Estimate at Completion Calculator (EAC)

The estimate at completion (EAC) is the forecast of what a project will finally cost, based on how it has performed so far.

This estimate at completion calculator takes your earned-value data — planned value, earned value, actual cost and the original budget — and computes all four standard EAC formulas side by side, each with its estimate to complete and its variance at completion, plus the to-complete performance index needed to still hit the budget.

Because each formula bakes in a different assumption about the rest of the project, seeing them together shows the full range of likely outcomes rather than a single guess. Enter your figures above and read the forecast the way a controller does.

What the estimate at completion calculator computes

Earned value management measures where a project stands today; the estimate at completion turns that status into a prediction of the final cost. The calculator derives the cost and schedule performance indices from your inputs, then applies each of the four EAC formulas. For every method it reports the estimate at completion itself, the estimate to complete — the money still to be spent — and the variance at completion, the gap between the forecast and the original budget. It also computes the to-complete performance index against both the original budget and the typical EAC, and it charts the four forecasts against the budget so the spread of outcomes is visible at a glance.

The reason there are four formulas rather than one is that forecasting requires an assumption about the future, and different situations call for different assumptions. Will the cost efficiency seen so far continue? Was the overrun a one-off that will not recur? Will schedule pressure keep dragging on cost? Or is the original baseline simply wrong, so the remaining work must be re-estimated from scratch? Each question corresponds to one formula, and choosing the right one is the real skill; the calculator does the arithmetic for all of them so you can compare and decide.

The inputs: PV, EV, AC and BAC

The forecast rests on four earned-value quantities. Planned value (PV) is the budgeted cost of the work scheduled by the status date; earned value (EV) is the budgeted cost of the work actually completed; actual cost (AC) is what has really been spent; and budget at completion (BAC) is the total baseline budget for the whole project.

From the first three the calculator derives the cost performance index, CPI = EV / AC, and the schedule performance index, SPI = EV / PV, which are the engines of the forecasts. If you already have a fresh bottom-up estimate of the remaining work, you can also enter it to drive the fourth method; leave it blank and the calculator simply omits that method.

These are the same measurements the Earned Value calculator uses, so the two tools share a common input and are designed to be used together.

Method 1: the typical-variance forecast

The most common forecast assumes that the cost efficiency seen so far will continue unchanged for the rest of the project:

EAC = BAC / CPI

If a project is running at a CPI of 0.8 — getting eighty cents of value per dollar — then the whole budget will buy only eighty percent of its planned value, so the final cost is the budget divided by that efficiency. Equivalently, EAC = BAC × AC / EV, a form with no rounding of the index. This is the right forecast when the cause of the current variance is systemic — a persistent productivity shortfall, a rate that was set too low — and will therefore persist. It is the default that most status reports quote, and it is the headline figure this calculator shows above the table. For a project at a CPI of 0.67 against a 4,000 budget, it forecasts a final cost of 6,000, a fifty-percent overrun.

Method 2: the atypical-variance forecast

The second forecast assumes the opposite — that whatever caused the variance so far was a one-off that will not recur, so the remaining work will be done at the originally budgeted rate:

EAC = AC + (BAC − EV)

Here the money already spent is taken as sunk, and the remaining budgeted work, BAC minus EV, is added at face value. This is the right forecast when the overrun had a specific, isolated cause — a one-time price spike, a single reworked deliverable — that has been resolved and will not affect the rest of the project. It always produces a more optimistic EAC than Method 1 when the project is over budget, because it assumes the inefficiency stops now. For the same example it forecasts 1,200 + (4,000 − 800) = 4,400, a much smaller overrun, reflecting the belief that the trouble is behind the team.

