Earned value management

Based on Wikipedia: Earned value management

The Question That Exposes Every Troubled Project

Imagine you're six months into a year-long project. You've spent exactly half your budget. Everything looks perfect on paper.

But here's the question that should keep project managers up at night: How much work have you actually completed?

If you've finished seventy-five percent of the work while spending only fifty percent of the budget, you're a hero. If you've completed just twenty-five percent? You're heading toward a spectacular failure that will cost twice what you planned.

The maddening thing is that traditional project tracking can't tell the difference between these two scenarios. Both show the same budget line. Both show the same timeline. The catastrophe hides in plain sight until it's too late to fix.

This is the problem that Earned Value Management, or EVM, was invented to solve. And its origin story begins in an unlikely place: the factory floors of early twentieth-century America.

From Stopwatches to Billion-Dollar Defense Programs

Frank and Lillian Gilbreth were an unusual couple. He was a bricklayer turned efficiency expert. She was one of the first women to earn a PhD in psychology. Together, they became obsessed with measuring work in ways that had never been attempted before.

Their key insight was something they called "earned time." Instead of just tracking how many hours workers spent on a task, they measured what those hours actually produced. A worker who completed a difficult task quickly had "earned" more time than one who took longer. This simple idea—that you can't understand performance without understanding output—would eventually revolutionize how governments manage their largest projects.

The concept percolated through industrial engineering for decades. Then, in the 1960s, the United States Department of Defense faced a crisis. Cost overruns on major weapons programs were spiraling out of control. Projects that were supposed to cost millions ended up costing billions. The Pentagon needed a way to see trouble coming before the money ran out.

Their first attempt was something called the Program Evaluation and Review Technique. It didn't work. Contractors found it impossibly burdensome, and everyone started creating their own variations, which defeated the purpose of having a standard system.

In 1967, the DoD tried again. They established thirty-five criteria that any cost control system had to meet. This framework, called the Cost/Schedule Control Systems Criteria (abbreviated C/SCSC), was the direct ancestor of modern earned value management.

The Three Lines That Tell the Truth

At its heart, EVM tracks three things. Understanding what each one measures—and how they relate to each other—is the key to understanding the entire methodology.

The first is Planned Value, often abbreviated PV. This represents how much work you planned to complete by a certain date, expressed in terms of budget. If your project has a total budget of one million dollars and you planned to be halfway done by June, your planned value in June is five hundred thousand dollars. Some people call this the Budgeted Cost of Work Scheduled, or BCWS, which is more descriptive but harder to remember.

The second is Actual Cost, or AC. This one is straightforward: it's the money you've actually spent. In the old terminology, this was called the Actual Cost of Work Performed, or ACWP.

The third—and this is where the magic happens—is Earned Value, or EV. This measures how much of the planned work you've actually completed, expressed in the same budget terms as planned value. If you've finished sixty percent of a million-dollar project, you've "earned" six hundred thousand dollars of value, regardless of how much you've actually spent or how long it took.

The older name for earned value was the Budgeted Cost of Work Performed, or BCWP. Notice how that phrase contains the word "budgeted" rather than "actual." That's crucial. Earned value isn't about what things actually cost—it's about what you planned them to cost when you assigned the budget.

When the Lines Diverge, Trouble Emerges

Plot these three values on a chart over time, and the relationships between them tell you everything about your project's health.

When the earned value line rises faster than the planned value line, you're ahead of schedule. You've completed more work than you planned to by this point. When earned value falls behind planned value, you're behind schedule.

When the earned value line sits above the actual cost line, you're under budget. You've gotten more work done than the money you've spent would suggest. When earned value drops below actual cost, you're over budget—the work is costing more than you planned.

The beauty of this system is that it captures both problems at once. Traditional tracking might show you spending money at exactly the right rate, creating the illusion of a healthy project. But if earned value is lagging, you know that money isn't producing results. You're burning through your budget without making proportional progress.

There's a subtlety here that trips up many newcomers. Being "ahead of schedule" in EVM terms doesn't necessarily mean you'll finish early. It means you've accomplished more work than planned by the current date. Whether that translates to an early completion depends on what work remains and whether the favorable trend continues.

