Construction cost overrun reporting an owner gets arrives last, usually two to four months behind the site. That is not concealment. It is the shape of how cost information moves through a project, and it is correctable at the reporting schedule rather than at the relationship.
The order people find out
A trade foreman knows first, on the day the work does not match the drawing. A superintendent knows within the week. A project manager knows when the request for information comes back and the answer implies a change.
The contractor project executive knows when the change order is priced, which may be a month later. The owner knows when it appears in a monthly report, which may be a month after that. A lender or an investor knows at the next quarterly report, which may be two months after that.
By the time the figure reaches the party whose money it is, the work has often been performed, because the alternative was stopping the job over an amount nobody had authority to argue about yet.
Nobody in that chain did anything wrong. Each step is a person waiting for the previous step to produce something reliable, and reliability takes time. The cumulative delay is the problem, and it is a design feature of the sequence rather than a failure within it.
Why construction cost overrun reporting to an owner lags
Three mechanisms, and they compound.
Pricing takes time. A change is identified, quoted by a subcontractor, marked up, reviewed and issued. That is routinely three to six weeks, and during it the item exists as a known problem with no number attached, which means it appears in no report. A project can carry a dozen of these at once and report none of them.
Reports show authorized cost. Most formats report executed change orders. Pending ones appear as a note if at all, and disputed ones frequently do not appear until resolved. The report is accurate and describes a smaller number than the project actually carries.
Allowance overages are not change orders. When an allowance item resolves above the carried figure, the difference is usually reconciled against contingency rather than issued as a change order. It moves cost without appearing in the change order total, which is the line everybody watches.
Add the three and an owner can be looking at a report showing $600,000 of change orders on a project that already carries $1.6 million of committed and probable movement.
The three lines that close the gap
Pending and potential change orders, with values. Every item identified but not yet executed, with an estimated value even where that estimate is rough and labeled as rough. This single line typically doubles the visible number and is the largest available improvement to any reporting format.
Allowance resolution against the carried figure. Each allowance closed, at what, against what was carried. The running variance predicts the remaining items better than any other available indicator, because allowances on a job tend to move in the same direction for the same reasons.
Remaining open exposure as a range. Built from the unresolved allowances, the uncapped exclusions, the unbought packages and the outstanding owner decisions. This is the forward looking line and the one no standard format contains.
All three come off documents the project team already maintains. None of them requires new work, and the request lands as a format preference rather than as a challenge if it is made at the start.
What each change is attributed to matters as much as the total, which is the ground covered in the cause written on a change order.
Why the delay is expensive rather than merely annoying
Late information removes options rather than just comfort.
A change identified before the work is performed can be questioned, priced competitively against an alternative, or deleted. The same change reported after the work is in place can only be paid for. The commercial position has evaporated and what remains is an accounting exercise.
Late information also removes the chance to change behavior. If owner directed changes account for 60 percent of movement and the owner learns this at month eighteen, the decisions that produced it have all been made. Learning it at month six changes what the next twelve months look like.
The third cost is credibility. An owner who reports quarterly to a board or an investor and is surprised twice stops being trusted on the third occasion, regardless of the underlying performance. The reporting failure becomes a governance failure one level up.
None of this argues for more reporting. It argues for three additional lines in the existing one, which is a different and much cheaper proposition.
It is worth being clear about what more reporting would cost, because the instinct after a surprise is to ask for it. A weekly cost report on a project that produces reliable numbers monthly is three additional weeks of estimates presented as facts, and it degrades the quality of the monthly one by consuming the time that made it reliable.
The useful change is not frequency. It is completeness at the same frequency, and specifically the inclusion of the items that are known and not yet priced. Those are the ones with commercial options still attached, which is the only category where earlier knowledge changes an outcome rather than merely the mood in a meeting.
A worked example
Illustrative figures. Not taken from any client project and not a quotation.
A $57 million project at month eleven. The monthly report shows executed change orders of $740,000 against a contingency of $2.4 million. The owner reports comfortable coverage to its investors.
The same project, with the three lines added. Pending and potential change orders $610,000 across nine items, four of them already built. Allowances resolved: six of twenty, averaging 19 percent above carried, with fourteen remaining carrying $3.9 million. Remaining open exposure $1.4 million to $2.9 million.
Committed and probable movement is therefore around $1.7 million, not $740,000, and the remaining allowances at the established 19 percent pattern imply another $740,000 on their own.
Nothing was concealed and every figure in the original report was correct. The project simply reported the one number that was fully settled and omitted the three that were not, and the gap between those two pictures is the entire contingency.
What to require in the reporting schedule
- Pending and potential change orders listed with estimated values, however rough, every month.
- Allowance resolution reported against the carried figure, with a running average variance.
- Remaining open exposure as a range, at least quarterly, with the three largest items named.
- Change orders split by cause with a running total for each cause since the start.
- The change order log required by the construction contract rather than requested informally.
- A standing instruction that a known item is reported before it is priced, not after.
- All of the above agreed at contract signature and written into the reporting schedule.
Item six is the one that changes the timing rather than the content. A project that reports a problem the week it is identified, with no number attached, gives the owner two months it did not previously have. Keeping that ledger current through the months after signature is what Costwitness was built to do.
What we do
We read the documents before signature so the reporting requirements are written into the contract rather than requested afterwards, and we set the baseline the later reports are measured against. The output names every open exposure with a range and a page reference, which is what makes a later movement attributable to a cause. The reading is the cost and change exposure assessment.
Questions people ask
Why do change order totals understate what a project is carrying?
Because most formats report executed change orders only. Items identified but not yet priced appear nowhere, pricing routinely takes three to six weeks, and allowance overages are usually reconciled against contingency rather than issued as change orders. All three move cost without moving the reported total.
Is the delay deliberate?
Almost never. Each step in the chain is a person waiting for the previous step to produce a reliable number, and reliability takes time. The delay is cumulative rather than intentional, which is also why it responds to a reporting format change and not to a conversation about trust.
What is the single most useful line to add?
Pending and potential change orders with estimated values, including items not yet priced. It typically doubles the visible movement, it comes off a log the project team already keeps, and it converts a report that describes the past into one that describes the next two months.
This is general information about construction contracts and is not legal advice.