Cash-on-Cash vs Cap Rate vs ROI vs IRR for STR

Four numbers claim to tell you the same thing about a short-term rental deal, and they disagree on purpose. Cap rate, cash-on-cash, ROI, and IRR each answer a different question. Pick the wrong one for the decision in front of you and you either pass on a good acquisition or overpay for a weak one.

For operators running 5 to 50 doors, the mistake is rarely arithmetic. It is category confusion: comparing your financed cash-on-cash against a competitor’s all-cash cap rate and concluding you are winning when you are not.

TL;DR: Cap rate measures the property before financing (NOI / value). Cash-on-cash measures your financed first-year return (annual cashflow after debt service / cash invested). ROI is cumulative over the whole hold and folds in the sale. IRR is ROI with the clock running, the discount rate accounting for when each dollar arrives. Cap rate screens a property before financing, cash-on-cash sizes the financed first-year return, and IRR frames the exit. Never compare a levered number to an unlevered one. Informational, not financial advice. Benchmarks are commonly cited, not guaranteed, and observed returns vary widely and are often lower.

---

The four metrics, defined

Each metric answers one question. Keep the question attached to the number and the confusion goes away.

Cap rate. Net operating income divided by property value. NOI is rental revenue minus operating expenses (cleaning, management, utilities, insurance, taxes, supplies) but before any mortgage. Cap rate asks: what does this property yield on its own, ignoring how I pay for it? Per Corporate Finance Institute (reviewed July 2026), it is the unlevered yield, the standard way to compare properties independent of any one buyer’s loan.

Cash-on-cash return. Annual pre-tax cashflow divided by total cash invested. The numerator is net of debt service: revenue minus operating expenses minus mortgage payments. The denominator is cash out of your pocket, down payment plus closing costs plus upfront rehab and furnishing, and it is not the property value and not the borrowed money. Per Corporate Finance Institute (reviewed July 2026), it asks: what first-year return am I getting on the actual cash I committed, given my financing? J.P. Morgan and Wall Street Prep define it the same way.

ROI. Used loosely in practice, but strictly it is cumulative over the entire holding period and includes the eventual sale and appreciation. Total gain (all cashflows plus net sale proceeds minus cash invested) divided by cash invested. It asks: across the whole hold, start to exit, what did I make in total? It does not tell you when the money arrived.

IRR. The internal rate of return is the discount rate that makes the net present value of all cashflows, including the eventual sale, equal to zero. Per Investopedia (reviewed July 2026), it accounts for the time value of money and the holding period, which cash-on-cash (a single-year snapshot) does not. IRR asks: what annualized rate did this deal earn once you account for when every dollar arrived? It needs a full cashflow timeline plus an exit assumption, so it is more work to compute.

---

Cash-on-cash vs cap rate vs ROI vs IRR: what each includes

The disagreements between these metrics are not noise. They come from four design choices: does it count financing, does it span the holding period, does it count the sale, does it discount for time.

Metric	Counts financing (mortgage)?	Spans holding period?	Counts sale / appreciation?	Time value of money?
Cap rate	No (uses NOI, pre-financing)	No (single-year)	No	No
Cash-on-cash	Yes (net of debt service)	No (single-year snapshot)	No	No
ROI (strict)	Yes	Yes (cumulative)	Yes	No
IRR	Yes	Yes	Yes	Yes

Read the table top to bottom and you are adding one dimension at a time. Cap rate strips out everything personal to the buyer. Cash-on-cash adds your financing but stays a one-year photo. ROI stretches across the hold and adds the exit but treats a dollar in year one the same as a dollar in year five. IRR adds the missing dimension, time, and is the only one that penalizes a return for arriving late.

None is more “correct.” They answer different questions, and a good acquisition memo carries at least two of them.

---

When an STR operator should use each

Deal screening: cap rate. When you are filtering candidate properties, cap rate ranks them on the same footing regardless of how each would be financed. It answers “which yields best on its own” before your loan terms muddy the picture. Fastest triage number.

Financed first-year return: cash-on-cash. Once a property survives the screen and you know your down payment, loan rate, and upfront costs, cash-on-cash tells you the return on the cash you are about to wire, the number that helps you judge whether the financed first-year return is competitive with your alternatives. Run your own numbers with the Nightlyroi short-term rental calculator for the cash-on-cash and ROI figures rather than trusting a listing’s pro forma.

Exit planning: IRR (and ROI). When you have a hold period and a sale-price assumption, IRR compares a five-year STR hold against a three-year hold or a different asset entirely, respecting that money returned sooner is worth more. ROI is the simpler cousin: the total multiple on your cash across the hold without annualizing it.

The how-to guide on calculating cash-on-cash for an Airbnb walks the single-year math line by line, and the pillar on cash-on-cash return for short-term rentals covers where operators most often distort the inputs.

---

Same property, four different numbers

Take one property and watch the metric you pick change the story. Assume a purchase price of 400,000 dollars, and the property produces 32,000 dollars of net operating income in year one (revenue after operating expenses, before any mortgage). These figures are illustrative, not a projection for any real listing.

Cap rate. NOI of 32,000 divided by value of 400,000 is 32,000 / 400,000 = 8%. That is the unlevered yield. Any buyer, all-cash or financed, sees the same 8 percent because it ignores financing.

All-cash cash-on-cash. Pay the full 400,000 in cash plus, say, 15,000 in closing and furnishing, so 415,000 invested. With no mortgage, annual cashflow equals NOI, 32,000. Cash-on-cash is 32,000 / 415,000 = 7.7%. Notice it lands right at the cap rate (the small gap is only the extra closing and furnishing cash in the denominator). Cap rate equals cash-on-cash when you buy all-cash. That is the identity to remember.