The Formulas That Forecast the Future

Once you have the three basic measurements, you can derive indicators that predict where your project is heading.

Schedule Variance equals earned value minus planned value. A positive number means you're ahead; negative means you're behind. Cost Variance equals earned value minus actual cost. Positive means under budget; negative means over budget.

But raw variances only tell you about the past. The power of EVM lies in its ability to forecast the future.

The Cost Performance Index, or CPI, divides earned value by actual cost. A CPI of one point zero means you're getting exactly the value you're paying for. Below one, you're overspending. Above one, you're getting a bargain. Research has shown that after a project is about twenty percent complete, the CPI tends to be remarkably stable. If you're running at a CPI of zero point eight, history suggests you'll probably finish the project at about zero point eight efficiency—twenty percent over budget.

The Schedule Performance Index, or SPI, divides earned value by planned value. Below one means behind schedule; above one means ahead.

These indices enable a calculation called the Estimate at Completion, or EAC. If your original budget was one million dollars and your CPI is zero point eight, a simple estimate would be one million divided by zero point eight, or one point two five million dollars. You're on track to exceed your budget by a quarter million.

More sophisticated formulas incorporate both cost and schedule performance, adjust for the nature of the remaining work, and account for management's expectations about future efficiency. But even the simple calculations provide early warning that traditional methods miss entirely.

The Navy's Twelve Billion Dollar Wake-Up Call

For two decades after its creation, earned value management lived in an analytical ghetto. Project managers saw it as a financial reporting requirement, something to hand off to specialists and forget about. The methodology had champions, but it also had critics who viewed it as bureaucratic overhead that didn't reflect reality on the ground.

Then came the A-12 Avenger II.

The A-12 was supposed to be the Navy's next-generation stealth attack aircraft, a sleek flying wing designed to operate from aircraft carriers. McDonnell Douglas and General Dynamics won the contract in 1988. The program was highly classified, which made oversight difficult.

By 1990, earned value metrics were telling a frightening story. The program was massively behind schedule and over budget. The contractors disputed the numbers, arguing that the classified nature of the work made traditional measurements unreliable. Navy program managers weren't sure whom to believe.

Secretary of Defense Dick Cheney decided to believe the earned value data. In January 1991, he canceled the A-12 program entirely—at the time, the largest contract termination in Department of Defense history. The decision triggered decades of litigation that wasn't fully resolved until 2014.

The cancellation sent shockwaves through the defense industry. Earned value management wasn't just a reporting exercise anymore. The Secretary of Defense had used it to kill a flagship program. Suddenly, everyone was paying attention.

Scaling Down: EVM for the Rest of Us

The earned value principles that govern multi-billion-dollar defense programs work just as well for a small team's quarterly project. The mathematics don't care about the number of zeros in your budget.

A lightweight implementation starts with defining the work. This usually takes the form of a work breakdown structure, which is exactly what it sounds like: a hierarchical decomposition of everything that needs to be done. Large projects might have thousands of elements in their work breakdown structures. A small project might just have a list of twenty tasks on a whiteboard.

Each piece of work gets assigned a planned value. For simple projects, this could just be the budgeted hours or the budgeted cost. The key is that every task has a measurable value before work begins.

The trickier part is establishing what EVM practitioners call "earning rules"—the criteria that determine when work counts as complete. For some tasks, this is obvious: you've installed the database, or you haven't. For others, especially long-running tasks, you need intermediate milestones or percentage-complete estimates.

A common approach for tasks in progress is the fifty-fifty rule. When you start a task, you claim fifty percent of its planned value. When you complete it, you claim the other fifty percent. This prevents both sandbagging (claiming no credit until everything is perfect) and over-optimism (claiming credit for work barely begun).

Other approaches include the twenty-eighty rule, which is more conservative, and the zero-one hundred rule, which gives no credit until a task is completely finished. The right choice depends on the nature of the work and how granular your task breakdown is.

Why Contractors Sometimes Hate It

Earned value management has a reputation problem in some industries. Critics argue that it creates perverse incentives, encourages gaming, and generates mountains of paperwork without improving outcomes.

