Earned Value Management

What is Earned Value Management?

Earned Value Management (EVM) is a project control technique that combines measurements of scope, schedule, and cost into a single integrated view of project performance. Instead of just asking "are we on budget?" or "are we on schedule?" separately, EVM lets you ask both at the same time, using three core numbers you track at a given point in time (the "status date"):

Planned Value (PV) — the budgeted cost of the work you scheduled to have done by now.
Earned Value (EV) — the budgeted cost of the work you have actually completed by now.
Actual Cost (AC) — what you have actually spent to get that work done.

Everything else in EVM — variances, performance indexes, and forecasts — is derived from these three values.

Step 1: Build Your Baseline

Before you can measure anything, you need a baseline to measure against. This means:

Break the project down into tasks or work packages (a Work Breakdown Structure).
Assign a budget (planned cost) to each task.
Schedule each task with a planned start and end date.
Lock this in as your baseline — the plan you will measure actual progress against.

Without a baseline, PV has no meaning, because PV is entirely defined by "what the plan says should be done by this date."

Step 2: Pick a Status Date and Gather Your Three Numbers

On a regular cadence (weekly or monthly), choose a status date and collect:

PV — sum the budgeted cost of all work that was scheduled to be complete by the status date.
EV — sum the budgeted cost of the work actually completed (or the % complete × task budget for partially done tasks).
AC — sum what was actually spent (labor, materials, etc.) on that work so far.

The honesty of your EV figure matters most. A common trap is marking tasks "50% done" out of habit rather than fact — this quietly corrupts every number that follows.

Step 3: Calculate Variances

Variances tell you, in currency terms, whether you're ahead/behind and under/over:

Schedule Variance (SV) = EV − PV — positive means ahead of schedule, negative means behind.
Cost Variance (CV) = EV − AC — positive means under budget, negative means over budget.

A quick way to remember it: EV is the anchor. Compare EV to PV for schedule, and EV to AC for cost.

Step 4: Calculate Performance Indexes

Indexes express the same information as ratios, which makes them easier to compare across projects or over time:

Schedule Performance Index (SPI) = EV ÷ PV — above 1.0 is ahead of schedule, below 1.0 is behind.
Cost Performance Index (CPI) = EV ÷ AC — above 1.0 is under budget, below 1.0 is over budget.

For example, a CPI of 0.90 means that for every R1 spent, you are only getting R0.90 of planned value back — a warning sign that costs are running ahead of progress.

Step 5: Forecast Where the Project Is Heading

EVM's real power is that it lets you forecast the future from current trends, rather than just reporting the past:

Budget at Completion (BAC) — the total original budget for the whole project.
Estimate at Completion (EAC) = BAC ÷ CPI — what the project is now expected to cost in total, assuming current cost performance continues.
Estimate to Complete (ETC) = EAC − AC — how much more money is needed to finish.
Variance at Completion (VAC) = BAC − EAC — how far over or under the original budget you expect to finish.
To-Complete Performance Index (TCPI) = (BAC − EV) ÷ (BAC − AC) — the cost efficiency required on all remaining work to still finish on the original budget.

If TCPI is much higher than your CPI so far, that's a strong signal the original budget is no longer realistic unless something changes.

Step 6: Track Trends, Not Just Snapshots

A single CPI or SPI reading only tells you where you are today. Plot PV, EV, and AC as curves over time (the classic "S-curve") and watch the trend in CPI/SPI over several reporting periods. A project drifting from 1.05 to 0.95 CPI over three months is a very different situation from one that has been steady at 0.95 the whole time.

Use the trend to decide when to intervene: re-plan, add resources, cut scope, or escalate — before the variance becomes unrecoverable.

Step 7: Report and Act

Turn the numbers into decisions:

Report SV/CV and SPI/CPI to stakeholders alongside the forecast EAC.
Investigate the root cause of any variance beyond your agreed threshold (e.g. ±10%).
Decide on corrective action: re-baseline, add resources, adjust scope, or accept the variance.
Re-forecast after any corrective action to confirm it moves CPI/SPI back toward 1.0.

