A pro forma is the spreadsheet an owner builds to decide whether to buy. Most of the ones we get sent are not wrong in a subtle way. They are wrong in the same handful of places every time, and the errors all run in the same direction: the model says the property makes more than it will.
That direction is not an accident. The model usually gets built after the owner has already decided they want the house. The inputs that survive are the ones that make the answer come out yes. Then the deal closes, the first slow season arrives, and the spreadsheet turns out to have been a persuasion document rather than an operating plan.
Here is how to build one that still holds up a year in, and where to check somebody else's before you trust it.
Almost every pro forma we review starts the same way: a nightly rate, an occupancy percentage, multiplied by 365. It is a reasonable place to start and a terrible place to stop, because that single line is doing four jobs at once and hiding all of them.
The arithmetic itself is fine. A rate of $200 at 70% occupancy is $51,100 across a year, and you should do that multiplication. Use your own figures, not those — they are placeholders chosen to keep the math readable. The problem is what the two inputs are standing in for. One number is carrying every weekend, weekday, holiday and dead week in the calendar. The other is carrying your entire assumption about demand, your pricing discipline, and how good the listing is at converting the people who see it.
Where the rate came from is the first question to ask of any model. If it came from looking at what comparable listings advertise, it is not a rate, it is an ask. Advertised rates are what hosts hope to get. Booked rates are what guests paid. The gap between them is the entire difference between a model that works and one that does not, and it is widest in exactly the markets that look most attractive on a listing-price scan.
The second failure is treating the top line as income. The number a booking platform shows against a reservation is not the number that lands in your bank account, and the difference is not small.
Money that arrives and then leaves again for somebody else was never yours. It still shows up in the gross, which is why owners who model from the gross consistently overstate what the property returns. Before anything is spent on operating the house, the top line has already been reduced by the channel's cut, by what you pay the cleaner regardless of what the guest was charged for cleaning, by payment processing, and by every amount collected from the guest that is passed through to a third party.
Pull your actual rates for each of those from the source rather than estimating them. Your channel publishes its fee schedule. Your processor states its rate. Your cleaner quotes a price per turn. None of these require guessing, and a model built on guesses at this layer will be wrong by a margin large enough to change the decision.
The third mistake is the most common one in optimistic models, and it is usually invisible because it looks like two separate reasonable assumptions sitting in two separate cells.
You cannot hold a high nightly rate and a high occupancy rate at the same time. Raising the rate costs you nights. Filling the calendar costs you rate. Any model that takes the strongest rate in the comp set and pairs it with the strongest occupancy in the comp set has combined two outcomes that no single property achieved, and then treated the combination as a forecast.
The honest unit is revenue per available night: what the property earns across every night it could have been booked, filled or empty. It collapses the two dials into the one number you actually get paid on, and it makes the trade visible. When you test a higher rate, revenue per available night tells you whether the extra dollars covered the lost nights. When you test a discount, it tells you whether the volume was worth it. Model that, and the rate-versus-occupancy argument settles itself.
A pro forma that produces one annual number can be right in total and still fail to describe the year you are about to have. Short-term rental demand is seasonal in every market we operate in, and the shape of that seasonality is the thing that determines whether you make it to the next peak.
Build the model by month. The annual total will barely change; what you gain is the trough. You will see the stretch where revenue runs below fixed costs, how long it lasts, and how much cash you need standing by before it starts. Owners rarely fail in the months the model is wrong about. They fail in the months the model never showed them separately.
Two adjustments belong in the monthly view. First, a ramp: a new listing with no reviews and no booking history does not perform like an established one, and the months before it does belong in the model at lower numbers. Second, any nights you intend to use the property yourself. Owner stays are a legitimate reason to buy. They are also revenue you will not earn, and they tend to land on exactly the dates that earn most.
The revenue line is where models go wrong by the largest margin, but the expense side is where they go wrong by omission. A short-term rental has operating costs a long-term rental does not, because you are running a small hospitality business rather than collecting rent.
| Line | Why it gets missed |
|---|---|
| Utilities at guest usage levels | In a long-term rental the tenant often pays them. Here you do, and guests do not run a house the way a resident does |
| Consumables and restocking | Paper goods, coffee, soap and laundry are per-stay costs, so they scale with success rather than with time |
| Linen and towel replacement | Washed after every stay, they wear out on a schedule measured in turns, not years |
| Furnishing refresh reserve | The furniture is a depreciating operating asset, not a one-time setup cost |
| Grounds, pest, pool and filter service | Recurring, seasonal, and easy to leave out of a model built in one sitting |
| Software and subscriptions | Pricing tools, locks, cameras, noise monitors and the listing stack each carry a monthly fee |
| Maintenance and repair reserve | A house turning over fifty-plus times a year breaks faster than one occupied by a single household |
| Management | If you are not paying a manager, you are the manager, and your hours are a real cost even when nothing leaves the account |
The last row is the one owners argue with, and it is the one that most often decides whether the deal was good. A model with no management line is not a model of a passive investment. It is a model with an unpaid employee in it. Price the work at what it would cost to hire out, then decide whether you want the job.
Insurance and the taxes that apply to short-term stays both belong in the model, and neither should be filled in from a blog post, a forum thread, or a figure a friend quoted for a different property in a different place. Both are specific to the address, the structure, how it will be used, and the jurisdiction it sits in.
Get a written quote from an insurance professional for the actual property and the actual intended use. Get the applicable tax treatment from a tax professional or directly from the jurisdiction. Put their numbers in the model. Anyone who hands you a confident figure for either without knowing the address is guessing, and you are the one who finds out.
A single-column pro forma produces one answer, which is the thing you wanted it to produce. Build three columns instead: the case you expect, a case where rate and occupancy both come in meaningfully below it, and a case where they come in above.
Then stop asking what the property returns and start asking what has to be true for it to stop working. At what revenue per available night does the property stop covering its fixed costs? How many months of the trough can you fund from cash on hand without a booking? If the first season comes in below plan, what is the move — reprice, extend minimum stays, hold? Decide that now. Owners who have not set a break point in advance tend to discover it in the middle of the slowest month of the year, which is the worst moment to be making that decision for the first time.
A pro forma that survives is not the one with the best bottom line. It is the one you can still open twelve months later, compare against what actually happened, and find the variance where you expected it to be.
If you are modeling a purchase in the Triangle or Southeast Florida and want a second read on the revenue line before you commit, we will put together a free revenue estimate for the specific property — what we would expect it to produce, and what we would want to see before we would believe it.
