Employee retention
Also called:retention rate, staff retention, workforce retention
It's the inverse framing of employee turnover, though the two aren't exact mirror images: retention holds one fixed group from day one, while turnover recalculates its base as headcount changes.
How the retention rate is actually calculated
Employee retention rate is the share of a workforce that stays for an entire measurement period, measured against who was already there when the period started:
Retention rate = (employees who stayed the full period ÷ headcount at the start of the period) × 100The starting headcount is fixed for the whole period. Anyone hired partway through doesn’t enter the calculation until the next period begins, which stops a hiring spree in month eleven from flattering the number. That’s the opposite convention from turnover, which recalculates its base as headcount moves: retention holds one group still and asks how many of them are left at the end.
Run it on a 200-person company that starts the year at that headcount and ends it with 178 of those original people still on payroll: 178 ÷ 200 × 100 = 89% retention, whatever hiring happened in between.
What the metric covers, and why it moves
The formula is deliberately generic. Run it over 12 months for the whole company, over one quarter for a single department, or over one manager’s reports, and the calculation doesn’t change, only the group it’s measured against.
That flexibility is also why the number is worth watching at all: it’s a lagging readout of decisions made earlier about management and fit, and the metric itself doesn’t cause anything. Gallup’s State of the Global Workplace: 2026 Report found global employee engagement had fallen to 20% in 2025, its lowest point since 2020, at an estimated $10 trillion in lost productivity. Disengaged people don’t automatically hand in notice, but they’re the pool a competitor’s offer is most likely to pull from, and a slipping retention rate is often the first place that shows up as a hard number rather than a hunch.
What it doesn’t cover
Two narrower terms get folded into “retention” that measure something more specific. First-year retention is the same calculation restricted to one cohort (new hires) and one fixed window (12 months from the start date). That’s useful precisely because it isolates hiring and onboarding quality from everything else that drives departures. A company-wide retention rate blends that signal with everyone who has been there for years, so a weak first-year figure can hide inside an otherwise healthy overall number.
A retention bonus isn’t a measurement at all. It’s one specific tactic, a payment tied to staying through a defined period, aimed at named individuals. Reaching for a retention bonus when the company-wide rate drops treats a symptom without asking what moved it. If the rate fell because pay slipped behind the market or a manager change went badly, the bonus keeps that one person around for a while and does nothing about the next person who fills the same seat.
Mercer’s 2025 survey puts average US voluntary turnover at 13%, which reads as roughly 87% retention across the same population, though the two aren’t exact mirror images since retention tracks one fixed group from day one and turnover’s denominator moves with headcount. Either way, compare the result against your own industry rather than a flat target: Mercer’s own range runs from 8.2% turnover in insurance to 26.7% in retail, so “healthy” means something different depending on the sector doing the measuring.
Where Join fits
Join is hiring software: retention itself lives outside the product, in whatever system holds the employee record after someone signs. What is inside it is the hiring trail behind that number, source channel, interviewer, and time-to-hire per role, so a falling rate can be traced back to how those people were brought in.
