Earned Value Management Example: The EAC Formula Nobody Picks Correctly

This article covers the full earned value management example — all the formulas, a complete worked calculation — but the EAC choice and TCPI get the most attention because those are the parts I find practitioners most uncertain about. CPI and SPI I’ll cover more quickly because there’s less to debate.

Why the Standard EAC Formula Gets Misused

The standard EAC formula divides the total budget by the current Cost Performance Index:

This assumes the project will continue running at whatever cost efficiency it’s running at now. If CPI is 0.82 today, it assumes CPI will stay at 0.82 for the remaining work. That’s the assumption. Most project teams apply this formula without stating that assumption explicitly, and sometimes without realizing they’re making it.

In many situations, that assumption is defensible. There’s decent evidence that CPI tends to stabilize once a project is past 20% or so of completion — I’ve seen this play out enough times that I take early CPI seriously as a predictor of final cost performance. But “tends to stabilize” is not the same as “will definitely remain constant.” A project where the first four months ran over budget because of a specific procurement failure — a supplier who’s now been replaced, a process that’s been fixed — is not well served by a forecast that assumes the overrun will continue at the same rate. In that case, the optimistic EAC is the right tool.

The problem is that the optimistic EAC requires you to argue that the past overrun was non-recurring. That argument takes effort. BAC ÷ CPI requires no argument. So teams reach for the easier formula, present a number that implies the overrun will continue forever, and nobody questions the assumption because it came from a formula.

I’m spending time on this upfront because it’s the thing I wish someone had flagged earlier in my career. The rest of the EVM framework is more straightforward. Let me cover the basics and then we’ll come back to EAC.

PV, EV, AC: What Earned Value Management Actually Measures

Earned value analysis adds a third variable to standard cost tracking. Most project cost reports compare what you planned to spend against what you’ve actually spent. The gap tells you whether you’re over or under budget — but not whether the work is getting done.

A project can be underspent because work is running efficiently, or because less work got done than planned. Those have completely different implications and standard cost tracking can’t tell them apart. That’s the problem EVM fixes.

Planned Value (PV) is the budgeted cost of the work that should be complete by the status date.

Earned Value (EV) is the budgeted cost of the work actually completed — not what it cost, what it was budgeted to cost. This is the number that makes the separation possible.

Actual Cost (AC) is total spend to date. Nothing more.

Once you have these three numbers, you can measure cost efficiency (how much budgeted work are we getting per dollar spent) and schedule efficiency (how much planned work are we completing per period) independently. That’s the whole system.

CPI — and the Thing About Percent Complete

Cost Performance Index:

CPI above 1.0: getting more than a dollar of work per dollar spent — under budget. Below 1.0: over budget relative to work completed. The interpretation is simple enough.

What I want to flag here is the percent-complete problem, because I think it’s underemphasized relative to how much it matters. EV is calculated from a percent-complete figure. That percent-complete figure is often estimated by the same people who are under pressure to show progress. The result is systematic overstatement — not always malicious, often just optimistic — and every EVM metric downstream gets skewed by it.

I once worked on a project where CPI was reported at 1.1 for three consecutive months. The project was visibly struggling. When we dug into it, percent-complete estimates had been consistently overstated across about a dozen work packages. The “healthy” CPI was a measurement artifact. By the time accurate progress was being reported, the project was significantly further behind than anyone had acknowledged.

So: CPI is only as reliable as the percent-complete inputs feeding EV. If those inputs are being managed rather than measured, the CPI is telling you what people want you to hear, not what’s happening.

Cost Variance (CV) is the dollar version: CV = EV − AC. Positive means under budget for work done, negative means over. I use CPI more than CV in practice because it’s scale-independent and easier to track as a trend, but CV shows up in every reporting template so you’ll need it.

SPI: Why I Use It Less Than I Used To

Schedule Performance Index (SPI = EV ÷ PV) measures schedule efficiency. Above 1.0: more work done than planned. Below 1.0: behind.

I used to track SPI closely. I’ve pulled back on it for one reason: it converges toward 1.0 as the project ends, regardless of actual schedule performance. Because planned value and earned value both ultimately reach the same total (all the work gets done eventually), SPI drifts toward 1.0 in the final project phases even on projects that are running months late. I noticed this on a late-stage project before I fully understood the math behind it, and it took a while to trust my own read of the situation over what SPI was showing.

I still look at SPI in the first half of projects. After that, I track milestone dates against the baseline schedule and don’t rely on SPI for late-phase schedule assessment.

Schedule Variance (SV = EV − PV) has the same limitation — it returns to zero at project end regardless of how late the project actually finishes. Positive SV: more work done than planned. Negative: behind. Use it early; stop trusting it late.

Full Earned Value Management Example 

In this earned value management example, we will focus on a water treatment plant upgrade. These numbers are modified but come from a real project — I’m using it partly because the CPI situation was genuinely complicated, which is more useful to work through than a clean example.

Total budget (BAC): $5,730,000. Planned duration: 14 months. Status at month 9.

CV = $3,184,000 − $3,890,000 = −$706,000. The work that’s been done cost $706K more than budgeted.

SV = $3,184,000 − $3,678,000 = −$494,000. The project is behind schedule by the equivalent of $494K of planned scope.

CPI = 3,184 / 3,890 = 0.818.

SPI = 3,184 / 3,678 = 0.866.

