Ask a project manager how the project is going and you’ll usually get a feeling, not a fact. “We’re roughly on track.” “Should be fine by next month.” “The team’s working hard.”
None of that tells you anything useful.
Earned Value Management fixes that problem. It takes three plain numbers you already collect and turns them into a straight answer: are you ahead or behind, and are you over or under budget? No opinions. Just arithmetic.
This guide walks through the whole thing — the three inputs, the formulas that matter, a full worked example, and the traps that catch people out.
What Earned Value Management actually is
EVM is a way of measuring project progress by comparing three things at the same time: what you planned to spend, what you’ve actually earned by finishing work, and what you’ve really paid out.
That third piece is where most tracking falls apart. Plenty of teams compare budget against spend and stop there. But spending money isn’t progress. You can burn through half your budget and have almost nothing to show for it. EVM catches that gap because it looks at completed work, not effort or cash.
The method came out of US government contracting decades ago, where sponsors wanted a hard number instead of a status meeting. It’s now standard in construction, defence, engineering, and any project where the money is big enough that someone will ask hard questions.
The three numbers everything is built on
Every EVM metric comes from these three. Get them right and the rest is easy.
Planned Value (PV)
The budgeted cost of the work you said you’d finish by today. It comes straight off your schedule and your cost baseline. If your plan says you should be 40% done and the total budget is $200,000, then PV is $80,000.
PV is your yardstick. Nothing else means much without it.
Earned Value (EV)
The budgeted cost of the work you have actually finished. Not the cash you’ve spent — the value of completed deliverables, priced at the original budget.
This is the number people fumble. EV uses budgeted rates, not actual rates. If a task was budgeted at $10,000 and it’s done, you’ve earned $10,000 of value — even if it cost you $14,000 to get there. That overspend shows up elsewhere.
Actual Cost (AC)
What you’ve genuinely paid: labour, materials, subcontractors, licences, everything charged to the project so far. Your finance system or a simple budget tracker should give you this figure.
The one rule that matters here: AC and EV must cover the exact same time period and the exact same scope. Mismatched cut-off dates are the single most common reason EVM reports come out wrong.
Before you can measure anything
EVM only works if the groundwork is there. Skip this and your numbers will be fiction.
You need a work breakdown structure. Work has to be broken into pieces small enough to call finished or not finished. Each of those pieces is a work package, and it needs a budget and a defined deliverable. “Design phase” is too vague. “Approved electrical schematics for floors 1–3” is measurable.
You need approved baselines. Your scope baseline locks down what’s included. Your cost baseline locks down the money and when it gets spent. Measuring against a plan that keeps quietly shifting tells you nothing.
You need control accounts. These are the points in the WBS where scope, budget, and schedule meet and where someone owns the result. A control account plan assigns that ownership. Without it, “the project is over budget” has no address — you can’t tell who needs to fix what.
You need credible estimates. Your Budget at Completion is the sum of your cost estimates across all work packages. If those estimates were guesses, EVM will just tell you precisely how wrong the guesses were.
The formulas, and what each one is really telling you
| Metric | Formula | Read it like this |
|---|---|---|
| Cost Variance (CV) | EV − AC | Negative = you’ve overspent |
| Schedule Variance (SV) | EV − PV | Negative = you’re behind |
| Cost Performance Index (CPI) | EV ÷ AC | Under 1.0 = poor cost efficiency |
| Schedule Performance Index (SPI) | EV ÷ PV | Under 1.0 = slower than planned |
| Estimate at Completion (EAC) | BAC ÷ CPI | Realistic final cost |
| Estimate to Complete (ETC) | EAC − AC | Money still needed |
| Variance at Completion (VAC) | BAC − EAC | Expected final overrun |
| To-Complete Performance Index (TCPI) | (BAC − EV) ÷ (BAC − AC) | Efficiency required from here on |
Two habits make these stick:
Variances are always earned value minus something else. Negative is always bad.
Indexes are always earned value divided by something else. Below 1.0 is always bad.
The indexes are usually more useful than the variances. A $15,000 overrun sounds alarming on a small job and trivial on a large one. A CPI of 0.82 means the same thing either way: you’re getting 82 cents of value for every dollar you spend.
A worked example
Let’s use an office fit-out.
The setup: Budget at Completion is $200,000 over ten months. It’s the end of month four. The schedule says 40% of the work should be finished by now.
The three inputs:
- PV = 40% × $200,000 = $80,000
- The team has actually finished 35% of the work, so EV = 35% × $200,000 = $70,000
- Finance shows $85,000 spent to date, so AC = $85,000
The health check:
- CV = $70,000 − $85,000 = −$15,000 (over budget)
- SV = $70,000 − $80,000 = −$10,000 (behind schedule)
- CPI = $70,000 ÷ $85,000 = 0.82
- SPI = $70,000 ÷ $80,000 = 0.88
So the project is behind on both fronts. Every dollar spent is returning about 82 cents of value.
