Attrition rate is the percentage of employees who leave your organization over a set period. You calculate it by dividing departures by average headcount, then multiplying by 100. For shift and multi-site teams, early exits in the first 90 days often matter as much as the annual headline number.
People and operations leaders track it to spot retention risk, compare locations, and estimate replacement cost before the roster falls apart. Everyday speech often treats attrition and turnover as synonyms, but published definitions are not always identical — so pick one rule and keep site reports comparable. Fair employee scheduling and clear time tracking help you see overload before people quit.
This page is about the employee metric. University dropout rates, clinical study attrition, military training washout, and customer churn use the same word differently and are out of scope here.
What is attrition rate?
In simple terms, employee attrition rate shows how quickly people leave relative to workforce size. If 12 people leave a team that averaged 100 employees over a year, the annual rate is 12%.
Most SMB and mid-market dashboards count all departures in the period — resignations, retirements, and involuntary exits — unless you intentionally report voluntary-only. Label what you include so site managers compare like with like. A voluntary rate of 8% and a total rate of 12% can both be true for the same year.
Replacing someone costs recruiting time, training, and lost productivity. For frontline roles, a practical planning range is often a few weeks to a few months of wages once you add overtime cover, hiring, and ramp-up — exact figures vary by role and market.
In hospitality, retail, logistics, and care, high attrition also breaks roster coverage and service quality. Track the rate monthly alongside absence, overtime, and time-to-fill. Pair the number with exit notes; the percentage alone never tells you why people left.
The rate is also not the same as headcount growth. You can grow headcount and still show high attrition if many people leave while you hire even faster. That is why average headcount in the denominator matters more than counting leavers alone.
- Usually count in the numerator: resignations, retirements, dismissals, redundancies, and end of fixed-term contracts when your policy includes them.
- Usually leave out: people who never started (offer declines), and internal moves unless you define attrition to include transfers out of the measured unit.
- Decide once for the denominator: heads vs FTEs, and whether employees on leave stay in headcount.
Attrition vs turnover vs retention
Two definitions show up again and again in HR writing:
- Everyday ops usage: Attrition rate and turnover rate both mean the departure rate for a period. Many teams use the words interchangeably in weekly reviews.
- Vendor “replacement” usage: Attrition means roles left unfilled after someone leaves; turnover means roles that are backfilled. Some HRIS and HR-software guides push this split.
Neither view is wrong — mixing them without a written rule breaks benchmarking. Document your definition, then brief franchisees and regional managers the same way every month.
For shift ops teams, we recommend the everyday formula — departures ÷ average headcount × 100 — and a separate count of open roles still unfilled after an exit. That keeps the leave rate comparable month to month without hiding resignations behind slow hiring.
| Metric | What it usually measures | Use it when… |
|---|---|---|
| Attrition rate | Departures ÷ average headcount × 100 | You need a clear leave rate over time |
| Turnover rate | Often the same formula; sometimes only replaced roles | You follow a vendor definition — document it |
| Retention rate | Share of employees who stay over a period | You want the inverse narrative (“% still here”) |
Retention is the flip side when definitions align: if annual attrition is 15% on a stable headcount, roughly 85% of the average workforce stayed. Exact math depends on starters and leavers — use cohort retention (for example, “% of January starters still employed at day 90”) when you care about new-hire quality rather than a blended annual rate.
People sometimes say “churn” in conversation. In product analytics, churn usually means customers lost. On board slides, prefer “employee attrition rate” so finance does not confuse people metrics with inventory or revenue turnover.
How to calculate attrition rate (formula + example)
The standard formula used across HR practice:
Attrition rate (%) = (Number of departures in the period ÷ Average headcount in the period) × 100
Average headcount is typically (headcount at start + headcount at end) ÷ 2 for that same period. Count people who left (voluntary and involuntary unless you filter). Do not subtract new hires from the numerator — new hires belong in the average headcount if they worked during the window.
