Margin analysis

The margin killers: where rental profits actually leak

By the Renting For Profit editors · 30 August 2026

Rental profit does not disappear in one place. It leaks out in five. This guide puts a number on each leak and names where the number comes from.

Rent is easy to see. It arrives monthly, or per booking, in one column. Costs are hard to see, because they arrive as dozens of small events across the year: some empty days, a paint job, a water heater, a burn mark nobody photographed, a fee that went up half a point.

Each of those events is too small to trigger a review. Together they are the difference between a property that pays and one that merely occupies capital. So we went looking for what each leak is actually worth, using published filings, government survey data and operator research rather than the ranges that circulate unattributed.

Leak oneVacancy, priced by the day

Vacancy is the only leak that costs you at full retail. Every other cost is a fraction of rent; an empty day is the whole thing. The institutional single-family operators publish this number, which makes them the cleanest available benchmark for how long a home actually sits between residents.

The same research isolates the part attributable to turnover rather than to a home being held off the market: roughly 10.7 vacant days per home per year, which it prices at about $862 of gross potential rent per home. Turnover rates themselves sat at 22.8 percent for Invitation Homes and 26.3 percent for AMH in 2025, both down sharply from the mid-to-high thirties and low forties a decade earlier. Fewer moves, but each move costing more days than it used to.

Two practical readings. First, if roughly a quarter of your doors turn in a year and each turn burns a month and a half, your realistic occupancy ceiling is several points below the one in your spreadsheet. Second, days are the cheapest thing to fix, because shortening a turn costs coordination rather than capital.

Short-term rentals have the same leak on a shorter clock. An unsold night is a vacancy day; a gap night created by a two-night minimum against a three-night hole is a vacancy day you chose. The arithmetic is identical, it just repeats fifty times a year instead of once.

Leak twoThe turnover itself

A turnover is not a cleaning bill. It is cleaning, plus paint, plus flooring and appliance repair, plus lost rent while all of that happens, plus the marketing and concessions needed to land the next resident. Counted whole, it is one of the largest single expenses a rental unit generates.

That number is worth sitting with, because it is roughly a month and a half of rent in many markets, spent to replace a resident you already had. It also explains why retention spending that looks generous in isolation is usually cheap: almost any concession is smaller than $3,872.

The mistake operators make here is budgeting only the visible slice. Cleaning and paint get a line item; lost rent and concessions do not, because they are absences rather than invoices. A turn that runs four days long does not show up anywhere in the ledger as a cost, which is exactly why it keeps happening.

Leak threeMaintenance that was deferred until it was not

Deferred maintenance is a loan you take from yourself at a bad rate. The work does not get cheaper while you wait, and the failure mode is that routine work becomes emergency work, which is priced differently.

On a per-square-foot basis, RapidEye's vacation rental maintenance cost research puts median rental maintenance at $0.90 per square foot per year, drawing on Belong's dataset of more than 15,000 work orders, with a typical range of $0.90 to $1.30. The line in that research that matters most for margin is not the median at all: it is that 32 percent of repair costs are tied to emergency maintenance. Roughly a third of the maintenance budget is being spent at the worst possible moment, on the vendor who was available rather than the vendor who was cheap.

That is the actual argument for inspecting on a schedule. Not tidiness, and not documentation for its own sake, but moving spend out of the emergency third and into the planned two thirds.

Leak fourDamage that never gets billed

Damage only becomes recoverable if somebody establishes, at the time, that the property was not like that before. Most of the time nobody does, and the cost silently reclassifies itself as maintenance.

The best population-level data on this comes from the UK, where the government surveys landlords directly. The English Private Landlord Survey 2024, fielded between 3 April and 19 May 2024 with over 9,000 landlords, found that at the end of the last tenancy 59 percent of landlords returned the deposit in full, 22 percent returned part of it and 15 percent returned none, including 11 percent who said the costs incurred exceeded the deposit.

Read that alongside the 11 percent whose costs exceeded the deposit and the shape of the leak is clear. Damage and cleaning are the two dominant reasons money gets withheld, they are the two things a deposit most often fails to cover, and they are the two categories where the evidence for a charge either exists on the day of move-out or never exists at all.

The equivalent in short-term rental is sharper still, because the window is hours rather than weeks and the next guest arrives before anyone has decided whose damage it was. If the condition of a property between two stays is not recorded, the cost does not disappear. It just moves from the guest's ledger to yours.

Leak fiveFee creep on the top line

The last leak is the one owners argue about most and measure least: the distance between what the guest or resident pays and what lands in your account.

On the short-term side, the platform fee is published and easy to check. Airbnb's own help documentation states that under the split-fee structure most hosts pay a 3 percent service fee, while under the single-fee structure most hosts pay 15.5 percent, with the remainder typically paying 14 to 16 percent. Which structure applies is not a detail; it is a twelve-point swing on gross bookings.

Fee creep is rarely one decision. It is a management percentage, plus a platform percentage, plus payment processing, plus a linen or supply markup, plus a maintenance coordination margin, each defensible on its own and none of them ever reviewed together. The exercise worth doing once a year is boring and effective: take twelve months of guest-paid or resident-paid totals, subtract twelve months of owner deposits, and divide. Whatever that percentage is, it is your real fee load, and it is almost always larger than the largest single fee you can name.

Putting the five together

The leaks are not equal in size, and they are certainly not equal in how hard they are to close.

If you only have appetite for one of these this year, take the fourth. Vacancy and turnover cost real money to compress, and fee stacks take negotiation. Recording condition at the turn costs a few minutes per property and it is the only one of the five that both reduces the leak and produces the evidence you need to bill somebody else for it.

A note on what we could not verify

Two figures we wanted for this piece were cut. Airbnb's widely-quoted claim about how many listings removed or lowered their cleaning fees after the total-price display launched could not be retrieved from Airbnb's own newsroom at the time of writing, so it is not here. Zillow's Consumer Housing Trends Report figures on photo documentation and deposit outcomes were likewise not retrievable at source, and are omitted rather than repeated second-hand. If either becomes checkable again, this page will be updated and the change dated.

Sources

1. Ministry of Housing, Communities and Local Government, English Private Landlord Survey 2024: main report, gov.uk. Fieldwork 3 April to 19 May 2024, over 9,000 landlords.
2. Multifamily Dive, Turnover costs hold steady at nearly $4,000 per resident, 11 October 2023, reporting Zego's 2023 Resident Experience Management Report (630 property managers, 250-plus unit communities).
3. Airbnb, Airbnb service fees, help documentation, checked 30 August 2026.
4. RapidEye, Single-Family Rental Turnover Statistics, compiled from Invitation Homes and AMH public filings.
5. RapidEye, Vacation Rental Maintenance Cost Statistics, including Belong's dataset of 15,000-plus work orders.