← All benchmarks HRRef/Methodology
Reference

Methodology

How every number on this site is sourced, defined, and kept current.

Policy
Raw vs. Derived

Every row in our dataset is tagged raw (published exactly as the source states it) or derived (computed by us from published source components, with the method disclosed). Of 210 data rows: 198 raw, 1 derived, 11 text-only source pointers or gap notes (no number republished).

Example derived rowMethod
Manufacturing (NACE C), EU27_2020 NETTUR_MEUR / EMP_NR (antall ansatte) i stedet for standardindikatoren NETTUR_EMP_TEUR (personer sysselsatt)
Definition traps
Read This Before You Compare Numbers

Fixed methodology boxes — see /methodology for the full policy.

F1 — "Annual average rate" ≠ annual turnover
BLS JOLTS reports an annual average rate: the mean of twelve monthly rates, each computed against that month's employment level. It is not the same as “what share of the workforce left over the year” — that would require summing separations against a rolling headcount. Treat JOLTS annual-average figures as a rate index, not a headcount-turnover percentage.
F2 — Quits ≠ voluntary turnover; separations ≠ turnover
“Separations” (JOLTS Table 20) includes quits, layoffs/discharges, and other separations (retirements, deaths, transfers) combined — it is not a clean “turnover” figure by itself. “Quits” (Table 22) is the closest published proxy for voluntary turnover, but includes some non-resignation exits reported as quits by employers. “Layoffs and discharges” (Table 24) is the closest proxy for involuntary turnover. None of the three are interchangeable.
F3 — Absence denominators are system-specific
BLS (CPS, reference-week self-report), ONS (LFS-based, UK), Eurostat (lfsi_abt_q, LFS-based, EU), and SSB/DST (StatBank, register-based) each define “absence” and its denominator differently. They are never combined into one comparable table on this site — each system's figures are shown separately, with its own denominator stated.
F4 — Vacancies, openings, and offers are three different metrics
“Job openings” (a stock, measured at a point in time), “vacancies” (used loosely as a synonym in some national statistics), and “offers extended/accepted” (a recruiting-funnel flow metric) measure different things. Offer Acceptance Rate on this site uses the NACE definition — accepted ÷ (accepted + declined) × 100 — not a vacancy-fill rate.
F5 — Internal mobility ≠ job-to-job movement
“Internal mobility” (moving to a new role inside the same employer) and “job-to-job flows” (published in some labor-market statistics as switching employers) are different phenomena measured by different instruments. No open, official, cross-industry internal-mobility benchmark was identified during sourcing — this site shows what vendor data exists (SmartRecruiters internal/referral hire shares) and states that gap explicitly.
Survey mechanics
JOLTS Benchmark Revisions & Size Classes

Annual benchmark revision: JOLTS estimates are re-benchmarked annually against Current Employment Statistics (CES), with a recalculation window reaching five years back. Earlier years' published figures can therefore be revised backward at each benchmark — a figure quoted from an older vintage may no longer match the current series. Compare year-over-year only within one vintage of the series database.

Establishment size classes: the establishment-size breakdown exists ONLY for Total Private in JOLTS — it is never published crossed with industry. hrref therefore never shows an industry+size combination as if it were an official BLS series.

License note
INE License Inconsistency (Spain)

INE's aviso legal — the authoritative license text — confirms CC BY 4.0 for statistical information where INE is the source. INE's own page footer, however, says CC BY-SA 4.0: an inconsistency on INE's side. hrref cites CC BY 4.0 per the aviso legal, records both versions in its source register, and follows INE's attribution requirements verbatim («Fuente: Sitio web del INE: www.ine.es», latest table update stated, no implied endorsement).

Denominators
What Each Metric Is Divided By

The denominator choice changes the number. We state it per metric rather than assuming one universal formula.

