The Fuzziness of Earned Value Management in Practice
I have never had the opportunity to manage a project where I needed to manage the budget. As a public librarian working in adult programming and outreach, I did manage our annual programming budget, but that's a different kettle of fish. The primary idea there was to pace ourselves with booking author events and other programs and, at the end of the fiscal year, procure needed equipment and other relevant items to prevent ending with a surplus.
When I went through formal education to learn project management principles, we learned about Earned Value Management (EVM) and how it helps to determine project health by evaluating the variances and performance of the project triangle: scope, schedule, and cost. Each side balances against the other, so that whenever one area is adjusted, the other two are affected. It's a useful visual; you can't add items to a project scope (extending the scope line of the triangle) and keep the triangle together without affecting either the timeline or the budget (or both).
We learned many equations for tracking variance and performance, but we were told about two measurements that, perhaps in the opinion of the instructor, were the most accurate indicators of project health, and those are Cost Performance Index (CPI) and To-Complete Performance Index (TCPI). These are the standard calculations for both:
CPI = (Budget at Completion • % Complete†) ÷ Actual Cost or
CPI = Earned Value ÷ AC
TCPI = (BAC - EV) ÷ [(BAC ÷ CPI) - AC] or
TCPI = (BAC - EV) ÷ (Estimate at Completion - AC)
†% Complete = EV ÷ BAC
CPI tells you where the budget is. A value greater than 1.0 means your project is under budget, and a value less than 1.0 means your project is over budget. When your project is over budget, the difference between CPI and TCPI can tell you what it will take to recover. If your project is over budget and your TCPI comes back greater than 1.0, either your budget needs to increase or your scope needs to decrease to get the project back on track, but I digress.
Do you notice something crucial about the above equations? You need to have a budget to calculate them, and as I have already confessed, I have never had a project budget to manage. OK, so there was this one time where I was working on a mobile app integration project and, in order to make the connection, we had to pay the vendor for access to their API, but even then, I didn't manage anything more than the justification and request for those funds.
Here's the thing, though: I don't think it's that uncommon in practice to manage projects and deal with every aspect of project management except for cost management and procurement management on a regular basis. But, it really depends on the industry you work in, the type of organizational structure your company has, the type of project you're managing, and a number of other variables.
Still, you have to find a way of tracking and reporting on project health, even if you don't have a budget, and I have tried a number of ways to show that effectively without a budget. Should I focus on the other edges of the triangle? Even Schedule Variance (SV) and Schedule Performance Index (SPI) require a budget to calculate, but I think Level of Effort (LoE) is a good place to start.
Even if you don't have a budget, you have a schedule that is based on the LoE to complete every task in every phase of work. If you want to take it one step further, you can consider the resources available and/or allocated to each task and estimate the cost of each resource. In my experience, though, that kind of reverse engineering only adds a layer of complexity that doesn't affect the final outcome when you don't manage the budget for your project.
| Task | LoE | % of Effort Complete | Hours Complete | LoE Variance |
|---|---|---|---|---|
| Task 1 | 40 hours | 40% | 16 hours | -20% |
| Task 2 | 60 hours | 40% | 24 hours | 0% |
| Task 3 | 56 hours | 54% | 28 hours | +7% |
| TOTAL | 156 hours | 44% | 68 hours | -2% |
In the table above, you can see three tasks that all started at the same time and are running concurrently. You can probably guess from the detail that we're 24 hours into our scheduled time for each task; Task 1 is one day behind, Task 2 is on time, and Task 3 is about half a day ahead of schedule. Overall, this group of tasks is 4 hours (or around 2%) behind schedule. Are we at risk of a schedule overrun?
It really depends on a few factors. Is Task 1—the only task behind schedule—on the critical path? No. Could you easily recover by temporarily reallocating resources from Task 3 to Task 1? Probably. What is the LoE Variance for the entire project, and how much time is left on the project timeline? If the project is on time or ahead of schedule, and/or you haven't exhausted the schedule contingency, it may recover without your intervention.
These are important questions, but there is one major question that, in this approach, determines how you report on the overall project status: How risk tolerant are the project sponsor and key stakeholders? When you examine this question, this approach becomes even more nuanced, but this question is critical for managing project communications, risks, and stakeholders.
Things that determine risk tolerance of your project sponsor and key stakeholders could include the organizational impact of the deliverables, how high profile the project is, whether external stakeholders know about it and have expectations for delivery, and many other factors. If you are unsure about the level of tolerance or you are presenting to a mix of people, I always think it's better to err on the side of caution and plan to share everything you know without prejudice and how you plan to address it.
Answer their questions, listen to their comments, adjust your approach the next time you provide a report, and repeat. It's an art that gets less fuzzy over time.