Financed cash-on-cash. Now finance it. Put 25 percent down (100,000) plus 15,000 upfront, so 115,000 cash invested. The mortgage on the borrowed 300,000 costs, say, 21,000 per year in debt service. Annual cashflow is NOI minus debt service: 32,000 - 21,000 = 11,000. Cash-on-cash is 11,000 / 115,000 = 9.6%. The financed number (9.6 percent) sits above the unlevered cap rate (8 percent) because the loan rate here is below the property yield, so leverage amplifies the return on the smaller cash committed.

IRR (sketch). Cash-on-cash of 9.6 percent is only year one. Hold five years, let cashflow grow modestly, then sell: the sale returns your equity plus any appreciation minus selling costs and the loan payoff. Feed the full timeline (115,000 out at year zero, roughly 11,000-plus per year in, a lump sum at exit) into an IRR and you get one annualized rate blending the operating years with the sale. It differs from the 9.6 percent first-year cash-on-cash precisely because it counts the exit and discounts for time, which is why IRR is an exit-planning tool, not a screening one.

One property, and depending on the question you get 8 percent, 7.7 percent, 9.6 percent, or an IRR different again. The number is not the deal. The question is.

---

The trap: comparing a levered number to an unlevered one

This is the error that costs operators real money in negotiations and portfolio reviews.

In the example above, the financed cash-on-cash was 9.6 percent and the cap rate was 8 percent. An operator who sees “9.6 versus 8” and concludes this property beats an 8-cap alternative is comparing two different things. The 9.6 percent is levered (it exists only because of the specific 25-percent-down, sub-8-percent loan). The 8 percent cap rate is unlevered (the property with no loan at all). Not the same quantity, so the comparison is meaningless.

Leverage cuts both ways. Had the loan rate been above the property yield, the financed cash-on-cash would drop below the cap rate, because debt service eats more than the borrowed money earns. Same property, leverage now dragging the return down. That is the point: cash-on-cash bakes in your financing, cap rate deliberately excludes it.

Two rules keep you honest:

1. Compare like with like. Cap rate to cap rate across properties. Cash-on-cash to cash-on-cash across financing scenarios. Never a levered metric against an unlevered one.
2. State the financing when you quote cash-on-cash. A cash-on-cash number without the down payment percentage and loan rate attached is unfalsifiable. “9.6 percent” means nothing until you add “at 25 percent down, roughly 7 percent loan rate.”

When a broker or a seller’s pro forma waves a high cash-on-cash at you, the first question is always: on what leverage? Change the down payment and the number moves without the property changing at all.

---

What counts as a “good” return

Benchmarks circulate freely in STR communities, and most are opinions dressed as thresholds. Treat them as orientation, not targets, and note that observed returns vary widely and are often lower than the figures people quote.

For rentals broadly, a cash-on-cash of roughly 8 to 12 percent is a commonly cited range among investors (per BiggerPockets and Griffin Funding, reviewed July 2026). A benchmark and an opinion, not a guarantee. STR operators sometimes target higher, but measured STR returns tend to run lower: around 6 to 7.5 percent across ranked top markets (per Mashvisor, reviewed July 2026).

The gap between the 8-to-12 aspiration and the 6-to-7.5 measured median is the whole lesson. Underwrite to what markets actually deliver, not to the pitch.

Two operating realities quietly move every one of these metrics on the STR side, and both belong in your NOI and cashflow before you compute anything:

• Platform commission. Airbnb’s host-only fee is around 15.5 percent for most hosts (roughly 14 to 16 percent) and is mandatory for hosts connected to property-management or channel software, which is most professional operators (per Airbnb Help Center article 1857, reviewed July 2026). Booking.com has no single published rate; commissions commonly sit near 15 percent across a broad 10 to 25 percent range (per Booking.com partner documentation, reviewed July 2026). Compute on gross bookings instead of net-of-commission revenue and you overstate every return.
• Occupancy. US STR occupancy averaged around 55 percent nationally through mid-2025 (54.9 percent in the first half), highly seasonal and varying widely by market (per AirDNA, reviewed July 2026). Not a projection for any single property. Plug your own submarket’s occupancy into revenue, not the national average.

These inputs feed the NOI and cashflow all four metrics sit on, which is why sloppy revenue assumptions corrupt cap rate, cash-on-cash, ROI, and IRR at once. Tax treatment matters too: how you file changes after-tax cashflow, covered in the Schedule E vs C tax guide. And if you are running these numbers to decide whether to add doors, the scaling from 10 to 30 properties guide covers when portfolio returns diverge from single-deal math.

---

The one-line rule for each

• Cap rate: screen properties on an even footing, before financing.
• Cash-on-cash: the financed first-year return on the cash you actually wire.
• ROI: total return across the whole hold, sale included.
• IRR: the same, annualized and discounted for when each dollar arrives.

Carry at least two of them into any acquisition decision, always compare like with like, and always state the financing behind a cash-on-cash number. Run the cash-on-cash and ROI yourself in the Nightlyroi calculator before you accept anyone else’s pro forma.

---

Disclaimer: This article is informational only and is not financial, investment, or tax advice. Return benchmarks cited here are commonly quoted opinions, not guarantees; real returns vary widely and are frequently lower. Figures in the worked example are illustrative and not a projection for any specific property. Verify current platform fees and market data with primary sources before underwriting. Information current as of July 2026.