In this article · 14 sections
Airbnb Cap Rate: How to Calculate It Properly (And Why Most People Get It Wrong)
The Airbnb cap rate error is gross booking revenue in the numerator. On a $485,000 three-bedroom that reads 13.4%. Calculated properly on NOI it is 4.4%. Same house, same year, triple the number.
Geo Pedro
STR Operator & Co-founder, Daystays Hospitality
Geo Pedro is a short-term rental operator and co-founder of Daystays Hospitality. He manages a multi-property STR portfolio and writes about the real numbers behind profitable hosting — deal analysis, occupancy strategy, and what the data actually shows.

An Airbnb cap rate is net operating income divided by purchase price, and almost every version circulating online puts gross booking revenue on top instead. On the $485,000 three-bedroom worked through below, the wrong method returns 13.4% and the correct one returns 4.4%.
Key takeaways
- Cap rate is net operating income divided by acquisition price, and CBRE's H1 2026 survey defines that NOI as gross income less operating expenses, which puts debt service, depreciation, and income taxes outside the calculation.
- Using gross booking revenue as the numerator on a $485,000 three-bedroom returns 13.4%, against a true cap rate of 4.4% once the 15.5% platform fee and $33,935 of operating expenses come out.
- Airbnb's single 15.5% host service fee costs 2.08 points of cap rate on this property before one operating expense is counted.
- Subtracting only the four obvious costs (cleaning, property tax, insurance, utilities) still overstates the cap rate by 55%, landing at 6.8% against the correct 4.4%.
- Cap rate is undefined on a leased arbitrage unit because nothing was acquired, so the denominator does not exist and cash-on-cash return on startup capital replaces it.
The formula, and the one substitution that breaks it
Cap rate equals net operating income divided by acquisition price, and the entire argument is about which number goes on top. The denominator is uncontroversial. The numerator is where a good deal and a bad deal swap places.
The institutional definition is not ambiguous. CBRE's U.S. Cap Rate Survey H1 2026, built from 3,600 cap rate estimates across more than 50 U.S. markets, states it in two lines: stabilized cap rates are the ratio of stabilized NOI to acquisition price, and the NOI calculation is gross income less operating expenses. Every appraiser and lender you will sit across from uses that. Short-term rental content mostly does not.
The substitution happens because gross booking revenue is the easiest number to find. It sits at the top of your earnings dashboard and it is what market estimate tools publish. Operating expenses live in eleven places, and platform fees are netted out before the money reaches you.
Getting the numerator right is a data problem before it is a math problem. MagicBNB's Property Detail carries a financial mini-card with net payout, NOI, expenses, and margin on one screen, plus expense breakdown by category, so the number you put on top of the cap rate is the one your bank statements already agree with.
The same property, three different cap rates
One house produces three defensible-sounding cap rates depending on how far down the stack you stop, and only the last one is real. The property: a three-bedroom asking $485,000, 205 booked nights at a $268 average nightly rate, 62 stays, a $165 cleaning fee charged to the guest.
- Gross booking revenue is $65,170: $54,940 of nightly revenue plus $10,230 of cleaning fees collected. Divided by $485,000 that reads 13.44%.
- Net payout is $55,069 after Airbnb's 15.5% host service fee takes $10,101. That fee alone is 2.08 points of cap rate.
- Operating expenses total $33,935: $8,990 cleaning paid out, $6,305 property tax, $3,720 utilities, $3,180 insurance, $2,750 repairs, $2,400 replacement reserve, $2,160 yard and pest, $1,860 supplies, $1,140 internet, $1,020 software, $410 permits.
- Net operating income is $21,134, and the correct cap rate is 4.36%.
- Stopping after only cleaning, tax, insurance, and utilities gives $32,874 and a 6.78% cap rate, which overstates the property by 55%.
A cap rate built on gross booking revenue is not an optimistic cap rate. It is a different number wearing the same name.
The platform fee line is not an estimate. Airbnb's Q2 2026 shareholder letter confirms the migration to a single 15.5% host service fee, replacing the split where hosts paid 3% and guests paid the rest separately. A model still carrying the old 3% host fee produces a 6.04% cap rate on this property instead of 4.36%, so it is running 1.68 points hot on every deal it has screened since.
What belongs in NOI, and what never does
Four things stay out of NOI permanently: debt service, depreciation, income taxes, and capital improvements. They are financing, accounting, and ownership decisions, and a cap rate that includes any of them stops describing the building and starts describing you.
Depreciation is the one operators argue about, usually because it is the largest line on their Schedule E. IRS Publication 527 settles it: depreciation is a capital expense, the mechanism for recovering your cost in an income-producing property, taken over the expected life of the property. A cost recovery mechanism is not an operating expense, and putting it in NOI double-counts the purchase price you already divided by.
The replacement reserve goes the other way. Mattresses, sofas, the grill, the hot tub cover: they wear out far faster on a short-term rental, and appraisers include a reserve in NOI for that reason. The $2,400 line above is 4.4% of net payout. Leave it out and the cap rate reads 4.85%, half a point of pure fiction.
