By the Unity Stays directors, Marcus Chong & Gadir Al-Khatib · Last updated 12 August 2026
In short: Assess a short-let deal by modelling its weakest month, not the average: if the low month covers mortgage, fixed costs and servicing, the deal survives a bad year. Subtract every cost between gross and net — commission, cleaning per changeover, linen, utilities, rates, software, management, maintenance and insurance — and treat inflated occupancy and undercounted changeovers as the two likeliest errors.
Start from the low month, not the average
Almost every short-let projection is built on a twelve-month average. Almost every short-let disappointment happens in a specific month.
Model the property on its weakest month first. If the weakest month covers the mortgage, the fixed costs and the servicing, the deal survives a bad year. If it only works on the average, you are relying on strong months to subsidise weak ones, which is fine until the strong months come in soft.
This single change in method removes most of the optimism from a model, which is why it is rarely done by anyone selling one.
The full cost list
Between gross booking revenue and money in your account, these are the lines. A projection missing more than one or two of them is not a projection.
- Platform commission on anything booked through Airbnb, Booking.com or Vrbo
- Payment processing, which is separate from platform commission and often forgotten
- Cleaning, costed per changeover — not per week, unless your average stay really is seven nights
- Laundry and linen, which on a busy property is a meaningful monthly figure in its own right
- Consumables — toiletries, coffee, cleaning materials, replacements for the things that walk
- Utilities, which you pay rather than the guest, and which short lets consume more of than tenancies
- Council tax or business rates, depending on how much the property is let and the nation it is in
- Broadband, which needs to be fast and unlimited rather than cheap
- Pricing software and any channel or PMS subscription
- Management, if you are not doing it yourself — and cost your own time honestly if you are
- Maintenance and replacement, at a higher rate than a tenancy because the property is used harder
- Insurance suitable for short letting, which costs more than a standard landlord policy
- Setup — furnishing, photography, compliance, listing build — amortised across the period you expect to hold it
The two assumptions that cause most of the error
Occupancy. It is the number most often inflated, usually by taking a strong operator's figure and applying it to a property that is not yet established. A new listing does not perform like an established one; it has no reviews, no ranking history and no repeat guests. Model the first six months well below where you expect to settle.
Changeover frequency. Cleaning is a per-stay cost, not a per-week cost. If your average stay is three nights, a fully booked month is around ten changeovers, not four. Get this wrong and the cleaning line in your model is out by a factor of two — which, on most properties, is larger than the entire management fee.
A worked structure you can copy
Work in this order and the model stays honest. Nightly rate, times nights available, times occupancy, gives gross. Take off platform commission as a percentage of gross. Divide let nights by average stay length to get changeovers, and multiply by the full cost of a changeover including linen. Subtract the fixed monthly costs — utilities, council tax or rates, broadband, software, insurance — as annual figures rather than monthly ones, so nothing gets rounded away. Subtract management. Subtract a maintenance and replacement provision. Only then subtract finance costs.
What remains is the figure worth comparing to what the property would earn on a standard tenancy. Compare like with like: a tenancy figure should have its own void allowance, agency fees and maintenance provision taken off before the comparison means anything.
Our income calculator does this arithmetic on the management side, using your numbers rather than guessing them from a postcode.
When the answer is no
A model that says no is doing its job. Short letting suits properties in places with year-round demand, run at a scale where the fixed costs are spread, by owners who can absorb a quiet quarter without distress.
It suits badly: properties whose numbers only work at peak, owners who need an exact figure every month, leasehold flats where the consents are doubtful, and anyone whose plan depends on occupancy assumptions they cannot evidence.
If your model lands in the second group, a standard tenancy is not a defeat. It is the right answer, and it is one we give to people who ask us for a projection more often than you might expect.
A general method for modelling costs, not investment or tax advice. Figures depend entirely on the specific property.