To fully understand earned value we must understand how the values are developed. Let's start with the Work Breakdown Structure (WBS). By definition the WBS is "a deliverable-oriented grouping of project elements that organize and define the total scope of the project". This in its most basic use is a list of all the tasks or activities that must be accomplished to deliver the project. A properly developed WBS is dissected into parts that represent a unit of work. Through the years the infamous "80-hour" rule has been handed down from generation to generation of project managers. I have even heard project managers that keep the unit of work down to 40 hours. Why is this? When you try to assess how you're doing - looking at 40 or 80 hours is much easier to assess than two months of work. If you can make better assessments of activity progress you will increase the accuracy of your reporting.
How far do you break down a task? I always say, "to the level that makes sense". For customer reporting you would go to a different level of detail than you would for management, and yet again a different level of detail for the employee who will do the work. The objective should be to break the work into parts that can be measured and identified that they are completed, and have not exceed 80 hours. Using common sense is best when determining this. Remember the WBS is a tool to help manage your project, it is not a tool to micro manage.
The lowest level that you dissect the WBS to is the "work package". By definition the work package is "a deliverable at the lowest level of the WBS". These are the activities that duration and costs will be assigned to. Once again, remember not to exceed the
Now that a WBS has been developed we can estimate the time and cost for each activity. Let's assess what we have so far. We have a WBS that consists of the project, its deliverables, tasks, and activities.
What else do you have? You can look to contracts, statements of work, requirements, and other organizational representatives that may have been involved in the development of the project plans (or whatever you may call them at this point). You also have subject matter experts and those that will do the work. Please tell me they are included! They have valuable inputs and need to be factored into the estimating process. You also may have access to project records of similar projects. If you do not, start archiving project records. This will help you to develop best practices as you develop in your organizations project management maturity.
Time estimating is your next step in this process. We have a very rough estimate that the work package can be done in less than 80 hours. Possibly there is a need for a more detailed WBS for the technical teams that will work on specific parts of the project. As you analyze each work package or activity, consider the time it would take under normal conditions, optimal conditions, and worst-case conditions. Estimates should be realistic, and there are many tools that can be used to minimize that risk that a project might run late.
When looking at tasks in a project, a strategic planning session should occur to determine task sequence. Think of this as a game of chess. Analyze the best sequence for the project tasks. This is a powerful use for the post-it note. They can be placed and moved at will. Once all the tasks have been laid out they can be placed into a scheduling software tool of choice. Remember, the software is a tool to help you with reporting; it is not project management. This is no more than saying that using a word processor makes someone an author. Understanding the skills and tools to manage projects is what makes you a project manager.
While estimating task duration all resources should be considered. Then all direct and (don't forget) indirect cost can be computed. Remember the factoring of project duration when considering costs. This will assist you in determining a reasonable and acceptable budget that you can manage your projects to.
Now that you have a schedule and a budget, you can determine the rate of spending. This is a time-distributed budget that is "the cumulative cost curve". This spending curve is developed based on the planned schedule. The schedule indicates when the tasks occur and the tasks have a cost associated with them. Now is when the 80-hour rule will become important. With the tasks being dissected to less than 80 hours, the validity of measuring the task to determine project progress is enhanced.
Key Earned Value measures are:
Budget at Completion (BAC)
Earned Value (EV)
Planned Value (PV)
Actual Costs (AC)
The cumulative cost curve can be used to develop the EV values. The entire project value is the budget at completion (BAC). Specific points along the curve are called planned value (PV). When you decide the appropriate point to measure your project progress (this could be weekly, monthly, quarterly, or any divisible factor) you can look at your cumulative cost curve and determine what should be done to that point in time and how much it should cost.
When monitoring project progress the schedule can be utilized to determine what work has been done. If 80 percent of the scheduled work is completed (again aren't you glad tasks are broken down to less than 80 hours?) then the value for that work is 80 percent of the cost (PV) given for that portion of the work. Here is an example; through one month we had tasks totaling $100,000, this would be the PV. When we measured the work completed we determine that only 80 percent is done. This results in an EV of $80,000.
