Employee turnover: the formula, a worked example and what to do when people keep leaving

Employee turnover is the share of people who left the company over a given period. The number won’t fix anything on its own, but without it you can’t see where you’re losing people or what it costs you. Below: how to calculate it, how to read the result, and what to do when it starts climbing.

The employee turnover formula

The basic turnover rate formula:

Turnover = leavers in the period / average headcount for the period × 100%

Take average headcount from your HR records. For a quick estimate, the average of the headcount at the start and end of the period will do, but if you have seasonal swings (retail before the holidays, construction in summer), average it month by month instead.

There are two ways to define the numerator — and you can’t mix them when comparing figures:

• Total turnover — every departure, including fixed-term contracts ending, retirement and redundancies.
• “Problem” turnover — resignations plus dismissals for misconduct. This is usually the number people mean when they say “our turnover is high”.

It helps to report voluntary departures as a separate line. If people leave on their own, the problem lies in working conditions, managers or expectations. If the company keeps dismissing people for misconduct, the problem is more often in hiring or onboarding.

A worked example

Say a retail chain in Almaty had 120 people on 1 January and 130 on 31 March. Average headcount for the quarter is about 125. During the quarter 18 people left: 14 resigned, 2 were dismissed for misconduct, and 2 reached the end of fixed-term contracts.

Multiplying the quarterly figure by four gives only a rough estimate. For the annual report, use actual data for the full 12 months.

Early turnover: the first 90 days

Overall turnover lumps experienced employees and new hires together. To see how well hiring and onboarding work, calculate early turnover separately:

Early turnover = leavers within 90 days / hires in the period × 100%

Work with cohorts: take everyone who started in a given month or quarter and see how many of them left before day 90. You can only run the numbers once the last person in the cohort has passed the 90-day mark.

Example: a call centre in Astana hired 25 agents in April. By the end of July, 7 of them had left — 4 within the first three weeks. Early turnover is 28%. The timing matters even more: if most people leave in the first few weeks, they either didn’t get what they were promised at the interview or were left to sink or swim.

Early turnover is expensive. The company has already paid for the job posting on hh.kz, spent recruiter and manager time on interviews, done the paperwork and trained the person — and has nothing to show for it yet. When a new hire walks out, all of that has to be done again.

You can estimate the cost from your own data. Say paid job ads and recruiter time come to KZT 80,000 per agent hired, and two weeks of training on a KZT 250,000 monthly salary add another KZT 125,000, plus buddy and manager hours. That’s over KZT 200,000 per person. Seven early leavers from the April cohort means more than KZT 1.4 million spent with nothing to show for it — before you count the dip in shift performance.

What level of turnover should worry you

There is no single norm that fits everyone. The “healthy turnover” figures floating around online rarely state the industry, country or method, so comparing yourself to them is pointless. A sales assistant and an accountant change jobs at very different rates, and a warehouse with seasonal hiring will always look worse than an office.

The warning sign isn’t an absolute value — it’s one of these three:

• Growth against your own track record. Quarterly voluntary turnover is up by a third on last year — time to dig in.
• Gaps between departments or managers. Two identical stores: one lost 3 people in six months, the other lost 11. The difference is almost always management.
• The wrong people are leaving. Top performers or key specialists leave more often than everyone else.

That’s why you should calculate turnover monthly, broken down by department, role and length of service. A single company-wide number once a year tells you almost nothing.

Why people leave in the first months

Have an honest conversation with new hires on their way out and you’ll hear the same reasons again and again:

• The job didn’t match the description. The interview was about “managing key accounts”; the reality is cold-calling a database.
• The pay didn’t match expectations. They were promised earnings of “up to KZT 600,000”, but in the first months they get a base salary with no bonuses and no idea when they’ll reach the promised figure.
• No support. The new hire got their logins and was left alone. Asking questions feels awkward, and mistakes get picked apart in public.
• Conflict with the manager. Unclear expectations, criticism without explanation, no regular check-ins.
• Workload. Shifts are understaffed, so the new hire is put on a full quota from day one.
• A better offer came along. Often a result of everything above: the person kept looking because they never felt the company wanted them there.

Most of these causes are within the company’s control. That’s the good news: you can move early turnover faster than you can move the salary market.

How to reduce employee turnover: four tools

Onboarding with a plan and named owners

The first 90 days should follow a clear plan: tasks before day one, on the first day, in the first week and month, probation goals, and a buddy. For a detailed breakdown of each stage and a ready-made checklist, see “New employee onboarding plan”. If you hire often, you can’t keep a plan like this running by hand — an onboarding module that launches the workflow automatically after a hire does it for you.

Regular feedback

A 20–30 minute 1:1 between manager and employee every one or two weeks. The agenda is simple: what went well, what’s getting in the way, what you need from your manager. Many resignations come as a “surprise” only to a manager who hasn’t talked to that person in a month.

Pulse surveys

A short anonymous survey of 3–5 questions once a month surfaces problems before the resignation letter does. Questions about workload, clarity of tasks, the relationship with the manager and eNPS — how likely people are to recommend the company as a place to work — work well. We cover how to calculate and read eNPS in a separate article.

Exit interviews

An exit interview is only useful if it isn’t run by the direct manager and if someone actually analyses the answers afterwards. The questions that tell you the most:

• When did you first think about leaving, and what happened at the time?
• What would have had to change for you to stay?
• Did the job match what you were told at the interview?
• How would you describe working with your manager?
• Would you recommend us to a friend? Why or why not?

Once a quarter, pull the answers into a table by reason and department. If three interviews in a row from the same department say the same thing, it’s no longer an opinion — it’s a task for that manager.

How to track it: the minimum setup

To calculate turnover every month instead of once a year by hand, recording five fields for each departure is enough:

Set the reason categories up front: “pay”, “manager”, “workload”, “job didn’t match expectations”, “relocation”, “career growth”, “personal circumstances”. Keep free text as a comment, or six months from now you won’t be able to group the reasons at all.

Where to start tomorrow

• Calculate voluntary turnover for the last 12 months, by month and by department.
• Calculate early turnover by hiring cohort for the last two quarters.
• Find the department or manager with the biggest deviation and talk to them.
• Introduce a leaver record and exit interviews with fixed reason categories.
• Launch a monthly anonymous pulse survey of 3–5 questions.
• Review the onboarding plan for the role with the highest early turnover.