The host problem
You want to calculate your occupancy rate. The month had 30 days. You
booked 18 nights. So occupancy is 60%, right?
Maybe. If your listing was available for all 30 nights, yes. But if
you blocked 5 nights for personal use or maintenance, your available
inventory was only 25 nights — and occupancy was 72%.
That 12-point difference is not a rounding error. It changes how you
read your performance, and it changes how you compare periods.
The definition: available
nights
Available nights means the calendar nights your listing was active
and bookable by guests.
Available nights excludes:
- Owner use blocks
- Maintenance closures
- Listing-snooze periods
- Any night you intentionally removed from the bookable calendar
Available nights does not equal calendar days in a month. It equals
only the nights a guest could have booked.
Why this matters for
every core metric
Occupancy rate, RevPAR, RCI, and Net RevPAR all use available nights
as the denominator. If you use the wrong denominator, every metric built
on top of it is wrong too.
Occupancy Rate = Live Booked Nights ÷ Available
Nights (not calendar nights)
RevPAR = Live Accommodation Revenue ÷ Available
Nights (not calendar nights)
If you block 8 nights in a 30-night month and use 30 as your
denominator, you’re dividing by a number that includes nights you never
offered for sale. Your occupancy looks artificially low. Your RevPAR
looks artificially weak. You might cut your price when the data actually
says you performed fine on the nights you sold.
Small example
April has 30 days. You blocked 6 nights — 3 for a personal trip and 3
for a cleaning and maintenance day. Your available nights: 24.
You booked 18 of those 24 nights. Your total accommodation revenue
was $2,700.
Using the correct denominator:
- Occupancy Rate = 18 ÷ 24 = 75%
- RevPAR = $2,700 ÷ 24 = $112.50
Using the wrong denominator (30 days):
- Occupancy Rate = 18 ÷ 30 = 60%
- RevPAR = $2,700 ÷ 30 = $90.00
The incorrect calculation makes your listing look like it
underperformed by 15 percentage points and produced $22.50 less per
night. Neither is true. You just used the wrong base.
Why this matters for
period comparisons
Comparing months with different block patterns is meaningless unless
you normalize for available nights. A March with 28 available nights and
a March with 20 available nights are not comparable if you use calendar
days as the denominator.
This is especially important when you change your blocking habits —
adding owner-use stays, reducing maintenance windows, or opening more
dates for a high-demand period. Your performance numbers will shift even
if nothing about your pricing or demand changed.
Always adjust for available nights before drawing conclusions from a
comparison.
What counts as an available
night
One quick test: could a guest have booked this night without any
manual override from you? If yes, it’s an available night. If no —
because you blocked it, snoozed the listing, or took the listing offline
— it’s not.
Airbnb’s calendar “blocked” dates typically show the nights you
removed from inventory. Use those dates to calculate your true available
nights, not the raw calendar count.
What to do this week
- Pull your calendar for last month.
- Count the nights you blocked for personal use or maintenance.
- Subtract those from the calendar total. That’s your available
nights. - Recalculate your occupancy rate and RevPAR using that number.