Quick answer
To stress-test a rental deal, change one assumption at a time (occupancy, nightly rate, expenses and interest rate), record cash flow and DSCR after each change, and find the value where cash flow reaches zero. That break point, and how far it sits from your base case, tells you how much room the deal has for error.
This guide is about method; for building the base case, see how to analyze a short-term rental deal. Every number below is an invented Example, not a forecast. This is general information, not tax, legal or investment advice.
Key takeaways
- A sensitivity table changes one input at a time and records what happens to cash flow and DSCR, so you can see which assumption the deal leans on most.
- The most useful number for each input is its break point: the value at which cash flow after the mortgage reaches zero.
- Test occupancy, nightly rate, variable costs, fixed costs and the interest rate, in steps you can defend from comparable homes, quotes and lender terms.
- A two-way grid of occupancy and rate shows how quickly small misses combine. Real downturns rarely move one input alone.
- Decide before you look at the results what margin of safety you need, and walk away from deals that only work at the base case.
In this guide
Why a single pro forma is not enough
A pro forma gives one answer from one set of guesses, and a stress test shows how wrong those guesses can be before the deal stops working.
Each line in a pro forma is an estimate: the occupancy you expect, the rate guests will pay, what insurance will cost next year, what your lender will charge. Each is uncertain, and a deal with healthy cash flow at the base case can turn negative when two or three of them miss in the same direction.
The Corporate Finance Institute describes sensitivity analysis as a way to analyze how the different values of a set of independent variables affect a specific dependent variable. For a rental, the dependent variable is usually cash flow after the mortgage, or the debt service coverage ratio (DSCR) your lender will test. The independent variables are your assumptions.
Done well, the exercise answers three questions:
- Which assumption does this deal depend on most?
- How far can each one miss before cash flow reaches zero?
- Is that margin wide enough given what I know about the market?
Set the base case first
You cannot stress a deal until the base case is honest, so build it from evidence, not from the listing.
For each input, write down where the number came from:
- Occupancy and nightly rate: comparable large homes nearby, not the seller's best year.
- Variable costs: platform fees, cleaning, supplies and repairs that rise with bookings.
- Fixed costs: property tax, insurance, utilities, internet, HOA dues.
- Loan terms: a written quote from a lender for a rental property, not a headline rate.
If the seller's pro forma is your only source, our guide on how to read a short-term rental pro forma shows what to check first. Taxes and insurance deserve their own deep look, which our guide on insurance and property tax shocks covers.
Here is the Example base case used through this guide.
Pick the inputs and the steps
Test the inputs that are both uncertain and large, and move each in steps you can defend.
Five inputs cover most of the risk in a short-term rental:
| Input | Why it is uncertain | Steps to test (Example) |
|---|---|---|
| Occupancy | New competition, demand shifts, ranking changes | 5 and 10 points lower |
| Average nightly rate | Discounting in slow months, new supply | 5% and 10% lower |
| Variable costs | Platform fee changes, cleaning rates, wear on a large home | 5 and 10 points of revenue higher |
| Fixed costs | Insurance renewals, tax reassessment, utilities | 10% and 25% higher |
| Interest rate | Rate at closing, an adjustable loan, a refinance | 0.5, 1.0 and 1.5 points higher |
Choose step sizes from evidence where you can. For occupancy and rate, look at the spread between the weaker and stronger comparable homes. For variable costs, check current fee pages: Airbnb's service fee page says most hosts on the single fee structure pay 15.5% of the booking, against 3% for most hosts on the split-fee structure, and that the single fee is mandatory for most hosts, including those who use property management software. If your base case assumed the lower fee, the step to the higher one is a real test, not a guess. Our guide on underwriting platform risk goes further into fee and ranking changes.
For interest, a public benchmark helps anchor the range. Freddie Mac's Primary Mortgage Market Survey showed a 30-year fixed average of 7.11% for the week of November 12, 2026. That survey is built from conventional loan applications, so a loan on a short-term rental may be priced differently. Use your lender's written quote as the base and test above it.
Move one input at a time
The core of the method is simple: change one input, hold everything else at the base case, and write down the results.
