What unfilled labor really costs your plant, and how to calculate it

Monday, 6:40 AM. You are standing at the whiteboard in the front office looking at six open squares on the shift grid.

You already know how today goes. You will call four people who are technically off and ask two of them to stay over. Line three runs at reduced rate. Your best setup tech, the one who has been there nine years, works his fourth Saturday in a row and says nothing about it, which is worse than if he complained.

At some point this week somebody will ask what the open positions are actually costing, and the honest answer is that nobody in the building knows.

So let’s fix that. Not with a statistic. With four numbers you already have on your own floor.

Fair warning: when we ran this model on a 140-person plant, our number came out lower than the one our whole industry likes to quote. We are showing you anyway, because a number you can defend beats a number that sounds impressive.

The $500-a-day problem

You have seen the figure. An unfilled position costs $500 a day. It is in staffing agency decks, HR software landing pages, and roughly half the content ranking for this topic.

We went looking for the study.

The oldest source anyone can actually reach is a 2016 article on Dice containing this sentence: “CEB (formerly Corporate Executive Board) estimates the cost of an open position at $500 per day.” That is it. CEB was acquired by Gartner in 2017. If a study exists behind that estimate, it is not published anywhere you can read it.

There is a second problem, and this one is a straightforward factual error made by pages currently ranking on page one for this exact search.

SHRM’s $4,129 figure is cost per hire. It measures what you spend to recruit somebody: advertising, screening, recruiter time, onboarding. It is fiscal year 2015 benchmarking data. SHRM’s more recent article puts it near $4,700.

Cost of vacancy is a different thing entirely. It measures what the empty seat costs you while it is empty. Overtime, lost output, quality, turnover.

Two pages ranking for “cost of unfilled positions” right now quote SHRM’s cost-per-hire number as if it were a vacancy cost, one of them as “$4,129 every 42 days” and the other as “$4,129 to $5,733 per month.” Those are not vacancy costs. They are recruiting budgets wearing a different hat.

Nobody starts a business because they love spreadsheets. But if you are going to defend a staffing line to a CFO, you need a number that survives one follow-up question. Here is how to build one.

The four costs of an empty seat

Every dollar an unfilled position costs you shows up in exactly four places. Three are on a report you already run. The fourth is the one that gets people.

A note on line one that saves people from overstating their case. The straight-time wages for those hours were always in the budget, because the work was always going to happen. The delay costs you the premium, not the whole wage. Counting the full wage doubles your number and gets you dismissed in the first five minutes of the meeting.

Line two is the one most models skip. Overtime does not deliver an hour of output for an hour of pay. BLS studies going back to the 1940s put productivity at 92% for a 50-hour week, 82% at 60 hours, and 78% at 70 hours. The National Electrical Contractor Association (NECA) conducted a 1962 survey that found six 10-hour days running at 84%. The Construction Industry Institute (CII) found that moving from a five-day to a six-day week costs 10% to 15% in efficiency. Different studies, different decades, same shape of curve.

Line four is where an unfilled seat stops being a linear cost. Turnover is contagious in a measurable way. Felps and colleagues, publishing in the Academy of Management Journal, found across 45 bank branches and more than a thousand departments that coworkers’ visible job-searching predicts an individual’s decision to quit, over and above every individual-level predictor. That research was done in banking and hospitality, not on a plant floor, and it does not give you a multiplier. But anyone who has watched two people leave in one month knows the mechanism is real.

And there is a cost we are deliberately not going to put a dollar figure on. Dembe and colleagues, in Occupational and Environmental Medicine, studied 110,000 job records over 13 years and found that jobs with overtime carried a 61% higher injury hazard rate than jobs without. With 12-hour days, 37% higher. For 60-hour weeks, the hazard rate is 23% higher. OSHA reports injury rates 18% higher on evening shifts and 30% higher on nights.

We are not going to monetize that for you. If somebody gets hurt on hour 11 of a Saturday your crew should not have been working, the cost is not a line item.

Running the numbers on a 140-person plant

Here is the model on a real-shaped operation. Every assumption is labeled. Swap in your own.

The plant: 140 production employees, six unfilled positions, eight weeks. Base rate $22 an hour, so overtime runs $33. The line produces 12 units per labor hour at a $4 contribution margin.