Quick Reference Cheat Sheet

Metric	Formula	Meaning
SV	EV − PV	Ahead (+) or behind (−) schedule, in Rand
CV	EV − AC	Under (+) or over (−) budget, in Rand
SPI	EV ÷ PV	Schedule efficiency ratio
CPI	EV ÷ AC	Cost efficiency ratio
EAC	BAC ÷ CPI	Forecast total cost at completion
ETC	EAC − AC	Estimated cost remaining
VAC	BAC − EAC	Forecast variance at completion
TCPI	(BAC − EV) ÷ (BAC − AC)	Efficiency needed on remaining work to hit BAC

The Scenario

You are running a software project made up of 10 modules. Each module is budgeted at $100,000. The plan was to finish everything in 10 months. In reality, the project took 12 months to finish, and the total actual cost came to $1.3 million. All 10 modules were eventually completed — nothing was left unfinished. You now need to work out the full EVM picture: were you on budget? On schedule? What do the numbers actually tell you?

Step A: Work Out BAC (Budget at Completion)

BAC is simply the total original budget for the whole project — before anything has happened. You get it by multiplying the number of modules by the cost per module:

BAC = 10 modules × $100,000 per module = $1,000,000

Step B: Work Out PV (Planned Value)

PV asks: "as of right now, how much work were we supposed to have finished, according to the original schedule?" Because we're looking at this project at its completion point, and the entire $1M budget was scheduled to be used by the original end date, PV at completion is the full budget:

PV = $1,000,000

Step C: Work Out EV (Earned Value)

EV asks a different question: "how much of the budgeted work have we actually completed?" — not how much we spent, just how much scope is genuinely done. Since all 10 modules were completed:

EV = 10 modules completed × $100,000 = $1,000,000

Notice EV and PV are identical here ($1M each) — that's the key detail that drives everything that follows.

Step D: Work Out AC (Actual Cost)

AC is simply what you actually spent, in total, to get the work done. The scenario tells us this directly:

AC = $1,300,000

Step E: Calculate the Variances (SV and CV)

Now plug PV, EV, and AC into the standard variance formulas:

SV = EV − PV = $1,000,000 − $1,000,000 = $0

CV = EV − AC = $1,000,000 − $1,300,000 = −$300,000

Reading these: SV of $0 says "zero schedule variance," and CV of −$300,000 says "we spent $300,000 more than the work was worth" — a clear cost overrun.

Step F: Calculate the Performance Indexes (SPI and CPI)

SPI = EV ÷ PV = $1,000,000 ÷ $1,000,000 = 1.0

CPI = EV ÷ AC = $1,000,000 ÷ $1,300,000 = 0.77

CPI of 0.77 confirms the cost overrun clearly: for every $1 spent, only $0.77 of planned value was actually earned back. That part matches what you'd expect. But look closely at SPI — it says exactly 1.0, meaning "perfectly on schedule."

Step G: Spot the Problem — Why Does SPI Say "On Schedule" When It Wasn't?

This is the important lesson in the whole exercise. The project actually took 12 months, not the planned 10 — that's a real two-month schedule slip. But SV and SPI both come out perfect (SV = $0, SPI = 1.0). Why?

Because SV and SPI are measured in money, not in time. By the time you reach the very end of the project, every module has eventually been completed — so EV has "caught up" to the full PV, no matter how late that happened. SV and SPI only tell you whether the budgeted value of completed work matches the budgeted value of planned work — they go quiet about lateness once all the scope is eventually delivered, even two months behind.

This is exactly why a separate technique called Earned Schedule (ES) was developed — it re-expresses schedule performance directly in time units (months, weeks) rather than money, so a genuine schedule slip like this one doesn't get masked at completion.

Step H: Forecast Forward — Estimate at Completion (EAC)

EAC asks: "based on how we've performed so far, what will the project actually end up costing in total?" There are several ways to calculate it, depending on what assumption you're making about the remaining work. Here they are, one at a time:

Method 1 — assume current cost performance continues:

EAC = BAC ÷ CPI = $1,000,000 ÷ 0.77 ≈ $1,300,000

Method 2 — assume remaining work goes exactly to the original plan:

EAC = AC + (BAC − EV) = $1,300,000 + ($1,000,000 − $1,000,000) = $1,300,000

Method 3 — the original estimate is no longer trustworthy, so remaining work needs fresh bottom-up estimating:

