A GC wins a $4.8M job on a clean quantity takeoff. Every square foot of board, every linear foot of pipe, every door is counted and priced. By week six, the drywall and framing crews are 22% over their labor budget, and the PM is explaining to the owner why the schedule is slipping.
The quantities were right. The hours were a guess.
That gap is what effort estimation closes. In contractor terms, effort estimation is the process of predicting the labor hours and crew-days required to put your measured quantities in place. This guide covers the labor-hour reality on a job site: where the numbers come from, how they break, and how to fix them before bid day.
Why Your Quantities Can Be Right and Your Bid Still Wrong
Labor is where bids bleed. Labor commonly runs 30% to 50% of installed cost, depending on the trade. A 10% miss on labor hours in a labor-heavy trade like finish carpentry or drywall can wipe out your entire net margin on that scope.
The broader productivity picture is worse. McKinsey's 2017 *Reinventing Construction* report found that construction labor productivity grew about 1% a year over two decades, against 2.8% for the total world economy. Rework adds to the damage: the Navigant Construction Forum's study of 179 projects put direct rework at 4% to 6% of project cost, and every one of those hours is an hour nobody carried in the bid.
Quantity Is Measured, Effort Is Predicted
A construction takeoff counts things that exist on the drawings. If you measure 18,000 SF of wall, it is 18,000 SF whether the estimator is you or the guy down the hall. Two careful people land within a percent or two.
Effort is different. The hours to hang and finish those 18,000 SF depend on crew skill, floor level, layout, sequencing, and who else is in the space. Two careful estimators can differ by 30% on hours, and both can defend their number. That is why effort carries far more uncertainty than quantity, and why it deserves more of your review time.
What Effort Estimation Means on a Job Site vs. in a PM Textbook
Software project managers measure effort in story points and person-months, using techniques like Delphi and planning poker. Budget-level cost estimation works in estimate classes and dollars per square foot. Both stop short of the crew-level math.
On a job site, effort is concrete. You need crew size, duration, a productivity rate, and the conditions the crew will work in. Everything else is derived from those four inputs.
The Core Formula: Quantity × Productivity Rate × Condition Factors
The working math is simple:
Labor hours = Quantity × Productivity rate (hours per unit) × Condition factors
Take 18,000 SF of 5/8" drywall, hang and finish to Level 4. Assume a hang rate of 0.011 hours/SF and a finish rate of 0.016 hours/SF. That is 0.027 hours/SF, or 486 base hours.
Now apply conditions. Upper-floor access and material hoisting add 5%. Trade stacking with MEP rough-in and ceiling grid adds 8%. Combined, that is a 1.134 multiplier, so 486 × 1.134 gives you about 551 labor hours.
With a four-person crew working 8-hour days (32 crew-hours a day), that is roughly 17 crew-days. At a $62/hour loaded rate, labor comes to about $34,200, or $1.90/SF. Change any one input and you can see exactly how the bid moves. For turning hours into dollars by trade, see our guide to labor unit costs.
Where Productivity Rates Come From
You have three real sources, and none is complete on its own.
Published data like RSMeans gives you a defensible, neutral starting point. The blind spot is that it reflects an average crew on an average job, which is rarely your crew on your job. Your historical job cost data reflects your actual people, your supers, and your markets, but it is only as good as your cost-code discipline. If field crews bury drywall hours under "general labor," your history is useless.
Sub input is the third source, and it is the most current. Subs know what their crews produce this month. The blind spot is incentive: a sub trying to win the job will quote an optimistic rate, and one protecting margin will pad it.
The best practice is to use published data to set a baseline, your history to adjust it, and sub input to challenge both.
Condition Factors That Quietly Move the Number
Condition factors are where most estimates get lazy. Typical adjustments look like this:
- Floor level and vertical access: 3% to 10% for upper floors without a dedicated hoist
- Restricted or occupied space: 10% to 25% for work in occupied buildings or tight phasing
- Weather exposure: 5% to 20% for exterior trades in extreme heat, cold, or wet seasons
- Trade stacking: 5% to 15% when multiple trades share the same floor
- Sustained overtime: a 50-hour week held for several weeks typically costs 10% to 15% in lost productivity per hour worked, a range studies have kept confirming since the Business Roundtable's 1980 report on scheduled overtime
Rework belongs here too. If your detail set is incomplete or RFIs are open at bid time, carry a rework allowance in hours, not just a vague contingency line.
Effort Estimation Techniques GCs Actually Use
Most effort estimation techniques from PM theory have a construction equivalent. The question is which fits the bid in front of you.
Bottom-Up Estimating by Work Package
Bottom-up estimating breaks a scope into work packages, then into tasks, and builds hours from the ground up. For drywall, that means hang, tape, finish, and sand by area type, each with its own rate and conditions.
It is the right method for lump-sum bids because it is the most defensible and the easiest to audit. The cost is estimator time. A full bottom-up labor build on a mid-size job can add 10 to 20 hours to your bid prep, which is why teams skip it when they are stretched.
Analogous and Parametric Estimating for Early Budgets
Analogous estimating borrows the hours from a similar past project. Parametric estimating scales them using a unit, like 0.9 labor hours per SF for a light commercial interior or 85 hours per unit for a multifamily trade package. Both are fast and work well for preconstruction budgets.
They break down when you bid from them. A parametric number cannot see that this job has 11-foot ceilings or a tight laydown area. For a data-driven version of this approach, see our guide to predictive cost estimating in construction.
