When to hire a hotel general manager: the hours you cover
· labor cost, labor planning, small hotels
There is no room count at which you hire a general manager. The number that answers the question is how many hours a week you cover yourself, split into hours where a person must be standing somewhere and hours where something merely has to get done — and that number is in no labor report at your property, because you are not on the payroll. At an invented 62-room interstate property it was 25 hours a week, 19 of them coverage. Handing the easiest 11 to the people already on the schedule costs nine hours of premium pay a week, or nothing at all, depending entirely on which of them you hand it to.
Why your own hours are missing from the labor report
A labor report is built from payroll, and you are not in payroll — so your hours are in no report, no labor percentage, no hours-per-occupied-room figure, and no scheduling decision that follows from them. This is not a bookkeeping error. An owner’s own time genuinely is not a wage expense. But it is capacity, and capacity is what you are measuring when you ask whether the property is staffed correctly.
It is a different problem from recorded hours landing in the wrong place: hours in the wrong bucket are at least countable. These are in no bucket at all, and the only person who can count them worked them and never wrote them down.
A week, written down
One ordinary week at the invented property. The last column is the part that matters.
| When | What | Hours | Kind |
|---|---|---|---|
| Mon 06:00–09:00 | Desk, before the weekday agent starts | 3 | Coverage |
| Tue 06:00–09:00 | Desk | 3 | Coverage |
| Tue 13:00–15:00 | Ordering, linen count, one interview | 2 | Admin |
| Wed 09:00–13:00 | Payroll, invoices, rate changes, OTA messages | 4 | Admin |
| Thu 06:00–09:00 | Desk | 3 | Coverage |
| Fri 05:30–07:30 | Breakfast set-up | 2 | Coverage |
| Sat 23:00–07:00 | Night audit, the auditor’s night off | 8 | Coverage |
| Total | 25 | 19 coverage, 6 admin |
Nineteen and six is the whole argument. The six admin hours can move in time: nobody is waiting at two on Wednesday for the invoices. The nineteen cannot — if nobody is at the desk at 06:15 on Monday, nobody is at the desk at 06:15 on Monday. Coverage hours price differently from every other hour, which is the arithmetic behind the 168-hour problem.
What happens when you hand the easy hours back
Take the 11 hours that look most transferable: three mornings of desk relief and the Friday breakfast set-up. The existing weeks are a senior desk agent at 38 hours, a weekday agent at 32, a weekend agent at 24, a night auditor at 40, and two room attendants at 34 and 30, one cross-trained on the desk. Three honest ways to place the 11:
| Option | Where the 11 hours go | Premium hours created | What it actually costs |
|---|---|---|---|
| A | All 11 to the senior desk agent (38 → 49) | 9 | The obvious choice, and the most expensive one |
| B | 6 to the weekday agent (32 → 38), 5 to the weekend agent (24 → 29) | 0 | A weekend-only agent drives in for a 3-hour Monday; the 38-hour week is two hours from the threshold, so any call-out longer than two hours is premium |
| C | 6 to the cross-trained attendant (30 → 36), 5 to the weekday agent (32 → 37) | 0 | Six hours leave housekeeping on two mornings, so the first checkout clean starts later on both |
A is what most properties do: the senior agent can open the desk unsupervised and is already awake. At the 40-hour weekly threshold this example assumes, that is nine premium hours every week — the largest consequence of the decision, and it appears nowhere in the conversation that produced it. Whether a daily threshold also applies depends on your state, which moves B and C too; the states with a daily overtime rule are worth checking before you commit a schedule.
The finding is worth more than the example: your own hours are free, and replacing them is priced at the margin of whoever is already closest to the threshold. You cannot see that margin unless every week-to-date sits in one place — start a free trial and the schedule and the hours recorded against it answer A, B or C before you offer anyone the shift.
The eight-hour Saturday audit is not in that table because nobody absorbs it: a single-person weekend overnight is the hardest shift at the property to fill, which is why the owner still works it.
What a general manager actually replaces
A general manager is an excellent answer to the six admin hours and a poor answer to the nineteen coverage hours. The mistake is assuming one hire buys both.
The arithmetic looks like it works: a 40-hour manager absorbs six hours of admin and has 34 left, more than the 19. In practice that is the arrangement where the new manager opens the desk at six, works the Saturday audit, and does payroll between interruptions — three jobs, one of them the one you hired for. It is the arrangement people leave.
Your hours are also elastic: you stay until the fire panel is reset, you answer the phone at two in the morning, and no rule in the building governs the length of your week. An employee’s hours are governed by rules yours are not — which rules apply to a salaried manager varies by state and by the job, and belongs with your accountant or an employment attorney rather than a blog post. That is why the swap is never hour-for-hour, and why “hire a GM and get my mornings back” often ends as a GM and a part-time morning agent: a different budget from the one that was approved.
The number a buyer will already have
If you ever sell, the hours stop being invisible immediately. The conventions for presenting a small business’s earnings treat owner labor explicitly — one adds the owner’s own compensation back into earnings, another deducts what a market-rate manager would be paid in its place — and which basis is used moves the number. Your accountant knows which one yours is on, and it is worth asking long before a sale: the answer is a price put on the hours no labor report of yours has contained.
Producing the number in two weeks
You do not need software for this, and it would be dishonest to pretend otherwise. A notebook works: for two weeks write down when you started, when you stopped, and one word — coverage or admin.
Two rules make it worth doing. Do not reconstruct it from memory: memory keeps the four-hour Wednesday block and drops the eleven fifteen-minute interruptions, and the interruptions are where owner-weeks go. And record the hours you think do not count — the walk-round at 22:00, the Sunday phone calls — because a replacement will either be paid for them or not do them.
What a time system changes is that it stops being a project. If you clock in the way everyone else does, your hours sit beside theirs in the same record and against the same schedule, and what happens to them at payroll is decided when punches are approved rather than never asked. The notebook answers the question once; the record answers it again next March, when the number has moved.
Where this doesn’t apply, and where it stays hard
If you are already absentee with a manager in place, the number is on the payroll and this is about somebody else’s property. If you take a wage from your own business — some ownership structures require it — then your hours are in payroll and the first section of this post does not apply to you, but the rest does: payroll will have your total and still not know which of those hours were coverage and which were admin. If a spouse or an adult child covers shifts, the invisible capacity is identical and the question is harder, because whether they are an employee at all depends on how the business is structured and on rules that vary — an accountant’s question, and not one to settle from a search result.
Two limits on the method. The coverage/admin split is cleaner on paper than in a corridor: the Wednesday block gets interrupted by the desk, which is part of why owner-weeks feel longer than they measure. And the number is an input, not a decision — whether the property can carry a manager’s salary is a question about your own statements, and no time system, ours included, answers it. What it can do is stop the 25 hours being a guess, which is the half of this decision nobody measures.
Start seeing what labor actually costs
Set up your property, put staff on the clock, and read tomorrow morning’s labor number.
14-day free trial · No credit card · Cancel anytime