The forecast:
- EAC = $200,000 ÷ 0.82 = $243,900
- ETC = $243,900 − $85,000 = $158,900
- VAC = $200,000 − $243,900 = −$43,900
- TCPI = ($200,000 − $70,000) ÷ ($200,000 − $85,000) = 1.13
That last figure is the one to bring to the sponsor. It says: to finish inside the original budget, the team has to run at 113% efficiency for the rest of the project, having managed 82% so far. That’s not a stretch goal. That’s a fantasy, and now you can prove it in month four instead of discovering it in month nine.
What to do once the numbers look bad
Numbers aren’t the point. Decisions are.
Find where it’s actually leaking. A project-level CPI of 0.82 is an average and averages hide things. Break it down by control account. You’ll often find two accounts dragging everything else down while the rest are fine.
Check whether risk caused it. Overruns usually trace back to something you either identified and under-planned for, or missed completely. Comparing the variance against your risk register — see our guide to risk management strategies — tells you whether this is a one-off or the start of a pattern.
Don’t quietly rebaseline. If scope grew, the baseline should change, but only through a proper change management process with approval attached. Adjusting the baseline to make the CPI look better is how projects fail silently.
Report it plainly. Sponsors don’t need eight indexes. They need to know the likely final cost, the likely final date, and what you’re asking for. A structured work performance report keeps the message consistent from one month to the next, which matters more than you’d think for building trust.
Three traps worth knowing about
SPI is a weak schedule measure near the end. As a project finishes, EV climbs to meet PV, so SPI drifts back toward 1.0 no matter how late you are. Finish three months behind and SPI still lands on 1.00 at completion. Always read it next to your critical path — the critical path tells you whether the delay actually pushes the end date, which SPI can’t.
Percent complete gets gamed. If teams self-report progress, work sits at 90% for weeks. Use fixed rules instead — 0/100 (nothing until it’s done), 50/50 (half at start, half at completion), or milestone-based credit. Pick the rule during planning and write it down.
EVM doesn’t measure quality. A project can post a CPI of 1.1 and be shipping work that fails inspection. Rework arrives later and destroys the numbers. Keep your quality management checks running alongside, because EVM will tell you that you’re efficient right up until it tells you you’re not.
Setting it up without drowning in spreadsheets
You don’t need enterprise software. A single spreadsheet with columns for period, PV, EV, AC and the calculated indexes will cover most projects. Plot the three curves on one chart and the story becomes obvious at a glance: if the EV line sits below both PV and AC, you’re behind and overspending, and everyone in the room can see it.
If you’d rather not build it from scratch, this walkthrough shows how to create an EVM dashboard in Excel in about five minutes, with the formulas already wired up.
Set a fixed reporting rhythm — weekly or monthly — and stick to it. EVM data collected occasionally is worse than none, because a trend from two data points is just a line between two dots.
Common questions
Does EVM work on Agile projects? Yes, with adjustments. Story points map to earned value once you assign a cost per point, and completed stories become your EV. Many Agile teams find velocity and burndown charts more natural for sprint-level tracking, and use EVM only at the programme level where finance needs a cost view.
What’s the difference between EVM and EVA? Earned Value Analysis is the calculation — running the numbers at a point in time. Earned Value Management is the whole system: the baselines, the control accounts, the reporting cadence, and the decisions that follow. EVA is a step inside EVM.
Is it worth it on small projects? Often not in full. On a six-week, two-person job, the overhead outweighs the benefit. A useful threshold: if the project is long enough that a problem could hide for a month before anyone notices, EVM earns its keep. If you’d spot trouble in a day, skip it.
Which is more urgent, a bad CPI or a bad SPI? Usually CPI. Cost efficiency rarely recovers on its own — if you’ve been running at 0.85 for three months, that’s your true rate, not bad luck. Schedule slippage can sometimes be pulled back by resequencing work. Cost overruns almost never reverse.
The short version
EVM works because it refuses to accept effort as evidence. It only counts finished work, prices it at the original budget, and compares that against both the plan and the real spend.
Start small. Get your work packages defined, get your baseline approved, pick a progress rule, and calculate PV, EV and AC for one reporting period. Once you’ve seen a CPI of 0.82 on your own project four months in, you’ll never go back to “we’re roughly on track.”
Internal links used (10)
| # | Anchor text | URL |
|---|---|---|
| 1 | cost baseline | https://pmstudyguide.com/cost-baseline/ |
| 2 | budget tracker | https://pmstudyguide.com/budget-tracker/ |
| 3 | work package | https://pmstudyguide.com/work-package/ |
| 4 | scope baseline | https://pmstudyguide.com/scope-baseline/ |
| 5 | control account plan | https://pmstudyguide.com/control-account-plan/ |
| 6 | cost estimates | https://pmstudyguide.com/cost-estimates/ |
| 7 | risk management strategies | https://pmstudyguide.com/risk-management-strategies/ |
| 8 | change management process | https://pmstudyguide.com/change-management-processes/ |
| 9 | work performance report | https://pmstudyguide.com/work-performance-report/ |
| 10 | critical path | https://pmstudyguide.com/critical-path-method-explained/ |
Bonus links also included in the body: quality management and the Excel EVM dashboard walkthrough.
