Airbnb Occupancy Rate Data (2026): How to Find It
Occupancy rate is the single most-asked-for number in short-term rental — it drives revenue estimates, purchase decisions, and pricing. Here's what it actually measures, why two sources rarely agree, and the three real ways to get it for any market.
What "occupancy rate" actually means
Occupancy rate is the share of a listing's available nights that were booked over a period — roughly booked nights ÷ available nights. The subtlety is in the denominator: some sources count only nights a host made available (available-based occupancy), others count all calendar nights including blocked ones (calendar-based). The two can diverge substantially for the same listing, which is why an "occupancy rate" number is meaningless unless you know how it was calculated.
For a market, the figure is an aggregate estimate across many listings — no one has Airbnb's private booking ledger, so every provider infers it from observable signals (calendar availability changing over time, minimum stays, pricing). Treat all of them as directional estimates, not ground truth.
Why two sources give you different occupancy numbers
- Definition: available-based vs calendar-based (biggest single cause of gaps).
- Sample: which listings are tracked, and whether inactive/blocked listings are filtered.
- Method: how "booked" is inferred from a calendar going unavailable (a block isn't always a booking).
- Recency: trailing-12-month vs last-30-day windows tell different stories.
None of this makes the data useless — it makes it something you validate against your own comps before betting money on it.
The three ways to get occupancy data
| Approach | What you get | Effort | Best for |
|---|---|---|---|
| 1. A market dashboard (e.g. AirDNA) | Occupancy, ADR, RevPAR with charts, comps, and market scores you read by eye | Low — sign in and browse | Investors & analysts researching by hand |
| 2. A data API (e.g. STRmetrics) | The same class of metrics as clean JSON your code can consume | Lowest for automation — one HTTP call | Developers & PMs wiring numbers into a model or app |
| 3. Scrape it yourself | Whatever you can parse from listing calendars over time | High — anti-bot walls, proxies, ongoing maintenance | Teams with scraping infra needing bespoke fields |
If you're researching a market by hand → a dashboard
For scouting where to buy, sizing a portfolio, or comparing neighborhoods, a dashboard is the right tool — you want to look at the data, not pipe it anywhere. AirDNA has one of the largest and most widely-cited STR datasets and covers both occupancy trends and buy/hold analysis. It's the default first stop for market research.
If your consumer is code → an API
If you need occupancy for 40 markets every morning inside a pricing model, an underwriting tool, or a feature in your own product, a dashboard is the wrong shape — you want an endpoint. A purpose-built data API returns the metric as JSON with one call.
Featured API — STRmetrics
curl -H "X-API-Key: YOUR_KEY" \
"https://strmetrics-api-production.up.railway.app/v1/market?city=austin-tx"
# example response shape (illustrative values):
# → { "occupancy": 0.63, "adr": 214.0, "revpar": 134.8, "revenue": 2984200 }How to actually use an occupancy number
- Check the definition before you compare two sources — available-based vs calendar-based.
- Validate against comps you can see: a handful of real listings in the exact submarket.
- Pair it with ADR — occupancy alone is half the revenue picture; RevPAR (ADR × occupancy) is the honest one.
- Watch the window — seasonality means a trailing-12-month figure hides the peak/trough you'll actually operate in.
How we pick
We weigh real-world value for STR operators and builders: data quality, how easily it fits a real workflow, transparent pricing, and time saved. AirDNA links are affiliate links (see disclosure) and we recommend it genuinely for dashboard research. STRmetrics is a sister project of ours (a data API, not one of the affiliate tools), so that link is a plain editorial one.