Cost Per Hire Calculator: Average Cost Per Hire, the Cost Per Hire Formula and 2026 Benchmarks

Add every internal recruiting cost, add every external recruiting cost, and divide by the number of hires you completed in the same window. Written out, the standard formula is:

Cost per hire = (total internal recruiting costs + total external recruiting costs) / total number of hires

Two rules make the result comparable to anyone else's. Use the same period for the numerator and the denominator, so a quarter of spend is divided by a quarter of hires. And count hires that actually completed, not offers extended or requisitions opened, because a role that fell through still consumed the spend.

Here is the worked example the calculator above starts with: a company that filled four roles in one quarter, running its own recruiting with no agency.

Notice the shape of it. Four fifths of the total is internal time, and none of that appears on an invoice. This is why companies that only count external spend report a cost per hire of a few hundred dollars and then cannot work out why the number bears no relation to anyone else's. If your figure looks remarkably low, the interviewer time line is almost always the reason.

To value that time honestly, use a loaded hourly rate rather than base salary. A hiring manager on $150,000 costs roughly $196,000 a year once payroll taxes and benefits are added, which is about $94 an hour rather than $72. The build-up for that multiplier is on our page covering the cost of hiring an employee.

The honest answer is that nobody publishes a current, authoritative US average, and the figure everyone quotes is older than most people realize.

$4,129 is the number you will see on nearly every page about this metric. It comes from the SHRM Human Capital Benchmarking Report, published in 2016 and built on fiscal 2015 data. It is a real, well-constructed figure. It is also more than a decade old, and it is quoted in article after article, calculator after calculator, with no date attached. Recruiting cost is mostly labor, and labor has not cost the same since.

So we restated it. Using the Bureau of Labor Statistics consumer price index for all urban consumers, the 2015 annual average index was 237.017 and the July 2026 index was 333.918, a factor of 1.409. Applying that to $4,129 gives about $5,817 in July 2026 dollars. That is our arithmetic, not a SHRM publication, and we show the inputs so you can check it. It is a floor rather than a ceiling, because consumer prices have risen more slowly than professional wages over that period, and cost per hire is disproportionately professional wages.

The gap between $5,475 for a normal role and $35,879 for an executive one is the useful part of that table. A single average across all hires is close to meaningless for planning, because the distribution is not remotely symmetric. One retained search can outweigh a year of everything else.

The arithmetic on agency fees makes the point on its own. A 20% contingency fee on a $100,000 role is $20,000 for one hire. That single line is roughly three and a half times the restated benchmark, before you have counted a minute of interviewer time. Any company using agencies for a meaningful share of hires will report a cost per hire several times higher than a company that does not, and neither number is wrong. The fee structures behind those figures are broken down on our page about recruiter fees.

The standard splits everything into two buckets, and the split matters because the two behave completely differently when you try to manage them.

External costs are money leaving the company for the purpose of recruiting: job advertising and board fees, sponsored job spend, agency and search fees, background checks and assessments, campus and events recruiting, candidate travel, and recruitment marketing. These are visible, invoiced, and easy to cut, which is why they get cut first even when they are not the problem.

Internal costs are the recruiting function itself: fully loaded recruiter and HR salaries, hiring manager and interviewer time, the allocated share of your applicant tracking system and sourcing tools, and employee referral bonuses. These are invisible, uninvoiced, and usually the majority of the total.

The exclusions are just as deliberate. Salary, signing bonuses and equity are compensation rather than recruiting cost. Relocation and immigration are reported separately. Onboarding, equipment and training sit outside the metric entirely. And the cost of leaving a role open, which is often the largest economic cost of a slow hire, is a vacancy cost rather than a recruiting cost. Excluding it is correct under the standard and also slightly perverse, because it means a process that takes twice as long can post a better cost per hire.

That is the metric's main blind spot, and it is worth naming rather than working around. Cost per hire tells you what the process cost. It says nothing about how long it took, whether the person is any good, or whether they stayed. Read it next to time to fill and to first-year retention, never on its own. We keep current benchmarks for the first of those on average time to hire a software engineer, and the arithmetic on the second is on the cost of a bad hire.

