Esc

↑↓ move↵ openIndex · Pagefind

How to run a cost-to-complete review: forecast what's left, line by line, before the job fades

A cost-to-complete review is where a job's margin is actually forecast. Cut off cost, refresh commitments, forecast each cost code from what is physically left, compare the estimate at completion with budget, and roll it up into projected margin. Here is each step, a weekly labor check, and a worked four-code example.

Charley Forey6 min readJob costForecastingWIP
On this page
  1. Step 1: Cut off cost first
  2. Step 2: Refresh committed cost
  3. Step 3: Forecast each cost code from what is left
  4. Step 4: Compare with the budget and read the flags
  5. Step 5: Keep pending changes and risks separate
  6. Step 6: Roll up and explain the change in margin
  7. Step 7: Assign actions and save the review
  8. A weekly check on self-performed labor
  9. Worked example: a small job, four cost codes
  10. Common mistakes
  11. How os.construction handles it
  12. FAQ

To run a cost-to-complete review, cut off cost at the period end, refresh committed costs, then forecast each cost code from what is physically left: open commitments, remaining quantities at the unit cost you’re actually getting, and remaining months of general conditions at today’s run rate. Add cost to date to get the estimate at completion, compare it with the revised budget line by line, and roll the lines up into projected margin against last month.

Cost to complete is the most important estimate a contractor makes after the bid. It sets the estimate at completion, which sets percent complete, earned revenue and profit on the WIP schedule. The free job cost report template has the columns and flags used below (download the .xlsx), and the simpler cost-to-complete worksheet works as a CSV.

Step 1: Cut off cost first

Before anyone forecasts what’s left, make cost to date complete for the period: post the payroll, subcontractor pay apps and supplier invoices that belong to it, and accrue material received but not yet invoiced. A forecast built on half-posted cost is wrong in both columns.

Step 2: Refresh committed cost

Update each commitment for executed subcontract and PO change orders. Note pending change requests from subs separately; they aren’t committed until they’re approved, but they’re exposure.

Step 3: Forecast each cost code from what is left

Work line by line with the PM and superintendent. The method depends on the kind of cost:

  • Subcontracted work: the open commitment (commitment minus invoiced to date), plus anything you expect to add.
  • Self-performed work: remaining quantity × the unit cost you’ve actually achieved so far, unless something specific will change it.
  • General conditions: remaining months on the current schedule × the current monthly run rate. See how to calculate general conditions costs.
  • Uncommitted scope: a current quote or estimate.

cost to complete = what is left to buy, build and run, priced today

estimate at completion = cost to date + cost to complete

Never enter budget minus cost to date as the forecast. It assumes the job will finish exactly on budget, which is the one thing the review is meant to test.

Step 4: Compare with the budget and read the flags

For each line, variance = revised budget − estimate at completion. A negative number is a projected overrun. Check two flags on every line:

  • Cost to complete below the open commitment. You’ve already promised more than the forecast allows. The forecast is understated.
  • Unit cost to date above budget with the forecast still at budget. The past is telling you something the forecast is ignoring.

Step 5: Keep pending changes and risks separate

Show pending owner change orders, pending sub change requests and known risks as their own lines or columns, not folded into the forecast. “Over unless the pending change is approved” is a different conversation from “over.”

Step 6: Roll up and explain the change in margin

Total the lines and compare projected margin with last month’s and with the margin at bid. Every move should be explained by a line: which code, how much, why. A drop in projected margin is margin fade, and the earlier it’s recorded the less it hurts.

Step 7: Assign actions and save the review

Close with owners and dates: chase a pending change, re-plan a crew, get a quote. Save a copy each period, since the trend in estimate at completion says more than any single month.

A weekly check on self-performed labor

Between monthly reviews, check self-performed labor every week. Compare hours spent with the quantity installed, so you know the unit cost or hours per unit you’re actually achieving, and compare it with the budget rate. Labor productivity is usually the first place an overrun shows, and the weekly number lets the superintendent fix it while there is still work left to fix it on.

Worked example: a small job, four cost codes

Illustrative numbersA made-up $2.3 million contract simplified to four cost codes, seven months into a twelve-month schedule that has slipped by one month.

