A thirty-bedroom hotel on the north Norfolk coast closes on the first Monday of November and reopens the week before Easter. The owner has done it for nine years because his father did it, and because November to March lost money the one year he tried staying open.
He has never worked out what the closure costs him. When he finally does, the figure that surprises him is not the rates or the insurance. It is that the head chef he rehires every March is the fourth head chef in four years, and each one takes until June to be any good.
Closing stops the revenue, not the costs
The first thing to do is list what continues, because operators consistently underestimate it.
Business rates keep running. A hotel that closes for the season is not an empty property in the rating sense, because it remains in your occupation with your furniture and fittings in it. Empty property relief, where it applies at all, is three months for most property and six for industrial, and it applies to genuinely unoccupied premises. Do not build a winter plan around it without talking to the Valuation Office Agency first.
Insurance is where closure actively costs you more. Most commercial policies have an unoccupancy condition that bites after a set period, commonly thirty consecutive days. Once it bites you may be required to notify the insurer, drain down the water system, maintain heating to a minimum temperature, inspect the building weekly and record it. Cover for escape of water, malicious damage and theft is frequently restricted or excluded on an unoccupied building. A burst pipe in a hotel nobody has been in for a fortnight is the classic winter claim, and it is the classic winter claim that gets declined.
Energy does not go to zero. Standing charges on gas and electricity run daily whether or not you use a unit, and you will be frost-protecting the building anyway. Your energy contract may also carry a take or pay clause against an estimated annual consumption, which is worth reading before you halve your usage.
Then the small recurring items that individually look trivial: the premises licence annual fee under the Licensing Act 2003, your music licence, the hotel television licence, the property management system subscription, the booking engine, the alarm monitoring contract, the lift maintenance contract, the accountant. Interest on the mortgage continues at exactly the same rate in January as in August.
If you run self-catering units alongside the hotel, there is a separate trap. In England, since 1 April 2023, a self-catering property is only assessed for business rates rather than council tax if it was available to let for at least 140 days and actually let for at least 70 days in the previous twelve months. In Wales the thresholds are considerably higher, at 252 days available and 182 days let. A long closure can push a unit back onto council tax, with a premium on top in some areas. These rules are set by the relevant government and have changed more than once, so check the current position each year.
The people are the expensive part and they do not appear on the closure budget
Fixed costs are visible. The recruitment cost of a seasonal model is not, because it arrives disguised as a normal March wage bill.
You have three broad choices, and all of them cost something. You can keep core staff on through the winter and pay them to maintain the building, which is the most expensive option in cash and the cheapest in capability. You can lay them off, which needs contractual authority. Without an express contractual right to lay off or put staff on short time, sending people home without pay is a breach of contract and potentially an unlawful deduction from wages. Where a lay-off is contractual, statutory guarantee pay is capped at a low daily amount, £38 a day for up to five days in any three-month period as at 2024/25, reviewed each April. After four consecutive weeks, or six in thirteen, an employee with two years' service can claim a statutory redundancy payment.
Or you can employ people on fixed-term contracts each season and accept that you are rebuilding the team every spring. That is the honest seasonal model, and it works in places where there is a pool of returning staff. It stops working when the local labour market tightens, because the good ones find something that pays in January.
The hidden cost of the third option is a quality curve. A team that reassembles in March is at its worst in the weeks it is being watched by the guests who will write the reviews that sell your summer. If your April scores are consistently below your July scores, that is what you are looking at.
Closure is also the only window you have for real work
The argument for closing is not purely financial. Some maintenance cannot be done around guests. Replacing a boiler, rewiring a floor, relaying a roof, refitting bathrooms, sanding and sealing the function room floor: each of these either takes rooms out of sale or fills the building with noise and dust.
If you trade all year, that work has to happen in the quietest weeks anyway, and you are now paying twice, once for the work and once in nights you cannot sell. A genuine closed period converts the disruption cost to zero. The discipline is to actually use it. A hotel that closes for five months and does £6,000 of painting has bought an expensive rest.
Finding the week that is worth opening for
The arithmetic is contribution, not profit. For any given week, take the revenue you believe you would take, and subtract only the costs that exist because you are open. Housekeeping hours and laundry. Breakfast and food cost. The utilities of an occupied building above frost protection. Commission on the bookings. The extra reception and kitchen cover.
Then compare that with the incremental cost of being open at all that week: heating the whole building rather than protecting it, the duty cover you need even at six rooms occupied, the fixed shift in the kitchen.
If contribution beats that incremental cost, the week is worth opening, even if the week does not cover its share of the mortgage. It never will. The mortgage is being paid by July.
Two things distort this in practice. Half term, Christmas and New Year, and any local event are not average winter weeks and should be modelled separately, because a single strong week can pay for two poor ones either side. And opening at all has an option value: a hotel that is known to be shut all winter stops receiving enquiries entirely, and the Google listing that says permanently closed for four months does not help you in April.
Where this shows up in the game
Hotels and holiday parks arrive in Business Baron with the Hotels & Leisure Track, an add-on to the full game. The seasons are real, and rates are charged on every property you occupy whether or not it is busy. The staff you let go in October are not necessarily there to hire back in March.