how much is enough?
George E. Heywood and Thomas J. Allen, PMP
The Project Controls functions, defined in the classical sense as estimating, planning and scheduling, cost control, risk analysis and related reporting functions, have been clearly demonstrated to provide cost benefits to a project. If embraced as a set of required functions in the project environment, the question frequently arises as to what resource level should be expended in order to receive those benefits.
We have found that there is a range of project control costs associated with increased probability of cost and schedule success. This range, measured by Project Control Costs as a percentage of Total Project Costs, is from 5 percent on smaller projects to less than 1 percent on multibillion-dollar projects. This band includes all project control costs on the project, whether they are incurred by owner, contractor, home office or field.
A properly conceived and implemented project controls strategy and plan will provide significant benefits to any project, regardless of project type, risk, contracting strategy, organizational strategy, or dollar value. While these benefits have been quantified as somewhere between 5 percent and 20 percent, certainly enough to welcome project controls functions as a “paying entity,” the question is certain to arise: How much cost did we incur to achieve those benefits? This is especially true in today's environment of heightened sensitivity to bottom line profits.
Thus, one needs to determine the minimum cost point at which project controls can provide both satisfactory information and the expected benefits to project managers for use in their control process. This would be defined as the point where project controls costs have been reduced to a level where overall project cost and schedule performance just begins to become erratic, and overruns previously not experienced start to become the norm. While this is subjective, any corporation with a history of projects, many of the major architect/engineer firms, and government agencies will be able to identify with this point if they have cut back on these functions to “economize.”
The “2 Percent Rule”: Fact or Blind Faith? For many years, a rule-of-thumb for project control costs as a percentage of total project cost has been 2 percent. This rule has often been blindly applied, regardless of project size or project implementation strategy. Worse yet, guidelines of any kind are often ignored. There has been more than nominal dissatisfaction with this hit-or-miss approach. Discussions with peer groups, academicians, and others across various industries indicate that project controls is an activity with some relatively fixed and some variable components. For example, developing a project estimate for a $10 million project will not take half the resources of an estimate for a similar $20 million project. Planning and scheduling time would also not be half. Certainly development and implementation of a project control system is not linear relative to project size. Resource burn associated with project control activities are not linear with overall project cost.
To determine if a band of acceptable cost could be developed, various projects of known cost and breakout of those costs were accumulated and discussions were held with peers across several industries. Results were compiled. There was some scatter within the data but definite trendlines could be seen. These results are shown in Figure 1. The upper graph, showing project costs to $1 billion, shows the overall envelope. The lower graph provides visibility to projects under $50 million, the range within which most projects in the United States fall. Trendlines were developed based on data points and a commercially available software package was used to develop lines of best fit.
Figure 1. Project Controls as a Percentage of Total Project Cost
The use of these graphs is left to the reader. Data points, while not as extensive as desired, do show trends supporting assumptions that were previously intuitive. Staying within the bandwidth and using proper project controls systems with trained staff should produce a level of project management information and awareness that will significantly improve the likelihood of project success.
Thomas J. Allen, PMP, has over 25 years of project-related experience. “Partially” retired, he is a part-time consultant and, during the winter, a full-time ski instructor. He is a Registered Professional Engineer in Oregon.
George E. Heywood is a senior consultant with Hawk Construction Consultants, Inc. in Anchorage, Alaska. He helped develop a Project Management and Control Manual and various scheduling models for Alyeska Pipeline Service.
PM Network • November 1996