# Cost per Hire and Time to Fill Explained

> The two standard hiring metrics were defined for salaried recruiting and mislead badly when applied to shift work. What each one measures, how to calculate it, and the third number that matters more than both.

Author: Kariaa Research
Published: 2026-07-30
Category: Benchmarks
Reading time: 10 min read
Canonical URL: https://www.kariaa.com/reports/cost-per-hire-and-time-to-fill

---

import { CompareColumns, FlowSteps, RankedList, Takeaways, FAQ } from "@/lib/report-charts";

Two numbers dominate every conversation about hiring performance. Both are worth
measuring. Both were defined for a kind of hiring that most frontline employers
do not do, and applied unchanged they produce confident conclusions in the wrong
direction.

## The definitions, precisely

**Time to fill is the number of days between a role being opened and an offer
being accepted.** It measures the whole process, including the part before any
candidate exists: writing the role, getting approval, sourcing, screening,
interviewing, deciding and closing.

**Time to hire is the number of days between a candidate entering the process and
that same candidate accepting an offer.** It measures only the part of the
process a specific person experiences.

The distinction is not pedantry, because the two numbers fail in opposite
directions. Time to fill counts sourcing delay and time to hire does not, so a
role that sat unposted for three weeks and then moved quickly shows an alarming
time to fill and an excellent time to hire. **A team optimising time to hire can
make its actual hiring slower**, because dropping candidates who need a day to
respond improves the average while extending the search.

**Cost per hire is the total of internal and external recruiting costs divided by
the number of hires in the same period.** The ANSI standard developed with SHRM
puts it plainly: add external costs such as advertising, agency fees, job board
subscriptions and background checks, add internal costs such as recruiter
compensation and referral payments, and divide by total hires.

<FlowSteps
  steps={[
    { label: "External costs", sub: "Ads, boards, agencies, checks" },
    { label: "Plus internal costs", sub: "Recruiter time, referral bonuses" },
    { label: "Divided by hires", sub: "Same period, all sources" },
    { label: "Equals cost per hire", sub: "One period, one number" },
  ]}
  note="The standard formula deliberately excludes the cost of the vacancy itself, which is why it understates frontline hiring so badly."
  caption="Cost per hire, as standardly defined"
/>

SHRM's benchmarking work puts average cost per hire in the region of several
thousand dollars. That figure describes the salaried roles that dominate its
sample, which is exactly why it makes a poor target for frontline hiring.

## Why both metrics mislead in shift work

The metrics were built for a hiring model with a specific shape: a small number
of expensive, individually-considered hires, where the recruiting cost is large
relative to the disruption of the vacancy.

Frontline hiring inverts almost every term.

<CompareColumns
  left={{
    title: "The hiring the metrics assume",
    items: [
      { label: "Few hires, individually considered" },
      { label: "Recruiting cost dominates" },
      { label: "One requisition, one hire" },
      { label: "The role waits for the right person" },
      { label: "Vacancy is an inconvenience" },
    ],
  }}
  right={{
    title: "The hiring most operators do",
    items: [
      { label: "Many hires, continuously" },
      { label: "Turnover volume dominates" },
      { label: "One requisition, repeated quarterly" },
      { label: "The shift happens regardless" },
      { label: "Vacancy is overtime or lost revenue" },
    ],
  }}
  caption="The metrics assume the left column. Most operators live in the right."
/>

Two consequences follow, and they are the reason a frontline operator can have
excellent-looking numbers and a serious problem.

**A low cost per hire is not necessarily good news.** Divide a modest recruiting
spend by a large number of hires and the result looks efficient. But the large
number of hires is the problem being measured, not the achievement. An employer
replacing eighty per cent of a workforce annually will report a flattering cost
per hire precisely because the denominator is enormous.

**Annual hiring cost is driven by hires per year, not by cost per hire.** Halving
cost per hire is a procurement exercise. Halving hires per year is an operations
exercise, and it is worth several times more. Since a large share of frontline
turnover occurs within the first ninety days, most of that denominator is
determined by how the hiring was done rather than by anything that happens later.

## The number that actually matters

For an operator whose success is measured in shifts covered, the decisive figure
is neither of the standard two.

**Cost of vacancy is what an unfilled role costs per day that it stays unfilled.**
It is the only one of the three that connects hiring to the operation, and almost
nobody calculates it, which is why hiring is chronically under-resourced relative
to what a delay actually costs.

It is straightforward to build for a specific role. Take the shifts that role
covers in a week, and for each one identify how it is currently being covered.
Overtime at a premium rate has a direct hourly cost. Agency or temp cover has a
rate you already pay. A shift that goes uncovered has a cost in lost revenue,
turned-away work, or service commitments not met. Add the load carried by the
people absorbing it, which shows up later as their turnover.

