The 30-second screen
Most listings deserve thirty seconds, not thirty minutes. The point of a screen is to throw out obvious losers fast enough that you still have energy for the two or three properties worth real work. Rules of thumb are perfect for this and terrible for anything else.
The one I still start with is the price-to-rent test. Monthly rent divided by purchase price, expressed as a percentage. The old shorthand said 1% or better, which in most metros today is a filter that returns almost nothing, so I treat it as a ranking tool rather than a pass or fail line. My sibling post does the one percent rule still work covers where it survives and where it does not.
The second is the expense shorthand: assume operating expenses will eat about half of gross rent, excluding the mortgage. It is crude and it is usually closer to the truth than the seller's expense sheet. There is more on its limits in the fifty percent rule explained. Run both in the rental deal screener and you can process a whole day of new listings in the time it takes coffee to cool.
The house we will run
Here are the assumptions, and I want to be clear that I am choosing them rather than quoting market data. A single-family rental at $350,000, renting for $2,600 a month, in a market where I am assuming property taxes of $4,200 and insurance of $1,900 a year. Financing at 25% down on a 30-year loan at 7%.
First the screen. Monthly rent of $2,600 against a $350,000 price is 0.74%, well short of the 1% shorthand. Gross scheduled rent is $31,200 a year. Half of that is $15,600, which is the expense figure the 50% shorthand would predict. Neither number kills the deal outright, but both are telling me this will be tight. Now the real work.
Income, after the part that does not show up
Nobody collects gross scheduled rent. Between turns, vacancy, and the occasional month a tenant leaves early, there is always a gap, and it belongs in the analysis rather than in your hopes. I will use 6%, which on $31,200 is $1,872, leaving effective gross income of $29,328.
Two habits here have saved me repeatedly. Underwrite to market rent rather than in-place rent when the current tenant is paying under market, because the day you raise it is the day they may leave. And do not add income lines you have not verified. Laundry, storage, pet fees, and parking are real in some buildings and imaginary in most pro formas.
The expense lines people forget
This is where analyses go wrong, and it is almost never through an exotic mistake. It is through omission. Here is the full list for our house, using percentages of gross rent for the variable lines.
Total operating expenses come to $12,778. Notice that is below the $15,600 the 50% shorthand predicted, which is a signal to double-check my own assumptions rather than to celebrate. Taxes and insurance are the two lines to verify from actual bills, since both can jump after a sale, and a reassessment at the new purchase price is the single most common reason a first-year budget breaks.
- Property taxes: $4,200, verified from the county, not the listing
- Insurance: $1,900, quoted for a rental policy rather than an owner-occupant one
- Property management: 8% of collected rent, $2,346, counted even if you self-manage, because your time is not free and a future buyer will price it in
- Repairs and maintenance: 5% of gross rent, $1,560
- Capital reserve: 6% of gross rent, $1,872, for the roof, the HVAC, and the water heater that all have finite lives
- Lawn, snow, and turnover cleaning: $600
- Bookkeeping, legal, and software: $300
NOI and cap rate
Net operating income is effective gross income minus operating expenses: $29,328 minus $12,778 equals $16,550. NOI deliberately excludes the mortgage, because it describes the property rather than the buyer.
Cap rate is NOI divided by price: $16,550 divided by $350,000 is 4.73%. Whether that is good depends entirely on the market, the asset class, and what safer alternatives pay right now, which is the subject of what is a good cap rate. What it definitely is, at this moment, is lower than the interest rate on the loan we are about to add. Hold that thought. You can rebuild all of this line by line in the rental property calculator or check just the yield in the cap rate calculator.
Now add the loan
Twenty-five percent down is $87,500, leaving a $262,500 loan. At 7% over 30 years, principal and interest run about $1,746 a month, which is $20,957 a year. Set that against NOI of $16,550 and annual cash flow is negative $4,407, before a single unexpected repair.
Two ratios describe the damage. Debt service coverage is NOI divided by annual debt service: $16,550 over $20,957 is 0.79. Lenders on investor loans typically want 1.20 to 1.25 or better, so this does not merely lose money, it likely does not fund. Check yours in the DSCR calculator, and if that loan type is new to you, DSCR loans explained covers how the underwriting differs from a conventional file.
