On a compact urban job, indirects at 25 percent of directs may be fine. On a remote marine terminal with a camp, a barge and a three-year schedule, the same ratio is wrong by a factor of two, and it is wrong in the direction that hurts.
The indirects are usually the least reviewed part of an estimate. The direct cost gets quantity checks, rate challenges and benchmark comparisons. The indirects get a percentage, inherited from the last job, applied in one line, and everyone moves on to argue about concrete prices. Yet on long, remote or phased projects, the indirects can approach or exceed the direct labour, and they are the single most schedule-exposed money in the estimate.
What lives in the indirects
Different industries name them differently: general conditions, site overheads, distributables, preliminaries. The content is the same. Field staff and supervision above the working foreman. The site offices, yards and temporary utilities. The shared construction equipment: the cranes on hire, the manlifts, the site trucks that belong to everyone and no one. The camp, the catering and the rotational travel where the job is remote. And the site services that keep it all running: scaffolding management, consumables, small tools, security, waste, snow clearing in a Canadian winter.
None of that is trivial money, and almost none of it cares how many cubic metres were poured this month.
Time-driven, not quantity-driven
That is the heart of it. Most indirects cost the same per month whether the crews are productive or not. The site office does not get cheaper because the piling went well. The camp feeds the same heads on a rain day. The tower crane invoices its month regardless of what it lifted. These costs are a function of duration, so they have to be estimated from the schedule, and they move when the schedule moves, not when the quantities do.
A percentage of directs gets this exactly backwards. It ties the time-driven money to the quantity-driven money, so the estimate silently assumes that every project has the same tempo. Projects do not have the same tempo. That is the whole problem.
Build them from the schedule
The build-up is not sophisticated. It is a small number of honest lists, each priced against the calendar.
- Staff: an organisation chart by month, priced at loaded rates. Not a headcount average, a chart, because the superintendent arrives before the crews and the quality lead stays after them.
- Temporary facilities: mobilise, monthly, demobilise, taken from the site plan. Offices, laydown, hoarding, temporary power and water.
- Construction equipment: the shared spread by month, from the method statements. What arrives, when it leaves, and what it costs to sit between picks.
- Camp and travel: heads on site by month, times cost per head-day, plus the rotation flights that a spreadsheet average always underestimates.
- Site services and consumables: from history, but converted to monthly rates, because they follow the calendar too.
A worked example, in both directions
Take the remote marine terminal, with direct costs of, call it, 30 million dollars over a twenty-month schedule. The habit says 25 percent, so 7.5 million. Now build it. The staffed organisation chart averages fourteen people at a loaded 16,000 dollars a month each: about 4.5 million across the job. Temporary facilities run 40,000 a month plus 300,000 to mobilise and demobilise: about 1.1 million. The shared equipment spread, barge included, prices at 120,000 a month: 2.4 million. The camp and rotational travel for the peak crews: about 2.6 million. Site services and consumables: call it 900,000. The schedule-built total is roughly 11.5 million, or 38 percent of directs. The percentage estimate just missed by four million dollars, and every dollar of the miss is time money that will be spent whether anyone budgeted it or not.
Now run the same discipline on a compact fourteen-month urban job with good access, no camp and a lean site team. The build-up comes to 18 percent. The bidder who priced 25 percent by habit is seven points fat on a competitive tender, and loses to the bidder who did the arithmetic.
The percentage is not conservative or aggressive. It is uninformed, and it misses in whichever direction the project happens to differ from the last one.
Indirects are what a month costs. Directs are what the work costs.
Where this goes wrong
The inherited ratio. Last project’s percentage arrives with none of last project’s context: its schedule, its access, its camp or its market. The number survives because deriving the real one takes a day, and the estimate is due Friday.
The flat staffing plan. An average headcount across the duration, ignoring that staff mobilise on a curve, peak with the crews and demobilise slowly. The average understates the peak months and overstates the tails, and the cash flow built from it is fiction.
The double-counted equipment. The crane priced inside the direct crew rates and again in the indirect spread. Or worse, in neither, with each estimator sure the other one had it. The method statements decide where shared equipment lives; write it down.
The frozen indirects. The schedule slips three months at sanction, the directs get updated, and the indirects ride through unchanged because they are “a percentage”. Every slipped month on the remote job above costs roughly 575,000 dollars in time money. Unpriced slip is how contingency gets eaten from the inside.
Why it matters at the end
When the schedule slips six months, an estimate with time-driven indirects can price the slip in an afternoon: months times the monthly burn, plus the extended equipment and camp. An estimate built on a percentage cannot price it at all, because the percentage was never about time.
The same arithmetic carries into disputes. Delay claims are argued largely in time-related costs, and the quantum comes straight from the monthly build-up: this staff, this spread, this camp, for this many additional months. An estimator who built the indirects from the schedule has already written the backbone of that analysis. One who wrote “25%” has homework to do, under deadline, with lawyers watching.
Indirects reward the same discipline as everything else in estimating: name the thing, put it against the calendar, and let the total be whatever the arithmetic says. The percentage is a summary of the answer, never a way to find it.
Emerald Group builds schedule-based indirect estimates for owners, engineering firms and contractors, and reviews indirects on estimates prepared by others. If your project is long, remote or phased and the indirects line is one percentage, get in touch.