For STR Operators
Occupancy Tells You One Thing. Margin Tells You Everything Else.
Definitional drift is what makes two operators compare cap rates and end up arguing about nothing. MagicBNB's 60+ metrics glossary holds one canonical definition each for cap rate, NOI, cash-on-cash return, DSCR, and RevPAN, injected automatically whenever a question touches them, so a deal underwritten in March and revisited in October is computed the same way both times. For the line-by-line build, see how to calculate net operating income on a short-term rental.
What counts as a good Airbnb cap rate
There is no published short-term rental cap rate survey, which is the honest answer nobody gives. CBRE covers office, industrial, retail, multifamily, and hotel. Short-term rental is not a category, so operators borrow a hotel range and import assumptions about scale and management that do not hold on a five-door portfolio.
Price it off the risk-free rate instead. The 10-year Treasury peaked at 4.67% in mid-May 2026 and sat near 4.6% in mid-July, per the same CBRE survey. A defensible screen is that yield plus 250 to 400 basis points, putting the entry range at 7.1% to 8.6%, wider where regulatory risk is live and narrower on a drive-to leisure property with three years of booking history.
The worked example fails that screen badly. At $21,134 of NOI it clears 7.1% only at $297,657, and 8.6% at $245,740. It is asking $485,000. That is not a negotiation gap, and the 13.4% number is the only reason anyone writes the offer.
Why cap rate is the wrong tool on a leased unit
Cap rate has no denominator on a leased arbitrage unit because you did not acquire an asset. Dividing NOI by furnishing cost produces a large percentage that measures nothing, since the number doubles if you buy cheaper sofas.
Use cash-on-cash return on total startup capital: security deposit, first and last month, furnishing, and the operating reserve that covers the ramp months. The screen most operators run is 20% or better on an arbitrage unit, because the risk profile is worse than ownership and no equity accrues underneath it. The full method is in our guide to cash-on-cash return for STR investors.
Running both structures through one tool keeps the comparison honest. MagicBNB's Property Analyzer splits into a purchase mode carrying loan terms, tax, insurance, and depreciation, and a lease mode carrying monthly rent, variable expenses, and platform fee percentage, each returning the metric that applies to that structure instead of forcing a cap rate onto a deal with no acquisition price.
Frequently asked questions
What is a good cap rate for an Airbnb?
7.1% to 8.6% is a defensible entry screen in 2026, built as the 10-year Treasury near 4.6% plus 250 to 400 basis points. No survey publishes short-term rental cap rates specifically, so anyone quoting you a precise national figure is quoting hotel or multifamily data with the label changed. Widen the spread in markets where the permit regime is under active review.
Do you use gross revenue or net income for cap rate?
Neither: you use net operating income, which sits between them. Gross booking revenue ignores the 15.5% platform fee and every operating cost, and net income after debt service and depreciation is too far down the stack. On the $485,000 example, gross reads 13.44%, NOI reads 4.36%, and the second one is the cap rate.
Does cap rate include the mortgage?
No, and this is the most common single error after the numerator problem. Cap rate deliberately excludes debt service so that two buyers with different loan terms value the same building identically. Once you want to know what your specific financing does to your specific return, the metric is cash-on-cash, not cap rate.
Should cleaning fees count as revenue in a cap rate?
Yes on the revenue side and yes on the expense side, netted honestly. The $10,230 of cleaning fees collected on our example is real income, and the $8,990 paid to cleaners is a real operating expense, leaving $1,240 of margin. Counting the collection and skipping the payment inflates NOI by $8,990 and the cap rate by 1.85 points.
Is a 4% cap rate bad for a short-term rental?
At 4.4% you earn less than the 10-year Treasury for taking property risk, illiquidity, and an operating business on top. It stays defensible when the thesis is appreciation or a repositioning that lifts NOI within eighteen months. It is not defensible as a stabilized yield play, and it should never be the number that surprises you after closing.
Recalculate one property this week
Take your best-performing door, pull twelve months of net payout rather than gross bookings, subtract every operating line including a replacement reserve, and divide by what you paid. Then compare it to the number you told yourself at purchase. Run the numbers in MagicBNB →
The gap usually sits between one and three points. Finding it on a door you already own costs an evening. Finding it on the next one costs $187,343.
About MagicBNB
MagicBNB is a portfolio intelligence platform for short-term rental operators running multiple doors. Its Net Payout source of truth drives one canonical calculation across profitability, listings table, property detail, trends, and reports, so an owner who challenges a number can be shown the path to it. The Smart transaction ledger categorizes every bank transaction with confidence bands and matches deposits against PMS payout records, and the Monthly Portfolio Report Builder turns 40+ column definitions into a PDF for owners and an Excel file for accountants in the same flow. See it at magicbnb.io.
Was this article helpful?
Free download
The STR Cash Leak Audit
A worksheet that walks every line where money quietly disappears from a portfolio: fee reconciliation, double-counted payouts, cleaning cost drift, and the expenses that never make it into your P&L. Download it on the next screen.
Instant download. Plus one operator-math email a week. Unsubscribe anytime.


