Earned hours are the budgeted man-hours for the work that was actually completed; actual hours are the man-hours the crew really spent, taken from the timesheet. Labour efficiency is the ratio between them: earned divided by actual. If an item was estimated at 100 man-hours and is 60% physically complete, it has earned 60 hours — regardless of whether the crew spent 45 hours or 90 getting there. That gap between earned and spent is the entire story of whether the job is making or losing money on labour.
The three numbers, untangled
Budgeted hours — the man-hours the work was priced at: the labour base of the contract, distributed across the BOQ items. This is fixed at the start and only changes through approved variations.
Earned hours — budgeted hours × physical progress, item by item. Earned hours grow only when work is physically completed. They cannot be earned by being busy.
Actual hours — what the timesheet says the crew spent, including the hours that produced nothing because the workface was blocked, the materials were late, or the permit never came.
Efficiency = earned ÷ actual. At 100% the job is running exactly as estimated. The number is only honest if earned and actual come from independent records: progress measured against the scope, hours recorded on the timesheet. Derive one from the other and the ratio collapses to a flattering constant.
Why hours spent tell you nothing on their own
A timesheet answers “what did the crew cost this week?” It cannot answer “what did the crew produce?” A 46-person crew burns roughly the same hours in a brilliant week and a blocked one. Judged by hours alone, every week looks like progress — which is exactly how a job arrives at 80% of budget spent with 55% of the work done, and everyone is surprised.
The reverse failure is just as common: progress reported as a percentage of hours consumed. “We’ve used 70% of the hours, so we’re 70% complete.” That is not a measurement; it is the assumption that efficiency is perfect, restated. Physical progress must be measured against the scope — quantities installed, systems completed — by people who see the work.
What weekly measurement changes
Measured monthly, a bad efficiency number is an autopsy: the hours are spent and the month is gone. Measured weekly, it is a diagnosis. Three habits make the weekly number worth having:
- Close every week into a numbered record. Earned hours, actual hours, efficiency — written down, per week, per section. Trends only exist if the weeks are comparable.
- Log production impacts with hours. Downtime, blocked workfaces, late drawings — recorded as events with an hours figure attached. The gap between earned and actual then splits into an explained part and an unexplained part, and those demand different conversations: one with the client, one with the crew.
- Keep extra work separate. Hours spent on unpriced extra work, silently booked against contract items, poison the efficiency of both. Separate items, separate earning.
The weekly efficiency figure is also the honest core of the progress report the client sees — and, when downtime caused by others is logged with hours, the evidence base for the claims that follow.
Frequently asked questions
What is a good labour efficiency for a subcontractor?
Efficiency is measured against your own estimate, so 100% means the work is going exactly as priced. Sustained values above 100% mean the estimate was conservative or the crew is genuinely outperforming it; values drifting below mean the job is burning more hours than it earns. The absolute number matters less than the trend and whether the variance is explained.
Is labour efficiency the same as productivity?
They are closely related but not identical. Productivity is output per unit of input (metres of cable per man-hour); efficiency compares earned hours to actual hours, so it already contains the estimate. A crew can be highly productive and still show poor efficiency if the estimate was aggressive — which is itself useful information.
How do you calculate earned hours without labour norms?
Distribute the total budgeted man-hours of the contract across BOQ items — by value, by tender build-up, or by experience — and let physical progress earn them: 40% progress on an item earns 40% of its budgeted hours. The distribution does not need to be perfect. It needs to be fixed, so week-to-week movement is real.
Is earned value tracking overkill for a small subcontractor?
The full EVM apparatus of a megaproject is. Earned hours are not: one number per week, produced from records you already keep — measured progress and timesheets. A subcontractor who knows their weekly earned-versus-actual position usually finds out about a losing job months before the accounts do.