Method 3: the cost-and-schedule forecast

The third forecast is the most pessimistic in the common case, because it assumes that both cost inefficiency and schedule pressure will drag on the remaining work:

EAC = AC + (BAC − EV) / (CPI × SPI)

The remaining work is divided not just by the cost index but by the product of the cost and schedule indices, so a project that is both over budget and behind schedule is penalised twice. The reasoning is that being behind schedule usually costs money to recover — overtime, extended overheads — so a low SPI compounds a low CPI. This is the right forecast when a schedule slip is expected to translate into extra cost on the work still to come. For the example, with CPI 0.67 and SPI 0.80, it forecasts 1,200 + 3,200 / (0.67 × 0.80) = 7,200, the highest of the three, a warning that if both problems persist the overrun could be severe.

Method 4: the bottom-up forecast

The fourth forecast abandons the formulas entirely when the original baseline can no longer be trusted:

EAC = AC + bottom-up ETC

Here the money already spent is kept, and the remaining work is re-estimated from the ground up — a fresh, detailed estimate to complete built by the team, ignoring the discredited baseline. This is the right forecast when the original estimate was based on assumptions that have proven wrong: a major scope change, a technology that did not work as expected, a market shift. No index can rescue a baseline that no longer describes reality, so the only honest forecast is a new estimate. This calculator computes Method 4 when you supply a bottom-up estimate to complete; enter your team’s fresh number and it adds it to the actual cost to date.

Choosing the right method

The four methods form a spectrum of assumptions, and the choice among them is a judgement about the future, not a calculation. Use Method 1 when the variance is typical and systemic — the normal default, and the one to reach for when in doubt. Use Method 2 when you are confident the cause of the variance was isolated and is now resolved. Use Method 3 when a schedule slip is expected to keep costing money as the project finishes.

Use Method 4 when the baseline itself is invalid and only a fresh estimate will do. In practice, wise controllers compute all four — as this calculator does — and treat the spread between the optimistic Method 2 and the pessimistic Method 3 as the plausible range of the final cost, reporting Method 1 as the central estimate unless there is a specific reason to prefer another.

The range itself is often more useful than any single number, because it makes the uncertainty in the forecast explicit.

Worked examples

Example 1 — the built-in project. The example loads a project with a planned value of 1,000, an earned value of 800, an actual cost of 1,200 and a budget of 4,000, giving a CPI of 0.67 and an SPI of 0.80. Method 1 forecasts 4,000 / 0.67 = 6,000; Method 2 forecasts 1,200 + 3,200 = 4,400; Method 3 forecasts 1,200 + 3,200 / 0.533 = 7,200. The plausible range for the final cost therefore runs from about 4,400 if the trouble is behind the team to 7,200 if both problems persist, with 6,000 as the central estimate. The variance at completion under Method 1 is 4,000 − 6,000 = −2,000, a projected overrun of half the budget. Press “Load example” to see the table and chart.

Example 2 — an on-track project. When a project is running at a CPI of exactly 1.0, Methods 1 and 2 both forecast the original budget, because there is no inefficiency to project forward and no variance to recover; the forecast simply equals the BAC. Method 3 also returns the budget when SPI is 1.0 as well. This is the reassuring case: all the formulas agree because there is nothing to disagree about. Divergence between the methods only appears once the project departs from plan, and the wider the divergence, the more the choice of method matters.

Example 3 — using the bottom-up method. Suppose the same project has had a major scope change that makes the original baseline meaningless, and the team re-estimates the remaining work at 3,500. Method 4 then forecasts 1,200 + 3,500 = 4,700, and the calculator adds this row to the table. Because it ignores the indices entirely, Method 4 can sit anywhere relative to the others; its value comes from the fresh engineering judgement behind the estimate to complete, not from any formula.

Estimate to complete and variance at completion

Two supporting figures accompany every EAC. The estimate to complete (ETC) is the forecast of the money still to be spent from the status date onward, simply the EAC minus the actual cost already incurred:

ETC = EAC − AC    VAC = BAC − EAC

The ETC is what a manager needs for cash-flow planning and for requesting the remaining funds, and it differs by method exactly as the EAC does. The variance at completion (VAC) is the gap between the original budget and the forecast final cost; a negative VAC is a projected overrun, a positive one a projected underrun. The calculator shows both for every method and colours the VAC green when the project is forecast to come in at or under budget and red when it is forecast to overrun, so the outlook of each scenario is immediately legible.