Both below 1.0. The project is over budget and behind schedule. At month 9 of 14, this is a serious position — there isn’t much schedule left to recover in, and if the percent-complete figures are being optimistically reported (which they were, it turned out), the real position was worse.

The Three EAC Calculations and When to Use Each

Back to the formula question I raised at the start. Three options:

Standard (assumes current CPI continues):

This is the right formula when there’s no credible argument that efficiency will improve. On this project, at month 9, with no specific identified cause for the overrun that was being fixed, this was the honest forecast. Overrun: $1,271,000.

Optimistic (overrun was non-recurring, remaining work at plan):

This is the right formula when you can specifically identify a one-time cause for the overrun — a procurement failure that’s been resolved, a weather event that won’t recur — and you can show the remaining work is genuinely unaffected. It requires an argument. If you can’t make that argument, you’re not using the optimistic formula because it’s appropriate; you’re using it because it produces a lower number.

Pessimistic (both cost and schedule inefficiency compound):

This is the right formula when being behind schedule is itself generating cost pressure — extended site overhead, premium rates to catch up, delay penalties. I’m honestly not sure I use this one enough. It feels pessimistic enough that there’s always a temptation to reach for the standard formula instead, but on projects where schedule pressure is compounding cost pressure, this is the more honest number.

ETC (Estimate to Complete) and VAC (Variance at Completion) follow from whichever EAC you choose:

Using standard EAC: ETC = $3,111,000 remaining. VAC = −$1,271,000 projected overrun. These are the numbers for a sponsor conversation — presented as projections based on a stated assumption, not as certainties.

TCPI: Asking the Question the EAC Doesn’t

EAC answers “where will we end up?” TCPI answers a different question: “what would we need to achieve to hit a given target?”

For this project, TCPI to meet the original $5.73M budget:

The remaining work must run at CPI 1.383. Current CPI is 0.818. That’s a 69% improvement requirement. Not impossible in some abstract sense — if scope gets cut significantly, or if the inefficiency was entirely measurement error — but it needs to be named as what it is: a very steep requirement that demands a specific plan, not general optimism.

TCPI to meet the standard EAC of $7,001,000:

That matches current CPI, which is what you’d expect — the standard EAC is built on the assumption that CPI stays at 0.818, so TCPI to meet it should equal 0.818. If these numbers disagreed, there’d be an inconsistency in the calculation.

Where I find TCPI most useful: in conversations where the project is still being measured against the original budget even though the standard EAC is showing a significant overrun. TCPI against the original budget makes explicit what efficiency the remaining work would need — and it makes visible whether that’s a realistic expectation or a number that will require an uncomfortable conversation to justify.

For a complete project status report that incorporates EVM, these metrics work best as trends over multiple periods rather than point-in-time readings. A CPI of 0.82 at month nine tells you something. Whether it was 0.91 at month six and has been declining, or whether it’s been 0.82 since month three, tells you more. 

 Earned Value Management Calculation Example FOR A RAILWAY CONSTRUCTION PROJECT

Let’s assume a 12-month railway project that is planned to spend $3M per month for a total budget at completion (BAC) of $36M. The project is 2 months along and according to the planned vs. actual cost figures is underspent by 50%.
We could assume therefore that the project is going well because it’s costing less than planned.

But after 2 months of work, 10 % of work has been completed so “EV” is 10% of the total $36M budget, which is $3.6M.

Budget at Completion (BAC) : $36M
Planned Value (PV) for 2 Months: $6M
EV: $3.6M
Actual Cost (AC): $3M

Schedule Variance (SV) : (Earned Value – Planned Value) = $3.6 – $6 = – $2.4 Behind the schedule
Cost Variance (CV): (Earned Value – Actual Cost) = $ 600K Under Budget
Cost Performance Index (CPI): (Earned Value / Actual Cost) = 1.2
Schedule Performance Index (SPI): (Earned value / Planned value) = 0.6
Estimate at Completion (EAC) : (Budget at Completion / Cost Performance Index) =$30M

Estimate to Completion (ETC): (Budget at Completion – Earned Value)/ Cost Performance Index = $27 M

Time to Complete: (Budget at Completion – EV)/ Schedule Performance Index =54 Months

By taking into consideration this performance, this railway project will be completed in 54 months with a $36 M budget. 

Earned Value Management Calculation Example FOR A hıghway CONSTRUCTION project

As the project manager overseeing the construction of a new highway, you’re tasked with ensuring its timely completion and within budget. With an initial plan of 18 months and an estimated cost of $100,000,000, the project has progressed for 9 months. At this juncture, expenditures have reached $42,000,000, and the completed work is valued at $55,000,000.

Planned Value (PV) = $50,000,000. 50% of the project duraiton has passed.

Earned Value (EV) = $55,000,000. The estimated value for the work that has been completed.

Actual Cost (AC) = $42,000,000. The mount of money has been spent.

CPI = EV / AC = 1,3 You are operating below the allocated budget for the project, suggesting a likelihood of completing it at a lower cost than originally anticipated.

SPI = EV / PV = 1,1 You are ahead of the schedule and likely to complete the earlier.

Fernando Aldena

I am a Civil Engineer in HES Consultancy Limited, experience as Director and Resident of Works and Technical, Financial and Administrative Project Audit. I have skills in the area of Procurement, Tenders and Contracting. I am PMP and PRINCE2 Certified. Monitoring and Control with the Earned Value Method.