Some of these criticisms have merit. When EVM is imposed as a compliance requirement without genuine buy-in, project teams often treat it as a box-checking exercise. They assign planned values that are easy to earn rather than values that reflect real work. They define earning rules that let them claim credit on schedule regardless of actual progress. The numbers look good while the project deteriorates.

There's also an inherent tension between earned value management and agile development methodologies. EVM assumes you can define the scope of work upfront and assign values to each piece. Agile embraces changing requirements and emergent scope. Adapting EVM to agile contexts is possible, but it requires thoughtful modification of traditional approaches.

And EVM can create adversarial dynamics between customers and contractors. If a contractor's payment depends on earned value metrics, disputes about what work is "really" complete become disputes about money. The A-12 litigation lasted more than two decades, with much of it focused on whether the government's earned value assessments were fair.

Yet when implemented thoughtfully, EVM provides something invaluable: an early warning system that surfaces problems while there's still time to fix them. The methodology has survived sixty years of reform efforts precisely because no one has invented a better way to answer the fundamental question: Are we getting what we're paying for?

The Defense Authorization Connection

Every year, the United States Congress passes a National Defense Authorization Act, which sets policies for the Department of Defense. These bills often include provisions about acquisition reform—changes to how the government buys weapons systems, technology, and services.

Earned value management requirements feature prominently in this landscape. Contractors on major defense programs must implement EVM systems that meet the criteria established in ANSI/EIA Standard 748. Government program managers must report EVM metrics to oversight bodies. When programs show unfavorable trends, those metrics trigger reviews, interventions, and sometimes cancellations.

Recent reform efforts have focused on making acquisition faster and more flexible without losing the visibility that EVM provides. This is a genuine tension. The detailed planning and measurement that EVM requires takes time. For rapidly evolving technologies, that time can mean fielding obsolete systems.

But the core insight of earned value management remains as relevant as ever. Whether you're building a stealth fighter or a smartphone app, you can't understand your project's health by looking at spending alone. You have to measure output. You have to know not just what you've spent, but what you've earned.

From Government Mandate to Industry Standard

Something interesting happened in the 1990s. The same era that saw government deregulation in many areas saw earned value management escape its government contracting cage.

In 1996, ownership of the EVM criteria transferred from the Department of Defense to the private sector through the adoption of ANSI/EIA Standard 748. The criteria were slightly simplified, from thirty-five to thirty-two, but more importantly, they became a commercial standard that any organization could choose to adopt.

The Project Management Institute incorporated EVM into its Project Management Body of Knowledge, first in 1987 and with increasing depth in subsequent editions. The construction industry became an early commercial adopter, attracted by EVM's ability to track progress on complex, multi-year projects with many interdependent tasks.

Australia codified earned value management in national standards. The European Union began requiring EVM on certain projects. What started as a Pentagon bureaucracy had become a global methodology.

The Sarbanes-Oxley Act of 2002, passed in response to corporate accounting scandals, gave EVM another boost. The law required publicly traded companies to maintain effective internal controls over financial reporting. For companies with significant project-based revenue, earned value management provided exactly the kind of objective progress measurement that auditors wanted to see.

The Limits of Measurement

Earned value management can tell you whether you're on budget and on schedule. It cannot tell you whether you're building the right thing.

A project can have perfect EVM metrics—on budget, ahead of schedule, all the curves exactly where they should be—and still fail completely because the requirements were wrong, or the market shifted, or the technology became obsolete. EVM measures execution, not strategy.

The methodology also struggles with highly uncertain work. Research and development projects, where breakthroughs can't be scheduled, don't fit neatly into the EVM framework. Neither do creative projects where the definition of "done" evolves as the work progresses.

And there's a deeper philosophical issue. Earned value management assumes that the value of work can be meaningfully quantified in advance. But some of the most valuable things humans create—scientific discoveries, works of art, revolutionary products—resist this kind of pre-specification. You can't assign planned value to a breakthrough you can't yet imagine.

Still, for the vast majority of projects—the ones where we know what success looks like and the challenge is achieving it efficiently—earned value management remains one of the most powerful tools available. It doesn't replace judgment, but it informs judgment with data. It doesn't guarantee success, but it surfaces failure early enough to do something about it.

That's worth quite a lot, whether you're managing a billion-dollar weapons program or trying to launch a new feature before the end of the quarter.