Cost per Hire

What cost per hire is

Cost per hire is the average an organisation spends to fill one position, computed by adding the internal and external recruiting costs for a period and dividing by the number of hires made in that period.

The definition is simple and the measure is unreliable in practice, for reasons that are all about what goes into the two halves of the fraction. Most of this page is about those reasons, because a cost per hire figure that nobody can reconstruct is worse than no figure at all: it gets compared, budgeted against and defended.

What goes into the calculation

• External costs. Job advertisements, job board and platform fees, recruitment agency fees, and the cost of screening and background verification.
• Internal costs. The time the HR team and the hiring managers spend on screening and interviewing, and the cost of the recruiting tools themselves.

The external half arrives as invoices. The internal half does not, and that asymmetry is the whole problem.

A worked example

An organisation spends SAR 60,000 on advertising and agencies in a quarter, carries SAR 30,000 of internal time cost, and makes 6 hires.

(60,000 + 30,000) divided by 6 gives SAR 15,000 per hire.

The cost that never reaches an invoice

The largest single item in cost per hire is usually not what is paid to an outside party. It is internal time: the manager’s hours spent screening applications and running interviews, and the team’s hours in follow up rounds. Because those hours generate no payment voucher, they drop out of the calculation in most organisations, and the reported figure comes out lower than the real one.

Using the same numbers, the size of that omission can be stated exactly. Counting external costs alone gives 60,000 divided by 6, or SAR 10,000 per hire. The full figure is SAR 15,000. So the omission is SAR 5,000 per hire.

How large is that? The answer depends entirely on what you divide by, and both divisions are defensible, so the basis has to be stated before the number rather than after it:

• As a share of the true cost of SAR 15,000, the omitted internal time is 33.3%. This is the right framing for “how much of the real cost are we blind to”.
• As a share of the reported cost of SAR 10,000, the correction is 50%. This is the right framing for “by how much does our number have to rise”.

Those are two different numbers describing one fact, and quoting either without saying which denominator it came from is how a cost per hire figure becomes an argument rather than a measurement. The same discipline applies to every ratio on this page.

Worth stating the same omission once more at the level a budget is actually argued at. SAR 5,000 per hire across 6 hires is SAR 30,000 for the quarter, which is the internal cost line in full. That is the arithmetic closing on itself rather than a new fact, and it is worth writing out because SAR 5,000 sounds like a rounding and SAR 30,000 does not, while the two are the same quantity described at different levels of aggregation. Which of the two gets quoted in a meeting decides how the omission is received, and that is a presentation choice rather than a measurement one.

Estimating the internal half is not difficult: hours multiplied by the hourly cost of each participant. Both inputs already exist in the payroll accounting records, so the reason this item is usually missing is not that the data is unavailable. It is that nobody was asked to collect it.

Why a single overall figure misleads

Computing one average across every role hides more than it shows. Hiring a senior accountant and hiring a customer service agent do not cost the same and are not comparable, so an average across both describes a position that nobody actually filled. A useful average is computed per job level or per department, and the level of aggregation should be stated wherever the number is quoted.

There is a second distortion that is easy to miss, and it works through the denominator rather than the numerator. A hire who leaves within a few months makes the organisation pay the cost twice: once for the hire that did not last, and again for the replacement. The first payment stays in the numerator of whichever period it fell in, while the lasting hire it was supposed to produce never arrives. For that reason cost per hire is read alongside turnover and alongside employee onboarding outcomes, never on its own.

The spending that does not show up as a hire at all

Every cost incurred on a vacancy that closes without an appointment, whether it was withdrawn, frozen or filled internally after the advertising had already run, sits in the numerator with no corresponding entry in the denominator. It therefore raises the apparent cost of every hire that did complete in the same period. This is not a reason to exclude it, since the money was genuinely spent on recruiting. It is a reason to know how much of the period’s total it represents, because a quarter with two abandoned searches and a quarter without them are not comparable even when they report the same cost per hire.

Three figures that are all called cost per hire

None of the three below is wrong. They answer different questions, and an organisation quoting one of them against a benchmark computed on another has compared two definitions and reported the result as a difference in performance.

• External spend divided by hires. The most widely reported version, and the one the worked example above shows running 33.3% below the true cost. Its appeal is that every input is an invoice, which is also precisely its defect.
• The whole recruiting budget divided by hires. This carries the recruiting function’s own salaries and tooling whether or not a vacancy ran, so the figure moves with the size of the team rather than with the difficulty of hiring, and it falls in a busy quarter for reasons that have nothing to do with efficiency.
• The cost of a filled vacancy, charged to that vacancy. Attributed to the role rather than averaged across a period. It is the only one of the three that answers what this hire cost, rather than what hiring cost us, and it is the one that requires the per vacancy record described at the end of this page.

How it differs from time to hire

Cost per hire measures money. Time to hire measures days, from the vacancy opening to the offer being accepted. They are distinct measures and they trade against each other: accelerating an appointment through a recruitment agency shortens the elapsed time and raises the cost.

That trade is the practical reason to keep both. Optimising either one alone produces a predictable failure. Driving cost down lengthens every search, and driving time down buys speed with agency fees that the cost figure will report next quarter, by which point the two decisions look unrelated.

What has to be recorded for the figure to be reconstructable

A cost per hire number is only as good as the ability to rebuild it later. That means recording, per vacancy: the external spend with its invoices, the participant hours with who spent them, whether the vacancy closed in an appointment, and the level or department the role belongs to. Everything else on this page follows from having those four, and no amount of care in the arithmetic compensates for not having them. The interview process that consumes most of the internal hours is covered under structured interview, and the impression that process leaves on the people who were not hired is covered under candidate experience.

Related terms

• Employee Onboarding
• Structured Interview
• Candidate Experience
• Competencies
• Salary Structure