Stop treating the month as one decision. A monthly occupancy figure hides the thing you need — which specific nights are empty, and why. Some are priced above the street, some are blocked by your own stay rules, and some were always going to sell late. A blanket discount fixes none of those and gives away the nights that were fine.
Each one is checkable against something real. None of them requires predicting demand, and the first four cost nothing but attention.
Mark every unbooked night in the next thirty and sort them into groups: midweek, weekend, orphan gaps between two bookings, near-term dates, and dates far enough out that nobody has booked them yet anyway.
The groups tell different stories. A single empty Saturday is usually a price. Repeated empty Sundays are usually a stay rule or a guest type. Nights wedged between two reservations are often not a demand problem at all — nobody could book them.
A rival's price next month explains nothing about why this Friday is empty. Read what comparable places ask for the same nights: similar area, similar size, similar guest count, similar cancellation terms, and check whether fees and taxes are inside the quoted number.
If you sit clearly above the alternatives on a weak night, you have found something you can act on. If you are already the cheapest and still empty, another discount is not the answer and will cost you the nights that were selling fine.
A two-night minimum leaves one-night gaps that no guest can reserve. Arrival-day restrictions do the same around weekends. Those nights read as no demand in your monthly average when they were never on sale.
Look at each gap from the guest's side: could someone actually search for and book it? If yes and the turnover is worth it, lower the minimum for those dates only. Leave peak dates alone — a rule that helps a Tuesday should not follow you into a sold-out Saturday.
An empty night tomorrow and an empty night in six weeks are different decisions. Before reacting to a far-out gap, check how far ahead your guests normally book; a calendar that always fills late is not a calendar in trouble.
Cancellation terms belong in the comparison too. A flexible rate competes against other flexible rates, and a cheaper non-refundable rate is a different product for a different guest. Compare like with like or the price gap you think you see is not real.
If your price is already competitive on the empty nights and guests still pick someone else, the problem is upstream of price. The first photo has to carry the single clearest reason to stay; the rest should explain the space, the beds, the bathroom and the amenities that decide a trip.
The title and opening lines should make location, capacity and what makes the place different obvious in seconds. Specifics beat superlatives, and a lower rate never fixes a guest's uncertainty about what they are booking.
Raising occupancy does not mean making everything cheaper. The nights your city fills on its own — an event, a holiday weekend, a school break — are the ones a flat rate sells at a Tuesday price.
On those dates the work is the opposite: check that no stay rule is blocking the arrival patterns guests actually want, confirm your best unit is open, and compare the exact dates before you move anything.
The mechanics are the same, but three things behave differently once you have more than one room type on sale.
Do not discount the whole month to fix four nights. Do not react to a far-out gap before your normal booking window has even opened. Do not compare your total against a rival's rate with the fees stripped out. And do not chase occupancy as a score — a calendar that fills instantly at an unchanged rate is telling you the rate is low, not that the strategy is working.
One more: we read asking prices from live listing pages on Booking.com and Agoda when you ask us to, and that is all we claim. Asking prices are what places are offering at the moment of the scan, not proof of what a guest eventually paid. We never forecast, and nothing we run changes your listing — you look at the comparison and decide.
Related: what counts as a good occupancy rate, minimum stay and orphan gaps, and why bookings slow down.