What is EVM?
Earned Value Management helps project managers determine whether a project is going according to plan. It provides data and metrics to analyze precisely where a project is going well and where it is potentially struggling. Then further forecasts out what the project will look like if the current pace continues.
EVM can be a tremendously valuable tool, especially for larger projects, where there is a lot of complexity and financial risk. These projects are also typically many months or years in duration and therefore also have the potential of schedule risk. EVM provides the tools to de-risk both cost and schedule.
For these big projects, project managers need reliable analytics and good data to stay on top of what’s going on and where to focus their attention. With the sheer volume of data produced by big projects, it helps to have the might of a powerful software system to crunch all that data. The system can then guide them to quickly pinpoint the troubled areas and give solid indications as to how to take corrective action.
In order to give reliable analytics, however, EVM requires good data. Which is another way of saying that “Quality Input results in Quality Output”. So, to get the most out of EVM, projects have to be setup to feed good quality data into the system so that it can run the algorithms on that data and produce meaningful reports & charts. This typically means spending more upfront effort in setting up the project plan (i.e. work breakdown structure, cost codes, budget, etc.) so that it is wired for EVM. What that means is that every activity or task on the project needs to be setup to easily track, at minimum, the following three data points:
- Actual cost
Simple EVM Example
Before jumping into the technical details of those data points, let’s have a quick look at what EVM is actually trying to accomplish. To do that, I’ll give a very simple everyday example of a common task. In this example there will be 3 main data points, and with that we can use EVM to forecast what’s remaining.
Here’s the example: Let’s say that you have to drive 2 hours to get to your mom’s house, which is 100 miles away. After driving an hour, you should be halfway there (50 miles), however when you check, you discover you’ve only gone 40 miles. Traffic has been slower than normal because of heavy rain.
So, with that small example, we can use EVM to figure out how long it’s going to take to get to mom’s house (i.e. what’s remaining). To reframe the example in EVM language: you’ve budgeted 2 hours to go 100 miles. After burning 50% of the budget (1 hour), you’ve only completed 40% of the task (40 miles). So:
- Budget (BAC) = 2 hours
- Actual (AC) = 1 hour
- Progress (PC) = 40% complete
Because I’ve used really easy numbers in that example, you could do the math in your head to figure out how much longer it’s going to take to go see your mom, but let’s go through the EVM exercise to see how it’s done by the software. To do that, I need to calculate two more variables: Earned Value (EV) and Cost Performance Index (CPI), so that I can calculate the final number, which is Estimate to Complete (ETC). ETC will indicate how much longer it will take given how far you’ve gone after an hour. Here’s how you figure out ETC:
- Earned Value (EV) = Budget * PC = 2*.40 = 0.80.
This means that even though we’ve traveled an hour, we’ve only earned .80 of an hour
- Cost Performance Index (CPI) = EV/AC = .8/1 = .80.
This means that we’re performing at 80%.
- Estimate To Complete (ETC) = (BAC-EV)/CPI = (2-.80)/.80 = 1.5
According to EVM therefore, it’s going to take a further 1.5 hours to get to mom’s house. Which means a half-hour later than planned – which, in EVM speak is called “Variance”. In this case, a negative variance.
It’s making some assumptions of course, like it’s assuming you’re going to continue to average only 40 mi/hr for the rest of the journey. That’s one key thing to understand about EVM – it’s only using your past performance to forecast future outcomes. Project controls professionals typically know this and take steps to override these algorithm-driven forecasts to account for other factors that the system doesn’t know about. Effectively what EVM is saying is, “If you keep going at this pace, this is how the rest of your task will play out.” Which is a legitimate thing to know. As stated by the Project Management Institute (PMI), “EVM uses the fundamental principle that patterns and trends in the past can be good predictors of the future.”
The PMI also succinctly bullets the main questions to be answered that are the primary drivers of using EVM:
- “How efficiently are we using our time?
- Are we currently under or over our budget?
- How efficiently are we using our resources?
- What is the remaining work likely to cost?
- What is the entire project likely to cost?