Benchmarking this metric is harder than it looks, because the number is easy to move without changing anything real. Three companies with identical recruiting operations can publish figures a factor of five apart purely through accounting choices.

• Whether interviewer time is counted. Leave it out and the example above drops from $4,875 to $3,825. Leave out all internal time and it drops to $975.
• How the denominator is defined. Counting offers accepted rather than hires who actually started inflates the denominator and lowers the number.
• Whether contract and temporary conversions count as hires. They are cheap to source, so including them pulls the average down sharply.
• How agency spend is allocated. Booking a retained search fee to the quarter it was invoiced rather than the quarter the hire started can swing a small company's figure by thousands.

The practical use of a benchmark is therefore not to match it. It is to notice when you are a long way from it and find out why. A cost per hire well above the restated $5,817 usually means agency dependence or a long, interview-heavy process. A figure well below it usually means the internal time is not being counted at all.

Comparing against yourself is more informative than comparing against an industry average. Track the number by role family and by quarter, on a consistent definition, and the trend will tell you something a national average never can. Segment engineering separately from support: they have different sourcing costs, different interview loads and different agency exposure, and blending them hides both.

Because internal time dominates the total for most companies, the levers that actually move the number are process levers, not procurement levers. Cutting job board spend by half moves a small line. Removing an interview round moves a large one.

Cut interview rounds that do not change decisions. Count how many candidates have ever been rejected at each stage. Any stage that rejects nobody is pure cost, and it usually costs several hundred dollars per candidate in loaded interviewer time. This is the single largest recoverable line in most processes.

Reduce the number of unqualified applicants rather than the number of applicants. Screening cost scales with volume, not with quality. Publishing the salary range, the location eligibility and the seniority in the posting itself removes most mismatched applications before anyone reads them, which is why pay-transparent listings tend to produce smaller and better applicant pools. Recent industry data puts open roles above 300 applications on average, so anything that filters at the source compounds.

Reduce agency dependence selectively, not universally. Agencies are expensive per hire and genuinely worth it for scarce senior roles where your own sourcing does not reach. They are poor value for roles that attract inbound applicants. Splitting the requisition list on that basis usually cuts more cost than any renegotiation.

Use channels with a fixed cost per listing rather than a percentage of salary. A flat listing fee has a cost per hire that falls as the salary rises; a percentage fee does the opposite. For a fully remote role, a specialist board reaches candidates who have already decided they want remote work, which shortens screening. Posting a role on our remote job board is $299 flat with the salary shown on every listing, and the full channel comparison is on what it costs to post a job.

Fix the requisition before you open it. Roles that get redefined halfway through burn an entire cycle of sourcing and interviewing that never appears as a hire, and that spend still lands in the numerator. Agreeing the level, the range and the must-haves before posting is free and removes the most expensive failure mode there is.

These are two different numbers and they answer two different questions. Mixing them up produces budgets that are wrong by an order of magnitude, which is why the same page usually should not try to cover both.

Cost per hire is a recruiting efficiency metric. It measures what your process costs to produce one hire, it runs in the low thousands, and you use it to manage the talent acquisition function. That is the number the calculator on this page produces.

Cost of hiring an employee is a budgeting figure. It covers what the person costs you once employed: salary, employer payroll taxes, benefits, equipment and onboarding. For a professional salaried role in the US it runs roughly 1.3 times base salary, so a $100,000 hire costs about $131,000 a year. If you are working out whether you can afford the headcount, that is the page you want, and the full build-up from Bureau of Labor Statistics and IRS figures is on cost of hiring an employee.

Cost of a bad hire is the third number, and it is the one that makes the other two look small. A $100,000 hire who leaves in month five costs roughly $36,700 to $38,700 in direct costs, or about $76,700 once agency fees on both searches are counted. Set against that, spending an extra $2,000 on cost per hire to improve the odds is obviously worth it, which is the strongest argument against optimizing this metric in isolation.

If you want the benchmark discussion and the formula variants in more depth, our article on cost per hire covers the reporting side, and what it costs to hire a remote employee covers the distributed-team version where multi-state registration enters the picture.