Is Airbnb Arbitrage Profitable? The Full Arithmetic

Is Airbnb arbitrage profitable? Sometimes, thinly, and almost never in the way the income screenshots suggest. The model works when a property fills enough nights at a high enough rate to clear a rent that arrives whether it fills or not — and the honest answer to "how much" is a range that includes negative numbers. This guide gives you the full arithmetic, the four costs that turn a healthy-looking gap into a loss, what a realistic year looks like, and the test that tells you whether a specific property clears before you sign anything.

Is Airbnb arbitrage profitable in 2026?

Short answer: it can be, on a property that fills well above half the month at a rate meaningfully above its long-term rent, in a market where short-term letting is permitted. On a typical single unit the realistic outcome is a few hundred dollars a month in good months and a loss in quiet ones, with the year decided by how many of each you get.

• What has to be true: a rent gap wide enough to survive four deductions, and enough demand to fill the calendar
• What kills it: empty nights, because the rent does not pause
• What is not the problem: the nightly rate, which most operators set correctly
• What decides the year: your three worst months, not your best one

The full arithmetic, with nothing left out

Most published examples subtract rent from revenue and stop. Here is what sits between them. All figures below are illustrative.

• Gross nightly revenue: $145 x 21 nights = $3,045
• Minus rent: −$1,900
• Minus platform fee at the host-only rate, roughly 15.5 percent = −$472
• Minus cleaning paid out, 8 turnovers x $75 = −$600
• Minus utilities, wifi, consumables: −$260
• Minus insurance and software: −$70
• Net for the month: −$257

Twenty-one nights at a decent rate, and the month is negative. That is not a pessimistic example; it is a normal shoulder month. The same property at 26 nights clears about $460. Five nights is the entire difference between a profitable month and a loss, which tells you exactly where the business is won.

What does an arbitrage operator actually earn?

Short answer: on one unit, realistically a few thousand dollars a year rather than a salary, and often less than that in year one once furniture is repaid. Published "arbitrage salary" figures usually describe operators running six to fifteen units, which is a different business with staff, systems and a very different risk profile.

• One unit, run well: low thousands a year after all costs, before your own time
• Year one: usually less, because furniture has to be repaid out of margin
• Five units: the numbers start to resemble an income, and coordination becomes a job
• The hourly truth: divide the annual net by the hours spent and compare it honestly to what those hours earn elsewhere

Meet Rafael: three units, one honest year

Rafael ran three leased units in a mid-sized city. All figures below are illustrative.

• Combined rent: −$5,400 a month, −$64,800 for the year
• Gross revenue: $96,300 across the three
• Platform fees, cleaning, utilities, insurance, software: −$24,900
• Net for the year: $6,600
• Furniture, repaid over two years: −$11,000 total, so −$5,500 attributed to this year
• Actual first-year result: about $1,100, on roughly 500 hours of work

Before: Rafael projected $2,000 a month from a spreadsheet built on peak-season occupancy. After: he had a real year, and the arithmetic said the business only became worth his time in year two, once furniture was paid off and he had review history. Why it matters: nothing went wrong in this example. This is what a normal, competently run first year looks like.

The four costs that decide profitability

Two of those four are inside your control and both are operational rather than financial. That is the actual finding: arbitrage profitability is decided by how well the calendar is run, not by how clever the lease was.

Is Airbnb arbitrage dead?

Short answer: no, but the easy version of it is. Rent has risen faster than nightly rates in many markets, supply has grown, and regulation has closed a number of cities to the model entirely. What is gone is the era of signing any lease and clearing a margin; what remains works only where the numbers are genuinely checked first.

The three things that changed are worth naming, because they tell you what to test.

• Rent moved: the input cost rose while nightly rates did not always follow
• Supply grew: more listings competing for the same nights compresses both occupancy and rate
• Regulation arrived: permit caps and primary-residence rules make the model unavailable in some markets regardless of the maths

How to test whether a specific property is profitable

Short answer: model the worst three months, not the average one. Take the rent, add the four deductions above, and work out how many nights at what rate you need just to break even. Then check by hand whether comparable listings nearby are actually achieving that in the quiet season. If they are not, the property does not clear, whatever the annual average suggests.

• Find your break-even nights: (rent + fixed costs) divided by (nightly rate minus per-stay cleaning share)
• Sample the real market: ten comparable listings, a window 45 to 75 days out, count unavailable nights — the method is in how to find Airbnb occupancy rates
• Check the shoulder season, not the peak: anyone fills in August
• Judge the market structurally: what separates a durable market from a busy one
• Add a reserve: three months of rent, because you will need it

How the platform fee changes the arbitrage sum

Short answer: it takes its share before you pay rent, and which share depends on a structure you do not choose. Airbnb runs a 3% split fee for most hosts and a 15.5% host-only fee for others, and connecting property-management software moves you onto the second automatically — which matters here, because operators running several units usually do exactly that.