Cash on cash is the other. With $87,500 down plus roughly $10,000 in closing costs, cash invested is $97,500, and negative $4,407 against that is negative 4.52%. Run your own structures in the cash on cash return calculator. The uncomfortable insight is that the property is not the problem here, the spread is. A 4.73% yield financed with 7% money produces negative leverage no matter how nice the kitchen is.
What would have to change
This is the step most people skip, and it is the one that converts a rejection into an education. Do not just say no, find the number that makes it a yes and then ask whether that number is plausible.
To reach a 1.25 debt service coverage ratio at this price and this loan, NOI would need to be about $26,196, which with this expense structure requires gross rent near $43,980 a year, roughly $3,665 a month. That is about 41% above the $2,600 I assumed. No amount of paint achieves that. Alternatively the price would have to fall dramatically, or the down payment would have to grow so large that the return on that additional equity becomes the real question.
Once the required change is that big, the answer is not to negotiate harder. It is to move on. I have watched investors spend three weeks trying to fix a deal that needed a 41% rent increase, and the three weeks were the real loss.
Cash flow is one of three returns
A negative cash flow number does not automatically end the conversation, because cash flow is only one of the three ways a rental pays you. The second is principal paydown. On our $262,500 loan at 7%, the first twelve payments include about $18,291 of interest and about $2,667 of principal, so the balance falls to roughly $259,834 by the end of year one. That $2,667 is a real gain in net worth funded by the tenant, and it grows every year as amortization tilts.
The third is appreciation, and this is where analysis turns into fiction. Assume 3% on $350,000 and you have added $10,500 a year, which single-handedly flips our losing deal into an apparent winner: negative $4,407 of cash flow plus $2,667 of paydown plus $10,500 of assumed appreciation is $8,760 of total return. But that appreciation number is an input I typed, not a fact I discovered. Property values fall in some years and in some markets. If a deal only works when the appreciation box is filled in, you are not investing, you are forecasting.
My rule is to underwrite to zero appreciation and treat any of it as a bonus. If you want to see what different assumptions do over a hold period, the property appreciation calculator will show you, and the exercise is more useful as a sensitivity test than as a projection.
The decision, and the mistake that taught me the checklist
The workflow in order: screen in thirty seconds, build income after vacancy, list every expense including the ones nobody bills you for monthly, compute NOI and cap rate to judge the property, then layer financing to judge your position in it, then check whether the deal survives with appreciation set to zero. This one does not survive step five, so it is a no.
My own checklist exists because of what I left off it. My first analysis had no capital reserve line at all. I reasoned that the roof was fine, the furnace was newer, and I would deal with big items when they arrived. They arrived in month fourteen, together, and I paid for them out of a line of credit that turned a mediocre year into a bad one. A capital expenditure reserve is not pessimism. It is the recognition that every component of a building has a lifespan and you own all of them.
Two closing notes. First, everything here is pre-tax. Depreciation can meaningfully change the after-tax picture on a deal like this, and I walked through that math in rental property depreciation explained. Second, all of this is arithmetic on assumptions I chose, not tax or investment advice, so confirm the tax treatment with a CPA and the rent assumptions with someone who actually leases in that zip code.
Questions people ask
Capital reserves, property management when self-managing, vacancy, and the tax reassessment that often follows a sale. In our example those lines total thousands of dollars a year. Leaving them out is what turns a break-even property into an apparent winner on paper.
It depends on the market and on what safe alternatives yield, but in this example it matters mainly because it sits below the 7% loan rate. When the cost of debt exceeds the property's unlevered yield, leverage reduces your return rather than amplifying it.
Only with eyes open and reserves in the bank. Negative $4,407 a year is a real bill you pay monthly, and the offsetting returns are $2,667 of principal paydown plus appreciation that may or may not arrive. If the deal only works with appreciation assumed, treat that as a warning.
Investor loan programs commonly look for 1.20 to 1.25 or higher, meaning NOI covers debt service with room to spare. Our example lands at 0.79, which not only loses money each month but would likely fail underwriting on a debt-service-based loan.
The screen takes under a minute. A complete underwrite, once you have verified taxes, insurance quotes, and market rents, usually takes 20 to 30 minutes. The slow part is not the arithmetic, it is confirming the inputs, and that is exactly the part worth doing.
