BudgetOverrun.com is an independent reference site. Not affiliated with any PM software vendor. Statistics sourced from published research and cited throughout.

How to Deal With a Project Cost Overrun

The project is already over budget. This is not about prevention -- it is about what to do next. A 7-step recovery playbook grounded in earned value, so the decision to recover, re-baseline, or de-scope is driven by numbers rather than optimism.

What to do when a project is over budget, in order

  1. Measure it -- compute CPI (EV / AC) so you know the true size of the overrun.
  2. Forecast it -- compute EAC (BAC / CPI) to see where the budget lands if nothing changes.
  3. Diagnose it -- separate scope growth, estimation error, and execution shortfall.
  4. Size the recovery -- compute TCPI to see the efficiency the rest of the work must hit.
  5. Decide -- recover, re-baseline, or de-scope. A TCPI far above your CPI means the original budget is gone.
  6. Rebuild and communicate -- re-baseline through governance and report the forecast honestly.

Run the numbers with the free EVM calculator (CPI, EAC, VAC, TCPI). New to these terms? Start with the earned value guide.

1

Measure the overrun precisely

What: Establish the exact size of the problem in earned-value terms before making any decisions.
How: Compute the Cost Performance Index: CPI = Earned Value (EV) / Actual Cost (AC). A CPI of 0.80 means you are getting 80 cents of planned value for every dollar spent. Also compute the Cost Variance (CV = EV - AC) for the dollar figure. Resist the urge to act on a gut feeling of 'we are over' -- a precise CPI is what every subsequent step depends on.
2

Forecast where the budget actually lands

What: Project the final cost so you are managing to the real endpoint, not the original number.
How: Calculate Estimate at Completion two ways. If the current inefficiency will continue: EAC = BAC / CPI. If the overrun was a discrete, non-recurring event: EAC = AC + (BAC - EV). The gap between the two is your forecast range. Then Variance at Completion (VAC = BAC - EAC) states the projected overrun in dollars -- the number your sponsor will actually care about.
3

Diagnose the root cause

What: Separate the three fundamentally different causes, because each has a different fix.
How: Classify the overrun as (a) scope growth -- the work expanded; (b) estimation error -- the original budget was never realistic; or (c) execution shortfall -- known work is costing more than it should. Scope growth is fixed with change control; estimation error usually forces a re-baseline; execution shortfall is the only one that responds to a genuine recovery push. Most real overruns are a mix, so quantify each contribution rather than blaming a single factor.
4

Calculate the recovery you would need

What: Convert 'we need to catch up' into a hard number so the decision is evidence-based.
How: Compute the To-Complete Performance Index: TCPI = (BAC - EV) / (BAC - AC). This is the cost efficiency the remaining work must hit to still finish on the original budget. Compare it to your actual CPI. A large body of earned-value research on completed contracts has found that the cumulative CPI tends to stabilise early in a project and rarely improves substantially afterwards -- so if the TCPI is far above the CPI you have been running, recovery to the original budget is almost certainly not realistic.
5

Choose the response: recover, re-baseline, or de-scope

What: Pick one explicit path instead of drifting while the overrun compounds.
How: Recover if the cause is execution and the required TCPI is close to 1.0 -- tighten productivity, remove blockers, and re-measure next period. Re-baseline if the estimate was wrong or scope has formally changed and the original budget is no longer meaningful. De-scope if neither the budget nor a re-baseline is acceptable to the sponsor: cut lower-value features to bring the remaining work back inside the money. Stopping the project is a legitimate fourth option when the forecast EAC destroys the business case.
6

Rebuild the plan and re-baseline formally

What: If you re-baseline, do it through governance so the new plan has authority and the old overrun stays visible.
How: A re-baseline is a sponsor-approved reset of the cost and schedule baseline, not a way to make a bad variance disappear. Document the reason, get written sign-off, and preserve the original baseline alongside the new one so the overrun remains on record for lessons learned. Reset the EVM measurement to the new baseline from the change date forward; never retroactively rewrite prior periods.
7

Communicate and capture the lessons

What: Report the forecast honestly and feed the failure back into how the next estimate is built.
How: Give governance the EAC range and VAC in dollars, the chosen response, and the required TCPI -- not a reassuring narrative. After the project, run a structured review: was the original estimate anchored to optimism rather than a reference class? That review is where a recovered project pays forward, by improving the estimates that prevent the next overrun.

The four numbers a recovery decision turns on

MetricFormulaWhat it tells you
CPI -- Cost Performance IndexEV / ACHow much value each dollar is buying. Below 1.0 = over budget for value delivered.
EAC -- Estimate at CompletionBAC / CPIWhere the total cost lands if current efficiency holds.
VAC -- Variance at CompletionBAC - EACThe forecast overrun in dollars. Negative = over budget.
TCPI -- To-Complete Performance Index(BAC - EV) / (BAC - AC)The efficiency the remaining work must hit to still finish on budget. Far above your CPI = recovery not realistic.

These are standard earned-value formulas (PMBOK). For a worked walkthrough of each, see the complete EVM guide, or enter your own numbers in the calculator.

Frequently Asked Questions

What should you do first when a project goes over budget?

Measure the overrun precisely before you react. Compute the Cost Performance Index (CPI = EV / AC) and the forecast final cost (EAC = BAC / CPI). Only once you know the size and trajectory of the overrun should you decide whether to recover, re-baseline, or de-scope. Reacting before measuring is the most common early mistake.

How do you forecast the final cost of an overrunning project?

Use Estimate at Completion (EAC). The most defensible formula is EAC = BAC / CPI, which assumes the cost efficiency seen so far continues. If the overrun was a one-off that will not recur, use EAC = AC + (BAC - EV) instead. Forecasting both gives a realistic range, and VAC = BAC - EAC shows the projected overrun in dollars.

What is TCPI and how do I know if recovery is realistic?

The To-Complete Performance Index is the cost efficiency the remaining work must achieve to hit the original budget: TCPI = (BAC - EV) / (BAC - AC). If work to date ran at a CPI of 0.85 and the TCPI required is 1.15, the team would have to become roughly 35 percent more efficient than it has been. A TCPI materially above 1.0 (rule of thumb, above about 1.10) is rarely achievable without changing scope or baseline.

When should you re-baseline instead of trying to recover?

Re-baseline when the original budget is no longer achievable and reporting against it hides reality. Triggers: the required TCPI exceeds roughly 1.10, the cumulative CPI has been stable and below 1.0 for several periods, or scope has formally changed. Re-baselining is a sponsor-approved governance decision, not a way to erase a variance -- preserve the original baseline so the overrun stays visible for lessons learned.

Related Resources

Updated 2026-06-13