When asked about where their project stands against initial budget and schedule, most project managers will have a pretty good idea. They’d be able to tell you something like, “We’re running quite close to budget”, or “We’re almost half done”. However, without the tools and tracking to provide sufficient substance to those statements, gut-feel assertions like that are often dangerous guesses that can lead to cost overruns and delays. Earned Value Management provides the tools and techniques to tell a project manager where he or she really stands in their project. It can not only report on how much over/under budget or ahead/behind schedule a project is; it can also inform a PM as to how a project is trending, so as to better predict schedule and cost remaining to complete their project. This provides the project manager a solid grounding on current status along with a good estimate of projected final results & timing.
Armed with better analysis tools like this, the cost savings to projects – along with the improved communication and control – are clear incentives to look at adopting EVM in almost any project organization.
Earned value management is much simpler than it sounds. It is simple because it’s something that we all do in everyday life – we just don’t usually use an elaborate name for it like EVM. Put simply, EVM is the technical term used for the process of getting quick and accurate answers to the “Where Am I At” questions on any project (construction, IT, or other). And we’re all continuously trying to assess where we’re at in most everything we do: whether it’s managing a complex construction project, or simply planning a birthday party.
The important “Where Am I At” questions in the world of EVM are,
- Are things going to plan?
- If not, how far off-plan are we?
- Where exactly are the critical difficulties occurring? And why?
- And, what can we do about it?
Why Use Earned Value?
What EVM does for you, is it enables you to quickly zero-in on good answers to these types of questions in minutes rather than it taking hours, days or weeks to compile. And then use those answers to apply any changes or corrections necessary to keep your project on budget and on schedule. At the end of the day, if you run projects with even a modest degree of complexity, you know you’re going to need continuous answers to these types of questions whether you’re familiar with EVM or not; so by regularly capturing EVM metrics, you’re really simplifying your job and reducing your total effort through small bits of regular tracking. This tracking then enables you to deliver answers to critical questions in a single-click; answers which you’re confident are accurate and based on current information.
Besides providing answers, adopting EVM can reduce the chances and sizes of project cost overruns and vastly improve the likelihood of profitable projects. Far from being unnecessary overhead that makes project management even more complex than it already is, EVM is widely considered to be a key and proven tool that’s essential for running consistently successful projects.
A Measurement of Progress and Performance
EVM is really all about two things: measuring project progress, and measuring project performance. So, why is it important to measure performance and progress? The short answer to that: is that it empowers you with early identification of project issues that may need to be addressed. So okay, why is that important? Well, if you wait until the end of your project to find out that, whoops, you’re $20 million over budget; it makes for a much trickier situation than if you had anticipated that this might happen when the project was only part way through. When you anticipate the results early on, you give yourself – and your project team – the opportunity to apply corrective action that can avoid a cost and schedule overrun. It also allows you to communicate the potential overrun to the various project stakeholders, partners, owners and subcontractors so that you can avoid conflicts or disputes, and even ask for more money to address unexpected costs.
The PMI describes it this way,
“EVM has been called management with the lights on because it can help clearly and objectively illuminate where a project is and where it is going—as compared to where it was supposed to be and where it was supposed to be going. EVM uses the fundamental principle that patterns and trends in the past can be good predictors of the future.
“EVM provides organizations with the methodology needed to integrate the management of project scope, schedule, and cost. EVM can play a crucial role in answering management questions that are critical to the success of every project, such as:
- 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?”
(source: Practice Standard for Earned Value Management, www.pmi.org)
A Uniform Way to Measure Project Progress
One of the key elements of EVM is to provide a uniform unit of measure for determining project progress. So, what’s the big deal about a uniform unit of measure? Let’s say you’re a bricklayer and your job is to lay 1000 bricks, and so far, you’ve completed 900 of those bricks. You can reasonably assess that your job is about 90% complete. Measuring this type project is simple in that it has a built-in unit of measure to determine progress: i.e. number of bricks. Most projects however, are much richer with diversity and include an array of activities such as: labor hours, equipment hours, lengths of pipeline, cubic meters of dirt, gallons of fluid, processing units, cu-yds. of cement, and a great variety of other materials, expenses, subcontractors, etc. to be measured. To figure out how far along a project is therefore; all these moving parts and activities require a common unit of measure to determine progress of the project as a whole.
It’s much easier to gauge the progress of an individual task than to gauge progress of the whole project. When you have hundreds or thousands of tasks within that project, each on its own unique timeline with varying degrees of resources and expense types, it becomes a bigger challenge to mash all these together to evaluate how well the whole project – or even any project summary level – is progressing along. You need a way to calculate apples to apples.
