Length-of-stay controls
A Socratic walk-through of length-of-stay controls — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why would a hotel with rooms standing empty refuse a two-night booking it could easily accommodate?
The booking page says thirty rooms are free on Friday. You ask for Friday and Saturday — two nights, money in hand — and the engine refuses: three-night minimum on arrivals this weekend. The hotel has chosen to earn nothing from a room it has just admitted is empty.
The reflex explanation is greed, but greed explains a high price, not a refusal. A refusal earns zero. So what would have to be true for turning your money away to be the better move?
Reasoning it through
REASONING #Start with what the hotel is selling. Its capacity is a night: a hundred rooms on Friday, a hundred again on Saturday, and Friday's unsold rooms cannot be carried forward. What you are buying is a stay — a run of consecutive nights that must all be free together. Supply is counted one way, demand arrives in the other, and almost everything strange about hotel booking rules falls out of that mismatch.
Given that, what does accepting your booking cost? Not the laundry; that is small, and I will come back to it. The real cost is what each room would otherwise have earned. Revenue managers call it displacement cost, or bid price, and the important feature is that it is worked out per night, because that is where scarcity lives. A night forecast to sell out carries a high bid price: releasing a room now forfeits a sale that was going to happen anyway at the full rate. A night forecast to run half empty carries a bid price near nothing, because the alternative to selling it really is nothing.
The decision rule then writes itself. Accept a stay when what the guest pays clears the sum of the bid prices of the nights it occupies. Watch what that sum does to a mixed stay. Take an illustrative weekend — this arithmetic is constructed to show the shape, not reported from any hotel. Friday's bid price is 40, because Friday will end near but not at capacity. Saturday's is 185, because Saturday is already oversubscribed and every remaining room will sell at the rack rate. Your two-night booking at 120 a night offers 240 against a threshold of 225, and scrapes through. Offer 100 a night and it is 200 against 225, so the hotel should refuse — even while showing you an empty room on Friday, because the Saturday room is the one you are consuming, and Saturday's room is worth more than you are paying for the pair.
So the empty rooms you can see and the room the hotel is protecting are not the same room. That is the whole trick. But if the test is a sum of bid prices, why express it as a blunt three-night minimum rather than simply quoting a higher price?
Because of who the peak-night demand is. The guests who must be there on Saturday — the wedding party, the festival crowd — have least choice about the date, and there are more of them than there are rooms. The hotel cannot sell them more Saturdays; it only has a hundred. What it can do is attach a condition to the scarce night: to have Saturday, take Friday and Sunday too. That converts an excess of demand for the one night nobody needs help selling into occupancy on the two nights that were going to sit empty. A minimum stay is not a price rise. It is a change in the quantity you are obliged to buy, aimed at the buyers who can least walk away.
The analogy
THE ANALOGY #Think of a theatre selling a three-night festival pass. Saturday's performance is the one everyone wants and the house is small, so instead of pushing Saturday's ticket price into the ceiling, the theatre stops selling Saturday singly. You may have it as part of the pass or not at all — and the Thursday and Friday houses, which would have played half empty, fill with people who came for Saturday.
The theatre knows exactly how many want Saturday, whereas the hotel is acting on a forecast that may be wrong; and a festival-goer holding a pass turns up on Thursday, while a guest forced into a third night may pay for it and simply arrive late.
Clarifying the model
THE MODEL #Three qualifications, because the tidy story is not the whole of it.
Much of what looks like revenue management is really labour arithmetic. Every stay costs one check-in, one check-out and one full turnover clean; spread over one night that is expensive, over five it is cheap. Small properties and holiday lets impose minimum stays for this reason and no other, which is why you meet them in a quiet cottage in February where no peak night exists to protect. The two stories produce the same rule and are constantly confused — but only one predicts that the rule will lift.
Second, the control is a bet on a forecast. If Saturday does not sell out after all, the hotel has turned away real money to protect a room nobody wanted. Revenue systems therefore reopen restrictions continually as the date approaches.
Third, this is the close relative of the shoulder-season price cliff, and the difference is worth naming. There the question was what the price of a room does when demand crosses capacity. Here the price may not move at all; what changes is what you are permitted to buy. Same underlying quantity — the bid price of the marginal room — but one lever acts on rate and the other on inventory, and a hotel reaches for the second when the first is at its ceiling.
How would you tell which story a given restriction is? The displacement account predicts that a restriction is keyed to a specific forecast-constrained date, is tighter for arrivals than for stays passing through, and is removed — often hours out — once the forecast for that night falls below capacity. A property whose minimum stay sits unchanged across a sold-out weekend and a half-empty one refutes it: that is a cleaning roster, or a channel contract, wearing a revenue manager's coat.
A picture of it
THE PICTURE #How to readEach bar is one night's bid price — what the hotel expects to lose later by giving a room away now — and the figures are illustrative, chosen to show the shape. Read a booking as a span rather than a point: a stay is acceptable when its total payment clears the sum of the bars it covers, so Thu-Fri sums to 55 and almost anything clears it, while any stay touching Saturday must carry that tall bar on its back. The tall bar is the room the hotel is refusing you; the short ones are the rooms you were shown standing empty.
What became clearer
WHAT CLEARED #Capacity is per night, purchases are per stay, and the gap between those units is where length-of-stay controls live. The room the hotel is protecting is never the empty one you were shown — it is the scarce night buried inside your request, whose bid price your rate failed to cover. And because peak-night demand is least flexible about dates, attaching a condition to it is the one way a hotel sells the nights either side. The refusal is not a rejection of your money; it is a claim that the same room, sold differently, is worth more.
Where to go next
ONWARD #- How airlines nest fare buckets across connecting itineraries — the same displacement arithmetic over a network rather than a calendar.
Key terms
TERMS #| Term | What it means |
|---|---|
| Displacement cost (bid price) | the revenue a hotel expects to forgo later by selling a room tonight; computed per night, and the number every acceptance decision is measured against. |
| Minimum length of stay (MinLOS) | a rule requiring a stay arriving on a given date to run at least so many nights. |
Every term the collection defines is gathered in the glossary.