EAC = AC + Bottom-Up ETC (you'd re-estimate remaining work item by item — not a fixed formula)

Method 4 — assume both cost and schedule performance will influence what's left:

EAC = AC + (BAC − EV) ÷ (CPI × SPI)
= $1,300,000 + ($1,000,000 − $1,000,000) ÷ (0.77 × 1.0)
= $1,300,000 + $0 = $1,300,000

In this particular scenario all methods land on the same $1.3M, because the project is already finished (BAC − EV = 0, so there's nothing left to forecast). In a project that's only partway done, these four methods will usually give you noticeably different answers — which is the whole point of picking the right one for your situation.

Step I: Work Out ETC (Estimate to Complete)

ETC answers: "from today, how much more money do we still need to spend?"

ETC = EAC − AC = $1,300,000 − $1,300,000 = $0

Makes sense — the project is already finished, so there's nothing left to spend.

Step J: Work Out VAC (Variance at Completion)

VAC answers: "compared to what we originally budgeted, how far over or under did we end up?"

VAC = BAC − EAC = $1,000,000 − $1,300,000 = −$300,000

Negative VAC confirms the project finished $300,000 over its original budget — the same overrun we already saw in CV, now expressed as a completion-level forecast rather than a point-in-time snapshot.

Step K: Work Out TCPI (To-Complete Performance Index)

TCPI answers a forward-looking question: "how efficient does the remaining work need to be, to hit a target budget?" There are two versions depending on which target you're aiming for:

TCPI (to hit original BAC) = (BAC − EV) ÷ (BAC − AC)
= ($1,000,000 − $1,000,000) ÷ ($1,000,000 − $1,300,000)
= $0 ÷ −$300,000 = 0

TCPI (to hit revised EAC) = (BAC − EV) ÷ (EAC − AC)
= ($1,000,000 − $1,000,000) ÷ ($1,300,000 − $1,300,000) = undefined (0 ÷ 0)

Both come out as edge cases here, simply because there's no remaining work left (BAC − EV = 0) — TCPI is far more meaningful mid-project, where the numerator and denominator are both still non-zero. Try re-running Steps H–K with, say, EV = $700,000 and AC = $900,000 (i.e. 70% done, partway through) to see TCPI behave more normally — that's a good next practice exercise once you're comfortable with this worked example.

Step L: A Note on Weighting CPI and SPI Together

One more nuance worth practicing: Method 4's formula (EAC = AC + (BAC − EV) ÷ (CPI × SPI)) assumes cost and schedule performance influence the remainder equally. In real projects that's often not true — maybe cost overruns are 60% likely to keep happening, while schedule slippage is only 10% likely to keep happening. In that case you can weight each index by how much you actually judge it will influence what's left:

ETC = (BAC − EV) ÷ [(CPI × 0.60) + (SPI × 0.10)]

The two weights (here 0.60 and 0.10) don't need to add up to 1.0 — each one independently reflects how much you expect that factor to keep affecting the project, not a fixed split between the two. A good way to build intuition here is to re-run this formula on a partway-done project (not this fully-finished example) with a few different weight combinations, and watch how sensitive the forecast is to your assumptions.

Practice material and worked examples in this section adapted from Management Yogi — credit to Satya Narayan Dash and contributing author Sathish Babu, PMP, for the original articles this practice scenario is based on.

The Scenario

A network upgrade project has a total budget of $500,000, planned to run over 12 months. You're now at the 6-month status date (halfway through the schedule). According to the plan, work worth $260,000 should have been done by now. In reality, your team has completed work worth $210,000, and has spent $240,000 getting there.

Step A: Identify the Three Core Numbers

BAC = $500,000

PV = $260,000 (what should be done by month 6)

EV = $210,000 (what is actually done)

AC = $240,000 (what has actually been spent)

Step B: Variances

SV = EV − PV = $210,000 − $260,000 = −$50,000 (behind schedule)

CV = EV − AC = $210,000 − $240,000 = −$30,000 (over budget)

Step C: Performance Indexes

SPI = EV ÷ PV = $210,000 ÷ $260,000 ≈ 0.81

CPI = EV ÷ AC = $210,000 ÷ $240,000 ≈ 0.88

Unlike the finished-project example, both indexes are genuinely below 1.0 here — this project really is behind schedule (SPI 0.81) and over budget (CPI 0.88), and both numbers agree with what a project manager would see on the ground: less work done than planned, and more money spent than that work was worth.

