Employee Turnover Rate: Formula, Calculator and How to Read the Numbers

Every company can tell you its turnover rate. Far fewer can tell you what the number means. I have worked on turnover in a bank, a TV group, a game developer and a software company, and each time the single company-wide percentage was the least useful line in the report.

This guide gives you the formulas — basic, monthly, annualized, voluntary and involuntary, first-year, the stability index — and then shows how to read the result so it leads to a decision.

If you only need the number, jump to the free turnover rate calculator.

Turnover rate shows the pace at which people leave a company. It is the number of employees who left during a period, divided by the average number of people who worked there in that period, shown as a percentage.

The minimum is 0% — nobody left. There is no maximum. A rate of 100% or 200% is possible when the same positions are refilled several times a year, and in businesses such as restaurants that is not unusual.

Most companies calculate it for a year. I count it monthly as well.

The example I use in the course: 20 people left during the year. There were 89 employees at the beginning of the year and 119 at the end.

Turnover = 20 ÷ [(89 + 119) ÷ 2] = 20 ÷ 104 = 19.2%

Two details decide whether the number can be trusted:

• The average. Start plus end divided by two is the quick version. If headcount moved a lot during the year, take the headcount for each month, add the twelve numbers and divide by 12.
• Who counts as a leaver. Decide once whether you include contractors, interns, people on probation and internal transfers, write the rule down and do not change it between reports. In my reports the calculation method is a separate section at the end, so anyone can recalculate and get the same result.

The formula is the same, only the period is one month: people who left in the month ÷ average headcount in the month.

Example: a company starts the month with 200 people and ends it with 204. Three people left. Average headcount is 202, so monthly turnover is 3 ÷ 202 = 1.5%.

One month proves nothing. The value is in the row: put 12 to 18 months next to each other and you will see whether turnover is flat or climbing.

Annual turnover — everyone who left in the 12 months ÷ average headcount for the year.

Annualized turnover is a projection: what the year will look like if people keep leaving at the current pace. The method is simple. Take the rate for the months you have, divide it by the number of months and multiply by 12.

Example: in nine months 27 people left a company with an average headcount of 300. Turnover for nine months is 9%. Annualized: 9% ÷ 9 × 12 = 12%.

This is how I compared an unfinished year with the previous one in my own reports. There are two honest ways to do it: compare the same months of both years, or annualize the current period and compare it with last year’s full number. Comparing three months of this year with twelve months of last year is the mistake.

Annualizing assumes the pace stays the same. If your business has a season, say so next to the number.

Split the leavers by whose decision it was and divide each group by the same average headcount:

• Voluntary turnover = people who resigned ÷ average headcount.
• Involuntary turnover = people the company let go ÷ average headcount.

In the example above, if 14 of the 20 leavers resigned, voluntary turnover is 14 ÷ 104 = 13.5% and involuntary is 6 ÷ 104 = 5.8%.

Why split them? Because they answer different questions. In a year with layoffs the total rate shoots up, while voluntary turnover may stay flat or even fall.

One warning from practice: the label in the HR system is not always the truth. A resignation “by the employee’s own decision” is sometimes a quiet agreement started by the manager. That is why I record the initiative in the exit interview as well, not only from the paperwork. And I add one more split that the system never has: do we regret this person leaving or not. Regretted turnover is the number management reacts to.

Take everyone who left during the period and count how many of them had worked in the company for less than a year.

Share of first-year leavers = leavers with under 12 months of tenure ÷ all leavers × 100%

If you want it as a rate for new hires, divide the people who left within their first year by the number of people hired in that same intake.

This breakdown points at a specific place. My rule from the course: when people with less than a year in the company make up half of all leavers, the problem is in recruiting and onboarding, not in pay or career growth.

Turnover counts those who leave. Stability counts those who stay.

Stability = employees who have worked more than one year ÷ all employees × 100%

To calculate it for last year, take the staff list as of December 31 and count the people who were hired before January 1 of that year — they worked the whole year and some time before it. Divide by the total headcount. (The textbook version of the stability index divides by the headcount a year earlier; in the course I use total headcount. Pick one and keep it.)

Why you need both numbers. Imagine a company of 100 people with a call centre of 25. During the year the whole call centre changed twice: 25 people left in spring, the 25 who replaced them left in autumn. Fifty leavers per 100 employees is a turnover rate of 50%, and it sounds as if half the company walked out. But 75 people out of 100 did not go anywhere — stability is 75%.

Stability removes the inflation caused by one department where the same seats are refilled again and again. If turnover is 50% and stability is also around 50%, the problem is everywhere. If turnover is 50% and stability is 75%, the problem has an address.

A percentage rarely moves a CEO. Money does. In the course I use a working average: one person leaving costs the company about six monthly salaries. It is my rule of thumb for a first estimate, not a universal constant — if you have your own numbers for hiring, onboarding and the months before a newcomer reaches full output, use them.

Cost of turnover = leavers × 6 × average monthly salary

With an average salary of $1,000 a month, each leaver costs about $6,000. Fifteen leavers a year is $90,000.

The same logic gives you the return on a retention programme:

ROI = (Benefits − Costs) ÷ Costs × 100%

If the programme brings leavers down from 15 to 10 a year, you keep five people: 5 × $6,000 = $30,000 saved every year. With a programme budget of $15,000, ROI = (30,000 − 15,000) ÷ 15,000 = 100%.

There is no universal “good” turnover rate, and I do not give one in the course. The company-wide number is an average of very different situations. It starts to speak when you cut it:

1. By department. If one unit runs far above the rest, most of your effort belongs there, not in a company-wide programme.
2. By position — when the same role exists in several departments.
3. By gender. A small gap means nothing. If one group leaves two or three times more often, check whether working conditions fit them.
4. By age — in groups (under 20, 20–25, 25–30), not year by year.
5. By length of service — the first-year check above.
6. By initiative — voluntary or involuntary.
7. By reason for leaving — from the exit interview.
8. By regret — whether you are sorry the person left.
9. By pay level — by grade, or by salary bands if you have no grades.

The pay cut once gave me the clearest finding of a whole report. At the TV channel, with 2,000 employees, I sorted all salaries in a spreadsheet from lowest to highest and split them into segments. The highest turnover sat in the lowest segment: people there were heavily underpaid at the start, and a market comparison confirmed it. At the next pay review we raised the floor — our own internal minimum salary — from about $300 to about $500. Retention in entry-level jobs improved a lot, and in absolute money it cost the company little.

Then look at the trend. Put 12–18 monthly rates on a chart and add a trendline in a spreadsheet; extend it six months forward. When I started building turnover analytics at Wargaming in 2016, the rate had been growing by 1% every year for five years in a row. Nothing dramatic in any single year — but project that line a few years ahead and it stops looking harmless.

The rate only tells you where to look. The reasons come from talking to people — start with the exit interview template. The full path — measuring, research, the business case, the retention programme and its results — is in my Employee Retention Course: Reduce Staff Turnover.