Introduction
Implementing Project Management systems into an Owner organization is largely an organizational change management exercise. In organizations with many different project sizes and types, the implementer would do well to consider introducing scalability into the design of the system in order to both prevent the inefficient use of company resources as well as to allay project managers’ often justified fears of excessive bureaucracy.
This paper will describe the role that scalability played in a project management implementation that the author led for Unocal Thailand Inc. (now Chevron Thailand Inc), starting in October 1997. It will also demonstrate the role that performance metrics play in re-aligning organizational behavior.
Results
Before we describe the tool, we will describe the results achieved by the implementation. Unocal Thailand went from being at the Ad Hoc stage of project management maturity (Fincher & Levin, 1997) to the Defined Stage over a three year period. During the same period, the Engineering and Construction group halved their wellhead platform cost, and the Information Technology group dramatically improved their project delivery to the organization. Although Departmental and Individual contributors actually made those improvements, we believe that the additional formality and structure imposed by the project management system and related coaching was a key enabler for making those improvements.
Background
In early 1996, it became apparent that although Unocal Thailand Inc (UT) appeared to manage its routine projects quite well, unique and/or large Engineering and Construction projects were running into trouble too often. After one false start on implementing a project management system into the Engineering and Construction organization, the author was asked by the UT President to implement a project management system not just for Engineering and Construction, but for the whole of the organization. At the time, UT employed about 1500 people, and the President noted that almost anything unique or different that was done in the organization could be treated as a project.
With this mandate in hand, the author traveled to his first PMI conference in Chicago to learn about what made a project management implementation successful. The answer that we found for our style of organization can be summarized as follows:
- Use coaching and mentoring to help people learn the necessary skills on the job
- Pay attention to cultural change issues – especially the “What's in it for me” (WIFM)
- Active Executive support is foundational
The Scalable Front End Loading Index (S-Felix©), was developed a part of responding to the fact that their was enormous diversity in the number, type, and importance of projects at UT, and yet all had to be managed under the system. The tool was developed to re-assure the project managers that they would only be asked to do things that would add value to managing the project, while at the same time ensuring that sufficient front end loading (FEL) was done to protect UT from a risk and cost management perspective. This message, and the resulting implementation, addressed peoples’ needs and made the implementation both smooth and effective.
The idea for developing the tool came from a paper given at the Chicago conference by A. J. Shenhar (1997). In it he proposed that the tools one should use on projects are a function of three distinct parameters – the degree of uniqueness, degree of complexity, and the pace of a project. This prompted the thought that the same concept could be used at a more detailed level to develop objective and scalable guidelines for the organization as to what would add value in a specific project. This final tool has eighteen questions that characterize the topology and type of the project. The answers to these questions drive a matrix addition scheme to turn on specific organization, planning and control tools relevant to the specific project. This tool has been empirically “tuned” against hundreds of projects over the intervening years in order to improve the suggestions the tools makes. However, there is still no substitute for the final application of good judgment!
The S-Felix tool has two components – a planning deliverables section and a design deliverables section. The selection of planning deliverables is driven by the answers to the eighteen questions. The design deliverables are chosen from a list which has been tailored to each main application area. The planning deliverables are worth 1/3 of the total score and the design deliverables are worth 2/3 of the score. This is in recognition of the fact that good cost estimates are mostly a function of adequate design having been done prior to implementation – and only secondarily related to the planning deliverables to ensure that the project runs smoothly. The S-Felix tool also generates a Project Complexity Index (PCI). This PCI is simply the percentage of the planning tools that are being required as compared to all of the tools that could be required. Small routine projects typically have PCI's in the 25% range, while large multi-year developments have PCI's in the high 80's.
How the S-Felix® tool is used
Initiation
During initiation, the project manager answers eighteen questions on an MS Excel® template that characterize his/her project. They then enlist the help of subject matter experts to determine what design deliverables are relevant to the specific project under consideration. Having done the assessment, they sit with the Project Owner or Sponsor (PO) to formally agree the planning and design deliverables that will be required for the project.
From a change management perspective, this step in the process ensured that the correct work was being identified to appropriately do the FEL, while bringing the PO into the process early. Prior to the implementation PO's often delegated projects in a haphazard way without clearly specifying goals and objectives. Implementing a project chartering process combined with the S-Felix assessment encouraged the PO's to be more proactive in their approach towards project initiation. At the same the Project Management Department collected the PCI's to maintain a view on the distribution of project complexity within the company on a monthly basis.
Another benefit from this approach was that many project managers – upon seeing an unfamiliar deliverable being required by the tool, would come ask the project management coaches for advice as to appropriate content and use of the deliverable. This allowed the coaches to have very productive and focused interactions where there were specific gaps in knowledge.
Planning
During the planning phase the project team developed the planning deliverables that were agreed upon in the previous phase. This encouraged appropriate use of the project management systems without any gratuitous bureaucracy.
