Someone in a forum says a two-bedroom turnover takes two and a half hours. You have seven properties, three of them two-bedrooms, and your crew is billing four. That number now sits in your head, and the next conversation with your cleaner starts from it.
It should not. The forum number describes someone else's flat, someone else's laundry arrangement, someone else's stairs and someone else's definition of when the clock starts. Held up against your crew, it is not a benchmark. It is a stick.
There is a real number for your properties, and it is worth having, not to push the crew down, but because scheduling, pricing a turnover fairly and promising a 15:00 check-in all depend on knowing it. You just have to measure it yourself, per property, with a definition you wrote down.
This article covers what actually drives turnover time, what to measure and where the clock starts, how to build a baseline from your own turnovers, how to tell a genuine outlier from normal variance, why the checkout-to-check-in gap matters more than the duration, and what happens to a crew that knows it is being timed.
What Actually Drives Turnover Time
Turnover time is not one variable. It is the sum of a dozen, and most of them are properties of the property rather than the person holding the mop.
| Driver | Why it moves the number |
|---|---|
| Bed count and bed type | Stripping and making dominates the clean; a sofa bed or a bunk with a ladder costs more than a double |
| Bathroom count | Bathrooms are the slowest room per square metre, and the second one rarely halves |
| Laundry on site or off | On-site laundry adds waiting; off-site adds a collection, a drop and a dependency on someone else |
| Access | Third floor with no lift, parking two streets away, a key collection: all of it before any cleaning happens |
| Length of the previous stay | A nine-night family stay leaves a different property than two nights for a couple |
| Who stayed | A toddler, a dog, six friends and a kitchen that got used are four separate multipliers |
| Restock depth | Consumables, welcome basket, linen sets, whether stock lives on site or in the boot of a car |
| Season | Sand, mud, wet coats, heating on, windows shut for a week |
| The last turnover | A property that was reset properly last time starts from a better place |
Read that list and the universal benchmark stops making sense. A one-bedroom with a washer-dryer in the cupboard and a lift is a different job from a one-bedroom with a laundry run and forty stairs, and no average covers both. A number that ignores the drivers is not describing work: it is describing an opinion about how fast people should move.
The same list is the reason a per-property number is achievable. The drivers are stable for a given property. Once you know what a turnover at number 14 costs, that estimate holds until something structural changes.
Measure Arrival to Clean-Ready, Not Time on Site
Most operations that do measure measure the wrong thing. "Time on site" is what a crew reports, and it quietly excludes travel, the linen pickup, the twenty minutes waiting outside because the guest was still in the shower at 10:40, and the trip to buy the toilet roll nobody restocked.
Measure arrival to clean-ready: from the moment the property is actually available (the real checkout, not the scheduled one) to the moment it is guest-ready and someone has said so. That is the number your schedule depends on, because it is the number that consumes the gap.
Then break it into parts, because a single total tells you nothing about what changed.
- Travel and access. Door to door, including parking and key collection.
- Wait. Time between arrival and being able to start. This is where late checkouts show up as a cost.
- Clean. The part everyone thinks they are measuring.
- Laundry handling. Loading, unloading, folding, or the collection and drop if it is off site. Note wall-clock separately from labour: a machine cycle occupies the property without occupying a person.
- Restock and reset. Consumables, linen sets, furniture back where it belongs.
- Report. The photo pass and anything flagged for maintenance.
An illustrative shape for a two-bedroom, one-bathroom flat with a washer-dryer on site: 15 minutes travel and access, 5 waiting, 95 cleaning, 20 handling laundry, 10 restocking, 10 on the report. Those numbers are invented to show the structure, not measured. If you use them as a target for your own crew you have done exactly what this article is arguing against.
The value of splitting it up is diagnostic. When a property goes from 150 minutes to 190, the total tells you there is a problem and the split tells you the laundry moved off site or the guest keeps checking out late.
Building Your Own Baseline
This takes a few weeks and a shared note. It does not need software.
- Pick one property to start. Do not roll this out across a portfolio on day one; the definition always needs a revision after the first fortnight.
- Write the definition down and give it to the crew. When the clock starts, when it stops, what counts as clean-ready. If two cleaners measure differently, you have two datasets and no baseline.
- Record ten to fifteen turnovers. Three is not a baseline, it is an anecdote. Fifteen gets you a usable median in most seasons.
- Record the context with each one. Previous stay length, number of guests, pets, actual checkout time, anything unusual. Without context you cannot explain your own outliers later.
- Take the median, not the average. One turnover after a party drags a mean upwards and makes every normal turnover look fast.
- Write down the spread as well as the middle. "Most turnovers land between 130 and 175 minutes" is more useful for scheduling than a single figure, because you schedule against the upper end and pay against the middle.
- Re-baseline when something structural changes. New sofa bed, laundry moved off site, a different crew, a property switched from couples to families. The old number is now describing a job that no longer exists.
What you end up with is not an industry benchmark and should never be presented as one. It is a description of your own property, which is the only thing that was ever going to be useful.
Outlier or Just Variance
Once you have a baseline, the temptation is to react to every number above it. Resist that, because normal variance is wide in this work. Two turnovers of the same flat by the same cleaner in the same week can differ by twenty or thirty minutes for reasons nobody could have predicted or prevented.
