Housekeeping Staffing: Rooms Per Attendant on an Island

Fourteen.

Call that the rooms-per-attendant figure your property plans against. It might be eleven at your place, or sixteen; the exact number matters less than where it came from, which is a building. Inside a building the distance between two rooms is measured in steps, the linen store is one lift ride from every floor, and a forgotten amenity costs forty seconds to fix.

Housekeeping staffing on a villa property is the same arithmetic with one term added, and that term is large enough to change the answer. Travel is not a rounding error hidden inside the cleaning time. On a spread-out property it is often the largest single line in the shift. What makes it invisible is the number everyone inherits: rooms per attendant has no term for distance in it at all.

What follows is the arithmetic with distance put back. Every figure below is a placeholder — mine, chosen to be plausible and round. Yours will be different, and finding out by how much is the whole point.

Start from minutes

A ratio hides its assumptions. Minutes don’t.

An eight-hour shift is 480 minutes. Take out the parts of it that are not cleaning: the morning briefing, drawing linen and amenities, returning the cart, breaks, the ten minutes at the end reconciling what happened. Call that 90 minutes. What is left — 390 minutes — is what reaches the villas.

Now price one job. In a corridor hotel it is the time in the room plus a walk of a minute or two. Say 26 minutes in the room and 2 minutes to the next door: 28 minutes a job, and 390 ÷ 28 gives you 13.9. There is fourteen, and you can see exactly which assumption produced it.

Put the walk back

The same 26 minutes of cleaning on a villa property buys you a very different day.

	Corridor hotel	Villa property
Shift minutes	480	480
Non-cleaning overhead	90	90
Minutes available for villas	390	390
Time inside the room	26	26
Travel to the next job	2	11
Recovery trips, amortized	0	4
Minutes per job	28	41
Jobs per attendant per shift	13.9	9.5

Two lines did all the damage.

The travel line is 11 because a list ordered by villa number — or by the order the checkouts came in — produces the property’s average hop. Two villas assigned to the same person can sit at opposite ends of the sand. Nobody chose that. It is simply what an unsorted list does once the rooms stop sharing a corridor.

The recovery line is the one that gets left out entirely. In a building, going back for the thing you forgot costs under a minute. On a villa property it is a full round trip to the store and back, and it is not rare — a wrong linen count, an amenity that was not on the cart, a bulb. One recovery trip every third villa at 12 minutes each amortizes to 4 minutes on every job.

Fourteen became nine and a half. Same people, same speed, same standard. Across ninety villas that is the difference between seven attendants and ten.

Neither line is a scheduling failure. They are the two places where a corridor assumption meets a kilometer of sand path, and they are two of the four assumptions worth checking in any housekeeping software built for a building.

Two levers cheaper than headcount

The table also shows where the money is, and it is not in hiring.

Order the work by geography instead of by villa number and the travel line falls from 11 toward 6. That alone takes minutes per job from 41 to 36, which is 10.8 jobs a shift instead of 9.5.

Stock the cart against the actual list — which requires knowing the list before the shift starts — and the recovery line falls toward 1. Now it is 32 minutes a job, or 12.2.

Neither of those is a hire. Together they are worth close to three villas per attendant per day in this model, and both are problems of sequence — what order the work comes in, and what is on the cart when it starts. Software touches those two lines. It does not touch the walk itself; only a map does, by putting a satellite linen store at the far end of the property.

The workload that never makes it onto the plan

Departures are visible. They are on a printout, they have times, and the whole morning is organized around them. Stayovers are not on anything.

But occupancy drives stayover service, and occupancy is the bigger number nearly every day. Six checkouts and eighty-four occupied villas is six turnovers and up to eighty-four service visits. A staffing model sized to the departure list is sized to the small half of the day.

This is why the two workloads are separate objects on our own housekeeping boards: a checkout raises that villa’s turnover by itself, and a separate morning pass raises one service job per occupied villa — skipping any villa that is departing that day, because that villa already has a turnover, and skipping any villa a manager has left out of the morning pass. The two never double-count the same villa, and neither of them depends on someone remembering to write the list.

Then the arithmetic bites a second time, harder. Stayover service is a shorter job — call it 14 minutes in the villa — but the walk to it is exactly as long as the walk to a turnover. Travel is a per-visit cost, not a per-minute one.

Worked with the assumptions above — minutes, and the percentages are arithmetic on them, not measurements from any property:

	Corridor hotel	Villa property
Time inside the room	14	14
Travel to the next job	2	11
Share of the visit spent walking	12%	44%

Small jobs are punished by distance far more than large ones. The practical consequence is that geographic batching matters more for stayover service than for turnovers, which is the reverse of how most properties plan: turnovers get the careful route because they have the deadline, and service visits get whatever order they fall in.

The villa that was not free yet

One more line the model usually omits. An attendant arrives for a stayover service and the guest has not gone to breakfast.

That is a real event with a real cost: the attendant comes back later, which is a second trip. If one stayover villa in five is occupied on first approach, add another 2 minutes to every service job in the model.

Whether that cost lands in your model or vanishes depends on what the system lets the attendant say. If the only two answers available are “done” and “still open”, the return trip is invisible and so is the reason for it, and you end up planning against a service round nobody has fully costed. Our boards take a pause with a reason attached, which is enough to make the second visit countable at all. For staffing purposes that is the whole of it: the return trip is a real hop and it belongs in the arithmetic, whether or not anything records it.

Getting your own numbers without timing anybody

Three of the four figures in the model are properties of the property, not of the people, and can be measured as such.

The travel number is a map exercise done once: walk the routes with a cart at a working pace and record the hops between adjacent villas on the paths people really use. That is an afternoon, and it produces a number you will use for years.

The recovery rate is a conversation. Ask three attendants how many times yesterday they went back for something, and what it was. The answers are usually specific and the fix is usually a stocking list.

The overhead number is one stopwatch on one shift start — the briefing, the linen queue, the wait for a key, and the radio round that reconstructs which villas are ready before anyone can be sent anywhere.

The fourth figure — time inside the villa — needs no stopwatch either: the boards’ own work screen shows the average minutes a job took, per person, over any dates you pick.

The arithmetic here is yours, done with a map, a notebook and a spreadsheet. Our boards hold the jobs and the job minutes; the staffing model is a decision you make from them.

What the number decides

Run the model and one of the three lines will dominate. Each points somewhere different.

If travel dominates — if the walk is more than about a third of the per-job total — the answer is geography before headcount. A satellite linen store, a route-ordered list, a cart that survives the whole run. Hiring an eighth attendant buys you an eighth person doing the same walk.

If time in the villa dominates, the conversation is about the standard and the equipment, and no scheduling change will move it.

If overhead dominates, the shift start is the problem: people waiting for a briefing, a key, or a list that someone is still writing. That one is embarrassing to find and cheap to fix, because the day can arrive on a phone before anyone reaches the office.

Headcount is the expensive answer to all three questions and the correct answer to only one of them. If your travel line comes out largest, the next conversation is with a map. If it is second and the ordering is the thing you cannot fix by hand, a walkthrough on your own villa list is a reasonable way to find out whether the route and the recovery rate move.