Staff turnover in hotels: calculating the rate and finding where people leave

Staff turnover rate is the number of people who left during a period divided by the average headcount for that period, multiplied by 100. Count leavers and headcount on the same dates, state the period, and report seasonal contract endings separately, or the number will describe the calendar rather than the hotel.

The payroll total looks the same as last season, but half the faces at the desk are new, and the supervisor is training again instead of running the floor. The hotel feels the loss every morning briefing and still cannot say which department is producing it. This article gives you the formula, the seasonal variants, the department and supervisor split, a cost method built from records you already keep, and the exit questions that point at causes. By the end you can run the calculation on last season and know where to act.

What is the hotel staff turnover rate and how do you calculate it?#

Turnover rate is the number of people who left during a period divided by the average headcount for that same period, multiplied by 100. It counts people, not positions, and it is only meaningful when the period and the headcount rule are written down next to it.

Count leavers by their last working day, not the day the paperwork closed, so the person lands in the period where the work actually stopped. Average headcount is normally the headcount on the first day plus the headcount on the last day, divided by two. If the hotel swings hard inside the period, use the average of the monthly headcounts instead and say which method you used.

Separate voluntary leavers from dismissals and from contract endings, because the three point to different decisions: recruitment, supervision, or season design. Exclude internal transfers between departments from the hotel total, but keep them in the department view, where a transfer out is a real loss for that team. Decide once how agency and outsourced staff are treated; if they never appear on your payroll, report them on a separate line rather than mixing them into the rate.

Write the definition into the report header: period, headcount rule, and what counts as a leaver. A rate without its definition cannot be compared with anything, including last year.

Which period and which headcount should the rate cover?#

Match the period to the decision you are making: monthly inside the season to manage, season-long for a seasonal hotel, and a rolling 12 months when you want a trend you can compare with last year. Short periods on small teams produce rates that look dramatic and mean nothing.

A department of eight that loses one person in a month shows a monthly rate of 12.5 percent; annualising that produces a figure no one should act on, so show the count next to the rate. Set a minimum denominator, for example ten people, below which you publish the number of leavers only and no percentage.

Headcount answers the people question; full-time equivalent answers the workload question. Use headcount for turnover and keep full-time equivalent for the staffing plan, so the two numbers are not confused in the same meeting. A rolling 12-month figure smooths the season and is the honest way to compare a resort with a city hotel, but it reacts slowly, so keep the monthly view for in-season control.

Fix the measurement dates, for example the first day of each month, and keep them for years; moving the date quietly changes the rate without anyone leaving. Record the reason for each exit in a short fixed list of categories, otherwise the split by reason becomes a pile of free text nobody can count.

How do seasonal hotels measure turnover differently?#

A seasonal hotel needs three numbers instead of one: people who left before their contract ended, contracts that ran to their planned end, and the share of last season's staff who signed for this season. The single annual rate mixes all three and describes the calendar rather than the hotel.

In-season leavers are people who leave before the contract end date, divided by peak headcount. This is the number that costs money, because the replacement arrives into a full house. The return rate is measured on a fixed date, such as the first of May, so that two seasons can be compared on the same basis.

Planned contract completion is not a loss; report it on its own line, or every well-run seasonal hotel will look like a failing one. Measure the return rate by department and by supervisor, because staff return to a manager and a shift pattern more often than to a brand. Ask one question at the end of the season: would you come back next year? The answer is an early warning that arrives months before the hiring window.

Keep the same three numbers for three seasons before drawing a conclusion; one season is a story, three seasons are a pattern.

Why do so many new hotel staff leave in the first 90 days?#

Early leavers usually leave the job they found, not the job they were offered: the shift pattern, the transport at night, the staff housing, or a supervisor who never spoke to them in week one. Measuring the first 90 days separately shows whether the problem is hiring or onboarding.

