Many years ago I took classes about project management. One of the classes was about a method of tracking project progress called Earned Value.
The intent of Earned Value is to set up budgets for tasks on a project. Progress on the tasks is tracked and is considered the earned value. The earned value is compared to the actual budget spent and the comparison is used to estimate the final cost and date of project completion. I've included an example of how progress might be tracked for different projects. The project occurs over ten units of time. Task completion and budget is tracked as a percentage over time. The three figures show the following:
• Figure 1, Budget Matches Progress - For this project the funds spent matched the progress exactly.
• Figure 2, Progress More than Costs - This project shows completion with less funds spent to complete the project
• Figure 3, Cost More than Progress - In this example, the project spends more to make progress and ends up over budget at completion.
In each case the expended funds and task completion both proceed in a linear fashion that is predictable. Early analysis of this type of progress it may even be possible to shift course and keep the project within budget while still successfully completing the tasks.
However, when using this technique in the real world, the predictions did not proceed as cleanly. And the budget predictions were much worse. The next figure shows a project where progress is uneven and sometimes goes backwards while the expended funds continues to grow.
I found it interesting but puzzling at how badly the project was earning value. The formulas from the start of the project were predicting large overruns and very late deliveries. I wondered why and I eventually realized that there are some assumptions about the data used when calculating earned value estimate to complete.
1. The initial budgets for tracking earned value are based on estimates for the final budgets. Earned value assumes that the estimates are good estimates for the final work and they have not been trimmed or cut before the start of tracking.
2. The formulas for predicting the final budget are based on the progress leveling off at the time the estimate to complete is calculated.
What I saw in the real world is the following:
1. An estimate had been provided for an engineering project that was developed in good faith by the engineers. However, the amount was more than the customer wanted to pay so the estimate was cut several times to a quarter of the original estimate. When earned value methods were used to track progress and predict the final budget and estimated date, they were off by large factors. When looking back, the projections were tracking how badly the initial budget cuts affected the progress on the project.
2. On a software development project, the project was dealing with new concepts and inexperienced developers. As the work progressed on the project, new things were continually discovered that had an impact on the final outcome. The estimates for completion were based on the level of work not getting any worse over the rest of the project. In reality, the development trend continued on a downward spiral and did not stop. In this case, it would have been better to extrapolate out the trend of failure as a slope instead of a plateau for the estimate of the final budget and due date.
I realized that accurate Earned Value tracking is based on at least two assumptions
1. The estimates used to create the budgets are as accurate as possible
2. The trend of progress for a project will level out and not get worse over time.
If these two assumptions aren't met than Earned Value tracks how bad the estimate was compared to the work performed on the project. An additional step could be added that if the trend continues downward the estimates for the project should be reviewed for accuracy and an updated estimate to complete generated. Otherwise what should be used as a tool gets used as a cudgel to punish the people doing the work on the project. Which seems like a poor use of a tracking tool.
In my work we used a method of tracking that is based on Earned Value. We set up budgets for different parts of the project and track progress each week. The data is also used to estimate how many hours will be required to complete the project. However, we keep the budgets flexible and continuously monitor the scope, making adjustments as we go so that we stay within schedule and budget for the project. The following is a list of items we implement to keep our projects within budget, schedule and maintain quality of the final product.
• Maintain infrastructure for efficient tracking of deliverables and progress - Our team has a lot of experience with our types of projects. We use this experience to build tools that improve our efficiency and help our work to be completed in a predictable amount of time.
• Buy-in from workers performing the tasks - When generating estimates we ask for input from the people who will be working on the project. The estimates are reviewed, not to cut them, but to make sure there is sufficient budget for the work while still staying competitive. When the project is started, the budgets for the project are reviewed with the team so they know what they have to work with. This open discussion builds trust that the team members input is important and necessary for defining the tasks for the project.
• Build trust with the team for early identification of issues - I've worked on too many projects where each team member worked to hide problems so they aren't forced into extra meetings or are disciplined for not making enough progress. On our teams we encourage people to identify issues early so they can be addressed as soon as possible. With this approach the solutions are usually less expensive and may be covered by other tasks that require less work.
• Avoid the Heroic or White Knight Management of a Project - For projects where issues are hidden until the last minute, a heroic effort is usually required by the team before the final delivery. This effort usually requires a lot of extra hours and time from many people to make the final deadline. This effort tends to be praised in companies and the individuals are recognized in award ceremonies for their valiant effort to fix issues and complete a project. Myself, I prefer to work on projects where issues are quietly identified and fixed. This method doesn't get a lot of recognition but it does lead to a higher quality of life outside of work.
• Retain flexibility in the budget and schedule, maintain the quality - Companies are focused on profit so if budget tracking shows an excess the amount is often marked for profit. Meanwhile, overruns are punished and team members are expected work extra unpaid hours to make up for it. On our projects we view budgets as flexible and extra hours in one budget are moved to another budget if it is running behind. Overall the project completes within budget and shows a profit by adjusting to circumstances as work is completed.
I have worked with these techniques for twenty years and our projects have had reduced stress in the area of budgets. These methods do not remove all stress and require discipline to implement. The techniques are also used on projects where the team has a lot of experience in estimating and performing on certain types of projects. If I were managing a software development team I would probably modify some items and keep a very close eye on the requirements and scope. However, these techniques can be helpful and reduce some of the issues that occur when managing projects while tracking progress.
Pictures by J.T. Harpster, prints of selected photos can be found at our Redbubble shop
Help support our work on Patreon, get access to short stories and news about Shell Creek Publishing.
- Log in to post comments