To do this, the typical units of measure used in projects are work-hours and cost. Although work-hours can seem adequate for labor-intensive projects (and are quite commonly used), it can actually bring about misleading results. It’s often the case that the person planned to complete a work-item or task may not be the actual person who completed the task. The differences in salary and overhead may not be accounted for by just utilizing hours (or time) as the common unit of measure. Cost, on the other hand is a much more reliable measurement since it normally forms the basis for the budgets on all contracts regardless of the nature of the work. For the purposes of this discussion, I’ll be describing EVM using cost & budget as the focus of measurement. The same principles apply for time and schedule.
So when talking about how things are progressing along, this “uniform unit of measure” translates into comparing Plan vs. Actual. In other words, “What did I plan to spend by this point in the project, versus what have I actually spent?”
Let’s say, for some example project, that on this date:
Planned cost to date was: $2,300,000
Actual cost to date is: $2,100,000
That means that for this project, the plan was to have spent $2.3m by this date, but only $2.1m has been spent. Looking at those numbers, it appears as though the actual costs are quite a bit lower than the planned costs. This seems to be good news. But, in this example, should the project manager be happy or concerned? Well, it’s impossible to tell until you look at the planned cost of the completed work. In other words, without an assessment of how much work has been completed, simply looking at the planned & actual costs alone is insufficient to know where things are truly at. And it’s Earned Value that provides that missing information to gain a complete picture of a project’s progress and performance. So, in this example, let’s say,
Earned Value = $1,800,000
What the earned value describes is the value associated with the physical amount of work that has been completed. When we look at this new earned value information in the above example, the project manager should be very concerned since the completed work is moving along at a slower pace than the costs. In other words, cost is $2.1m but progress is only $1.8m – so they’re spending more than they’re doing. Earned Value is calculated by taking percent complete multiplied by the planned estimate (for a quick example: a $100 task that’s 50% complete would have an earned value of $50).
At any point in time in a project, there are three concurrent progress values that can be measured: the Planned Work, the Actual Work and the Value of the Actual Work. Putting these three things together provides a full appraisal of where a project is at; and this lays the groundwork for earned value management.
Consider an example where a project team was able to achieve a higher-than-expected rate of productivity and managed to complete 80% of a project well ahead of schedule. Everyone was quite pleased until it was discovered that they did this by consuming 110% of the allotted budget. And they still had another 20% to complete! Obviously this part was not well received and the project came in over budget; and because they didn’t use EVM practices to uncover this issue early, they didn’t find all this out until late in the project.
As mentioned, the three essential metrics to gather for EVM are: Plan, Actual and Earned Value. Here’s a brief explanation of each. I’ve included the official acronyms for your reference. These acronyms are worth knowing from an academic point of view, but don’t let them clutter your understanding.
Plan (PV or BCWS): The plan is the original or baseline estimate of schedule and cost of all the tasks, activities, work-items, etc. that are laid-out in your Work Breakdown Structure.
Actual (AC): The actuals are the actual costs that are tracked as the project progresses. Any labor, equipment, materials, commitments, accruals, expenses that are costs on a project should be tracked on a regular basis (daily or weekly). Also known as Actual Cost of Work Performed.
Earned Value (EV or BCWP): EV is the assessed value of your project according to the work completed. Also known as Budgeted Cost of Work Performed.
Tracking these three items enables the project manager to perform a continuous analysis of project performance and progress. Tracking these is really quite easy once you’re setup with a decent system that supports it all.
Measuring Project Performance
So, what does it mean to measure project performance? Let’s say for example a project team makes a plan to complete a $100,000 job in 2 weeks. After a week they’re only 20% complete, and they’ve exhausted 50% of the budget. As a result, they only have a week left – or, half the time – to finish 80% of the work on the remaining budget. Statistically speaking in this case, they’re under-performing according to the original plan.
Alright, so maybe this is expected in certain cases, but let’s imagine that it isn’t okay in this particular instance. With these three numbers, let’s do some earned value performance analysis:
Estimate (PV) = $100,000
Actual (AC) = $50,000
Earned Value (EV) = $100,000 * 20% = $20,000
Cost Performance Index (CPI) = EV/AC = 0.4 or 40%
The cost performance index (CPI) – also known as Operating Efficiency – is an indicator of how well you’re project team is performing to date. This is important because it enables you to envisage how well your project is going to complete in terms of cost and schedule if your team continues to perform at this rate. Operating at a 40% CPI is like using 3.2 out of 8 cylinders in your car, but using all the gas that 8 cylinders would. For this example then, continuing to perform at this rate would result in a significant cost overrun. So then, how much of an overrun? Let’s do some more calculations to figure that out. What every project manager, owner, or otherwise interested stakeholder wants to know is, Estimate At Complete. EAC is likely the most desired number out of any EVM exercise. It truly represents “The Bottom Line”.