Forecasts by line

  • Concrete (self-performed): 2,000 CY at a budget of $210/CY ($420,000). 1,100 CY placed at a cost of $264,000, which is $240/CY. Remaining: 900 CY × $240 = $216,000.
  • Structural steel (sub): committed $598,000 including approved changes; invoiced $410,000. Remaining: the open commitment of $188,000.
  • Drywall (sub): committed $540,000; invoiced $120,000; open commitment $420,000. The sub has also submitted an $18,000 change request for added soffits, which the GC expects to pass to the owner. It’s shown as pending, not in the forecast.
  • General conditions: budget $380,000 for 12 months; $236,000 spent in 7 months. The schedule now shows 6 months left, and the run rate over recent months is $33,000. Remaining: 6 × $33,000 = $198,000.
Cost code Revised budget Cost to date Cost to complete Estimate at completion Variance
Concrete $420,000 $264,000 $216,000 $480,000 −$60,000
Structural steel $610,000 $410,000 $188,000 $598,000 +$12,000
Drywall $540,000 $120,000 $420,000 $540,000 $0
General conditions $380,000 $236,000 $198,000 $434,000 −$54,000
Total $1,950,000 $1,030,000 $1,022,000 $2,052,000 −$102,000

Pending: $18,000 drywall change request, expected to be covered by an owner change order.

Margin

At bid Projected now
Contract $2,300,000 $2,300,000
Cost $1,950,000 $2,052,000
Margin $350,000 (15.2%) $248,000 (10.8%)

The job has faded 4.4 points. Two lines explain it: concrete is running $30/CY over budget, and the one-month slip adds time to general conditions. If the PM had entered budget minus cost to date for general conditions, the forecast would have been $380,000 − $236,000 = $144,000, which is $54,000 short of six more months at the current rate.

The forecast matters for revenue too. Using cost-to-cost, percent complete is $1,030,000 ÷ $2,052,000 = 50.2%, and earned revenue is about $1,154,500. Using the stale budget instead, $1,030,000 ÷ $1,950,000 = 52.8%, and earned revenue would be about $1,214,900: roughly $60,400 recognized early. The cost-to-cost post covers the other ways that number misleads.

Margin bridge from bid to projectedMargin at bid of 350,000 dollars falls by 60,000 for the concrete overrun and 54,000 for extended general conditions, rises 12,000 for steel buyout savings, and ends at a projected 248,000.350,000−60,000−54,000+12,000248,000AT BIDCONCRETEGEN CONDSTEELPROJECTEDMARGIN 15.2% AT BID, 10.8% PROJECTED
Every point of fade should trace to a cost code. Illustrative.

Common mistakes

  • Forecasting before the cutoff. Missing invoices make both cost to date and the forecast wrong.
  • Cost to complete = budget − cost to date. It hides every overrun until the line goes over.
  • Ignoring the unit cost you’re achieving. If the first half cost $240/CY, the second half probably won’t cost $210.
  • General conditions at the remaining budget after a slip. Time costs money every month.
  • Folding pending changes into the forecast. Keep them visible so someone chases them.
  • One person’s review. The PM forecasts; someone else should challenge it before it reaches the WIP.

How os.construction handles it

On the Budget’s Forecast tab, forecast to complete defaults to projected budget minus projected costs, and the PM can type their own number on any line; estimated cost at completion and projected over/under sit beside it, with pending changes kept apart from approved numbers. In Accounting, the Job Cost Forecast app takes a dated forecast per cost code, and a different person approves it before it feeds the WIP. See Job cost reports and cost to complete. The free estimate at completion calculator runs the line math in your browser.

FAQ

How do you calculate cost to complete in construction?

Forecast it by cost code from the work that is actually left: open subcontract and purchase order balances, remaining quantities at the unit cost you're really achieving, remaining months of general conditions at the current run rate, plus likely costs not yet committed. Don't back into it as budget minus cost to date.

What is the difference between cost to complete and estimate at completion?

Cost to complete is what's left to spend. Estimate at completion is cost to date plus cost to complete: the total the job is now expected to cost. Comparing the estimate at completion with the revised budget shows the projected overrun or underrun on each line.

How often should you review cost to complete?

A full review by cost code every month, at the period cutoff, because it feeds percent complete, earned revenue and the WIP schedule. Self-performed labor is worth a quick weekly check of hours and quantities, since productivity problems show up there first.

Why does cost to complete affect revenue?

Under the cost-to-cost method, percent complete is cost to date divided by estimated total cost. If cost to complete is understated, the estimated total is too low, percent complete is too high, and the job recognizes revenue and profit it hasn't earned yet.

What is a red flag in a cost-to-complete review?

A cost to complete lower than the open commitments on that line, a unit cost to date well above budget with the forecast still at budget, general conditions forecast at the remaining budget after a schedule slip, and a projected margin that holds steady while cost codes go over.

Share this article

Written by

Charley Forey

Founder of os.construction. We're building the operating system for construction. Project Financials follows job money, and company Accounting covers the books and controls. AI agents remain planned. We write about how money and data actually move on a job, and we build in public.

Apply to the founding program →Start a free trial →
Open file