PageDenominator
Overall Turnover Rateannual separations / mean of 12 monthly employment levels (BLS annual avg rate)
Voluntary Turnover (Quits)annual quits / mean of 12 monthly employment levels (BLS annual avg rate)
Voluntary Turnover (Quits)annual layoffs and discharges / mean of 12 monthly employment levels (BLS annual avg rate)
Absenteeism Rateemployed full-time wage and salary workers absent from work in reference week (CPS definition)
Absenteeism RateONS LFS-based sickness absence definition; 'official statistics in development' caveat
Absenteeism RateONS LFS-based sickness absence definition
Absenteeism RateLFS survey: persons employed absent in the reference week — NOT an employer-reported sickness-absence rate
Cost Per HireSHRM 2025 Benchmarking Survey, n=2,371, fieldwork Jan–Mar 2025
Cost Per HireCIPD survey respondents: n=246 (senior) / n=207 (other)
Offer Acceptance Rateaccepted offers / (accepted + declined offers) × 100 (NACE definition)
Internal Mobility Rateinternal hires (respectively, referrals) as a share of all hires
Revenue Per Employeenet turnover / mean number of persons employed (Eurostat NETTUR_EMP_TEUR)
Revenue Per Employeenet turnover / number of employees (NETTUR_MEUR / EMP_NR) — alternative denominator, NOT the standard indicator
Time to HireMedian, not average (SmartRecruiters' own precision)
Span of Controldirect reports per supervisor (Honolulu audit definition)
Job Openings Ratejob openings / (employment + job openings) x 100 (BLS JOLTS definition)
Job Openings Ratehires / employment x 100 (BLS JOLTS definition)
Job Openings Ratequits / employment x 100 (BLS JOLTS definition)
Freshness
Update Cycles, By Source

Verified 2026-07-26. Quarterly-cycle sources (Eurostat, Culture Amp) are checked most often; annual sources are re-verified at each release.

PublisherDatasetUpdate cycleVerified
AshbyAshby Offer Acceptance Rate benchmark — methodology landing page (230K+ applications 2021-2024)unknown2026-07-17
CIPD (Chartered Institute of Personnel and Development)Resourcing and Talent Planning 2024annual2026-07-17
City and County of Honolulu, Office of the City AuditorSpan of control audit (definition: direct reports per supervisor)unknown2026-07-17
City of PortlandSpan of Control Update (memo), April 2026 versionunknown2026-07-17
Culture AmpCulture Amp Employee Engagement Benchmarks (All Industries, Retail, New Tech) — January 2026 vintagequarterly2026-07-17
Danmarks Statistik (DST)DST StatBank absence tables LSK11/LSK12/LSK13/LSK14/LSK15unknown2026-07-17
EurostatStructural business statistics overview — sbs_ovw_act (NETTUR_EMP_TEUR / NETTUR_MEUR / EMP_NR indicators)annual2026-07-17
EurostatLFS series — lfsi_abt_q (absence from work, quarterly, LFS-based)quarterly2026-07-17
GallupState of the Global Workplace 2026 Report (2025 data)annual2026-07-17
Instituto Nacional de Estadística (INE)Encuesta Trimestral de Coste Laboral (ETCL) — tablas 6047/6048/6064 (puestos de trabajo vacantes)quarterly2026-07-19
Instituto Nacional de Estadística (INE)Estadística de Flujos de Población Activa (EFPA) — tablas 66257/66262 (flujos entre situaciones de actividad)quarterly2026-07-19
LinkedIn Talent Solutions (cited via HR Dive / iCIMS)LinkedIn Talent Blog, Feb 2024 — internal-mobility formulation (cited via secondary sources)unknown2026-07-17
McKinsey & CompanyOrg design archetypes — span of control 3-5 (hierarchical) to 15+ (flat) direct reportsunknown2026-07-17
NACE (National Association of Colleges and Employers)Professional Standards 2024 (PDF), p. 68: definition of Offer Acceptance Rate (OAR)unknown2026-07-17
Office for National Statistics (UK)Sickness absence in the UK labour market: 2025annual2026-07-17
SHRM (Society for Human Resource Management)2025 Benchmarking Survey — Cost per Hire (press release)annual2026-07-17
SmartRecruiters (with Lighthouse Research & Advisory)2025 Hiring Benchmark Report / country articles (e.g. United States Benchmark Recruiting Metrics)annual2026-07-17
Statistisk sentralbyrå (SSB)SSB StatBank sickness-absence tables 12452 (sex + occupation) and 14341 (region + sex)unknown2026-07-17
U.S. Bureau of Labor StatisticsJOLTS Table 20. Annual total separations rates by industry and region, not seasonally adjustedannual2026-07-17
U.S. Bureau of Labor StatisticsJOLTS Table 22. Annual quits rates by industry and region, not seasonally adjustedannual2026-07-17
U.S. Bureau of Labor StatisticsJOLTS Table 24. Annual layoffs and discharges rates by industry and region, not seasonally adjustedannual2026-07-17
U.S. Bureau of Labor StatisticsCurrent Population Survey, Table 47. Absences from work of employed full-time wage and salary workers by occupation and industryannual2026-07-17
U.S. Bureau of Labor StatisticsJOLTS series database — Job Openings, Hires, and Quits rates, seasonally adjusted, by industry and establishment size classmonthly (annual benchmark revision against CES, 5-year recalculation)2026-07-19