Here is the Example deal with each input moved on its own. The base row is the same in every case.
| Change from base (Example) | Revenue | NOI | Cash flow | DSCR |
|---|---|---|---|---|
| Base case | $100,375 | $43,244 | $7,317 | 1.20 |
| Occupancy 50% (-5 points) | $91,250 | $37,312 | $1,386 | 1.04 |
| Occupancy 45% (-10 points) | $82,125 | $31,381 | -$4,545 | 0.87 |
| Rate $475 (-5%) | $95,356 | $39,982 | $4,055 | 1.11 |
| Rate $450 (-10%) | $90,338 | $36,719 | $793 | 1.02 |
| Variable costs 40% of revenue | $100,375 | $38,225 | $2,299 | 1.06 |
| Variable costs 45% of revenue | $100,375 | $33,206 | -$2,720 | 0.92 |
| Fixed costs +10% ($24,200) | $100,375 | $41,044 | $5,117 | 1.14 |
| Fixed costs +25% ($27,500) | $100,375 | $37,744 | $1,817 | 1.05 |
| Interest 7.50% | $100,375 | $43,244 | $5,486 | 1.15 |
| Interest 8.00% | $100,375 | $43,244 | $3,620 | 1.09 |
| Interest 8.50% | $100,375 | $43,244 | $1,722 | 1.04 |
Two things stand out. First, occupancy and variable costs do the most damage per step: a 10-point occupancy miss or a 10-point rise in variable costs takes this Example from positive to negative. Second, the interest rate matters less in this Example than many investors expect, because a fixed-rate loan locks the payment once you close. The interest risk is mostly before closing, or later if you plan to refinance or carry an adjustable loan.

Find the break points
A break point is the value of one input at which cash flow after the mortgage reaches zero, and it is the single most useful number in a stress test.
Steps tell you what happens at a few chosen values. A break point tells you how far the deal can slide before it costs you money each month. For the Example deal, holding everything else at the base case:
| Input (Example) | Base case | Break point | Room before cash flow reaches zero |
|---|---|---|---|
| Occupancy | 55% | About 48.8% | About 6 points |
| Average nightly rate | $500 | About $444 | About 11% |
| Variable costs | 35% of revenue | About 42.3% | About 7 points |
| Fixed costs | $22,000 | About $29,300 | About 33% |
| Interest rate | 7.00% | About 8.95% | About 2 points |
Read this table as a ranking. In this Example, the deal is most fragile on occupancy: six points is a small miss for a home in its first year, a year with new competition nearby, or a year when the listing loses ranking. Fixed costs have the most room, but a 33% jump is not unheard of when insurance is repriced.
The deal analyzer at /investors/tools/deal-analyzer shows break-even occupancy directly in its results. For the other inputs, find the break point by trial: lower the rate, or raise a cost, until cash flow after the mortgage reads zero, then note that value.
Combine the misses: a two-way grid
Real downturns rarely move one input at a time, so the next step is a grid that moves two inputs together.
Occupancy and rate are the natural pair, because a soft market usually means fewer bookings and lower prices at once. The Corporate Finance Institute notes that a data table shows the effect of up to two variables at once, which is why two-way grids are the standard format. Here is cash flow after the mortgage for the Example deal.
| Cash flow (Example) | Rate $500 | Rate $475 | Rate $450 |
|---|---|---|---|
| Occupancy 55% | $7,317 | $4,055 | $793 |
| Occupancy 50% | $1,386 | -$1,579 | -$4,545 |
| Occupancy 45% | -$4,545 | -$7,214 | -$9,883 |
A 5-point occupancy miss with a 5% rate cut, both modest on their own, turns this Example negative. That is the lesson of the grid: small, related misses add up faster than the one-way table suggests.
You can extend the idea with a combined downside row. For example, take occupancy 50%, rate $475 and fixed costs 10% higher together. In this Example, revenue falls to about $86,688, NOI to about $32,147, and cash flow to about -$3,779 a year, with DSCR at 0.89. Ask whether you could carry that from savings for a year or two.
Build it in the free tools
You can build the whole table with the free calculators and a notebook or spreadsheet.
A simple way to work:
- Open the deal analyzer and enter your base case. Note cash flow after the mortgage, DSCR, NOI and break-even occupancy.