• Split fee, 3%: most hosts; the guest pays a separate fee on top
• Host-only, 15.5%: hosts on management software and certain listing types, per the Airbnb Help Center
• On a $3,000 month: the difference is roughly $375, which on an arbitrage margin is often the whole profit
• What to do about it: nothing, except model the fee you will actually pay rather than the one you would prefer

Does a second unit make it worth doing?

Short answer: financially yes, operationally only if the first one is already clearing. A second lease doubles the fixed cost immediately and doubles the revenue only if it fills as well as the first, which a new listing with no reviews will not do for two to three months.

• Month one of unit two: full rent, partial occupancy, no review history
• What scales well: your systems, your cleaner, your message templates
• What scales badly: your attention, and any cost charged per listing
• The honest threshold: add a unit when the current one clears its rent in the quiet months, not when it clears in the peak

This is the point at which most arbitrage operations either become a business or stall. Two units run casually earn less than one run properly, because the empty nights multiply faster than the revenue does.

Myths about arbitrage profitability

Myth: the margin is the rent minus the nightly revenue.

Reality: the platform fee, the cleaning per turnover, utilities and empty nights sit between them. In the worked example above they turned a $1,145 gross gap into a $257 loss.

Myth: more units means proportionally more profit.

Reality: revenue scales, coordination scales faster, and a per-listing software bill scales with it. The third unit is usually easier than the second; the sixth is where operators either build systems or stall.

Myth: a good month proves the model.

Reality: a good month proves the market has a peak. The quiet months prove the model, and they are the ones nobody screenshots.

What a realistic three-year picture looks like

Short answer: year one pays for the furniture, year two is the first real profit, and year three is decided by the lease renewal rather than by anything you do. That last point is the one operators underestimate: your input cost resets when the landlord chooses, and a rent increase lands on a margin with nothing to absorb it.

• Year one: furniture repaid, review history built, occupancy below what the listing will eventually achieve
• Year two: the first clean year, with the same rent and a listing that now ranks
• Year three: renewal — a rent rise of a few percent can remove most of the margin, and you have no asset to offset it
• What protects you: a longer lease agreed at the start, or a rent-review clause you have read

An owner facing the same market gets a rising asset value alongside the squeeze. An arbitrage operator gets only the squeeze, which is the clearest single argument for treating this as an operating business rather than an investment.

Mistakes that destroy arbitrage margins

• Modelling on peak occupancy. Twelve months of rent against four months of demand is the single most common way this fails.
• Ignoring turnover count. Two properties with identical revenue can differ by thirty cleans a year, which on a thin margin is the whole profit.
• Treating furniture as spent rather than owed. It has to come back out of margin before anything is profit.
• Paying per listing for software. A per-property fee takes its share whether the month was good or not, which on this model is a fixed cost against a variable income.
• Scaling before the first unit clears. A second lease multiplies a loss just as efficiently as it multiplies a profit.

What actually moves the number

Short answer: nights, not rate. Because the rent is fixed, every extra night sold drops almost entirely into margin, while a higher nightly rate only helps on the nights that actually sell. That makes the levers that fill a calendar worth more here than on any owned property, and it is why operators who are good at occupancy make this model work when the spreadsheet says it should not.

• Gap nights — the one and two-night holes between bookings are the cheapest revenue you will ever add: how to fill them automatically
• Minimum stay — a three-night minimum in shoulder season rejects the demand that exists: the nine levers, ranked
• Response speed — feeds placement, which feeds bookings: how to hold 100 percent
• Review velocity — compounds over months: how review scores work
• Your benchmark — what counts as a good occupancy rate for your market shape

Software pricing matters more on this model than on any other, for a structural reason: a per-listing fee is a fixed cost charged against a variable income, on a margin that is already thin. BnBGenius is $10 per month flat for any number of units, with the first 500 messages free, and it automates the three things that fill calendars — replies around the clock, gap-night and extension offers through the Upsell Engine, and reviews posted inside the window. It does not set your rate and it does not make an unviable lease viable.

Before any of this, two things have to be true: the landlord has consented in writing and short-term letting is permitted where the property is. Both are covered in is rental arbitrage legal, and the model itself is explained in what Airbnb rental arbitrage is. If the arithmetic here has put you off leasing, best places to buy an Airbnb covers the ownership route instead.