Cost to Complete: The Number That Tells You If You Are Still Making Money

The commercial manager and the project manager are looking at the same job and they are four hundred thousand dollars apart on where it finishes.

Neither of them has made an arithmetic error. The PM built his number from the remaining work, package by package. The commercial manager took the budget and scaled it by percentage complete. Both are legitimate methods. On a project with an unresolved variation and a trade running hot, they give different answers, and nobody in the room can say which one is right.

What is cost to complete in construction?

Cost to complete (CTC) is the money still required to finish all remaining work on a project. It includes forecast labour, materials, subcontractor commitments, plant and every other expense not yet incurred.[1]

On its own it is a half number. The one that matters is what it implies about the finish, and that requires adding what you have already spent.

Committed cost tells you what you have promised. Actual cost tells you what you have paid. Only EAC tells you where the job lands, and it is entirely dependent on the quality of the CTC underneath it.

EAC = Actuals to date + CTC

The only number that says where the project finishes. Everything else is history.

Standard project controls

How do you calculate cost to complete?

Three methods are in common use. They are not interchangeable, and the difference between them is the four hundred thousand dollars in the opening scene.[2][3] Software that forecasts cost typically asks you to pick one of them at setup, which is a decision most teams make once and never revisit.[5]

Bottom-up is the one to run the job on. It is the only method that forces someone to look at each package and say what is genuinely left, which is also why it is the one that gets skipped when the month-end is tight.

Top-down is a legitimate cross-check. If your bottom-up and top-down numbers diverge sharply, that gap is information: usually it means either a package is under-forecast, or the percentage complete is optimistic.

Why the three methods disagree

Because each one carries a different hidden assumption about the remaining work.

Top-down assumes the rest of the job performs to budget. On a project where the first forty percent ran ten percent over, that is a heroic assumption dressed up as arithmetic.

Earned value assumes the cost performance index holds. That is reasonable on a stable, repetitive build and unreliable the moment conditions change, which on construction is most months.

Bottom-up assumes somebody has actually reviewed each package. Where that review degrades into copying last month’s figure, the method still produces a number, and the number is now fiction that looks like diligence.

A forecast that does not change is not stable. It is stale. If a CTC has carried the same value for three months on an active project, nobody is forecasting. They are transcribing.

3 methods

Bottom-up, top-down and earned value. They answer the same question with different assumptions, and the gap between them is diagnostic.

Project controls practice

What makes a cost to complete forecast wrong

Four things, in rough order of how often they bite.

Unapproved variations sitting outside the forecast. Work is instructed, the subcontractor starts, the variation is not yet approved, so it is in nobody’s number. The cost is real and accruing while the forecast says the package is on budget. This is the single most common source of a late surprise, and it is why variation and progress claim discipline is a cost control issue, not an admin one.

Percentage complete measured by spend. If you calculate progress from money spent, then a package running over budget reports as further advanced than it is. The overrun disguises itself as progress. Progress has to come from the work, not the ledger.

Prolongation not carried. Site overheads, supervision and plant hire continue for as long as the job does. If the programme has slipped four weeks and the forecast does not carry four weeks of preliminaries, the number is wrong by a knowable amount. Where the slip is compensable, that is an extension of time question as well as a cost one.

Invoices in the approval queue. Cost that has been incurred but not yet processed is invisible to a ledger-based forecast. Where invoice approval takes six weeks, your actuals are structurally six weeks light, and every EAC built on them is optimistic by the same margin.

What good looks like

• Cost to complete is forecast bottom-up, per package, by the person who owns that package

• A top-down number runs alongside it as a cross-check, and a divergence is investigated rather than averaged

• Approved and pending variations both appear in the forecast, with pending ones flagged rather than omitted

• Percentage complete comes from measured work, never from spend

• Programme slip flows into preliminaries automatically, so prolongation is carried without anyone remembering

• Invoices in approval are visible as accrued cost, so actuals are not artificially light

• The forecast moves month to month, and a package that has not moved is a question rather than a comfort

The test of a forecasting process is not whether the final number was right. It is how early it started being right.

The Australian context

Australian builders are running this calculation under real pressure. The Australian Bureau of Statistics records more than $60 billion in construction work done nationally each quarter, and the sector continues to carry the highest share of corporate insolvencies of any industry.[4]

Insolvency in construction is rarely a company that did not know it was losing money. It is usually a company that found out too late to do anything, on a fixed-price contract where the recovery levers had already expired. The gap between “we are over” and “we can still act” is measured in weeks.

That is what makes cost to complete a governance number rather than a finance one. It is the earliest reliable warning a builder gets, and its usefulness decays fast. A forecast produced three weeks after month end describes a project that no longer exists, which is the same structural problem covered in why monthly PDF reports arrive too late.

Where Plexa fits

Plexa builds cost to complete from the commitments the project has actually made. Subcontracts, purchase orders, approved variations and progress claims all live in the same platform, so committed cost is a live figure rather than a monthly reconstruction.

Because the forecast sits next to the budget, the EAC and the variance against budget update as commitments are made rather than when someone assembles a report. A variation approved on Tuesday is in Tuesday’s forecast. An invoice sitting in the approval queue is visible as accrued cost rather than absent from it.

Package owners forecast their own remaining cost against their own commitments, which is what makes bottom-up sustainable rather than a month-end scramble. Where a package has not been touched, that shows as an ageing forecast rather than passing silently as a repeated number.

The commercial manager and the project manager in the opening scene are not really four hundred thousand apart on the maths. They are apart on which packages carry the unapproved variation and whether the programme slip is in the preliminaries. When both are reading the same commitment record, that conversation takes ten minutes and produces one number.

Frequently asked questions

What is cost to complete in construction? The money still required to finish all remaining work on a project, including forecast labour, materials, subcontractor commitments and plant. Added to actuals to date, it gives the estimate at completion, which is the forecast final cost.[1]

What is the difference between CTC and EAC? Cost to complete is the remaining money. Estimate at completion is the total forecast final cost of the job. EAC = actuals to date + CTC. CTC answers “what is left”, EAC answers “where do we finish”.[1]

How do you calculate cost to complete? Three methods. Bottom-up sums each package’s committed and uncommitted remaining cost. Top-down takes budget × (1 − % complete) adjusted for known variances. Earned value uses (BAC − EV) ÷ CPI. Bottom-up is the most reliable for running an active project.[2][3]

Why do two forecasts for the same project disagree? Usually because they used different methods with different hidden assumptions, not because someone made an error. Top-down assumes remaining work performs to budget; earned value assumes current cost performance continues; bottom-up assumes each package has genuinely been reviewed.

How often should cost to complete be updated? At minimum monthly, and it should change. A forecast that carries the same value for three consecutive months on an active project is being transcribed rather than forecast.

Related reading

The margin pressure this number exists to detect is covered in why construction profit margins keep getting tighter. The reporting lag that blunts it is the subject of why monthly PDF reports arrive too late. And the invoice backlog that makes actuals look light is explained in why construction invoices take six weeks to approve.

If you want to see a live cost to complete on your own projects, book a 30-minute demo with the Plexa team.