A support agent we were brought in to replace had been in the seat for five months. On paper the hire looked fine - relevant experience, a clean interview, good references. In practice, the team lead was rewriting half of every shift handover, two senior agents had quietly started covering the gaps, and customers were getting confident answers that were wrong. By the time the company asked us to find a replacement, the salary they had paid was the smallest part of the bill. The expensive part was the five months the role was never actually filled.
That is what the cost of a bad hire hides. The headline figure everyone quotes - a bad hire costs 30% of first-year salary - is the floor, not the ceiling. On a small team the real number sits well above it, and most of it never shows up on an invoice.
Start with the number everyone quotes - then keep going
The U.S. Department of Labor puts the cost of a bad hire at 30% of the employee's first-year salary. For a $50,000 role, that is $15,000. Treat it as the entry point, because it mostly covers direct replacement and misses everything downstream.
The wider sources fill in the rest. The Society for Human Resource Management estimates that replacing an employee runs 50% to 200% of their annual salary. CareerBuilder put the average loss per bad hire near $17,000, climbing past $240,000 for senior and specialized roles. A Center for American Progress review of 30 studies found the median turnover cost lands around 21% of annual salary.
The range is wide for one reason: the cost depends on the role, the seniority, and how long the person stayed before anyone acted. Every extra month they sit in the seat, every line below gets bigger. The published averages also blend industries and company sizes, so a single number is close to useless until you break it into parts.
The five costs inside one bad hire
Cost of a bad hire is not one number. It is five, and only the first one is obvious.
| Cost | What it is | Rough size on a $50,000 hire |
|---|---|---|
| Replacement recruiting | You pay to hire, then pay again to replace | $4,000–7,000 per search, billed twice |
| Salary for low output | Months of full pay for a fraction of the work | $8,000–16,000 over four months |
| Onboarding written off | The training time of everyone who ramped them | $3,000–4,000 |
| Manager drag | Managers spend around 17% of their time on underperformers | ~$3,800 across one quarter |
| Team slowdown | 85% of HR leaders say one bad hire dents the whole team; nearby turnover risk rises ~15% | Hard to invoice, easy to feel |
Add only the visible lines and a $50,000 hire that lasts four months has already cost $25,000 to $30,000 - before you count the work that never got done. That is why the 30% figure is a floor.
Why a bad hire costs a 20-person company more than a 200-person one
The generic statistics are written for enterprises, and they understate the damage for a company that is still small, because the math concentrates.
At 20 people, one hire is 5% of your headcount. At 200, the same hire is 0.5%. Same mistake, ten times the concentration. On a five-person pod, one underperformer changes the team's daily reality by roughly 20%, and the people who absorb the slack are your strongest operators - the ones you least want pulled off their own work.
There is a second small-company tax. At 20 people the hiring manager is often a founder or a first-time lead with no time to coach a struggling hire back to standard. So the problem either festers or ends in a second search, and both outcomes cost more than they would at a company with a real management layer.
On a tight runway, the cost is not only money. It is time. The role you hired to fix is still broken, the calendar moved by a quarter, and a small team rarely has a quarter to spare. A bad hire at Series A is not a line item. It is a delay you feel across the whole company.
Run the formula on your own role
You do not need a consultant to estimate this. The cost of a bad hire breaks into four parts you already know:
Bad-hire cost = (recruiting cost × 2) + (salary paid for low output) + (manager hours spent managing the problem) + (the months the role still is not done)
Run it on a real seat. A $50,000 support hire, noticed at month three, replaced by month six: about $5,000 to re-run the search, roughly $12,000 in salary for thin output, close to $4,000 of a manager's quarter spent supervising the problem, plus six months where tickets piled up and the queue the role was meant to clear kept growing. The visible bill clears $20,000. The invisible one - churned customers, a drained pod - usually runs higher.
The early signals, before it shows up in the numbers
The cost compounds while you wait for proof. By the time a bad hire is obvious in the metrics, you have already paid for three or four months of it. The cheaper move is to read the early signals, which are operational long before they are financial:
- The manager is quietly redoing the new hire's work instead of reviewing it.
- The rest of the team has started routing around the person rather than through them.
- A ramp that should take six weeks is still going at twelve.
- The person is visibly busy, but the queue or the backlog is not moving.
- You hear about problems from customers or other teams, not from the hire.
One of these is a coaching conversation. Three of them at once is a hire that is already costing you more than the salary line shows.
A bad hire is a process failure, not bad luck
After enough searches, the pattern is hard to miss: bad hires are almost never random. They come from two root causes, and both are preventable.
Wrong skills. The resume matched the keywords, the work did not. This gets through when nobody tested the actual job during screening - no work sample, no scenario, just a conversation about experience.
Wrong fit. The skills were real, but the working style or the role expectations were not. This gets through when the brief was vague and the screen was rushed to fill an urgent seat.
Both slip through for the same reason. The role was urgent, so a step got cut. When we inherit a search, the previous process almost always optimized for speed-to-offer over fit-to-role. A warm body in the seat felt like progress. It was the first line of the bill.
How we cut the bad-hire cost
Our recruiting practice is built to remove the two things that produce bad hires: the rush, and the keyword-match screen.
We charge a flat success fee - typically $1,000–$3,000 for support and operations roles - not a percentage of salary. We are not paid more for placing a pricier candidate, and there is no clock that rewards filling the seat fast over filling it right. The only incentive is a hire that stays.
We screen against the real role, not the resume. Because our team staffs support, operations, and tech seats directly, we test for the work the person will actually do in week one - not the words on the page. That is the step that catches a mismatch before it becomes your problem.
And ask any recruiting partner the same question before you sign: what happens if the placement does not last? The bad-hire risk should sit with the recruiter who made the match, not the client who trusted it. That is the standard we hold our own work to - a flat-fee model only pays off for us when the hire stays.
A bad hire is expensive because it is quiet. The salary is the only line you see clearly. The rest - the manager's lost quarter, the pod slowing down, the role still open - adds up in the background until it is the biggest number on the page. The cheapest way to pay it is to not make the hire wrong in the first place.
Weighing whether a role is worth handing to a recruiting partner? Tell us the seat you need filled and we will give you a flat quote and a realistic time to shortlist - typically a first qualified list in 7–10 business days. Book a call and get a quote →