Employer guideRead at source, 8 September 2026
The Cost of a Bad Hire: Why the Famous Number Is Not Real, and What to Use Instead
Almost every article on this subject opens with the same claim: that the U.S. Department of Labor puts the cost of a bad hire at 30% of the employee's first-year earnings. We went looking for that document. Every trail led to another article citing another article, and none of them to a Department of Labor publication. We are not going to repeat a figure we cannot source. What follows instead is the arithmetic — six costs a bad hire actually generates, all of which sit in records you already have — plus a calculator that turns your own numbers into a figure you can defend in a meeting.
Why we will not give you a headline figure
There are three numbers in circulation on this topic, and each has a problem.
| The claim | Where it is usually attributed | The problem with using it |
|---|---|---|
| A bad hire costs at least 30% of first-year earnings | The U.S. Department of Labor | The citation trail runs through staffing and HR articles rather than to a Department of Labor publication. We could not locate a primary source. |
| A bad hire costs about $15,000 | An employer survey from the 2010s | Self-reported estimates from survey respondents, not measured costs, and a decade out of date on wages. |
| A bad executive hire costs several times salary | Consulting firm research | Even where the method is sound, it is about senior hires. It says nothing about a $19-an-hour role. |
The deeper problem is that a percentage of salary is the wrong shape for the answer. Most of what a bad hire costs a frontline employer does not scale with the wage: the advertising spend is the same whether the role pays $16 or $26; the manager's time is the manager's time; the empty seat costs whatever the empty seat costs. A percentage-of-salary rule flatters low-wage roles and understates exactly the hires where the damage is largest.
The one number worth knowing about yourself
Not the cost of a bad hire in general — the cost of your last one, worked out from your own records. It takes about twenty minutes with a payroll export and a recruiting invoice, and unlike the borrowed figure it survives being questioned. Everything below is built to make that twenty minutes structured.
The market a bad hire happens inside
Before the arithmetic, the context that decides how much a mistake costs: how quickly you can replace someone.
In July 2026 the Bureau of Labor Statistics reported that “the number of job openings was little changed at 7.3 million,” that “hires and total separations both changed little at 5.1 million,” and that within those separations “quits (3.1 million) and layoffs and discharges (1.7 million) were little changed.”
Two things follow. First, 1.7 million layoffs and discharges a month is the scale of employment relationships ending at the employer's initiative — bad hires are not rare events, they are a routine cost of doing business, and treating each one as a scandal is the wrong response. Second, with 7.3 million openings against 5.1 million hires, the replacement is not instant. The empty seat is part of the cost and for most employers it is a larger part than the recruiting fee.
Source: U.S. Bureau of Labor Statistics, Job Openings and Labor Turnover Survey, July 2026, released 1 September 2026. Read at source 8 September 2026.
The six costs, and where to find each one
Every figure below exists in a system you already own. None of them requires an estimate from a survey.
What it cost to fill the seat the first time
Advertising spend, agency fee, referral bonus, job board subscription apportioned to the role. Look at the invoice, not at an average cost-per-hire figure. If you ran ads, your platform reports the spend against the campaign.
Wages and payroll taxes across the whole tenure
Hourly rate times hours actually worked, plus employer payroll taxes and any benefit cost that accrued, which the employee cost calculator will assemble if you do not already carry a loaded rate. This is the largest single line and it is the one people forget to count, on the grounds that they “would have paid someone anyway” — but you paid it and got the shortfall in item four.
Onboarding and training time
The hours a trainer, supervisor or buddy spent, valued at that person's fully loaded hourly cost, plus any orientation, licensing or equipment cost that does not transfer to the next person.
The productivity shortfall
The honest version of this is not a formula. Ask the manager one question: across the whole tenure, what fraction of a competent person's output did you get? Half? A third? The number is a judgment, but it is a judgment by the person who watched it happen, which beats a coefficient from a consultancy.
Manager and administrative time
Supervision beyond normal, coaching conversations, documentation, the disciplinary process, the exit. Managers routinely underestimate this. Ask for hours across the whole period, not a weekly average.
Doing it again, and the gap in between
The second recruitment cost, plus the days the seat sat empty multiplied by what a filled seat contributes. For a billable or revenue-generating role that is a margin figure; for a support role it is the overtime you paid other people to cover.
Work out your own number
Fill in whatever you know and leave the rest at zero. Nothing is sent or stored — it runs in your browser.
Bad hire cost model
What the model deliberately leaves out
There are real costs here that we have not put a number on, because any number we invented would be worse than the honest omission.
The effect on everyone else. A bad hire on a small team makes other people's shifts harder, and some of them leave because of it. That is real and it is expensive. It is also not measurable from your records, so counting it would mean guessing.