Step D: Forecast EAC — Four Ways

Method 1 — current cost performance continues:

EAC = BAC ÷ CPI = $500,000 ÷ 0.88 ≈ $568,182

Method 2 — remaining work goes exactly to original plan:

EAC = AC + (BAC − EV) = $240,000 + ($500,000 − $210,000) = $530,000

Method 3 — both cost and schedule performance influence what's left:

EAC = AC + (BAC − EV) ÷ (CPI × SPI)
= $240,000 + ($500,000 − $210,000) ÷ (0.88 × 0.81)
= $240,000 + $290,000 ÷ 0.713 ≈ $240,000 + $406,732 ≈ $646,732

Now you can see the three methods genuinely disagree — $568,182 vs $530,000 vs $646,732 — which is exactly the kind of spread that didn't show up in the finished-project example. The choice of method matters a lot mid-project.

Step E: ETC, VAC, and TCPI

Using Method 1's EAC ($568,182) as the working forecast:

ETC = EAC − AC = $568,182 − $240,000 ≈ $328,182 still needed

VAC = BAC − EAC = $500,000 − $568,182 ≈ −$68,182 (forecast overrun)

TCPI (to hit original BAC) = (BAC − EV) ÷ (BAC − AC)
= ($500,000 − $210,000) ÷ ($500,000 − $240,000)
= $290,000 ÷ $260,000 ≈ 1.12

A TCPI of 1.12 means the team now needs to be 12% more cost-efficient than planned on all remaining work just to still land on the original $500,000 budget — a realistic, useful number, unlike the 0/undefined result we got in the finished-project example.

Original practice scenario written to illustrate EVM calculations mid-project, where BAC − EV is still non-zero and the forecasting methods genuinely diverge.

The Scenario

A data migration project has a budget of $300,000 over 8 months. At the 4-month status date, the plan expected $150,000 worth of work to be done. Your team has actually completed $180,000 worth of work, spending only $160,000 to get there. This is the "good news" scenario — worth practicing so you recognize it, not just the overrun cases.

Step A: Core Numbers

BAC = $300,000

PV = $150,000

EV = $180,000

AC = $160,000

Step B: Variances

SV = EV − PV = $180,000 − $150,000 = +$30,000 (ahead of schedule)

CV = EV − AC = $180,000 − $160,000 = +$20,000 (under budget)

Step C: Performance Indexes

SPI = EV ÷ PV = $180,000 ÷ $150,000 = 1.20

CPI = EV ÷ AC = $180,000 ÷ $160,000 = 1.125

Both indexes above 1.0 confirms good news on both fronts: 20% more schedule progress than planned, and every $1 spent is returning $1.125 of planned value. Worth practicing the instinct that "above 1.0 is good, below 1.0 is bad" for both SPI and CPI equally.

Step D: Forecast EAC

EAC = BAC ÷ CPI = $300,000 ÷ 1.125 ≈ $266,667

A forecast total cost of $266,667 — below the original $300,000 budget — is the mirror image of the overrun scenarios: the project is trending to finish under budget if this performance holds.

Step E: VAC and TCPI

VAC = BAC − EAC = $300,000 − $266,667 ≈ +$33,333 (projected underspend)

TCPI = (BAC − EV) ÷ (BAC − AC) = ($300,000 − $180,000) ÷ ($300,000 − $160,000)
= $120,000 ÷ $140,000 ≈ 0.857

A TCPI below 1.0 (0.857) means the team could actually afford to be less efficient than before on the remaining work and still land inside the original budget — the opposite reading from Example 2's TCPI of 1.12, which demanded better-than-before performance just to stay on budget.

Original practice scenario written to illustrate a project performing ahead of schedule and under budget, for contrast against the overrun scenarios above.

The Scenario

A construction fit-out project has a budget of $800,000 over 10 months. At the 5-month status date, planned work value was $400,000. The team has actually completed $460,000 worth of work — ahead of plan — but has spent $520,000 to get there, by paying overtime and rush-order material fees to move faster. This is the trickiest case to practice: the two indexes point in opposite directions.