- Use Copy a link to this deal to save the base case. Every input is stored in the link.
- Change one input to its first step. Note the same four results. Copy the link for that row.
- Put the input back and move to the next one. Repeat for every row in your table.
- Check the analyzer's own stress test panel, which shows revenue, cash flow, cash-on-cash and DSCR at 10 and 20 points below your planned occupancy.
- Look at the month-by-month view and the cash reserve for slow months for the worst row. A deal that survives on a yearly total can still need a large reserve in the quiet season.
- Use the DSCR calculator to check what the lender will see for any row, and the cash-on-cash return calculator if you want that measure for each case.
The analyzer spreads your yearly occupancy across busy and quiet months, so its results will not match a flat model like the Example above to the dollar. That is fine. What matters is that every row uses the same model, so the differences between rows are fair.
The limits of a sensitivity table
A sensitivity table is a tool for finding weak points, not a prediction, and it has blind spots you should know.
- Inputs move together. A recession can cut occupancy, cut rate and raise some costs at once. One-way rows understate that, which is why the grid and the combined row matter.
- The base case can be wrong. If the base occupancy came from the seller's best year, every row inherits the error. Stress tests do not fix bad inputs.
- Step sizes are judgment calls. A 10-point occupancy drop may be mild in one market and severe in another. Tie each step to evidence.
- Tail events are outside the table. A suspended listing, a local ban on short-term rentals, a storm that closes the home for months: these are not small moves in one input. Treat them as separate scenarios with their own question: how long could I carry the home with no income?
- Results only hold inside the range you test. The Corporate Finance Institute makes the same point: the analysis applies within the boundaries you set.
None of this makes the exercise less useful. It means the table should change your questions, not end them.
What to do with the results
Decide before you build the table what margin you need, then let the results decide the deal, not the other way round.
Some ways investors use the results:
- Walk away when the break point sits inside the range of normal outcomes, such as a break-even occupancy close to what weaker comparable homes book.
- Renegotiate the price using the table: a lower price lowers the loan and the debt service, which moves every break point.
- Change the financing: a larger down payment or a different loan structure can widen the margin, at the cost of more cash in.
- Hold a larger reserve, sized to carry the worst realistic row for a stated number of months.
- Fix the weakest input where you can: firmer insurance quotes, a cleaning contract with set rates, or a plan for more than one booking channel.
If you want to talk about investing with us in large group homes, that conversation starts at /investors. Nothing in this guide is an offer or a projection of returns.
The bottom line
A stress test turns one hopeful pro forma into a map of where the deal breaks. Set an honest base case, move occupancy, rate, variable costs, fixed costs and interest one at a time, find each break point, then combine the likely misses in a grid. If the deal only works when every guess goes right, it does not work. Build your table in the free tools before you make an offer.
Frequently asked questions
What is a sensitivity table for a rental property?
It is a table that shows how cash flow, DSCR or another result changes when you move one assumption, such as occupancy or the interest rate, while holding the others fixed. It tells you which assumptions the deal depends on most.
What is a break-even point in a rental deal?
It is the value of one input at which cash flow after the mortgage reaches zero. For example, a break-even occupancy is the share of nights the home must book to cover every cost and the loan payment, with everything else as planned.
How is a sensitivity table different from a scenario analysis?
A sensitivity table moves one or two inputs and holds the rest. A scenario analysis describes a specific situation, such as a recession year or a fee change, and moves several inputs together. Use both: the table finds the weak points and scenarios test realistic combinations.
How much should I lower occupancy in a stress test?
There is no single rule. Use what comparable homes actually book as the base, then test steps such as 5 and 10 points lower, and find the occupancy where cash flow reaches zero. The size of the gap between the base and that break point is what matters.
Can I build a sensitivity table with the JekkyStay tools?
Yes. The deal analyzer at /investors/tools/deal-analyzer shows break-even occupancy and an occupancy stress test, and you can change one input at a time, note the results and copy a link to each version. The single calculators help check DSCR and cash-on-cash for each row.
Does a stress test tell me whether to buy?
No. It shows how much room for error a deal has under your assumptions. Whether that room is enough depends on your reserves, other income and risk tolerance. This is general information, not investment advice.