EAC versus re-baselining

A forecast overrun raises a question the EAC itself cannot answer: should the project keep measuring against its original budget, or should the budget be formally reset? These are two different responses and it is important not to confuse them. The EAC is a forecast; it leaves the baseline intact and simply predicts where the project will land relative to it, preserving the honesty of the variance history.

Re-baselining, by contrast, is a governance action that replaces the original budget with a new one, usually because scope has changed or the original estimate was proven invalid — the same circumstance that calls for the bottom-up Method 4 forecast.

The danger is re-baselining to hide an overrun: if a project simply resets its budget to the current EAC whenever it slips, every period will show zero variance and the early-warning value of earned value is destroyed. Sound practice keeps the original baseline and reports the EAC against it for as long as the baseline remains meaningful, and re-baselines only through formal change control when scope genuinely changes.

The forecast tells you the truth about the current plan; re-baselining changes the plan, and the two should never be quietly merged. Reading the VAC on this page as a forecast against the original budget, rather than as a trigger to reset that budget, keeps the distinction clear.

The to-complete performance index

Alongside the forecasts, the calculator answers a sharper question: how efficiently would the remaining work have to be done to still hit a target? That is the to-complete performance index (TCPI). To finish within the original budget it is the remaining budgeted work divided by the remaining money:

TCPI (to BAC) = (BAC − EV) / (BAC − AC)

To finish within the forecast EAC instead, the denominator becomes the EAC minus the actual cost. The power of the TCPI is as a reality check on the forecast. If a project has been running at a CPI of 0.67 but the TCPI needed to hit the original budget is 1.14, the remaining work would have to be done far more efficiently than any work so far — a clear signal that the budget is no longer achievable and the forecast overrun is real. When the required TCPI drifts more than about five to ten percent above the current CPI, most controllers treat the original budget as lost and manage to the EAC instead. The calculator reports the TCPI against both the budget and the typical EAC.

How to read the results

The headline shows the typical forecast, Method 1, as the central estimate. The metric grid gives the CPI and SPI that drive the formulas — colour-coded so a troubled index is obvious — and both to-complete indices.

The method table is the heart of the output: one row per formula, each with its EAC, its ETC and its VAC, so you can compare the optimistic, central and pessimistic forecasts directly and read the plausible range from top to bottom. The chart plots the forecasts against the budget bar, making the overrun visible as the height of each EAC bar above the BAC.

Export the table to CSV for a forecast report or save the page as a PDF for a review. Read the range, not just the headline: the distance between Method 2 and Method 3 is the honest measure of how uncertain the final cost still is.

How EAC relates to the earned-value dashboard

The estimate at completion is the forecasting half of earned value management, and it pairs with the status half. The Earned Value calculator in this silo reports where the project stands now — the variances and the CPI and SPI — and shows a single headline EAC.

This tool takes those same indices and expands the forecast into the full family of methods, which is the natural next step once the status is understood. The workflow is to compute the status first, see the CPI and SPI, and then come here to turn them into a defensible forecast of the final cost under each assumption.

The two calculators deliberately share the PV, EV, AC and BAC inputs so you can move between them without re-entering data, and together they cover the whole of earned-value cost analysis.

Forecasting cautions

Every EAC is a projection, and projections carry assumptions that can fail. The single biggest caution is that the CPI-based forecasts assume the past predicts the future, which is most reliable once a project is well underway — past roughly the fifteen-to-twenty-percent-complete mark — and least reliable at the very start, when a few early data points can swing the indices wildly.

Early in a project the formulas can produce alarming or rosy forecasts on thin evidence, so treat them cautiously until enough work has been earned to stabilise the indices. A second caution is that the SPI, and therefore Method 3, becomes unreliable near the end of a project, where the schedule index drifts back toward one regardless of how late the project is; a Method 3 forecast late in the project can understate the true cost of recovery.

