How I run short-term rental turnovers between guests

In short-term rentals the whole operation compresses into the few hours between one guest leaving and the next one arriving. A typical listing publishes an 11 a.m. checkout and a 3 or 4 p.m. check-in, so there are roughly four hours to clean, restock, inspect, and fix whatever the last stay broke. There is no slack in that window, so I plan it like a scheduled job with a named owner and a deadline rather than a chore someone will get to.

The window is fixed, so the plan has to be

A long-term turnover gives you weeks. You can hold a unit off market, wait on a painter, and reschedule a carpet crew twice. A short-term turnover gives you an afternoon, and the deadline is a guest who is already on a plane. That single difference shapes everything else. My long-term turnover workflow is a different animal →

The booking calendar drives the schedule, not requests. Every morning I pull the next few days of checkouts and check-ins and turn them into assigned turnover jobs, each with a property, a window, and one person responsible. When a booking changes, the schedule changes with it the same day. Turnovers get missed when someone assumes a cleaner already knows about a reservation that was made yesterday.

Treat the gap between checkout and check-in as a scheduled job with an owner, a checklist, and a deadline.

Cleaning is only part of what a turnover has to catch

It is easy to think of a turnover as cleaning with some restocking attached. The complaint data suggests a wider job. In a survey of short-term rental guests compiled by ConsumerAffairs, poor internet connections were reported by 39 percent of guests and poor property maintenance by 38 percent, both ahead of lack of cleanliness at 33 percent. Poor parking and difficult landlords rounded out the top five at 23 percent each.

Two of the top three problems are things a cleaner will walk straight past unless the checklist says to stop. So the turnover checklist includes an actual wifi test on a device, not a glance at the router lights. It includes running the shower until the water is hot, checking that the air conditioning responds, and switching on the appliances the listing advertises. Catching a dead router at noon is a repair. Catching it at 4 p.m. from a guest message is a refund and a review.

A checklist that produces evidence

A checklist that only produces a tick box tells me someone pressed a button. I ask for photos at fixed points in every turnover: the made bed, the bathroom, the kitchen counter, the empty bin, and the entry area. They are timestamped, they take about a minute, and they settle almost every argument that comes later.

Evidence also protects the arriving guest from the last one. Damage found during a turnover has to be reported before check-in, because a scuff or a stain that goes unrecorded quietly becomes the next guest's problem and, eventually, the owner's cost. The same discipline that keeps work orders from disappearing applies here. More on how I keep maintenance tickets from getting dropped →

Restocking runs on par levels, not shopping trips

Consumables are inventory, and inventory works best with a set par level per property. Coffee, paper goods, soap, dishwasher tablets, and bin liners each have a target quantity, and the turnover restocks back up to that number instead of guessing. Whoever runs the turnover reports anything below par, and the resupply happens on a schedule rather than as an emergency run after a guest reports there is no toilet paper.

Linens follow the same logic. Three sets per bed keeps the system honest: one on the bed, one in the wash, one on the shelf. With two sets you eventually hit a same-day turnover with everything still damp, and that shortage turns into a late check-in.

Same-day turnovers need their own plan

Back-to-back bookings are where the schedule earns its keep. I flag them a few days out and treat them differently from a turnover with a free day on either side. That means confirming the cleaner the night before, sequencing that property first in the day's route, and deciding in advance which tasks can move to the following morning if something goes wrong.

Late checkout requests deserve a real answer rather than a reflexive yes. When a guest is arriving that afternoon, granting an extra hour eats the buffer that absorbs surprises. I either say no kindly and explain why, or say yes and immediately tell the cleaner the window moved. What I try never to do is grant it and let the schedule find out on its own.

When a turnover genuinely will not finish in time, the arriving guest hears from me before they reach the door, with a specific new time and something useful to do in the meantime. Guests forgive a delay they were warned about far more readily than one they discover standing outside with luggage.

Turnover cost mostly sits in the nightly rate now

The economics have shifted in a way that rewards predictable turnovers. Airbnb reported in its 2023 fall update that over 260,000 listings lowered or removed their cleaning fees that year, on top of nearly 3 million listings that already did not charge one. By the company's fourth quarter 2023 results, almost 300,000 listings had removed or lowered cleaning fees since early 2023, and nearly 40 percent of active listings no longer charged one at all. Lower cleaning fees had been the most requested improvement when Airbnb surveyed roughly 3,300 people in its community that May.

When the fee comes off the checkout page, the cost stays in the business. It moves into the nightly rate, which means a turnover that runs long, needs a second visit, or triggers a partial refund comes directly out of margin. Consistency is worth more here than raw speed.

Where AI helps and where it does not

I use AI for the parts of turnover coordination that are repetitive and text-heavy. It reconciles the booking calendar against the assigned schedule and flags which turnovers have nobody attached. It drafts the guest arrival message and the cleaner's job brief so both are specific about time and address. It summarizes a week of checklist notes so I can see which property keeps producing the same problem.

What it does not do is decide whether a unit is guest ready. That judgment stays with a person looking at photos against a standard. More on the AI stack behind this →