Execution and Control
Prior to beginning the implementation of the project, the Project Owner reviews the S-Felix score, which is attached to the Authority for Expenditure (AFE). The PO exercises their judgement as to how much FEL is necessary before approving the AFE. On small routine projects, where the project manager has a lot of experience, a PO could approve a project with a 30% score and not be exposing the company to much of a risk of loss. On the other hand, to approve an important, expensive, and unique project with less than a 75% score on the FEL would not be prudent.
From a change management perspective, the organization's dollar-weighted average S-Felix score was tracked and published on a monthly basis. It was also connected in the early years to the organization's balanced scorecard and pay for performance. This showed the company's seriousness about doing adequate planning and signalled that project management was not to be “flavour of the month”. The specific target that was set for the organization was a 70% or better score on a dollar-weighted average of all projects over $70,000. Shown in Exhibit 1 below is the initial year of S-Felix scores as a cumulative weighted average. It is important to note that prior to implementing this measurement, benchmarking suggested that the average project S-Felix score was about 35. So the impact on behaviour was immediate.
Exhibit 1: History of first (partial) year of implementing S-Felix into Unocal Thailand, showing cumulative dollar-weighted average score and dollar amount of AFE approvals.
It is important to note that the first projects were approved by the Project Owner as being at or near the target rate of 70%. In practise, the FEL was not that complete, but the Project Management Department used the disconnect as a coaching opportunity rather than as a club to penalize the project. As the motivation of the group to help project people succeed rather than punish them for gaps became apparent, it reinforced the program. The PM Department consciously chose not to be “project cops”, but rather “project coaches”. Note how quickly the organization responded
In the first year of implementation, the display in Exhibit 1 was used to convey progress in meeting the organizational goals related to FEL. By the second year, we realized that we were perfectly positioned to spot which projects were going ahead in advance of AFE approval, and so we added that metric to the graph above.
Close-out
During close-out the actual cost vs the AFE'd cost was collected, and those data were used to understand whether there was any relationship between the amount of front end loading and the variations in the project cost. We also looked at whether there was any relationship between the project complexity index and the S-Felix scores. As it turned out, because of the scalability of the method, project managers were consistently doing 75-80% of what was suggested – limiting the correlations that could be drawn from the data but confirming the inference that the scalability in the system was working quite well.
Organizational Change Management Context
We have concluded that implementing project management in a PM-immature organization is 30% Project Management and 70% Organizational Change Management. Our model, shown in Exhibit 2, attempts to illustrate that successful organizational change is holistic in nature. Implementing Project Management starts on a foundation of expectations of change – hopefully at the most senior levels in the organization. Because everyone that touches projects in the organization has to change to some degree, those expectations expressed by senior managers and reinforced through metrics and scorecards form a basis on which the change effort is built. The S-Felix metrics were used as to reinforce expectations in the organization, and to reward that change when it occurred.
People must have the necessary knowledge to behave in a different way, and consequently training is a very important component of PM implementation – as is the expectation that people will attend and learn. Techniques such as how to track open issues, or what should be in a communication plan are the practical expression of the information contained in the training. We found that the project management coaches were a fundamental part of translating the training into practise by showing project managers the techniques when they needed them on the job. This real life on-the-job training to reinforce the concepts taught in the training, and to translate them into practise, helped the Project Managers and Owners tremendously.
Exhibit 2: Organizational change triangle
The next higher level in the triangle is Systems. Systems allow the organization to collectively understand the way they are to behave in the future. They do not typically work in isolation of the other elements, but are a necessary component. The project management system was developed using A Guide to the Project Management Body of Knowledge (PMBOK® Guide) as a guide. It consisted of a set of procedures as well as an annual audit procedure and tool that was adapted from some early published work on project management maturity (Fincher & Levine, 1997).
At the top of the triangle are the tools that are required to implement the rest of the triangle. The S-Felix is one of those tools. Just as putting project management software on peoples’ desks does not make them project managers, using the S-Felix approach by itself does not improve an organization's approach to project management. Using a scalable approach to project management in the context of the rest of the triangle however, will increase an organization's likelihood of having a readily accepted, efficient and effective approach to project management.
The final element on the triangle is “practise”. It is important for senior executives contemplating the implementation of project management to realize that these skills and behaviours are not acquired over night. They required consistent application of constructive pressure on the organization to change. The length of time will depend on both the size and complexity of the organization, as well as the cycle time of an average project. Project Managers learn by doing, and lengthy cycle times will extend the period of learning.
Acknowledgements
Unocal Thailand gave me the privilege of building my own team of project management coaches to assist in the development and implementation of the PM system. Without them, we would not have been successful. They are Mike Sypsomos, currently Director Project Management, Chevron Thailand; Mark Schaaf, Project Management Coach, Chevron, and Burt Campbell, whereabouts unknown.