Imagine this conversation:
Owner: “How are things going?”
Project Manager: “We’ve run into a few snags and we’re running a bit behind”
Owner: “How much behind? Are we over budget?”
Project Manager: “Uh, yes. We’re running into some budget overruns”
Owner: “So, how much? What’s the final cost looking like?”
There it is right there. “What’s the final cost.” That’s exactly what everyone wants to know: the bottom line. When you’re watching the local news, and they tell you about your local reconstruction project cost overruns, they typically tell you two numbers: Planned and EAC. That’s what makes news. What was planned, and what the anticipated final cost is now. The thing is, this information is only available if you track and perform EVM analysis.
For the above example:
Estimate At Complete = $250,000
I’ll show you how this is calculated a bit further down. But it’s telling me that this $100,000 project has now ballooned into a $250,000 project after it’s only about 1/5 of the way through! Don’t you think it’s preferable to know that now rather than waiting until the end?
Estimate At Complete
Here’s how EAC is calculated. If you’re not a math person and you’d rather just let some software application figure all this out for you, well I don’t blame you – you might want to skip these details. But here they are anyway.
First, we need to know Estimate to Complete (ETC). This describes how much is left to spend from this point on. Then we simply add ETC to the current actual cost to get the final cost. It looks like this:
ETC = (PV-EV)/CPI
EAC = AC + ETC
Plugging in the numbers from the example above:
ETC = (100000-20000)/0.4 = $200,000
EAC = 200000 + 50000 = $250,000
In the above example I used the current performance trend to calculate the end result. In other words, the project team has operated at 40% efficiency, so that calculation assumes they’re going to continue to perform at that rate for the remainder of the project. Is it fair to make that assumption? Well it turns out it is – and I’ll get back to that in a moment. I could’ve made the assumption that the poor performance is all done with and things are going to go swimmingly to plan from here on; which means using 100% efficiency. That would look like this (I’ve highlighted the key number in green):
ETC = (100000-20000)/<span style="color: #000000; background-color: #00ff00;">1.0</span> = $80,000
EAC = 80000 + 50000 = $180,000
So, from an optimist’s point of view this project could end up being only $80,000 over budget rather than $150,000 over budget. However, this is unlikely. Statistically speaking, most projects continue on at the same performance rate once the project hits about 15-20% complete. So you can generally make the assumption that unless you do something to change current performance, it’s probably going carry on like it is. In other words, using that 0.4 CPI above is the most likely measurement, and unless there’s some sort of intervention, it’s going to come in at a final cost of $250,000.
This is one of the big reasons that EVM works so well. You can make very good projections of where a project is going based on where it’s been. And as you continue to leverage those historical trends, you’ll be able to constantly provide accurate readings of the future.
According to a study done by the US Defense Acquisition University (www.dau.mil), once a project is 20% or more complete, the CPI will vary at most 10% from its current value for the remainder of the project. It’s for this reason that many project managers will hesitate from publishing forecasted budget numbers until the project is at least 20% complete (this would be considered an industry best practice).
When a project isn’t quite on budget, the answer most stakeholders and project managers want to know is, “By how much?” The word variance sounds a bit like a stats class I took in university, but Cost Variance (CV) in EVM, is the calculated value that tells you how much over or under budget a project or task is. When someone says “We’re 10% over budget”, that’s a variance of -10%. A negative variance indicates over budget, a positive variance indicates under budget. Schedule Variance (SV) is identical except that it refers to being ahead or behind schedule. Cost variance is most often expressed as a percent, however many project managers calculate it as an absolute number.
It’s calculated by subtracting the current cost (AC) from Earned Value (EV or EV). Here are the calculations:
CV = EV – AC
Or, as a percent:
CV = (EV – AC)/EV
Plugging in the numbers from the above example, variance looks like this:
CV = (20000 – 50000)/20000 = -150%
This indicates that this project is 150% over budget. Let’s hope you have better performance on your projects!
Change in Variance
An important metric to monitor is the change in variance over time. What you want to watch out for is any wild swings in project performance. When things change suddenly, there may be reason to question the numbers that are feeding the results. A change in performance is, of course, a perfectly valid thing and could be a result of anything like: adding resources, a backlog of project costs appearing all at once, an unexpected breakthrough, etc. Nevertheless, when you come across an abrupt change in variance, it’s worth investigating so as to understand why. Looking at Figure 1, the Cost Variance has swayed from 16% over budget to 3% under budget the following week, only to dive down to 41% over budget the week after that. It looks a bit suspicious.
What You Need to do to Implement EVM
Below I’ve laid out eight essential steps for establishing a plan to implement earned value management in your next project. Hopefully, you’re already executing on many of these steps, so it’ll be a matter of adding one or two extra pieces to get the full value of EVM.