A genuine outlier has three properties. It sits well outside the spread you recorded, not just above the median. It has a cause you can name: a blocked drain, a checkout at 12:30, a guest who moved all the furniture. And when the cause is removed, the number returns.
Normal variance has none of those. It wobbles, it has no story attached, and it reverts on its own. Acting on it produces a crew that feels watched and no change in the number.
Two practical rules. Do not act on one data point; act on a run, three consecutive turnovers above your upper bound, or a shift in the median over a month. And before concluding anything about a person, separate the property from the cleaner: compare the same property across different cleaners, and the same cleaner across different properties. Very often the "slow cleaner" is the person who always gets the third-floor flat with the off-site laundry.
One outlier type is worth flagging separately, because it is not a cleaning problem at all: the turnover that ran long because the crew found something broken. That is a maintenance event and often a conversation with the previous guest. See handling damage without losing the review for what to do with it, and keep it out of your cleaning numbers.
The Gap Is the Real Constraint
Here is where the benchmark obsession does the most damage. Hosts optimise the duration of the turnover when the thing that actually breaks is the gap.
Checkout at 10:00, check-in at 15:00 gives you five hours. That sounds comfortable against a two-and-a-half-hour clean, right up until you account for what else is inside it: a late checkout you granted at 11:00, a crew doing three properties in the same window with travel between them, a laundry cycle that has to finish before beds can be made, and the handyman who can only come at 13:00. A turnover that fits on paper and not in practice is exactly how the 15:00 check-in slips.
The scheduling question is not "how long does a turnover take" but "how many turnovers can this crew complete in this window, in this order, given travel and laundry." Those are different questions and only the second one predicts a late check-in. When the answer is tight, the fixes are structural rather than motivational: stagger your check-in times, move the laundry, add a second crew for the back-to-back days, or stop granting late checkouts on changeover days. The mechanics of that day are covered in managing back-to-back bookings.
The other honest fix is the promise itself. If your true clean-ready time on a busy Saturday lands at 15:40, publishing a 15:00 check-in buys you a guaranteed complaint every changeover. Either fix the gap or move the promise.
What Timing Does to a Crew
Now the uncomfortable part, and the reason to be careful with everything above.
A crew that knows it is being timed gets faster. It does not get faster by inventing efficiency. It gets faster by cutting what nobody checks, and what nobody checks is precisely what guests notice later: under the beds, behind the toilet, the inside of the kettle, the oven, the top of the fridge, the balcony, whether the spare bulb is still in the drawer, whether the remote batteries work. None of it is visible at 15:00. All of it is visible by day three, when the guest opens the oven or makes tea.
This is a predictable consequence of measuring, not a failure of your crew. A number becomes a target, and a target gets met from wherever the slack is. The slack is the invisible work.
Three things reduce the damage. Define what "done" means in enough detail that time cannot be taken from the wrong place: that is the job of a written standard rather than a stopwatch, which is why a cleaning SOP your crew can follow comes before any timing exercise. Pair timing with a quality check, so speed and standard are measured together rather than speed alone; turnover inspection photos are the usual mechanism. And never publish a league table of cleaners by minutes. Ranking people by a number that is mostly a property of the property is how you lose the good ones.
Be honest with the crew about why you are measuring. "I need to know whether the schedule is realistic and whether you are being paid fairly for number 14" is a different project from "I think you are slow," and crews can tell which one they are in within about a week.
Honest Limits
The biggest one is in the section above: measuring changes the thing you are measuring. A baseline built while the crew knows it is being timed is already slightly faster than reality, and the difference has come from somewhere. Treat your first fortnight of data as contaminated and expect the numbers to settle.
Your baseline is an estimate, not a fact. Fifteen turnovers gives you a middle and a spread with real uncertainty around both, and that uncertainty is largest for exactly the properties you care about: the ones with variable stay lengths and mixed guest types. Do not put two decimal places on it or build a pricing model that assumes it is exact.
Benchmarks from other operations are not comparable and will not become comparable. The definition of the clock differs, the drivers differ, and what counts as clean-ready differs. Anyone publishing a turnover standard across the industry is averaging things that should not be averaged, and any number you read (including the illustrative ones here) should be treated as a shape rather than a target.
A baseline also creates no capacity. Knowing that Saturday needs seven hours of work in a five-hour window tells you the schedule is broken; it does not fix it. The fix is hiring, staggering, or changing the promise, and all three cost something.
Finally, to be plain about the tooling: Welco is not a cleaning platform. It does not schedule cleaners, time turnovers, run inspections or verify a finished room. Turno, Breezeway and Properly are built for that, and this article's measuring can be done in a shared spreadsheet without buying anything.
Where Welco Sits
Welco is a WhatsApp AI assistant for vacation rental hosts, and its part in the turnover is upstream of the clean. A verified guest confirming they have left, or reporting a broken blind on the way out, turns "the property is available at some point after 10:00" into a known start time and a task the crew can see before they arrive. That is the input your baseline depends on, and it is the part most operations get from a guess. The wider picture is in managing cleaning crews across multiple properties. If you want to see how the checkout confirmation and the escalation actually land, request a demo.