A third is that all four formulas assume the earned value has been measured honestly; an inflated EV flatters every forecast. Used with these cautions in mind, the EAC is a powerful early-warning tool; used blindly, it can mislead.

The EAC and the funding decision

Beyond reporting, the estimate at completion drives real decisions about money and continuation. When the EAC exceeds the budget, someone must decide what to do: absorb the overrun from contingency, request additional funding, reduce scope to bring the forecast back within budget, or in the worst case stop the project. The ETC is the figure finance needs to release the next tranche of funding, and the VAC is the figure the sponsor weighs against the project’s remaining value.

A well-supported EAC — one backed by the method most appropriate to the situation and cross-checked against the TCPI — is what makes that conversation rational rather than political. This is also why the range across the four methods matters: presenting a sponsor with a single number invites false precision, whereas showing that the final cost will land somewhere between the optimistic and pessimistic forecasts, with a central estimate, frames the decision honestly.

The best project controllers use the EAC not to defend the original budget but to give leadership the earliest possible truthful picture of where the project is heading, while there is still time to act.

The EAC in agile and hybrid projects

Completion forecasting adapts to agile delivery as readily as the rest of earned value. In an agile project the budget at completion is the total funding, the earned value is the budgeted value of the story points accepted as done, and the CPI follows directly; the Method 1 forecast, budget over CPI, then predicts the funding the backlog will actually consume at the current rate of delivery.

Because agile teams measure velocity continuously, the CPI-based EAC can be recomputed every sprint, giving an early and frequently updated forecast of whether the product can be finished within its funding. Hybrid projects, which fund and forecast at the programme level while delivering iteratively, use the EAC at that level in exactly the traditional way.

The bottom-up Method 4 also fits agile naturally, since re-estimating the remaining backlog is a routine agile activity; the fresh estimate to complete is simply the forecast cost of the remaining prioritised work. Whatever the delivery model, the question the EAC answers — what will the whole thing finally cost — is universal.

Where the estimate at completion is used

The EAC is the central forecasting number on any project that tracks earned value, and it is a contractual reporting requirement on large government and defence programmes, where a formal estimate at completion is submitted each period and scrutinised for realism against the TCPI and the independent estimate.

On commercial projects it drives the funding conversation: the ETC tells finance how much more the project will need, and the VAC tells the sponsor whether the project is heading for an overrun in time to act. It is also a staple of project-management certification, and the four EAC formulas, the ETC, the VAC and the TCPI on this page are exactly what the PMP and CAPM examinations expect candidates to compute and to match to the right situation.

Anywhere a project must answer the question “what will this finally cost?”, the estimate at completion is the answer, and computing all four methods is how a professional answers it honestly.

A step-by-step walkthrough of the example

Compute the whole forecast by hand once and the calculator’s output becomes transparent. The example project has planned value 1,000, earned value 800, actual cost 1,200 and budget 4,000. First the indices: CPI is earned value over actual cost, 800 / 1,200 = 0.667, and SPI is earned value over planned value, 800 / 1,000 = 0.80. The remaining budgeted work is the budget minus what has been earned, 4,000 − 800 = 3,200.

Now the four methods. Method 1 divides the whole budget by the cost efficiency: 4,000 / 0.667 = 6,000. Method 2 keeps the actual cost and adds the remaining work at face value: 1,200 + 3,200 = 4,400. Method 3 divides the remaining work by the product of both indices, 0.667 × 0.80 = 0.533, giving 1,200 + 3,200 / 0.533 = 1,200 + 6,000 = 7,200.

For each, the estimate to complete is the EAC minus the 1,200 already spent — 4,800, 3,200 and 6,000 respectively — and the variance at completion is 4,000 minus the EAC — −2,000, −400 and −3,200.

Finally the to-complete index to still hit the budget is the remaining work over the remaining money, 3,200 / (4,000 − 1,200) = 3,200 / 2,800 = 1.14, well above the 0.67 achieved so far, confirming the budget is out of reach. Every figure in the table is reproduced by this arithmetic.

