An estimate priced at the start of a job is a snapshot of costs and conditions on the day it was written. Materials move, programmes slip, and instructions change, and none of that waits politely for you to notice. If nobody goes back and checks the estimate against reality as the job runs, the first time anyone finds out it was wrong is at final account, which is the worst possible time to find out.
Reviewing cost estimates during a project isn't about doubting the original price. It's about keeping it true as the job it describes keeps changing, and about knowing where you stand at any point without having to rebuild the numbers from scratch.
Build reviews into the programme
A cost review is not a crisis response, it is routine maintenance. Set fixed points in the programme, monthly or at key stages, where you sit down and compare what was priced against what is actually happening on site. That includes checking material costs against current supplier pricing, checking labour hours against actual progress, and checking provisional sums against what has really been instructed.
Tie the reviews to something that already happens. Many contractors line them up with the monthly valuation, since the measuring is being done anyway. Others pick natural break points: substructure complete, watertight, first fix done. What matters is that the date is in the diary before the job starts, not added when things already feel tight.
What reviewing cost estimates during a project should cover
A good review is not just adding up invoices. It checks whether the original assumptions still hold, whether variations have been priced and agreed rather than just built, and whether contingency has been eaten into faster than the programme suggests it should be. Structuring the review along NRM2 lines makes it far easier to see exactly which part of the job is running hot.
A practical checklist for each review:
- Assumptions. Read back the qualifications and assumptions in the original estimate. Are any of them no longer true?
- Quantities. Compare quantities complete against quantities priced, element by element.
- Rates. Check recent supplier and subcontractor invoices against the rates in the estimate for the same items.
- Variations. Confirm every instructed change has been measured, priced and agreed, not just built.
- Provisional sums. Check what's been instructed against each one, and whether any are close to running out.
- Contingency. Compare how much has been used against how far through the programme you are.
- Forecast. Update cost to complete and compare the total against the current agreed figure.
A review that only happens under pressure is not a review. It's damage control.
Contractual obligations do not pause for a busy week
Most contracts carry clear obligations around notifying cost movement and agreeing variations before they are built. Skipping a review because the job is busy does not remove that obligation, it just means you find out about a breach later, usually when it is harder and more expensive to fix.
Check the notice provisions in the contract you've actually signed rather than assuming. Different forms, and different amendments to them, handle notices and time limits differently, and a regular review is a good moment to confirm nothing has slipped past a deadline.
The mistakes that keep repeating
The same errors turn up job after job:
- Reviewing against the wrong baseline. Comparing costs to the original estimate instead of the latest agreed figure, so approved variations look like overspends.
- Leaving provisional sums alone. Letting provisional sums roll on unchecked until they're spent twice over.
- Reviewing only on request. Treating a review as something to do when a client asks for one, rather than as part of running the job.
- Looking at totals only. A job can be on budget overall while one element is badly over and another hasn't been started yet.
That last one catches people out more than it should. Say the groundworks are well over, but the fit-out hasn't begun, so its budget is untouched. The total looks fine. It isn't. The overspend is real and the fit-out still has to be paid for. Only an element-by-element review shows that.
The same applies to timing. An element that's slightly over budget early in the job is a warning. The same overspend found after the element is finished is a loss. Reviewing cost estimates during a project gives you the warning while there's still work left to adjust, whether that's re-sequencing trades, revisiting a specification with the client, or tightening orders on the elements still to come.
What the review needs to check against
A review is only as good as the estimate underneath it. If the original price was a lump sum with a few trade headings, there's not much to review against; you're comparing actual costs with a guess. If it was measured to NRM2 with quantities and rates for each element, every line can be tested.
That's the kind of estimate we produce: every element measured off your drawings and priced on current UK rates, local to the job, for a fixed fee agreed before we start. See a sample estimate and you'll see the level of detail we'd expect a proper review to be checked against. For reviewing a price before it goes out, rather than during the job, see why estimates need checking before submission.
A rule of thumb for any job on site
If you can't remember the date of the last cost review on a live job, it's overdue. Reviewing an estimate regularly is not extra admin bolted onto a job; it's how you keep control of the number you're responsible for. Catch drift early and it's a conversation. Leave it to final account and it's a dispute.