A foreman writes the day’s installed quantities in the daily log. A payroll clerk codes each worker’s hours to a cost code from the timecards. Divide the quantity by the hours for the same cost code and the same day, and the result is labor productivity in units per labor hour.
The Bureau of Labor Statistics (BLS) defines labor productivity as output per hour, the efficiency with which goods and services are produced from labor hours. BLS measures whole industries, with output counted in inflation-adjusted dollars. On a job the output is a physical quantity, such as square feet of board hung or cubic yards of concrete placed.
The definitions
A 2011 paper in the Alexandria Engineering Journal by Mostafa Shehata and Khaled El-Gohary collects the measures contractors use at the project level, drawing on earlier research by Thomas and colleagues. The authors note that there is no standard definition. They describe these forms:
- Output per work-hour, such as square feet per labor hour. A higher number is better.
- The unit rate, which is the inverse, work-hours per unit of output, where a lower figure means faster work. Estimators usually think in this form, because an estimate multiplies quantity by a unit rate to get hours.
- The performance factor, the estimated unit rate divided by the actual unit rate. A value above 1.0 means the crew beat the estimate.
The paper also describes earned hours. Each unit completed earns the hours the estimate allowed for it, and earned hours divided by actual hours works out to the performance factor. The logic matches earned value in SPI and CPI, counted in hours.
Cumulative productivity, as the authors define it, is all work hours charged to an activity divided by the total quantity installed to date. They say its main use is to judge how the work is going as a whole and to predict the final rate when the activity finishes.
Where each input lives
- Quantity installed. This comes from the daily log. Each entry needs a number and a unit for each cost code the crew worked, such as 1,600 SF of board on level 3.
- Labor hours. These come from timecards coded to the same cost codes as the log.
- Estimated unit rate. Divide budgeted hours by budgeted quantity for the cost code.
The paper describes a research unit at Dundee University that also divides output by available time, which removes unavoidable delays such as meal breaks and weather. Job cost runs on total paid time, because payroll charges the job for every paid hour.
A worked example
The figures below are an example with round numbers. A drywall contractor’s estimate carries 40,000 square feet of board hanging at 0.020 labor hours per square foot, or 50 square feet per labor hour, for a budget of 800 labor hours. A crew of four hangers works eight-hour days, and a fifth hanger joins on days 4 and 9.
The crew ran below the estimate rate through day 10
Square feet of board hung per labor hour, example drywall crew
| Workday | Hangers | Labor hours | SF installed | SF per labor hour | Earned hours | Cumulative SF per labor hour | Cumulative performance factor |
|---|---|---|---|---|---|---|---|
| 1 | 4 | 32 | 1,200 | 37.50 | 24 | 37.50 | 0.75 |
| 2 | 4 | 32 | 1,500 | 46.88 | 30 | 42.19 | 0.84 |
| 3 | 4 | 32 | 1,700 | 53.13 | 34 | 45.83 | 0.92 |
| 4 | 5 | 40 | 2,000 | 50.00 | 40 | 47.06 | 0.94 |
| 5 | 4 | 32 | 1,600 | 50.00 | 32 | 47.62 | 0.95 |
| 6 | 4 | 32 | 900 | 28.13 | 18 | 44.50 | 0.89 |
| 7 | 4 | 32 | 1,700 | 53.13 | 34 | 45.69 | 0.91 |
| 8 | 4 | 32 | 1,800 | 56.25 | 36 | 46.97 | 0.94 |
| 9 | 5 | 40 | 2,100 | 52.50 | 42 | 47.70 | 0.95 |
| 10 | 4 | 32 | 1,600 | 50.00 | 32 | 47.92 | 0.96 |
After 10 workdays the crew has hung 16,100 square feet in 336 labor hours, or 47.9 square feet per labor hour. The estimate allowed 322 hours for that quantity, so the performance factor is 0.96. The activity is 40% complete by quantity and 42% through its hour budget.
Day 6 accounts for most of the shortfall. In the example, the crew waited half the day on a board delivery and hung 900 square feet in 32 hours. The other nine days produced 15,200 square feet in 304 hours, exactly the estimate’s 50 per hour.
Baseline productivity
The paper quotes Thomas’s method for baseline productivity, the rate a crew reaches on days with few or no disruptions:
- Count the workdays that make up 10% of the total.
- Round up to the next odd number, with a minimum of five. Call it n.
- Pick the n days with the highest daily output.
- Take the median, the middle value, of those days’ productivity.
In the example, 10% of 10 days is one, so the minimum of five applies. The five highest-output days are 9, 4, 8, 3 and 7, and the median of their rates is 53.1 square feet per labor hour. At that rate, 16,100 square feet would have taken about 303 hours, so roughly 33 of the 336 hours went to disruption.
The authors also report a critique by Ibbs and Liu, who call the method subjective and point out that the days with the highest output can be the days with the biggest crew. Days 4 and 9 in the example had five hangers.
A related idea, the measured mile, compares an unimpacted stretch of the same job, with the same resources, to the impacted stretch. Shehata and El-Gohary call it the approach courts and boards accept most widely for labor productivity loss claims. A measured mile depends on daily quantities and hours recorded by cost code at the time the work happened.
Forecasting hours at completion
The cumulative rate gives the forecast. At 47.9 square feet per labor hour, the remaining 23,900 square feet take about 499 hours, for 835 hours at completion against the 800 budgeted.
| Forecast basis | SF per labor hour | Hours for remaining SF | Labor hours at completion | Hours over 800 budget | Overrun at $60 per hour |
|---|---|---|---|---|---|
| Cumulative rate through day 10 | 47.9 | 499 | 835 | 35 | $2,100 |
| Days 1 to 10 without day 6 | 50.0 | 478 | 814 | 14 | $840 |
| Baseline (Thomas method) | 53.1 | 450 | 786 | -14 | -$840 |
At an example loaded rate of $60 an hour, the 35-hour overrun costs about $2,100. If the crew holds the 50 per hour it reached outside day 6, the activity finishes at 814 hours. Carry the hour forecast into the project manager’s cost-to-complete on the WIP schedule, so a labor overrun shows up as margin fade in the month it starts.
Your rate next to the national figures
On September 28, 2026, BLS updated its productivity measures for four construction industries through 2025. Productivity fell in all four in 2025: 4.5% in single-family residential, 7.4% in multifamily residential, 12.1% in industrial building and 1.8% in highway, street and bridge construction. Industrial building had grown 21.1% in 2024.
Those four industries account for about 12.5% of construction sector employment, according to BLS. Their hours exclude work done by subcontractors, which BLS counts as a purchased input. BLS also notes that productivity growth estimates are typically lower, often by a wide margin, when subcontractor hours are included. A unit rate by cost code counts physical units and the hours of the crew that installed them, so it carries neither adjustment.
How to start on your jobs
Pick one self-performed activity with a clear unit, such as board hung or concrete placed, on one active job. Have the foreman log quantity by cost code every day, confirm that timecards use the same codes, and compute the daily and cumulative rates each week next to the estimate’s unit rate. After 10 workdays, run the baseline steps to see how much of the gap to the estimate comes from a few disrupted days.