Next we can determine the money spent to complete the work accomplished. This is referred to as actual costs (AC). Based on the previous example we might assume that the work accomplished is $80,000, but after all check is cleared we might have spent $85,000. Thus our actual cost for the $80,000 worth of work was $85,000.
What does this all mean? You have a tool that provides a three dimensional view of the project. Typical project variance has been determined by looking at the difference between the planned and the actuals. Earned value lets you look at the planned values (PV), actual expenditures (AC), and the actual work completed (EV).
The previous example showed us that we have a schedule variance of negative $20,000. This is calculated by subtracting the PV of $100,000 from the EV of $80,000. This identifies that we did not do all of the work we said that we would ($100,000). We are behind the planned schedule. We can also determine a rate of efficiency by dividing EV by the PV. This results in a schedule performance index of .80.
Also, the previous example showed us that we have a cost variance of negative $5,000. This is calculated by subtracting the AC of $85,000 from the EV of $80,000. This identifies that the work accomplished did not cost what we estimated the work to cost. We have gone $5,000 over budget. We can also determine a rate of efficiency of spending by dividing EV by the AC. This results in a cost performance index of .94.
For effective project management, identifying the acceptable variance is a must. What if the acceptable variance is 10%? We can determine that our budget estimating is acceptable. But we also can see that our schedule variance is unacceptable. We had better data for the cost of work than we did for the time to do the work. This is what we need to manage to. We can now better control our projects.
We can also predict where our project cost will end up. There are several estimates at completion (EAC) formulas. These formulas can be used to determine the upper estimates and the lower estimates. By dividing the BAC by the CPI we can get a center estimate. If our BAC for this project is $1.5M, we can assume that with a cost performance index at .94 we will have an EAC of $1,595,744. If we subtract this value from the BAC we can determine the Variance at Completion (VAC) as -$95,744. Remember what I said about acceptable variance? If we have an acceptable variance of 10%, we would be within tolerance ranges.
What does this all mean? We have a tool, EV, which allows us to identify how much we should spend based on how much work we said we would do. But as all plans go, these are subject to change. We must determine acceptable ranges of variation when we first plan our project. How does an organization determine an acceptable variance range? First determine the level of difficulty for the project. Second determine the level of experience of those working on the project. Last determine how the organization assesses variance. I have seen organizations set variance level as; +/- 10%, +/- 25%, +/- 50%, +10/-5%, +25%, -10%, and +75/-25. The different ranges should be articulated in the beginning and should be based upon the level of confidence that you can assign to your estimating efforts.
What have we learned? Earned Value is a control and monitoring system used to assess project performance. You can measure schedule progress, task progress, and cost progress. EV can be used to predict project performance and determine where the project will end up. Corrective actions can be taken to possibly put the project back on track or to stay within acceptable tolerance. We have identified that the values we use in EV are based on the WBS, schedule, estimates, and cumulative costs. Connecting these aspects will make EV an easy and valuable tool. Go ahead give it a try!
Wayne Brantley , MS Ed, PMP, CRP, CPLP is the Senior Director of Professional Education for the Villanova University (www.VillanovaU.com). Wayne has taught and consulted project management, quality management, leadership, curriculum development, Internet course development, and return on investment around the world to Fortune 500 companies. He has over 26 years experience from the Air Force as a project manager for AF technology training and curriculum development programs. Wayne has developed numerous AF and corporate training programs, classroom, multimedia, and Internet based programs. A dynamic presenter and trainer, he has spoken at numerous conferences such as; the Project Management Institute (PMIÂ®), the International Society for Performance Improvement (ISPI), and the American Society of Training and Development (ASTD) annual conferences. He is a doctoral candidate with Nova Southeastern University specializing in Computer Information Technology. University.