Cost-per-Hire Is an Allocation Problem: What Gets Excluded, Why Firms Cannot Compare, and a Decomposition FP&A Will Accept

Cost-per-hire looks like division: recruiting spend over hires. In practice the numerator is an allocation exercise, and most published figures are produced by allocating only the costs that are easy to see. The result is a metric that understates the true cost of hiring, varies with accounting choices more than with recruiting efficiency, and is almost never comparable across organisations. This piece covers the gap between advertised and fully-loaded cost, the costs that are silently excluded, why cross-firm comparison fails, and a decomposition that a finance team will accept.

Advertised versus fully-loaded cost

The advertised version of cost-per-hire typically counts external spend that passes through a recruiting budget line: job board fees, agency placement fees, background checks, recruiting technology licences, and sometimes events. These are invoiced, coded, and easy to sum.

The fully-loaded version adds internal costs: the time of the people who do the work, overheads attributable to the recruiting function, and costs carried in other departments’ budgets. These are larger, harder to measure, and usually omitted.

The two versions can differ by a multiple. That gap is the reason the metric is so easily used to make a recruiting function look efficient: the cheapest way to lower advertised cost-per-hire is to move work from external vendors to internal staff whose time is not counted.

The silently excluded costs

Three categories account for most of the gap.

• Internal recruiter time. Recruiter and coordinator salaries, benefits, and on-costs are often held in an HR operating budget and not allocated to hires. Where they are allocated, the allocation basis (hires, requisitions, or hours) is rarely stated.
• Hiring-manager and interviewer hours. Every interview loop consumes hours from managers and panel members whose cost sits in their own department. For senior and specialist roles, interviewer time can exceed recruiter time. It is almost never counted.
• Agency retainers and fixed fees. Contingency fees attach naturally to a hire. Retained-search fees, exclusivity retainers, and master-service minimums are paid whether or not a hire results, and are frequently booked to a general line or amortised inconsistently.

Smaller items follow the same pattern: referral bonuses booked to payroll, relocation booked to the hiring department, assessment platform licences held by IT, and the cost of requisitions that never fill.

Why cross-firm comparison fails

Cost-per-hire is not comparable across organisations because each firm makes different allocation choices, and those choices dominate the number. Hiring mix compounds the problem: a firm hiring mostly high-volume frontline roles will report a far lower figure than one hiring mostly specialists, at identical efficiency.

To illustrate — numbers constructed for demonstration, not measured: two firms with identical recruiting operations, one counting only external spend and one including recruiter salaries and interviewer time, could report cost-per-hire of roughly 3,000 and 9,000 respectively. A benchmark mixing both is meaningless.

The defensible use of cost-per-hire is internal: trended over time under a frozen definition, and segmented by role type.

A decomposition finance will accept

FP&A teams reject recruiting cost figures for two reasons: they do not reconcile to the general ledger, and the allocation logic is not stated. The decomposition below addresses both.

1. External direct costs. Agency contingency fees, job advertising, background checks — attributable to a requisition. Reconciles to specific GL accounts.
2. External fixed costs. Retainers, platform licences, employer-brand spend — allocated across hires by a stated basis. Reconciles to GL.
3. Internal recruiting labour. Recruiting team fully-loaded compensation, allocated by requisition or hire with a stated basis (recruiter_hours_per_req if time data exists, otherwise hires).
4. Internal interviewing labour. Interview hours from the scheduling system, multiplied by a standard hourly cost by level. An estimate, labelled as one.
5. Unrecovered cost of unfilled requisitions. Spend on cancelled requisitions, reported separately rather than buried in the per-hire figure.

Each component is reported separately, with its GL mapping or its estimation method, before any total is computed. Finance can then accept the components it can reconcile and challenge the estimates it cannot, rather than rejecting the whole figure.

What we cannot claim

A fully-loaded cost-per-hire is a cost measure, not a value measure. It says nothing about the quality of hires, and minimising it can easily increase the cost of a bad hire downstream. The interviewing-labour component is an estimate whose precision depends on scheduling data quality. What the decomposition provides is a figure whose construction is visible. The data always wins over the narrative — once every cost is allowed into the data.