Customer and client damage. In frontline roles this can dwarf every line above — a lost account, a bad review cycle, a safety incident. Where you can attribute a specific loss, add it. Where you cannot, do not manufacture one.
Manager attention that had a better use. Genuine, and unquantifiable without a fiction about what the manager would otherwise have achieved.
Which means the figure the calculator gives you is a floor, and you should present it as one. A defensible floor persuades people; an impressive number that falls apart under one question does not.
What actually reduces the number
More candidates, not better interviews. The single strongest predictor of a bad hire in frontline recruiting is a shortlist of one. When one person applies, the process stops being selection and becomes acceptance. Almost everything else on this list is downstream of that.
Say the hard parts in the ad. The shift, the weekend rota, the physical requirement, the pay. Every unpleasant surprise you defer to week two is a resignation you have scheduled in advance.
Ask the same questions, in the same order, and write the scores down. Structured interviews are not bureaucracy; they are the only way two candidates are compared on the same basis. Our interview questions to ask candidates guide covers the method.
Use the first thirty days as a real checkpoint. Most bad hires are visible in week two and exited in month five. The cost is roughly linear in tenure, so the whole saving is in acting early. Our onboarding entry covers what to have in place.
Consider a temp-to-hire route where it fits. It converts a hiring decision into an observation period, at the cost of a margin. Our temp-to-hire entry covers the trade.
Frequently asked questions
How much does a bad hire cost?
There is no credible general figure, and the widely quoted claim that the Department of Labor puts it at 30% of first-year earnings does not trace to a Department of Labor publication. The cost varies enormously with tenure, wage, how long the seat then sits empty and how much manager time the situation consumed. The only defensible number is one built from your own records: recruiting cost, wages and payroll taxes paid, onboarding time, the output shortfall, manager hours, and the cost of hiring again.
Is the 30% of salary rule for bad hires accurate?
It is unverifiable and structurally misleading. Following the citations leads to staffing and HR articles quoting each other rather than to a primary source. More importantly, most of what a bad hire costs a frontline employer does not scale with the wage at all: the advertising spend, the manager's time and the empty seat cost roughly the same whether the role pays $16 or $26 an hour, so a percentage-of-salary rule understates the cost on exactly the roles where it hurts most.
What are the costs of a bad hire?
Six that can be looked up: the recruiting cost to fill the role the first time; wages and payroll taxes paid across the whole tenure; onboarding and training time valued at the trainer's cost; the shortfall between the output you paid for and the output you got; extra manager and administrative hours including the exit process; and the cost of recruiting again plus the days the seat sits empty. Costs to teammates and to customers are real but not reliably measurable, so they belong as a stated caveat rather than an invented figure.
How long does it take to spot a bad hire?
Managers usually know within the first two or three weeks, and most bad hires are exited several months later. Because nearly every line of the cost is proportional to tenure, the saving available is almost entirely in acting on what the manager already suspects. A genuine thirty-day checkpoint, with the manager asked a direct question rather than an open one, is the cheapest control available.
What is the most common cause of a bad hire?
Having only one candidate. When a single person applies, the process stops being a selection and becomes an acceptance, and no amount of interview technique compensates. The second most common cause is a job advertisement that omitted something the candidate would have refused — the weekend rota, the physical requirement, the actual pay — which produces someone who takes the job and leaves once they find out.
Should the cost of a bad hire include the salary paid?
Yes, at least in part. The common objection is that the wage would have been paid to someone regardless, but that is the point: you paid for a level of output and received less. Counting the full wage and then separately estimating the share of competent output you actually received is the cleanest way to handle it, and it is what the model on this page does.
How do you calculate the cost of an unfilled position?
Multiply the days the seat is empty by what a filled seat contributes per day. For a billable or revenue-generating role that is the daily gross margin on that person's work; for a support role it is usually the overtime and agency cover paid to other people, which appears directly in payroll. With 7.3 million openings against 5.1 million hires in a single month, replacement is rarely quick, so this line is often larger than the recruiting fee.
What is the cost of bad hiring?
The cost of bad hiring is the sum of six things you can look up in your own records: the advertising and recruiter cost of filling the seat, the wages and payroll burden paid for output you did not get, onboarding and training time, the shortfall in that person's output while they were there, the cost of running the whole search again, and the days the seat sits empty in between. Nearly every line is proportional to tenure, which is why the same mistake caught in week three costs a fraction of the same mistake caught in month eight.
Most bad hires are a shortlist-of-one problem.
When one candidate applies, you hire that candidate. We write and run the recruitment ads that put your open roles in front of people already scrolling Facebook and Instagram, and report what each applicant costs with the management fee inside — so the choice is a choice.
Book a Demo