Analyzing Airbnb comps before you buy: how to read three listings honestly

Three listings, four data points each. The framework underwriters and seasoned hosts use to estimate revenue without buying an AirDNA subscription.

You don’t need an AirDNA subscription to underwrite your first short-term rental. You need three real listings in the immediate area, four pieces of data per listing, and the discipline to pick comps your property could realistically replace. Once you’re past a deal or two and underwriting at volume, paid data like AirDNA earns its keep — but for the property in front of you today, the free comp method below is enough. AirDNA

This post is the framework. The Comp Analyzer tool does the math; this is the part that tells you which three listings to paste in.

The 60-second version

Find three active listings of similar bed count, similar square footage, within roughly half a mile (urban) or 2 miles (rural/resort) of your subject property.
For each, capture: nightly ADR (last 30 days), occupancy (last 90 days), cleaning fee, total reviews.
Apply a discount to the median: take 80% of median ADR × 75% of median occupancy. That’s your underwriting revenue baseline.
Stress test: drop occupancy another 10%. If the deal still works, it’s a buy. If it only works at the optimistic numbers, walk.

The discount looks pessimistic. It isn’t — it accounts for the gap between operator skill at scale and a brand-new listing’s first 6 months. Most under-performance in year 1 isn’t market collapse; it’s slow ramp.

Picking the three comps (this is the whole game)

Bed count must match. A 3-bedroom and a 2-bedroom + den are not comparable. Guests filter by bedrooms.

Square footage within ±20%. A 1,200 sf 3-bed and a 2,400 sf 3-bed produce different ADRs even at the same bed count.

Distance:

Urban (Phoenix, Nashville, Charlotte): 0.5 mile radius
Suburban: 1 mile
Resort/rural (Sedona, Asheville, Smokies): 2–3 miles or “same submarket” (e.g., same valley, same lake)

Activity:

Must have at least 12 reviews to be a real comp. New listings have unstable numbers.
Must show bookings in the last 90 days (look at the calendar — unavailable nights = bookings).

Avoid as comps:

Listings owned by the subject’s seller (they may be inflated to support the asking price)
“Unique experience” listings (treehouses, A-frames, glamping) unless you’re buying a similar unique
Listings with paid promotion / featured-listing badges

What “nightly ADR” actually means

ADR = Average Daily Rate, net of platform fees, gross of cleaning fee. On Airbnb’s calendar view, you can read approximate ADR by:

Look at the last 30 days of unavailable (booked) nights
The nightly price for an upcoming booked-equivalent date is your proxy
If the listing varies weekday/weekend, average them weighted by weekend share (typically 28%)

The Comp Analyzer tool accepts the nightly rate you can see and the cleaning fee separately. Always input the rate net of guest service fees — that’s what the host actually receives.

What “occupancy” actually means

90-day occupancy = unavailable nights / 90.

Look at the listing’s calendar for the next 90 days. Count nights blocked. A “minimum-stay-3-nights” listing showing 60 of next 90 booked is at 67% occupancy for underwriting purposes — that’s good in most markets.

Two gotchas:

Host-blocked nights (owner personal use) show as unavailable but aren’t bookings. If you see large multi-week blocks in the calendar, mentally exclude them.
Future-loaded calendar: a brand-new listing might show high occupancy for the next 60 days that doesn’t reflect last-12-month performance. Cross-check with review velocity — a listing with 12 reviews and 60 future bookings is real; one with 3 reviews and 60 future bookings is suspicious.

The 80% × 75% discount, explained

Once you have median ADR and median occupancy across your three comps, discount both:

ADR × 0.80 — your property will not be the top-priced unit on day 1. Reviews are the moat; you don’t have them yet.
Occupancy × 0.75 — first 6 months of a new listing typically run 15–25% below stabilized. Algorithms favor proven units.

Together, that’s annual revenue × 0.60 of median comp expectation. If the deal works at 60% of comp median, you have a deal. If it requires 95% to clear, you don’t.

Worked example

Three Asheville 3-bed comps:

Comp A: $285 ADR, 76% occupancy
Comp B: $310 ADR, 72% occupancy
Comp C: $260 ADR, 80% occupancy

Medians: ADR $285, Occupancy 76%.

Discounted underwriting:

ADR: $285 × 0.80 = $228
Occupancy: 76% × 0.75 = 57%
Annual revenue: $228 × 365 × 0.57 = $47,433

If the deal needs $65,000 to cash-flow at 25% down, walk. If it cash-flows at $45,000, you have margin. The discount is the safety net.

What this connects to

Comp Analyzer — paste three listings, see the discount math run automatically.
Cash-on-Cash Return Calculator — what the discounted revenue translates to in actual return.
Market Score Tool — if your three comps disagree wildly, the market may be in flux — check the macro view.
DSCR Loan vs. Conventional for Airbnb — your discounted revenue is what the lender will underwrite to anyway.

The cheapest way to lose $40,000 on a short-term rental is to underwrite to the optimistic median. The second cheapest is to underwrite without comps at all. The discount method is the floor — pay attention to anyone selling you a deal that requires optimistic occupancy to work.

Before you pull comps on a property you’re serious about, grab the free 47-point pre-purchase checklist — comp discipline is one line on it, and the other 46 are the things buyers forget until after they’ve signed.