Step 1: Define the work by establishing the hierarchical work breakdown structure (WBS) of the project, and dividing the project into manageable chunks of effort.
Step 2: Assign a value (cost estimate) to each activity or task in the WBS. This is the Planned Value (PV) of the project. Estimating at the bottom-up level is preferred so that task values can accumulate and roll-up to their parent levels.
Step 3: Define Earning Rules for each activity in the WBS. Earning rules are an objective way to determine the percentage of work completed on the activity. A full discussion of earning rules is beyond the scope of this paper, however it’s important to stress the significance of the upfront effort in defining complete and the various stages of completeness of a task.
Step 4: Schedule the activities. Assign start and end dates for all tasks so that you can anticipate when the schedule and costs for those tasks are going to occur. Without scheduling the dates, there’s no way of lining up the estimate against the actual and EV.
Step 5: Track the Actual Costs. When your project is underway, you’ll need to regularly enter costs as they occur on the project. This could be: labor, equipment, materials, commitments and expenses. Costs should be tracked daily or weekly.
Step 6: Update the Schedule with Percent Complete. Using the earning rules described above – along with any overrides you may have as a smart project manager who knows something that a software application doesn’t – enter percent complete for any task that has undergone some activity.
Step 7: Forecast Regularly. Creating a new project forecast every week or every two weeks is a key way to break up the project trend reporting into incremental chunks. Creating a forecast is a way to save a snapshot of all the project costs, changes, earned value, percent complete, etc. so that you can look back to see critical project trends.
Step 8: Report. Create charts and reports on the KPIs generated from your EVM calculations.
Percent Complete by a Human or a Computer?
Not everyone agrees on whether a computer should calculate percent complete on a project activity, or a human should provide the assessment of percent complete. A computer is going to perform calculations only on the numbers that it knows. So for example, if you’re doing excavation on a construction project you’ve excavated and moved 6,000 of the 10,000 cubic yards of dirt that you’re contracted to; you’re obviously 60% complete, right? Well, according to the software you would be. But what if that’s not the case? What if, for instance, you excavated the easy area first? And the last 40% is going to take 70% of the effort? Then you need to tell the software that you’re only really 30% complete.
On the other hand, computers are entirely objective, which can be a good thing since people can have highly subjective perceptions of percent complete. People tend to be overly optimistic and often leave out key considerations when entering a value for completeness. This is why sometimes that last 10% takes 90% of the time.
Which is correct then, a human or computer? There’s no real right answer to that, but here are four good practices to follow when going about determining percent complete:
- Establish earning rules for each task. As mentioned above, state the criteria for progress on every task as a part of the initial project planning. You’ll be able to use these criteria as milestones for objectively entering percent complete.
- Use multiple, objective sizing attributes as indicators or benchmarks. Every project workpackage will have one or more measurable units of progress and consumption. For example, X of Y materials have been consumed, or Y of Z hours, or W of X items completed.
- Work in a team. Working in a team environment where more than one person is responsible for subjective calls on a task’s percent complete can help reduce individual bias.
- Update percent complete regularly. When you commit to weekly or bi-weekly updating of progress information, it’s less likely that you’ll suffer from any sort of momentary lapse of reason.
- Be transparent and report your numbers. When you issue reports on a regular basis, you and your team will be nudged into a transparency bubble which will (in time) encourage more accurate conclusions.
In the example above, the project resources were operating at a very low performance rate (0.4 CPI), and if left alone, the total project would result in a serious cost overrun. While knowing that information is important, actually doing something about it is where the next level of purpose in EVM takes hold. Earned Value Management provides a ton of details and reporting to enable a project manager to drill down to pinpoint the areas that are causing the troubles in a project. What to do about the issues is entirely dependent on the project and the nature of the issue. But having access to that information so quickly and accurately is often where the bulk of the effort needed is in applying corrective measures. The point is to take action. Leverage that good information and do what’s necessary to get the CPI back up and over 1.0.
The End Result
While it’s true that EVM appears to be a luxury of big, well-funded projects, the value that it brings to any project that has a degree of complexity, is enormous. If you`re running projects that go beyond mentally calculating time, budget, cost and progress all together in your head to report on project status; then there’s no question you’ll benefit from utilizing EVM techniques. Knowing where you’re at – and sharing good answers to that with all project stakeholders – is surprisingly important. I say ‘surprisingly’ quite sarcastically, because the truth is, everybody wants to know what’s going on. Getting decent answers to project performance, anticipated end results, and how much over or under budget, are pretty common questions to demand answers to. And, as custodians of project costs and controls, we all have an obligation to deliver that information quickly and accurately.