Six positions at 40 hours is 240 unfilled labor hours a week. Now break it down.

1. Overtime premium. You cover 60% of the gap, 144 hours, with overtime. The premium is $11 an hour. 144 × $11 = $1,584 a week

2. Productivity you paid for and did not get. Spreading 144 overtime hours across the crew puts your covering group at roughly a 50-hour week, which the BLS curves say runs at 92% productivity. Eight percent of those hours, about 11.5 of them, produce nothing. 11.5 × $33 = $380 a week

3. Output you never made. The other 40% of the gap, 96 hours, simply does not get covered. 96 × 12 units × $4 = $4,608 a week

4. Attrition you caused. Say two of your 140 leave over those eight weeks because of the schedule. At $22 an hour, annual wages run $45,760. Center for American Progress research, a meta-analysis of 31 case studies, puts replacement cost at about 20% of salary for workers under $50,000, so roughly $9,152 each. $18,304 ÷ 8 weeks = $2,288 a week

Weekly total: $8,860. Across eight weeks: $70,880.

Divide by six positions and 40 working days and you get $295 per unfilled position per day.

Which is a good deal less than $500. We could have picked assumptions that got us over $500 without straining, and every one of those choices would have been defensible on its own. That is exactly why the borrowed number is worthless: the answer is entirely a function of your margin, your rate, and how much of the gap you cover. A plant with a $12 contribution margin lands somewhere very different from one with a $4 margin.

Run yours. The output is a number you can put in front of a CFO and survive the follow-up question.

When temp labor is not the answer

We place about 250 people a week. We are still going to tell you the cases where calling us is the wrong move.

Skip the temps when the gap is under about three weeks. Screening, orientation, and the learning curve eat most of the value. Overtime is genuinely cheaper for a short bridge.

Skip them when the role takes more than a few weeks to become productive. A CNC programmer or a certified welder on a critical weld path is a search, not a fill. Direct hire is the right instrument.

Skip them when your supervisors have no capacity to onboard. Temps who get a badge, a vest, and a vague gesture toward the line will underperform and leave, and you will conclude staffing does not work. What did not work was dropping people into a building with nobody assigned to them.

Temps are the right answer when the work is learnable in days rather than months, when the gap runs six weeks or longer, when your core crew is already past 45 hours, or when you want to evaluate people before committing to permanent headcount. That last one is why temp-to-hire exists.

How to take this to your CFO

The mistake is walking in with a complaint. Walk in with a variance.

1. Lead with the weekly number, not the annual one. $8,860 a week is a decision. $460,000 a year sounds like a scare tactic and invites an argument about your assumptions.
2. Show the four lines separately. A CFO will trust a model they can attack one line at a time far more than a single headline figure.
3. Put the alternative next to it. Six temps at a bill rate against $8,860 a week of vacancy cost is a comparison. A staffing invoice on its own is just an expense.
4. Name what you did not count. Say plainly that you left injury exposure and quality escapes out because you could not price them honestly. That one sentence is what makes the rest of the model credible.

Two pieces of context worth bringing with you. BLS JOLTS data for July 2026 shows durable goods manufacturing carrying a 5.2% job openings rate, the highest of any major sector, while manufacturing’s quits rate sits at 1.4%, below the 1.9% all-industry figure. Plants are not bleeding people faster than everyone else. They cannot fill the seats they have. That reframes the whole conversation away from retention and toward supply.

And locally: per the Utah Department of Workforce Services, Utah manufacturing grew 0.6% year over year in July 2026 while construction grew 3.0% and professional services grew 4.5%. Your competition for an hourly worker is not the plant across the valley. It is a construction crew that pays more.

What it looks like when the grid is full

Picture the same Monday, eight weeks from now. The whiteboard has no open squares. Line three runs at rate. Your nine-year setup tech works five days and goes home, and the Saturday conversation does not happen.

Nothing dramatic changed. You ran a four-line calculation, took a number to a meeting, and got a decision made in one conversation instead of three.

The plants that are not short in November are not the ones with the biggest budget. They’re the ones that stopped arguing about the staffing line and started measuring the cost of not having one.

