Measuring attrition and hiring speed without misleading denominators
8 min read
Lesson objectives
- Calculate attrition using a defined denominator and period.
- Distinguish time to fill from time to hire.
- Inspect segment differences before recommending action or claiming causation.
Please sign in to use all the tools
Once you sign in, the reading tools are yours: larger or smaller text, typeface, line spacing, focus mode, highlighting and bold, and notes saved to your account and gathered in My study.
When a manager says 'attrition reached 18%', do not interpret the number yet. Ask what was counted and when. It might mean 18 departures from a workforce of 100 during a year, but internal transfers and temporary contracts may have been treated differently. A percentage without numerator, denominator, period and organizational unit cannot support a sound decision. This lesson uses an explicit operational definition rather than claiming a universal one: period departures divided by the simple average of headcount at the start and end of the period, multiplied by 100. Set the definition before comparing teams and do not change it after seeing the result.
Time to fill measures the period from an approved vacancy opening to offer acceptance or start date, depending on the employer's definition. Time to hire follows an individual candidate from entering selection to accepting an offer. If the vacancy opens on day 1, the chosen candidate enters on day 20 and accepts on day 35, time to fill through acceptance is 34 days when dates are subtracted, and candidate time to hire is 15 days. An inclusive convention would yield 35 and 16 days; use one convention consistently. A long fill time may arise from delayed vacancy approval or weak sourcing before a candidate appears, so faster interviews alone may not fix it.
Read measures by meaningful segments. Suppose 15 of the department's 21 leavers were in their first year, while first-year workers averaged 60 of the 210 people. Their quarterly rate is 15/60 = 25%, compared with 6/150 = 4% for others. The overall 10% conceals the difference, but it does not prove onboarding caused departures. Job demands, season, supervisor or contract type might differ. Review exit feedback, onboarding records and shift patterns, while protecting confidentiality in small groups. Show group sizes alongside percentages so a large rate based on two cases is not mistaken for a stable trend.
Turn the diagnosis into a testable decision. If several exit interviews mention unclear role expectations, HR can revise the job description and add a structured expectations conversation in week one, then compare similar cohorts after equal follow-up periods. Do not attribute improvement solely to that action if pay, managers or labor conditions also changed. Track a leading measure, such as completion of the conversation; an outcome, such as 90-day retention; and an early quality measure, such as performance. Screening candidates faster could improve a speed figure while increasing subsequent departures, so read speed with quality.
An organization-wide average can move opposite to within-team trends when workforce mix changes. Suppose a 100-person service team has 4% departures and a 20-person temporary project team has 20%. That represents four expected departures in each, roughly 6.7% of 120 combined. Next period the project team grows to 80 and its own rate improves to 18%, while service remains 100 at 4%. The illustrative expected departures are 14.4 plus four, about 10.2% of 180. The aggregate rises though neither team's rate worsens; the higher-rate team gained weight. Fractional departures are explanatory expectations, not literal people in a real report. Inspect group rates, sizes and contract changes before blaming a retention policy. A standardized comparison can hold group weights fixed, but must disclose those weights instead of passing the result off as the observed rate.
When team sizes change quickly, examine time exposure instead of relying only on opening and closing headcount. Imagine a team with twenty people for most of a month and another twenty joining on its last day. The simple average is thirty, overstating how many people were exposed to departure over the month. A consistent daily average can help when data are available; otherwise explain that the simple denominator is approximate. Choose a method that fits the question and retain its definition in the report. Present the departure count, denominator and period together. Readers can then reproduce the calculation and recognize the effect of a small population. If one departure separates two teams, case review and trends over several periods matter more than a league table.
Lesson summary
Calculate attrition using a defined denominator and period. Inspect segment differences before recommending action or claiming causation.