Step A: Core Numbers

BAC = $800,000

PV = $400,000

EV = $460,000

AC = $520,000

Step B: Variances and Indexes

SV = EV − PV = $460,000 − $400,000 = +$60,000 (ahead of schedule)

CV = EV − AC = $460,000 − $520,000 = −$60,000 (over budget)

SPI = EV ÷ PV = $460,000 ÷ $400,000 = 1.15

CPI = EV ÷ AC = $460,000 ÷ $520,000 ≈ 0.885

Step C: Reading the Mixed Signal

This is worth sitting with: SPI (1.15) says the project is genuinely ahead of schedule — that part is real, not an artifact like in Example 1. But CPI (0.885) says it's buying that speed at a real cost premium. This is a classic "crashing the schedule" pattern — paying overtime, rush fees, or extra resources to pull work forward, which shows up exactly this way: good SPI, bad CPI, at the same time.

Step D: Forecast EAC — Cost-Only vs Combined

Method 1 — cost performance only:

EAC = BAC ÷ CPI = $800,000 ÷ 0.885 ≈ $903,955

Method 2 — combined cost and schedule influence:

EAC = AC + (BAC − EV) ÷ (CPI × SPI)
= $520,000 + ($800,000 − $460,000) ÷ (0.885 × 1.15)
= $520,000 + $340,000 ÷ 1.018 ≈ $520,000 + $334,047 ≈ $854,047

Notice Method 2's forecast ($854,047) is lower than Method 1's ($903,955) — because it gives some credit for the strong schedule performance offsetting the cost overrun, whereas Method 1 (cost-only) assumes the current cost inefficiency is the only thing that matters going forward. Which one you trust depends on whether you believe the good schedule performance is sustainable, or was a one-time push.

A good practice exercise here: decide for yourself which EAC method you'd actually report to a client in this situation, and be ready to explain why — that judgment call is a real part of using EVM, not just plugging numbers into formulas.

Original practice scenario written to illustrate a project with opposing schedule and cost signals, a pattern common in schedule-crashing situations.

Project Data Extraction Tool

Download the tool to help extract project data for use in your own EVM calculations.

Download ProjectDataExtractionTool.zip

Earned Value Management

EVM Detailed Image (Click to open and then click image to expand and contract it)

What is Earned Value Management?

Step 1: Build Your Baseline

Step 2: Pick a Status Date and Gather Your Three Numbers

Step 3: Calculate Variances

Step 4: Calculate Performance Indexes

Step 5: Forecast Where the Project Is Heading

Step 6: Track Trends, Not Just Snapshots

Step 7: Report and Act

Quick Reference Cheat Sheet

Practice Example: Working Through a Real EVM Scenario

The Scenario

Step A: Work Out BAC (Budget at Completion)

Step B: Work Out PV (Planned Value)

Step C: Work Out EV (Earned Value)

Step D: Work Out AC (Actual Cost)

Step E: Calculate the Variances (SV and CV)

Step F: Calculate the Performance Indexes (SPI and CPI)

Step G: Spot the Problem — Why Does SPI Say "On Schedule" When It Wasn't?

Step H: Forecast Forward — Estimate at Completion (EAC)

Step I: Work Out ETC (Estimate to Complete)

Step J: Work Out VAC (Variance at Completion)

Step K: Work Out TCPI (To-Complete Performance Index)

Step L: A Note on Weighting CPI and SPI Together

Practice Example 2: A Project Only Partway Done

The Scenario

Step A: Identify the Three Core Numbers

Step B: Variances

Step C: Performance Indexes

Step D: Forecast EAC — Four Ways

Step E: ETC, VAC, and TCPI

Practice Example 3: Ahead of Schedule, Under Budget

The Scenario

Step A: Core Numbers

Step B: Variances

Step C: Performance Indexes

Step D: Forecast EAC

Step E: VAC and TCPI

Practice Example 4: Mixed Signals — Ahead of Schedule but Over Budget

The Scenario

Step A: Core Numbers

Step B: Variances and Indexes

Step C: Reading the Mixed Signal

Step D: Forecast EAC — Cost-Only vs Combined

Project Data Extraction Tool