Short answer: a good employee turnover rate is commonly cited as somewhere around 10% a year — but that number is a rule of thumb, not a law, and the variance between industries is wider than the rule of thumb itself. Hospitality and retail routinely run several times that and are considered healthy doing it. Government and parts of financial services run well under it. Before you decide whether your number is good, you have to know what you're counting, who you're comparing against, and which departures you'd actually have wanted to prevent.
Benchmarks in this space come mostly from a handful of sources: the U.S. Bureau of Labor Statistics (which publishes monthly separations data by sector through JOLTS), SHRM, and the Work Institute's annual retention reports. They measure slightly different things on slightly different cadences, so any single "average turnover rate" you see quoted has a methodology hiding behind it. Treat the ranges below as orientation, not a scoreboard.
First: What Are You Actually Counting?
Most arguments about whether a turnover rate is good are really arguments about definitions. Three splits matter more than anything else.
Voluntary vs. involuntary
Voluntary turnover is people choosing to leave. Involuntary turnover is terminations, layoffs, and performance exits. Lumping them together produces a number that's almost impossible to act on — a 20% rate made up of a restructuring looks nothing like a 20% rate made up of quiet resignations, but the spreadsheet can't tell the difference. Split them from the start.
Regretted vs. non-regretted
Inside voluntary turnover there's a second split that almost nobody tracks and everybody should: would you have re-hired this person? Regretted attrition — strong performers you wanted to keep — is the only number where "as low as possible" is genuinely the right goal. Non-regretted departures, including people who were a poor fit and left on their own, are often the system working. A 12% overall rate that's 3% regretted is in far better shape than an 8% rate that's 7% regretted.
Turnover vs. attrition
Some organizations use "attrition" for roles that don't get backfilled and "turnover" for roles that do. It's not a universal convention, but if your leadership team uses both words interchangeably in the same meeting, pin down which one you mean before anyone sets a target.
If you haven't settled on a formula yet, start with how to calculate employee turnover rate, or drop your numbers straight into our employee turnover rate calculator. If your reporting lives in a spreadsheet, the turnover rate formula in Excel walks through the monthly and rolling-12-month versions.
Headcount Turnover, Defined
"Headcount turnover" is the plain-language version of the standard formula: separations during a period divided by average headcount during that period, expressed as a percentage. The average part matters. Using starting headcount inflates your rate if you grew; using ending headcount deflates it. Average headcount — usually (beginning + ending) / 2, or a monthly average if you're growing fast — is the version that compares cleanly to published benchmarks.
One consequence worth internalizing: rapid hiring mechanically lowers your turnover percentage even when the same number of people are walking out the door. A 40-person team that loses 6 people looks like 15%. That same team after doubling to 80 loses 6 people and looks like 7.5%. Nothing improved. Watch the raw count of departures alongside the rate, especially during growth.
Is a High or Low Turnover Rate Better?
Lower is generally better, right up until it isn't. Very low turnover — think low single digits, sustained over years — is worth investigating rather than celebrating. It can mean you've built somewhere people love working. It can also mean:
- Nobody can leave. Depressed local job markets, visa dependencies, or highly specialized skills keep people in seats they'd rather vacate. That's retention by constraint, not by choice.
- Underperformance goes unmanaged. If involuntary turnover is near zero, the most likely explanation isn't that everyone is excellent — it's that difficult conversations aren't happening.
- New thinking never arrives. Some healthy churn imports skills, questions defaults, and opens promotion paths for people below. Teams with zero movement often develop a "we've always done it this way" problem that no offsite fixes.
The useful target isn't a minimum. It's a healthy band for your industry with the regretted-attrition slice as small as you can make it.
Benchmarks by Industry
Industry is by far the biggest driver of what "normal" looks like. BLS separations data consistently shows accommodation and food service at the top by a wide margin, with retail trade and arts/entertainment also running high — high-churn, high-seasonality, high-part-time-share sectors where annual rates in the double digits or well above are structural rather than symptomatic. At the other end, government, utilities, and parts of finance and insurance sit at the bottom, often in the single digits. Professional services, manufacturing, tech, and healthcare land in the broad middle, with healthcare having drifted upward in recent years.
