charts indicate the most frequently mentioned factors for successful and unsuccessful projects.
It is interesting to note that projects that are being delivered late and/or over budget are being deemed successful. Specifically, 29% of successful projects were delivered late and/or over budget. This statistic suggests that some organizations place greater emphasis on the benefits delivered by the project than they do to the cost and timing of delivery. This reminds us of the importance of defining project success criteria. CIOs should insist that during the initiation phase: (a) there is an appropriate definition as to when the project will be completed, and (b) the criteria for determining project success are clearly identified. This is especially important for enterprise projects, where the results can have a significant effect on the IT department’s relationship with its clients.
In the short term, it is important to take a look at projects in progress. Our data suggests that there is a high probability of success if you have:
– processes and procedures in place;
– effective communications and decision-making models;
– the PM position staffed early; and
– an engaged sponsor.
If the project is lacking any of these characteristics then it may be at risk and you may need to take action.
In the longer term, CIOs should create an environment that facilitates continuous improvement in IT project delivery. This environment can include the collection of project metrics and the identification of best practices within the organization. The data can easily be graphed to facilitate distribution to project stakeholders.
Using these tools, the CIO can transform the knowledge and experience of project managers into an asset that protects the organization’s investment in IT projects and delivers shareholder value.
Al Taylor is a systems professional with extensive experience managing IT projects. He is currently on assignment at one of Canada’s leading life insurance companies.
Factors mentioned most frequently by respondents whose projects were deemed a success:
PM planning and procedures: Mentioned by 29.3% of respondents
Communications: Mentioned by 21.1% of respondents
Knowledgeable staff: Mentioned by 18.3% of respondents
Committed and motivated resources: Mentioned by 15.5% of respondents
Project sponsor influence and motivation: Mentioned by 14% of respondents
Teamwork: Mentioned by 11% of respondents
Well-defined requirements: Mentioned by 8.5% of respondents
Senior management support: Mentioned by 7% of respondents
Factors mentioned most frequently by respondents whose projects were deemed a failure:
Sponsor issues: change of sponsor, lack of involvement: Mentioned by 25% of respondents
Poor/changing requirements: Mentioned by 17% of respondents This confirms the need for a strong scope statement in the charter and effective change-control procedures.
Organization change, e.g. mergers: Mentioned by 17% of respondents CIOs need to be aware of project risk introduced by organization change. Some projects should be influenced by organization change and perhaps even terminated. Others need to be protected from organization change. Project teams may need help to steer the project through periods of organization change.
No project plans with goals/strategies for success: Mentioned by 17% of respondents
About the Project Metrics Survey
An interest in project metrics and concern over the state of IT project management contributed to the development of the survey that generated the data for this article. The survey was conceived by Al Taylor and produced with the assistance of CIO Canada, and its editor, David Carey, and editor-in-chief, John Pickett.
Additional assistance was provided by the Project Management Institute (PMI), which provided a temporary link from their research portal to the site.
We would like to thank all of those who have participated in the survey, as well as our focus group members, who provided invaluable help in creating and fine-tuning the questionnaire: Ed Boctor, Dr. Elizabeth Moxley-Paquette, Stephen Wong, and Cary Zweig.
The survey itself, as well as its ongoing statistical findings, continue to appear on the Web at www.itprojectmetrics.com. The site will continue to collect data for at least several more months. Additional responses would be appreciated.