<RankedList
  items={[
    {
      label: "Overtime premium on covering staff",
      note: "The most direct component and the easiest to obtain, since it is already in payroll. Usually the largest single line for shift-based roles.",
    },
    {
      label: "Agency or temporary cover rate",
      note: "Charged at a multiple of the internal rate. Where used, this alone frequently exceeds the entire cost of hiring a permanent replacement.",
    },
    {
      label: "Revenue not earned",
      note: "Clients not served, capacity not sold, work declined. Invisible in payroll, and often the largest number of the four.",
    },
    {
      label: "Load transferred to remaining staff",
      note: "The slowest and most expensive component. It surfaces months later as further turnover, and it is why a long vacancy in a small team compounds.",
    },
  ]}
  caption="The components of cost of vacancy, per day the role stays open"
/>

Once that daily figure exists, every hiring decision becomes arithmetic rather
than argument. A sourcing channel that costs money but closes a search two weeks
sooner is either obviously worth it or obviously not, and the comparison is
between two numbers rather than between a known cost and a vague benefit.

**Recruiting spend is habitually compared against cost per hire, which is a
budget line, rather than against cost of vacancy, which is the thing being
bought.** That comparison is the single most common reason a role stays open for
five months to avoid an expense worth a fraction of the delay.

## Measuring your own, without a system

None of this requires analytics software. It requires four dates recorded per
role and a spreadsheet.

<FlowSteps
  steps={[
    { label: "Role opened", sub: "The day cover was needed" },
    { label: "First candidate", sub: "Reach is measured here" },
    { label: "Offer accepted", sub: "Time to fill ends" },
    { label: "First shift worked", sub: "The gap nobody measures" },
  ]}
  note="The interval between offer accepted and first shift worked is where a large share of frontline hires are lost, and it is outside both standard metrics."
  caption="Four dates that make a hiring process measurable"
/>

The interval between the first and second date is the most diagnostic and the
least recorded. **A long gap before the first candidate appears is a reach
problem; a long gap afterwards is a decision problem, and they have entirely
different remedies.** Most operators know only the total, which is why the usual
response to a slow search is to interview faster, which addresses the second
interval when the first is the one failing.

The fourth date belongs on the list because both standard metrics stop at
acceptance, and a meaningful share of frontline hires are lost between
acceptance and the first shift. A process that measures only to the offer will
record those as successes. That gap is examined in [How to Reduce Interview
No-Shows](/reports/how-to-reduce-interview-no-shows).

## What to do with the numbers

Three questions are worth more than any benchmark comparison, because they are
answerable from your own four dates.

How many days pass before a first qualified candidate appears? If it is most
of the total, the search is failing at reach, and no amount of interview
discipline will fix it. That is the subject of [How to Hire Hourly
Employees](/reports/how-to-hire-hourly-employees).

What share of hires leave inside ninety days? That share is the part of your cost
per hire denominator you are paying for twice.

What does a day of vacancy cost in this specific role? Until that number exists,
every decision about hiring spend is being made against the wrong comparison.

<Takeaways
  items={[
    "Time to fill measures the whole process from role opened to offer accepted. Time to hire measures only one candidate's journey. Optimising the second can make the first worse.",
    "Cost per hire is external plus internal recruiting costs divided by hires in the period. The standard formula excludes the cost of the vacancy itself.",
    "A low cost per hire can be a symptom rather than an achievement: a large denominator means high turnover, not efficiency.",
    "Annual hiring cost is driven by hires per year, not cost per hire. Reducing early turnover is worth several times more than reducing per-hire spend.",
    "Cost of vacancy per day is the figure that connects hiring to the operation, and almost nobody calculates it.",
    "Record four dates per role. The gap before the first candidate is a reach problem; the gap after is a decision problem, and they need different fixes.",
  ]}
/>

<FAQ
  items={[
    {
      question: "What is the difference between time to fill and time to hire?",
      answer: "Time to fill is the number of days from a role being opened to an offer being accepted, including sourcing. Time to hire is the number of days from a specific candidate entering the process to that candidate accepting. Time to fill counts sourcing delay and time to hire does not, so a team optimising time to hire can make its actual hiring slower.",
    },
    {
      question: "How is cost per hire calculated?",
      answer: "Add external recruiting costs such as advertising, job boards, agency fees and background checks, add internal costs such as recruiter compensation and referral payments, and divide the total by the number of hires made in the same period. This is the ANSI standard formula developed with SHRM. It deliberately excludes the cost of the vacancy itself.",
    },
    {
      question: "Is a low cost per hire good?",
      answer: "Not necessarily. Cost per hire divides recruiting spend by number of hires, so an employer with very high turnover reports a flattering figure because the denominator is large. In frontline work the large number of hires is usually the problem being measured rather than evidence of efficiency.",
    },
    {
      question: "What is cost of vacancy and how do I calculate it?",
      answer: "Cost of vacancy is what an unfilled role costs per day it stays open. Calculate it by identifying how each shift that role covers is currently being covered: overtime premium on existing staff, agency or temporary cover rates, revenue not earned from work that cannot be taken, and the load transferred to remaining staff which later shows up as further turnover.",
    },
    {
      question: "Which hiring metrics should a small operator actually track?",
      answer: "Four dates per role: the day cover was needed, the day the first qualified candidate appeared, the day an offer was accepted, and the day the first shift was actually worked. The gap before the first candidate indicates a reach problem, the gap afterwards a decision problem, and the final interval catches hires lost between acceptance and start, which both standard metrics miss entirely.",
    },
  ]}
/>

---

*Written by the team at **Kariaa**. Learn more at [www.kariaa.com](https://www.kariaa.com).*