The night with the prices on it was ten days out, not three

Four nights read, and one of them had the prices

Our rate reader opens a city's public Hostelworld page every morning, takes the listings on it, and asks what a dorm bed costs on four fixed nights: three days out, ten days out, seventeen days out, and twenty-eight days out. The same four every day, spread across the next month. Then it writes down the one of those four nights that the most listings in that city carried a price for.

On 8 October it read Panama City. The night that came back was 18 October. Ten days out. Twenty-five listings on the city page, seven of them carrying a price for that night. Median USD 15.75 a bed. Cheapest 9.00. Dearest 36.18. Public Hostelworld prices, one guest, one night, read that morning. Nothing modelled, nothing estimated.

The number worth an operator's attention there is not the 15.75. It is the 18 October.

Where the prices in a city actually sit

Because the reader always asks the same four lead times, the night it reports back is a measurement in its own right. In Panama City on 8 October, more of the city's listings showed a price for ten days out than for three days out. That is the whole finding, and it is worth being precise about what it does and does not say.

It does not say the closer night was sold out. A listing can carry no price for a night because the beds are gone, because the property closed that date, because the dorm is held for a group, or because the page simply does not surface it that day. We read prices, not reasons. What we can say is where the visible prices in that city were on that morning: further out than most owners assume.

And it varies by city, which is the part that matters. On 6 October the same reader landed on three days out in Lisbon, Porto and Oaxaca, and on twenty-eight days out in Barcelona, where 33 listings were on the page and 8 carried a price for the night of 3 November at a median of USD 28.14. On 7 October it landed on ten days out in Tulum, where 4 of 31 listings were priced for 17 October. Same engine, same four lead times, four different answers about where a city's prices live.

What the spread costs, in arithmetic

Take the Panama City row apart. Cheapest 9.00, median 15.75. The gap is USD 6.75 a bed a night. On 25 beds that is USD 168.75 a night, and USD 5,062.50 across a 30-night month, between sitting at the city floor and sitting at the city median. That is arithmetic on a price gap, not a forecast of your revenue, and I am not claiming anybody will move from one to the other.

Now the part that catches people. Say you sell 15 of 25 beds at the median: 15 times 15.75 is USD 236.25 for the night. Now fill the house at the floor: 25 times 9.00 is USD 225.00. The full house takes less money than the sixty-percent house. To match 236.25 at a 9.00 bed rate you would need to sell 26.25 beds, which is more beds than the property has. There is no occupancy number that rescues that rate.

The dearest bed in the same city on the same night was 36.18: 4.02 times the cheapest, 2.30 times the median. One city, one night, seven prices, a four-fold range. None of that makes any price wrong. The dearest listing may have a better location, or may know something about 18 October that the cheap one does not. The spread is the fact; what you do about it is the decision.

Ten days out is a standing price, not a last-minute cut

This is why the lead time matters more than the median. A bed at 9.00 for tonight is a decision about tonight: beds are empty, something beats nothing, and you can reverse it tomorrow. A bed at 9.00 for a night ten days away is not that. Ten days out, nothing has gone wrong yet. The night is still open, the demand has not arrived, and nobody has to discount anything. A price that low that far out is usually not a decision at all. It is a standing rate that nobody has looked at.

Those two get confused constantly, and they are opposite problems. Discounting late is a tactic you can measure. Being cheap early is a leak that runs every night of the month unnoticed, because there was no moment where you chose it.

What we do with it at our own hostel

I run a 46-bed hostel in Oaxaca, and it is the property our engine has been pricing since August. It does not hold one rate. It rewrites five lead-time bands every morning: same day, one to three days out, four to seven, eight to twenty-one, and twenty-two and beyond. Ten days out sits in the eight-to-twenty-one band, and that band gets its own price, read against what the other hostels in the city are asking for that same night.

Across the 32 unattended runs we logged between 14 August and 5 September, the moves were 19 deepen, 13 hold, and 0 lift. More interesting is what it did when it checked its own discounting. Its read of our data: an 8 percent cut ran at 33.7 percent average occupancy over 23 days, while holding at 0 percent ran at 42.0 percent over 5 days. Deeper discounting correlated with fewer beds sold, not more, so it held instead of cutting further. That is our property, our season, one read of a small sample. Yours may say the opposite. The point is that it was read rather than assumed.

A note on how we publish this: aggregates only, never a hostel's name beside its price. Seven prices is a small sample and we say so, so every figure above sits inside one city, on one night, read in the same hour.

Nightfill does this read every morning and rewrites rates off the back of it, five lead-time bands a day, inside a floor and a ceiling you set. If that is the job you want off your desk, the three tiers are here: view pricing. Nightfill is independently operated and results vary by property, market and season.

The same read, for your property

You do not need to buy anything to check this on your own listing. Paste your Hostelworld listing at nightfill.co/#read and the same reading for your own hostel arrives by email in about ten minutes: your cheapest dorm bed against the other hostels in your city, night by night across the next 30 days, your rank from cheapest, and the nights you sit above your city's median.

When it lands, do not read it top to bottom. Read down the lead-time column instead, and find the night furthest out where you are the cheapest hostel in your city. That is the night nobody made a decision about.