Rather than reprint the table here, we keep one maintained in employee turnover rates by industry — use that to find your sector's band, then compare yourself to it instead of to a generic 10%.
Benchmarks by Company Size
Size shifts the picture in ways that are easy to misread:
- Under ~50 people: the percentage is statistically noisy. One departure out of 25 is 4%; three is 12%. Neither is a trend. At this size, look at raw counts and reasons, not rates.
- 50–500: the rate starts meaning something, and this is where turnover often rises — the founding-team glue weakens, management layers appear, and the people who joined for a 30-person company find themselves at a 300-person one.
- 500+: aggregate rates stabilize and become genuinely comparable year over year, but they also start hiding everything interesting. The company-wide number can hold flat while one department quietly bleeds out. Segment by department, manager, and tenure or you'll miss it.
Benchmarks by Role and Tenure
The single most diagnostic cut is tenure — specifically first-year turnover. A large share of voluntary departures happens within the first twelve months, and those exits point at things you control directly: hiring accuracy, role clarity, onboarding, and whether a new hire's first month felt like joining a team or filing paperwork. If a disproportionate chunk of your annual turnover is people with under a year of service, that's a hiring-and-onboarding problem wearing a retention costume. Recognizing new hires in their first 90 days covers the cheapest intervention available here.
Role matters too. Frontline, hourly, and entry-level positions turn over faster than senior or specialized ones almost everywhere — so a blended company rate that mixes a call center with an engineering org tells you nothing about either. Segment before you judge.
From a Number to an Action Plan
Knowing your rate is the easy half. Here's the sequence that turns it into decisions.
1. Price it
SHRM commonly puts replacement cost at 50–60% of the departing employee's annual salary, and the figure climbs steeply for senior or specialized roles. Worked example: a 100-person company at a $65,000 average salary with 15% annual turnover, costed at 50% of salary, is spending about $487,500 a year on replacement. Run your own version in the cost of employee turnover calculator, and see what it costs to replace an employee for the line-item breakdown.
2. Segment it
Slice by department, manager, tenure band, and voluntary/involuntary. Turnover is almost never evenly distributed — the company average is usually one or two pockets dragging an otherwise-fine organization upward.
3. Find out why, before they go
Exit interviews are a postmortem. Stay interviews, engagement pulses, and behavioral signals catch the problem while it's still fixable. The Work Institute has reported that roughly three in four voluntary departures were preventable — which is either depressing or encouraging depending on when you look.
4. Fix the cheap things first
Recognition is the clearest example of a low-cost lever with real evidence behind it. Gallup and Workhuman research has found employees who feel inadequately recognized are roughly twice as likely to say they'll quit within a year, and Deloitte/Bersin has linked strong recognition cultures to up to 31% lower voluntary turnover. We did the arithmetic on that trade in recognition vs. turnover: the retention math.
5. Set a band, not a number
Pick a target range appropriate to your industry and size, track regretted attrition separately, and review quarterly. A single annual figure with no segmentation behind it is a vanity metric with a spreadsheet attached.
Where Recognition Fits
Full disclosure — this next bit is about our own product. Propsly is a Slack-native peer recognition tool: anyone types /props to give a teammate points and a public shout-out, every give lands in a feed channel, and the leaderboard does the rest. Free for unlimited users at 200 props per person per month; the $50/month flat Pro tier adds analytics and automated rewards. The reason it belongs in a post about turnover rates is the data trail: recognition activity is one of the few leading indicators you get for free, and a team whose props volume falls off a cliff is often a team about to show up in next quarter's numbers. We wrote about using that as an early warning system for attrition.
So: is your turnover rate good? Compare it to your industry, not to 10%. Split voluntary from involuntary and regretted from not. Check whether it's concentrated in the first year. And then go look at what the number costs you — that's usually the argument that gets something done.