Use a cohort measure: take everyone who started in a given month and count how many are still there at day 30, 60 and 90. Compare cohorts to see whether a change in onboarding actually moved anything. Check the promises first: the shift pattern in the advert, the duties described at interview, the meal and housing arrangements, and the transport home after a late shift.

Assign one named person for the first week, with a written list of what the newcomer must practice before facing a guest alone. Count supervisor time in the first two weeks as a real cost, because a floor supervisor retraining a replacement is not running the floor. Look for clusters: several departures in the same fortnight after a supervisor change, a menu change or a peak week is a signal about the job, not about the people. Report early leavers as their own line in the monthly report, so they are not buried inside an annual figure that looks acceptable.

A structured first week is the cheapest intervention available; see our seasonal staff onboarding checklist by department for a department-by-department version.

Where does turnover concentrate, by department and by manager?#

The hotel-wide rate is an average of very different teams. Run the same formula per department, then per supervisor, and the losses almost always sit in two or three places rather than spread evenly.

Use each department's own average headcount as the denominator; dividing department leavers by the hotel total produces a share, not a rate, and the two get mixed up in reports. Show count and rate side by side, and suppress the percentage where the team is small, so a single departure does not become a headline.

Group by span of control rather than by department name: a housekeeping team of sixty with three floor supervisors can hide one span losing most of its people. Look at timing as well as volume: departures clustered in the first weeks after a supervisor change, a rota change or the peak week point at the working conditions of that span. Cross-check with the staffing plan, because a team that loses people and covers the gap with overtime shows its cost in hours rather than in the rate.

Take the result to the person who can act on it, which is the department head and the general manager together, not the HR file. If you are unsure who owns which decision, our hotel organisation chart guide sets out the roles, and our staffing plan guide helps you compare the losses against the workload the team is carrying.

What does turnover cost, counted from your own records?#

You can build a defensible cost per leaver from four records the hotel already keeps: recruitment spend, paid training hours, overtime used to cover the gap, and supervisor hours spent alongside the new person. The number does not need to be perfect; it needs to be traceable.

Recruitment covers advertising spend, any agency or recruiter fee, and the interview hours of the managers involved, valued at their hourly cost. Onboarding and training covers the new hire's paid training hours plus the trainer's hours, uniform, name badge, materials, and any certification course paid by the hotel. Vacancy cover covers overtime hours and extra-shift payments in the gap between departure and replacement, plus any outsourced or agency cover bought in. Ramp-up covers the weeks until the new person handles a full station alone, counted as supervisor hours spent working alongside them.

Leave out anything you cannot source; a rough figure with the records behind it survives a budget meeting, a borrowed benchmark does not. For example, and this is illustrative only: if your records show twenty manager hours and thirty paid training hours per new hire, multiply each by your own hourly cost and add the advertising spend for that vacancy. The total per leaver is the number you want; do not divide it by anything.

Which exit questions actually tell you something?#

The useful exit questions are about the job as it turned out, asked by someone other than the person's own supervisor, and asked before the last week. A single question about why someone is leaving produces a polite answer that cannot be counted.

Ask about the first week, the shift pattern, the equipment and materials, the feedback they received, and whether they would return next season. Have HR or a manager from another department conduct it; people rarely tell their own supervisor what actually drove the decision. Run it in the last two weeks, not on the last day, and record the answers in the same fixed categories used in the turnover report so they can be counted.

Add stay interviews at day 90 and mid-season, which give the same information while the person is still there and can still be kept. Close the loop: pick one change per season, tell staff what changed because of the interviews, and re-measure the same cohort next season. Treat the answers as evidence about the rota and the supervision, not about the character of the people who left. If the answers keep pointing at the rota, our hotel shift planning guide covers how to rebuild it fairly.

What to do next#

Start with one number and one period: run the formula on last season, split it by department and supervisor, and add the first-90-day cohort. Then take the two departments where the losses sit, read their exit answers, change one thing in the rota or the first week, and measure the same cohort next season. The rate is only useful if it points at a place and a person who can act.