Worked example (café with shifts)
A café starts January with 40 employees and ends with 44. During the year, 10 people leave. Average headcount = (40 + 44) ÷ 2 = 42. Attrition rate = (10 ÷ 42) × 100 ≈ 23.8%.
| Input | Value |
|---|---|
| Start headcount | 40 |
| End headcount | 44 |
| Departures | 10 |
| Average headcount | 42 |
| Attrition rate | 23.8% |
In a spreadsheet, put start headcount, end headcount, and departures in three cells, then compute average headcount and the rate. Avoid counting the same person twice if they left and were rehired in the same period unless your policy says otherwise — write rehire rules down for auditors.
Monthly vs annual and common caveats
Monthly rates look small. Annualize carefully, or compare same-length periods. A 2% monthly rate is not automatically a 24% annual rate if the same people cannot leave twice — treat annualization as a planning estimate, not exact math. Seasonal businesses should compare year-over-year months, not a single peak month against an annual peer benchmark.
Mergers, store openings, large seasonal cohorts, and unpaid leave policies can distort averages. Segment by location, role type (hourly vs salaried), and tenure when the overall rate hides a hotspot. If two people share one FTE slot, decide whether you count heads or FTEs and stay consistent.
Some organizations prefer a midpoint method or a daily average headcount from the HRIS. Those approaches are fine for precision; for most SMBs, start/end averages are good enough if everyone uses the same rule.
When you present the rate to leadership, show it next to headcount and hiring volume. A rising rate with flat hiring means the team is shrinking. A rising rate with aggressive hiring can still grow headcount while burning training capacity — those two stories need different actions.
Types of attrition (voluntary, involuntary, early)
Voluntary attrition is when employees choose to leave — new jobs, relocation, burnout, childcare conflicts, or better schedules elsewhere. It is the main lever for culture and ops improvements because managers can influence many of the drivers.
Involuntary attrition covers dismissals, redundancies, and end of temporary contracts when you count them. Separating voluntary from involuntary shows whether exits are controllable. A spike after a site closure is a different story from a spike after rota changes.
Early attrition (new-hire attrition) tracks exits in the first 30–90 days. For shift teams, it often signals broken onboarding, unclear rotas, missing equipment, or first-week overload. Fixing early exits usually moves the annual rate more than broad engagement programs alone. Report it as its own KPI next to overall attrition.
You may also track internal attrition — transfers out of a site or department into another unit. That can look like “loss” for the sending manager even when the company retained the person. Decide whether internal moves count in site-level attrition before you rank locations.
A practical early-attrition checklist for shift employers:
- Did the first published rota match what was promised in hiring?
- Was a buddy or trainer scheduled on the first peak shifts?
- Were access, uniforms, and system logins ready before day one?
- Did overtime or short-notice changes appear in week one without explanation?
Some teams also mark exits as regrettable or non-regrettable (high performers you wanted to keep vs expected turnover). Keep that judgment layer separate from the core rate so the math stays auditable, and use clear criteria — two managers can disagree on the same person.
What is a good or high attrition rate?
There is no single “good” attrition rate. Industry, wage level, contract mix, and local labor markets all move the baseline. Still, managers need a practical reading guide — not a law.
- ~5%: Often healthy for stable office or specialist teams over a year — confirm you are not freezing hiring so hard that the rate looks low while remaining staff burn out.
- ~10%: A common watchpoint in HR guides. Investigate by team if it is rising or concentrated in new hires. It can be fine when tenure is long and exits are planned retirements.
- ~20%: Usually high for steady white-collar teams; still common in high-churn hospitality or retail unless you segment correctly. Treat it as a prompt to dig into causes, not a moral judgment of the brand.
- ~25%+: Often a structural problem for multi-site operators if it persists year over year — dig into which roles and tenures drive it.
- ~80%: Extreme for most employers — often a data error, a micro-team, or a site in crisis. Recheck headcount and period length before presenting it.
Compare yourself to your own trailing twelve months and peer sites first. External “industry averages” vary wildly by source and job family. Call-center, nursing, and restaurant averages are not interchangeable with corporate HQ benchmarks.
Also read the rate with tenure mix: a 12% rate driven by planned retirements is a different problem from 12% driven by day-60 resignations. Split early attrition and average tenure of leavers before you set one company-wide target.