Reading the forecast chart

The chart places the budget as the first, neutral-coloured bar and each EAC forecast beside it in the accent colour, so the overrun is the visible height of the forecast bars above the budget bar.

In the example the budget bar stands at 4,000 while the three forecast bars rise to 6,000, 4,400 and 7,200, and the picture tells the story at a glance: even the most optimistic forecast sits above the budget, so some overrun is almost certain, and the spread between the bars is the uncertainty in how large it will be.

Presenting this chart to a sponsor communicates the forecast far more forcefully than a single number, because it shows both the expected overrun and the risk around it. When the forecast bars cluster tightly the outlook is fairly certain; when they fan out widely, the final cost genuinely depends on which assumption about the remaining work turns out to be right.

Statistical and regression forecasts

The four formulas on this page are the standard deterministic EACs, but they are not the only way to forecast. On large programmes with many periods of history, analysts sometimes fit a statistical or regression model to the trend of the cumulative CPI, or blend the index-based EACs with a bottom-up estimate in a weighted independent estimate at completion.

Research on completed projects has found that the simple CPI-based forecast, EAC = BAC / CPI, is rarely beaten for accuracy once a project is more than about a fifth complete, because the cumulative CPI tends to stabilise and even worsen slightly rather than recover — the empirical rule that the cumulative CPI does not improve by more than a few points after the twenty-percent mark.

That finding is a strong argument for taking the Method 1 forecast seriously rather than hoping for recovery, and for treating optimistic Method 2 forecasts with caution unless there is concrete evidence the variance was truly a one-off. The deterministic formulas this calculator computes are, for most projects, as accurate as far more elaborate models.

Watching the forecast trend over time

A single EAC is a snapshot; the real value comes from watching it move across reporting periods. Recompute the forecast every period and plot the Method 1 EAC over time. A forecast that holds steady period after period is the signature of a project under control, whether it is on budget or forecasting a stable overrun.

A forecast that keeps climbing is the danger sign: it means each period reveals more inefficiency than the last, corrective action is not working, and the eventual overrun will be worse than any single reading suggested. Because the CPI tends to stabilise as a project matures, a still-rising EAC late in a project is especially alarming.

Controllers therefore report not just the current EAC but its trend, and a graph of the EAC against the budget over successive periods is one of the most informative single charts in project reporting. This calculator gives you each period’s forecast; keeping the series is how you turn snapshots into an early-warning trend.

Common mistakes when forecasting the EAC

The most common mistake is defaulting to the optimistic Method 2 forecast because it produces a comfortable number, when the evidence points to a systemic variance that Method 1 or 3 captures better; hope is not a forecasting method. A second is forecasting too early, when a handful of data points make the indices unstable and the EAC swings wildly from period to period; wait until enough value has been earned for the indices to settle.

A third is ignoring the TCPI reality check and reporting an EAC that quietly assumes the remaining work will run far more efficiently than any work so far. A fourth is using Method 3 late in the project, where the SPI has drifted toward one and no longer reflects the real schedule position, producing a falsely reassuring forecast. A fifth is forecasting against a stale baseline after a scope change, when only the bottom-up Method 4 gives an honest number.

Avoiding these keeps the forecast defensible, which matters because an EAC is often the number a sponsor uses to decide whether to continue, re-baseline or cancel a project.

A short glossary of forecasting terms

BAC (budget at completion) — the total baseline budget. EAC (estimate at completion) — the forecast total cost of the project. ETC (estimate to complete) — the forecast of the money still to be spent, EAC − AC. VAC (variance at completion) — the forecast budget difference, BAC − EAC. CPI — cost performance index, EV/AC, the value earned per unit of cost. SPI — schedule performance index, EV/PV. TCPI — to-complete performance index, the efficiency the remaining work must achieve to hit a target. Typical variance — a systemic variance expected to persist (Method 1). Atypical variance — an isolated variance not expected to recur (Method 2). Bottom-up ETC — a fresh, detailed re-estimate of the remaining work (Method 4). Keeping these straight is most of what it takes to read any completion forecast with confidence.