Three-Point Estimating for Uncertain Scopes
Three-point estimating suits scopes with unknown conditions, like a renovation where you cannot see behind the walls. You estimate an optimistic (O), most likely (M), and pessimistic (P) number, then weight them: (O + 4M + P) ÷ 6.
Say you are estimating demo and patch of existing plaster walls. Optimistic is 40 hours, most likely is 60, and pessimistic is 120 because of possible lath damage and hidden asbestos abatement coordination. The weighted result is (40 + 240 + 120) ÷ 6, or about 67 hours.
Notice that the weighted number sits above your "most likely." That is the point. It forces the downside into the bid instead of hoping it does not show up.
Labor Hour Estimation: A Worked Example on a 40-Unit Multifamily Job
A GC estimating a 40-unit apartment building in Fargo might see a 25% spread between the lowest and highest drywall bids, and half that spread is scope and productivity assumptions, not rate. One sub assumes a dedicated hanging crew and a separate finish crew. Another assumes one crew doing both. Neither is wrong on paper, but they imply very different hours.
A Cincinnati-based estimator put it this way: "I stopped comparing bid totals years ago. I compare hours per unit. If two subs are 25% apart on hours per unit and the same on rate, one of them is wrong about the building."
Reverse-Engineering a Sub's Number into Hours
Start with each sub's total, back out material, then divide the remainder by an assumed loaded billing rate. Suppose the material for board, mud, tape, and fasteners is about $62,000.
Sub A bids $212,000. That leaves $150,000 for labor and overhead. At $60/hour, that is 2,500 hours, or 62.5 hours per unit. With a six-person crew at 48 crew-hours a day, that is about 52 crew-days.
Sub B bids $268,000. That leaves $206,000, or roughly 3,433 hours, which is 86 hours per unit and about 72 crew-days. The $56,000 gap between them is almost entirely about hours, not rate.
Checking the Result Against Your Own History
Now compare those numbers with your own job history. Say your last three comparable three-story walk-ups ran 78 to 84 hours per unit.
Sub A is about 23% leaner than your history. That is either a smarter crew, a different scope, or a bid they will try to recover through change orders. Sub B is slightly above your range, which is easier to live with. Your options are to adjust (add hours or a buyout allowance to Sub A's number), clarify scope (ask both subs to confirm finish level, hang method, and number of mobilizations), or walk. If Sub A cannot explain its crew plan in a five-minute call, treat the low number as risk, not savings.
Construction Labor Productivity: Why Your Rates Drift and How to Recalibrate
Rates in an estimating database do not stay right. Materials change: lightweight board, new adhesives, and prefabricated panels all shift hours. Labor changes too, as experienced tradespeople retire and crews get newer.
The long-term trend is not encouraging. Paul Teicholz's analysis of Bureau of Labor Statistics data showed construction labor productivity declining over several decades while non-farm industries gained. Even if you question the method, the signal is clear: you cannot assume yesterday's rate describes today's crew.
Close the Loop with Field Actuals
The fix is a routine, not a software purchase. At project closeout, the PM pulls actual hours per cost code and compares them to the estimate. The estimator owns updating the database, and the super reviews the explanations for anything off by more than 10%.
Then run a 30-day check on the next job. Early actuals, once a crew has built a rhythm, tell you whether your updated rate holds. Without that check, you often fix last job's problem and miss this job's.
Get the Superintendent into the Review
Your superintendent will spot a bad crew-day number faster than anyone in the office. The problem is nobody sends them the estimate before bid day.
Make it a 20-minute review. Spend 5 minutes on the schedule and crew-day totals by trade, 10 minutes on the three largest labor lines and the rates behind them, and the last 5 on site conditions the estimate assumed: access, staging, working hours, and weather exposure. Ask one question for each line: does this crew size and duration match how you would actually run the work?
Write the super's changes into the estimate with a note on why, so the next estimator inherits the reasoning, not just the number.
This is where historical data pays off. Struvia learns from your company's past jobs, bids, and estimates, so its pricing reflects your own history and gets sharper with every job you feed back in. New accounts start with 600 free credits, no card.
Frequently Asked Questions
What is effort estimation in construction?
Effort estimation is predicting the labor hours and crew-days needed to install your measured quantities. Quantity is measured from the drawings; effort is predicted from productivity rates and adjusted for site conditions, crew experience, and schedule.
How do you calculate labor hours for a construction estimate?
Multiply the quantity by a productivity rate in hours per unit, then adjust for condition factors such as access, height, weather, and crew experience. Do it by work package rather than by trade total, so each rate matches the work it describes.
Where do construction productivity rates come from?
The best source is your own field actuals by cost code from closed-out jobs. Published cost databases and sub quotes are useful starting points and cross-checks, but they describe an average crew, not yours.
Which effort estimation technique should a GC use?
Use bottom-up estimating by work package once you have drawings and quantities. Use analogous or parametric estimating for early budgets, and three-point estimating for scopes with real uncertainty, like renovation work behind existing walls.
How do you keep labor productivity rates accurate?
Compare actual hours to estimated hours by cost code at every closeout, update the database, and have the superintendent explain anything off by more than 10%. Then check early actuals on the next job to confirm the updated rate holds.
Accurate quantities only get you halfway to a good bid. The other half is a labor-hour estimate built from your own rates, adjusted for the job in front of you, and reviewed by the person who will run it. To spend less time on quantities and more on effort, upload a plan set free and let Struvia handle the takeoff.
*Reviewed by Baylor Jeppsen, Construction Estimating Expert and Founder of Struvia.*