If you operate across US and UK sites, keep the formula the same. Interpretation of “good” still depends on local labor markets and contract norms — use one global template with local commentary rather than different math per country.
Why attrition rate matters for shift teams
Shift operations feel attrition immediately: open shifts, last-minute agency spend, and managers covering the floor. Unlike desk teams, you cannot quietly absorb a vacancy for weeks without guests, patients, or customers noticing.
High attrition also damages schedule fairness — remaining staff inherit unpopular shifts, which can trigger more resignations. That feedback loop is why roster quality and attrition belong in the same ops review. When you publish rotas late or change them daily, voluntary exits often follow even if pay is competitive.
Concrete cost channels for hourly sites include overtime premiums, agency or temp markups, manager time spent recruiting instead of coaching, and slower service while new hires ramp. A café that loses two baristas in peak season may spend more on short-term cover than the wages of those roles — even when the annual percentage looks “only” mid-teens.
Attrition also drains site knowledge: closing routines, regulars’ preferences, safety habits that never made the handbook. That loss shows up as longer tickets, more mistakes, and heavier load on the people who stay — which can push the next round of resignations.
Use the rate with related metrics: absence patterns, overtime hours, time-to-fill for frontline roles, and early attrition. Together they show whether the problem is hiring, onboarding, or day-to-day working conditions. A site with low attrition but sky-high overtime may simply be one resignation away from crisis.
Multi-site operators should rank locations by attrition and early attrition, not only by sales. A profitable site with 40% annual attrition may still destroy manager time and brand reputation. Put the rate on the same ops dashboard as late-published rotas and overtime hours so the link stays visible.
How to reduce attrition (practical levers)
Start with diagnosis: short exit interviews or pulse questions by site and shift pattern. Ask about rota notice, overtime surprises, and first-week readiness — those answers map faster to ops fixes than a generic “culture” score. Then pull levers you control week to week:
- First 90 days: Structured onboarding, buddy shifts, and a clear first-month rota — not a sink-or-swim first weekend. Confirm equipment, access, and training before the first peak shift.
- Fair scheduling: Publish rotas early, limit back-to-back close-then-open (“clopening”) where possible, and balance unpopular shifts. Tools for employee scheduling make fairness visible to managers and staff.
- Workload visibility: Track hours and overtime honestly with time tracking so chronic overload shows up before people quit. Hidden overtime is a silent driver of exits.
- Manager capacity: Frontline managers need time for check-ins; attrition often clusters under one supervisor, not “the brand.” Coach or redistribute load when one team’s rate diverges.
- Clean exits: Orderly offboarding protects remaining teams and your employer brand. See our guide to offboarding software for process patterns (tools vary; pick what fits your size).
Pay and benefits still matter — especially when competitors raise wages — but schedule predictability and respectful onboarding often explain why two sites with similar pay show different attrition. Pilot changes on one site first, then keep measuring early attrition for 90 days so you know whether the lever worked.
Ordio helps shift employers with scheduling, time tracking, absences, and employee files — the operational backbone behind retention. It is not a dedicated attrition-analytics HRIS or predictive people-science platform. Pair ops data with your HR system when you need deeper people models. For a broader metrics toolkit, see HR analytics software and related onboarding software patterns. Retention also ties to your lived employee value proposition — the promise people experience on the rota, not only on the careers page.
Summary
Attrition rate is the share of employees who leave over a period, usually calculated as departures ÷ average headcount × 100. Decide whether you treat it as synonymous with turnover or use a “not replaced” definition — then stick to one rule.
Interpret 10% or 20% in context, segment early attrition for shift teams, and fix scheduling and onboarding before chasing vanity engagement scores. Track voluntary and involuntary exits separately when leadership needs controllable resignations in view.
Consistent measurement beats a perfect industry benchmark. For frontline teams, ops discipline — fair rotas, transparent hours, solid first 90 days — is often the fastest way to move the number. if you want to see how Ordio supports multi-site scheduling and time tracking behind that ops discipline.