- How much will we be under or over budget at the end?”
While that simple example is only meant to shed light on the idea of EVM, clearly a major construction project will have thousands of very complex tasks and deliverables. This is where the system comes into play. By taking all that data across the entire project, running the calculations, and summarizing it to the many levels of the work breakdown structure (WBS), the project controls team can gain key insights into the health of the entire project.
The Three Main Inputs for EVM Calculations
As discussed above, the three primary input values required for calculating all EVM metrics are:
- Budget. Or, budget at complete (BAC). This refers to the total cost budgeted for the activity. The budget can be broken down to core resources such as hours, materials, equipment, expenses, etc. However, it is typically expressed in a monetary cost amount.
- Actual Cost (AC). This refers to the total current expenditures recognized on the project at a point in time.
- Progress. Or, Percent Complete (PC). Progress is determined by evaluating each activity’s physical completions at a point in time. This can be measured in a variety of ways from actual production quantities to using rules of credit or other method.
These values are always input at the lowest level of the work breakdown structure and rolled up to the higher levels of the WBS.
Time-Phased Budget for Planned Value
On large projects it’s important to merge cost with time to visualize the project’s budget over a timeline. This shows the budgeted cash flow over time and allows project controls to both forecast project spend as well as isolate planned spend at any point in time. That ‘point-in-time’ figure is known as Planned Value (PV). Planned value is a metric that shows, on any certain date, what was supposed to happen according to the plan. This is critical as it enables project controls to compare what was planned against what actually occurred.
Looking at the partial S-Curve below (click to enlarge), you can see the budget line in the blue, with the actual cost line in yellow and earned value in brown. Each point on the budget curve represents planned value at that point in time. In this example, clearly the project’s activities are not keeping up with what was planned!
Real-time Actual Cost
Costs on a project can be incurred from various sources including, Labor, Equipment, Materials, Expenses, Fees, Services, etc. Costs are incurred daily as activities are carried out on the project – such as labor timesheets, materials received, equipment hours, subcontractor progress achieved and so on. Regardless of the type, costs are usually expressed as transactional entries in the system that occur on a particular date. Costs can originate from various locations and systems as well. For example, companies may choose to use 4castplus for all procurement activities and thus capture all vendor costs directly in the system; and use a different system for capturing labor expenses. Other system’s costs can be imported into 4castplus to ensure full cost capture in one place. In order to accurately represent cost compared to planned value and earned value, having full cost capture that’s up-to-date is critical for the metrics to be meaningful. Clearly it’s also key to have costs recognized on the project in real-time as this gives project controls the most current information to make timely and informed decisions.
Establishing percent complete is likely the most nuanced of the three inputs. In 4castplus this is accomplished by submitting progress measurements on a regular incremental basis. This enables the project controls team to establish rules around how each activity is going to be measured, and then as work occurs on that activity, the system will use the inputs to calculate its physical progress. Inputs could be, for example, labor hours, deliverables, subcontractor progress, weighted steps, etc. There are many methods for how to compute physical progress, including simply inputting it directly into the system.
Progress measurements are incremental. Which means that they work a bit like a period close in an accounting system. The intent is to establish an iterative reporting period (such as monthly) and submit progress measurements at the end of that period. Once submitted, the system will take a complete snapshot of all the data at that point in time. Project controls can then create a new progress measurement for the next period, and so on.
The progress measurement will establish percent complete for every activity and WBS level on the project. With this, the full suite of EVM metrics can be calculated, such as Earned Value, CPI, ETC, CV, and so forth.
More EVM Metrics
One nice thing about EVM is that, using the 3 inputs, the system can then generate a broad variety of analytics for helping project controls gain a clear understanding of where the project is at, and where it is in need of a bit of help. Of course, you don’t have to use all the metrics for EVM to be useful – if you only use CPI, CV and ETC, you’re mostly covered. Those three metrics give you the essentials of where each activity is at and where it’s potentially going. Once you’re a bit more seasoned, other metrics like TCPI, FAC, SPI, etc. can provide tremendous insight.