Estimate at Completion (EAC) is the forecasted total cost of a project from start to finish, based on actual performance to date. It's the number that answers the question every project board asks: "What will this actually cost when we're done?" Not what you budgeted. Not what you hoped. What the data says. EAC is part of the earned value management framework and it's arguably the most argued-about figure in any progress meeting.
EAC is part of the earned value definitions glossary. For the full breakdown of all three forecasting metrics together, see the EAC, ETC and TCPI page.
The Core Formula
The most widely used EAC formula on UK construction projects:
Where:
- BAC = Budget at Completion (your approved total budget)
- CPI = Cost Performance Index (cost efficiency ratio: EV / AC)
But that's only one of four formulas. Each one makes a different assumption about the future, and picking the wrong one gives you a forecast that's technically correct but commercially useless.
The Four EAC Formulas: A Decision Framework
Here's how to decide which formula to use. This isn't academic. It's the decision you make every month when you sit down to produce a forecast.
START HERE | Is CPI stable (3+ months)? / ╲ YES NO | | Was variance a Is project < 20% one-off event? Complete? / ╲ / ╲ YES NO YES NO | | | | Formula 2 Has scope Formula 4 Formula 4 (One-off) changed (Bottom (Bottom fundamentally? -up) -up) / ╲ YES NO | | Formula 4 Are cost AND (Bottom-up) schedule both affected? / ╲ YES NO | | Formula 3 Formula 1 (CPI x SPI) (BAC/CPI)
The Four Formulas
| # | Formula | Assumption | When to Use |
|---|---|---|---|
| 1 | EAC = BAC / CPI | Past cost efficiency continues | CPI stable for 3+ months |
| 2 | EAC = AC + (BAC - EV) | Remaining work runs to original budget | One-off event caused the variance |
| 3 | EAC = AC + (BAC - EV) / (CPI x SPI) | Both cost and schedule inefficiency continue | Over budget AND behind programme |
| 4 | EAC = AC + Bottom-Up ETC | Fresh estimate from first principles | Scope change, early stage, or original assumptions are dead |
On a £40M highways package, I'd calculate all four and present the range. The board doesn't want a single number. They want the range of outcomes and your assessment of which is most likely.
Worked Example: £12M Project With CPI of 0.91 and SPI of 0.85
Worked ExampleScenario: A £12M NEC4 Option C rail signalling renewal project. Month 9 of a 16-month programme. The commercial team runs the monthly numbers.
Given:
- BAC = £12,000,000
- EV = £5,460,000 (45.5% earned)
- AC = £6,000,000
- CPI = 0.91 (£5,460,000 / £6,000,000)
- SPI = 0.85 (£5,460,000 / £6,424,000 planned)
Formula 1 (BAC / CPI): EAC = £12,000,000 / 0.91 = £13,187,000 Forecast overrun: £1,187,000
This assumes the current cost efficiency of 0.91 continues for the remaining 54.5% of work. If CPI has been stable since month 5, this is your most reliable estimate.
Formula 2 (AC + remaining budget): EAC = £6,000,000 + (£12,000,000 - £5,460,000) = £12,540,000 Forecast overrun: £540,000
This assumes the overrun was a one-off and the remaining £6.54M of work will be delivered at budgeted cost. Only valid if you can point to a specific event (say, a subcontractor re-procurement that's now resolved) and honestly say it won't happen again.
Formula 3 (CPI x SPI composite): EAC = £6,000,000 + (£12,000,000 - £5,460,000) / (0.91 x 0.85) EAC = £6,000,000 + £6,540,000 / 0.7735 = £14,456,000 Forecast overrun: £2,456,000
This is the conservative view. Both cost and schedule inefficiency continue to compound. If the project is genuinely behind programme (SPI of 0.85 means you've done 85% of what you planned by this point) and there's no recovery plan in place, this is the honest number.
Formula 4 (Bottom-up): The package managers re-estimate remaining work at £7,100,000. EAC = £6,000,000 + £7,100,000 = £13,100,000 Forecast overrun: £1,100,000
Summary for the board:
| Formula | EAC | Overrun | Assumption |
|---|---|---|---|
| BAC/CPI | £13.19M | £1.19M | Current efficiency continues |
| One-off | £12.54M | £0.54M | Remaining work at budget rate |
| CPI x SPI | £14.46M | £2.46M | Cost and schedule issues persist |
| Bottom-up | £13.10M | £1.10M | Fresh estimate from package managers |
The most likely outturn sits between £13.1M and £13.2M. Formula 1 and the bottom-up estimate converge, which builds confidence. The CPI x SPI figure of £14.46M represents the worst case if the team doesn't address the programme slippage.
Why EAC Matters in Construction
Without EAC, cost forecasting is opinion. A commercial manager's gut feel that "we'll probably come in about £500K over" has no defensible basis. EAC, calculated from earned value data, provides an arithmetic foundation.
On NEC4 Option C contracts, EAC directly feeds the pain/gain share calculation. If EAC exceeds the target total of the Prices, the Contractor absorbs a share of the overrun. If EAC is below target, the Contractor shares the saving. Getting this forecast right isn't just good project management. It's the difference between a bonus and a loss.
I've sat in final account meetings where the difference between an EAC produced from earned value data and a traditional CVR forecast was £2.3M. That's not a rounding error. That's the gap between a data-driven forecast and one built on optimism.
Common Mistakes
Using EAC before you have enough data. CPI is volatile in the first two to three months. An EAC based on month-2 CPI is fiction. Wait for three months of stable data, or use a bottom-up estimate instead. I watched a graduate present a £6M overrun forecast based on two months of data. The commercial director nearly choked on his coffee.
Never updating EAC. I've seen teams calculate EAC once at month 4 and not touch it for six months. EAC should be recalculated every reporting period. Conditions change. CPI shifts. Compensation events adjust BAC.
Presenting only one formula. A single EAC gives false precision. Present the range. Explain the assumptions. Let the decision-makers judge which scenario to plan around. If your commercial director sees a single EAC figure, the first question will be "what's the range?" Have the answer ready.
Forgetting to adjust BAC for scope changes. On NEC4, every implemented compensation event should adjust BAC. If BAC is stale, your EAC is measuring against the wrong target. That's not forecasting. That's fiction.
How Gather helps. Gather's AI reads your site diaries daily and maps progress against your cost-loaded programme, giving you accurate earned value data without manual spreadsheet updates. Book a demo to see it working on a live NEC4 project.
Frequently Asked Questions
What's the difference between EAC and ETC?
EAC is the total forecast cost from project start to finish. ETC is what's left to spend from today: ETC = EAC - AC. The finance director cares about ETC because it's the cash still required. The project board cares about EAC because it's the final outturn compared against the original budget.
How often should EAC be updated?
Monthly, aligned with your cost reporting and progress measurement cycle. On high-risk projects or during periods of significant change, fortnightly updates are worthwhile. The critical thing is consistency: use the same formula month to month unless you have a documented reason to switch.
Which EAC formula is most accurate?
The BAC/CPI formula is the most widely validated across industries. Research consistently shows it produces reliable forecasts once three to four months of CPI data is available. But "most accurate" depends on whether the underlying assumption holds for your specific project. If something fundamental has changed, a bottom-up ETC is more honest than any formula.
Does EAC account for scope changes?
Only if you update BAC when scope changes are approved. On NEC4, every implemented compensation event should adjust BAC. If BAC is stale, EAC is measuring against the wrong target and the forecast is meaningless.
.webp)
