Summarise with AI
How to calculate it
The standard formula measures survivors of a starting population:
Retention rate = (employees at end who were also there at start ÷ employees at start) × 100
Start with 40 people, and 34 of those same 40 are still there twelve months later, and retention is 85%. Note what the formula ignores: anyone hired during the period. That is the deliberate difference from turnover.
Why it is not simply 100 minus turnover
Turnover is calculated against average headcount across the period and counts every departure, including people who joined and left within it. Retention is calculated against the starting population and ignores mid-period hires entirely.
In a stable business the two roughly complement. In a growing or seasonal one they diverge sharply — a venue that hires 30 casuals for summer and loses most of them can post respectable retention and alarming turnover in the same year, and both numbers are correct.
Quote which method you used. Most disagreements about these figures are definitional rather than factual.
What to measure in a casual workforce
Neither headline figure means much where casuals drift in and out without formally resigning. More useful cuts:
- 90-day retention — the share of new starters still working after three months. The sharpest signal you have about onboarding and rostering
- Retention by tenure band — under 90 days, 90 days to a year, over a year
- Active-casual retention — of casuals who worked in the first month, how many still worked in the last
Key takeaways
- Retention measures survivors of a starting population; turnover measures departures against average headcount
- They are not simply inverses, and diverge most in growing or seasonal businesses
- 90-day retention is the most actionable single cut
- State the method whenever you quote a figure
Track who stays and who leaves against the rosters they actually worked, not against a headcount snapshot.
Try RosterElf free