In the early years of project management, the United States government dis- covered that the project managers in the contractors’ firms were functioning more as project monitors than as project managers. Monitors would simply Project Management Metrics, KPIs, and Dashboards:
A Guide to Measuring and Monitoring Project Performance by Harold Kerzner Copyright © 2011 John Wiley & Sons, Inc.
record information and then pass it along to higher levels of management for consideration. Project managers were reluctant to take any action as a result of negative information because they were either lacking sufficient authority to implement change or did not know what actions to take. The result was often customer micromanagement of projects. Unfortunately, as projects became larger and more complex, government micromanagement became exceedingly difficult.
Determining the true project status became difficult, as shown humor- ously in Figure 3-1. On some larger projects, it became questionable as to who was controlling costs. This is shown in Figure 3-2.
The solution was to get the contractors to learn and implement project management rather than project monitoring. Project managers were now expected to perform the following:
◾ Establish boundaries, baselines and targets for performance
◾ Measure the performance
◾ Determine the variances from the baselines or targets
◾ Develop contingency plans to reduce or eliminate unfavorable variances
◾ Obtain approval of the contingency plans
◾ Implement the contingency plans
Figure 3-1 Determining Project Status
Time? Cost?
Performance?
?????
69
3.1 PROJECT MANAGEMENT METRICS: THE EARLY YEARS
◾ Measure the new variances
◾ Repeat the process when necessary
For this to work, metrics would be needed. The government then created the Cost/Schedule Control Systems and later the Earned Value Measurement System (EVMS).1 The metrics that were established as part of these systems allowed us to determine project status, at least what we thought was the status. As an example:
◾ The project
◾ A budget of $1.2 million.
◾ Time duration is 12 months.
◾ Production requirement is 10 deliverables.
◾ Time line
◾ Elapsed time is six months.
◾ Money spent to date is $700,000.
◾ Deliverables produced: 4 complete and 2 partial.
With the EVMS metrics, we were able to reasonably determine status.
The metrics helped us determine both present and an often questionable prediction of the future. The metrics provided an early warning system that Figure 3-2 Who Controls Costs?
Do You Control Costs? OR Do Costs Control You?
1. In 1967, DOD Instruction 7000.2 identified 35 Cost/Schedule Control System Criteria (C/SCSC). In 1997, DOD Regulation 5000.2-R identified 32 Earned Value Measurement System (EVMS) criteria.
allowed the project managers sufficient time to make course corrections in small increments. The metrics emphasized prevention over cures by iden- tifying and resolving problems early. Metrics can function as risk triggers such that the impact of downstream risks can be minimized.
The benefits of using these metrics now became abundantly clear:
◾ Accurate displaying of project status
◾ Early and accurate identification of trends
◾ Early and accurate identification of problems
◾ Reasonable determination of the project’s health
◾ A source of critical information for controlling projects
◾ Basis for course corrections
The prolonged use of these metrics led us to the identification of sev- eral best practices that had to take place:
◾ Thorough planning of the work to be performed to complete the project
◾ Good estimating of time, labor, and costs
◾ Clear communication of the scope of the required tasks
◾ Disciplined budgeting and authorization of expenditures
◾ Timely accounting of physical progress and cost expenditures
◾ Frequent, periodic comparison of actual progress and expenditures to schedules and budgets, both at the time of comparison and at project completion
◾ Periodic reestimation of time and cost to complete the remaining work For more than 40 years, these metrics have been treated as the Gospel.
There are limitations to the use of the metrics in EVMS, however:
◾ Time and cost are basically the only two metrics. Most of the other met- rics being reported are derivatives of time and cost.
◾ Measurements of time and cost can be inaccurate, thus leading to faulty status reports.
◾ Quality and the value of the project cannot be calculated using time and cost metrics alone.
◾ Completing a project within time and cost does not imply that the proj- ect is a success.
◾ Time and cost information can be fudged.
◾ Unfavorable metrics do not necessarily mean that we are in trouble.
◾ Unfavorable metrics do not provide us with information for corrective action.
◾ Customers and stakeholders do not always understand the meaning of these metrics.
◾ Others.
71
3.3 UNDERSTANDING METRICS
The conclusion is clear. In today’s project management environment where projects are becoming more complex, the metrics of EVMS, by themselves, may not be sufficient for managing projects. EVMS does not address some fundamental issues that can lead to project failure such as unrealistic planning, poor governance, low quality of the resources, and poor estimates. We are not claiming that EVMS does not work. Rather, that additional metrics will be needed. We need to establish metrics that cover the big picture, namely business value to be delivered, quality of the results, effort, productivity, and team performance to name just a few. In another example, in IT projects today there is a push for the establishment of code- related metrics. Also a grouping of metrics rather than a single metric may be needed to accurately determine performance.