How to use this calculator

Enter four numbers in consistent money units: the planned value, the earned value, the actual cost and the budget at completion. Optionally add a bottom-up estimate to complete to drive Method 4; leave it blank to see only the three formula-based forecasts, which is the usual starting point for a quick read. Press calculate, or press “Load example” to populate a worked project.

Read the headline Method 1 forecast, then the CPI and SPI in the grid, then the method table for the full range of EAC, ETC and VAC across the four methods, and finally the to-complete indices to judge whether the budget is still reachable. Export the forecast to CSV or PDF for your report, and note the plausible range from the optimistic to the pessimistic method rather than reporting a lone figure.

Recompute at each reporting period and watch how the forecasts move: a stable set of EACs across periods is a sign of a predictable project, while forecasts that keep climbing are a warning that the trouble is not being contained.

Keep each period’s four-method table rather than only the headline, too: a change in which method looks most credible — a variance that once seemed atypical proving systemic after all — is itself a signal, and only the full set of forecasts preserved over time makes that shift visible.

Five worked examples of estimate at completion

Example 1: the budget at completion

A project’s total budget (BAC) is 100,000. So far EV = 40,000, AC = 50,000, giving CPI = 0.80. The EAC methods below all forecast the final cost from this position.

Example 2: EAC assuming the cost trend continues

EAC = BAC ÷ CPI = 100,000 ÷ 0.80 = 125,000. If the current cost overrun continues at the same rate, the project finishes 25,000 over budget.

Example 3: EAC assuming the overrun was a one-off

EAC = AC + (BAC − EV) = 50,000 + (100,000 − 40,000) = 110,000. This assumes remaining work runs at the original budgeted rate, so only the past variance carries.

Example 4: EAC using both cost and schedule

With SPI = 0.90, EAC = AC + (BAC − EV) ÷ (CPI × SPI) = 50,000 + 60,000 ÷ 0.72 = 133,333. This is the most pessimistic method, used when both cost and schedule are slipping.

Example 5: the variance and the index to complete

VAC = BAC − EAC = 100,000 − 125,000 = −25,000. The to-complete performance index TCPI = (BAC − EV) ÷ (BAC − AC) = 60,000 ÷ 50,000 = 1.20 — the efficiency now required to still hit budget.

Three expert tips for estimate at completion

Pick the method that matches the cause

Use BAC/CPI when the overrun is systemic, AC + (BAC − EV) when it was a one-time event, and the CPI×SPI form when schedule pressure is driving cost. State your assumption.

Reality-check the TCPI

If the to-complete index needed to hit budget is far above 1.0 (like 1.20), meeting the original budget is usually unrealistic — re-baseline rather than pretend.

Forecast a range, not a point

Report EAC as a range across the methods (here 110,000–133,000). A single figure hides how much the outcome depends on whether the overrun repeats.

Frequently asked questions

Sources, disclaimer and editorial transparency

This calculator applies the four standard EAC formulas: BAC/CPI (typical variance), AC + (BAC − EV) (atypical variance), AC + (BAC − EV)/(CPI × SPI) (cost and schedule), and AC + a bottom-up ETC, together with ETC = EAC − AC, VAC = BAC − EAC, and TCPI = (BAC − EV)/(BAC − AC) or (BAC − EV)/(EAC − AC). This is consistent with standard project-management references such as the PMBOK Guide and the practice standard for earned value management. This calculator and guide are created and reviewed by the OpsCalculators team; see our Editorial Policy for how each tool is researched, built, and tested.

Results are accurate estimates for planning and education, not certified project-controls advice. Index-based forecasts assume performance to date is representative and are least reliable very early in a project; read Method 3 cautiously near completion, where SPI drifts toward one. OpsCalculators.com is operated by MAFHH INTERNATIONAL LTD. Your data is processed in your browser and never stored; see our Privacy Policy.