Frequently Asked Questions

How do you calculate the cost of an unfilled position?

Add four numbers: the overtime premium you pay to cover the gap, the productivity lost on hours worked past 40, the output you never produced valued at your contribution margin, and the replacement cost of anyone who quits because of the workload. Count only the overtime premium rather than the full wage, because the straight-time hours were always going to be paid. The result is specific to your margin and your wage rate, which is why a borrowed per-day figure is not usable.

What is the difference between cost per hire and cost of vacancy?

Cost per hire is what you spend to recruit and onboard someone: advertising, screening, recruiter time, and orientation. SHRM benchmarking puts it near $4,700. Cost of vacancy is what the empty seat costs you while it stays empty, through overtime, lost throughput, quality, and turnover. They measure entirely different things, and several widely read articles quote SHRM’s cost-per-hire figure as though it were a vacancy cost. If someone quotes you a vacancy cost in the low thousands per month, check which one they actually measured.

Does an unfilled position really cost $500 a day?

There is no traceable study behind that number. The oldest reachable source is a 2016 article attributing it to CEB, a research firm acquired by Gartner in 2017, and no published CEB or Gartner document containing the figure can be located. Actual vacancy cost depends on your contribution margin, wage rate, and how much of the gap you cover with overtime. A worked example on a 140-person plant at a $4 contribution margin produced about $295 per position per day, and a plant with a higher margin would land well above $500.

Is it cheaper to pay overtime or hire a temp worker?

For gaps under about three weeks, overtime is usually cheaper, because screening and the learning curve eat most of the value of a short assignment. Past roughly six weeks the math flips, for three reasons: the overtime premium is a permanent 50% surcharge on every hour, productivity at a 50-hour week runs about 92% and at 60 hours about 82% per BLS studies, and jobs with overtime carry a 61% higher injury hazard rate per Dembe and colleagues in Occupational and Environmental Medicine. The hybrid answer that works for most plants is a small temp buffer used to cap overtime at a sustainable level rather than eliminate it.

How much does it cost to replace an hourly manufacturing worker?

Roughly 20% of the worker’s annual salary, per a Center for American Progress meta-analysis of 31 case studies. At a $22 hourly rate, which is about $45,760 a year, that is roughly $9,150 per departure in recruiting, training, and lost productivity. That research dates to 2012 and draws on studies published between 1992 and 2007, so treat it as a floor rather than a current figure. The number does not include the productivity ramp of the replacement, which in a skilled production role can run several months.

What percentage of manufacturing jobs are unfilled right now?

Manufacturers reported an average of 4.13% of jobs unfilled in the NAM Manufacturers’ Outlook Survey for the first quarter of 2026, with nearly one in four reporting 5% or more. BLS JOLTS data for July 2026 puts the manufacturing job openings rate at 4.4%, and durable goods at 5.2%, the highest rate among the major sectors listed. Notably, manufacturing’s quits rate of 1.4% is below the 1.9% all-industry rate, which means the problem is filling positions rather than keeping people.

Does mandatory overtime increase workplace injuries?

Yes, and the effect is well documented. Dembe and colleagues, analyzing 110,236 job records over 13 years in Occupational and Environmental Medicine, found jobs with overtime carried a 61% higher injury hazard rate than comparable jobs without, with a clear dose-response relationship: 12-hour days raised the hazard rate 37% and 60-hour weeks raised it 23%. OSHA separately reports injury rates 18% higher on evening shifts and 30% higher on night shifts compared with day shifts.

How long does it take to fill a manufacturing job?

There is no reliable published benchmark specific to manufacturing, since the main public series tracking vacancy duration is no longer produced. SHRM’s 2025 benchmarking puts median time-to-fill across all roles at roughly 45 days. In practice, an hourly production or material handling role with an approved wage and defined screening requirements moves far faster through a staffing partner with an active local pool, often in days rather than weeks, while a skilled or certified role behaves like a search and should be planned as one.

Ready to run this on your own floor?

Want help running this on your operation?

We will walk your floor, build the four-line model with your actual numbers, and tell you honestly whether temporary staffing is the right answer for your gap. We fill about 250 positions a week across the Salt Lake Valley, Utah County, and Davis and Weber counties.