Ask a landlord with one or two properties whether the investment is working, and most will do the same sum in their head. Rent comes in, the bond repayment goes out, and whatever is left over gets called profit. If that number is positive, the property is working. If it is not, something is wrong.
That sum answers a different question to the one being asked. It tells you whether this month's cash flow was positive. It says nothing about whether the property is actually a good investment, because it leaves out almost every cost of owning it.
Gross yield is the number everyone quotes. Net yield is the number that matters.
Gross yield is simple: annual rent divided by what you paid for the property, times 100. It is the number in the property portal listing, the number an agent quotes when they are trying to sell you something, and the number most landlords carry around in their head as "my return."
It is also the least useful number in property investment, because it assumes the property has no running costs. No rates, no levy, no insurance, no maintenance, no vacant months, no fee to whoever manages it, and no tax on the income. A property that looks like an 8% investment on gross yield can easily be a 4% or 5% investment once those costs are counted. The gap is not rounding error. It is the entire difference between a property that is quietly building wealth and one that is quietly leaking it.
Net yield fixes this. Same formula, but the annual rent is reduced by everything it actually costs to hold the property first. Bond repayments are deliberately left out of both calculations, because yield measures what the property itself earns, independent of how you financed it. What the bond does is turn that yield into a cash flow number, which is a separate question, and the one most landlords accidentally answer instead.
A worked example, with real numbers
Say you own a two-bedroom sectional title unit, bought for R1,300,000, currently renting for R11,000 a month. The bond balance sitting against it is R850,000, on a 20-year term.
Gross yield: R11,000 × 12 = R132,000 a year. R132,000 ÷ R1,300,000 = 10.2%. On paper, a strong result. Above 8% gross is generally considered a solid South African residential number.
Now the running costs, month by month:
| Line item | Monthly amount |
|---|---|
| Body corporate levy | R1,450 |
| Municipal rates | R980 |
| Landlord insurance | R280 |
| Maintenance reserve (1% of value, annualised) | R1,083 |
| Management fee (10% of rent) | R1,100 |
| Vacancy allowance (5% of rent) | R550 |
| Total | R5,443 |
That is R65,320 a year the gross yield calculation never mentioned. Take it off the R132,000 in rent and the property actually earns R66,680 a year against a R1.3 million purchase price. That is a net yield of 5.1%, roughly five percentage points below the number on the listing. Five points is not noise. On this property it is the difference between R132,000 and R66,680 a year, which is most of what the "profit" was supposed to be.
Now bring the bond back in, because this is where the mistake in the opening paragraph actually happens. At the current prime lending rate of 10.5%, following the Reserve Bank's rate hike at the end of May, that R850,000 bond costs roughly R8,486 a month. Rent minus bond: R11,000 - R8,486 = R2,514. Positive. Looks like R2,500 a month in your pocket, and plenty of landlords stop right there, satisfied.
Except the levy, rates, insurance, maintenance reserve and management fee still have to come out of that R2,514, and they add up to R4,893 a month before you have even accounted for a vacant month. R2,514 minus R4,893 is minus R2,379. The property that looked like it was paying you R2,500 a month is actually costing you close to R2,400 a month, and that is before tax on the portion that is taxable, and before a single month of vacancy has happened.
This is the specific mistake. Positive cash flow after the bond is not proof of a good investment. It is proof that the bond, on its own, is smaller than the rent. Every other cost of ownership is still sitting there, uncounted, until a levy increase, a geyser failure or a two-month vacancy forces it onto the table all at once.
The costs landlords underestimate, and why
None of the six line items above are exotic. They are the ordinary cost of owning property in South Africa, and most landlords already know each one exists. What gets missed is adding them up in one place, every month, against the actual rent collected rather than an assumed round number.
The rental market context makes the gap worse than it looks. According to PayProp's Rental Index for the third quarter of last year, the national average rent reached R9,286, up 4.9% year on year, while the share of tenants in arrears climbed to 17.2%, the highest level in a year. TPN's vacancy data puts the national vacancy rate at roughly 5.4% to 5.6%, which works out to around three weeks empty every year even in a stable market. A landlord who has not budgeted a vacancy allowance is not avoiding that cost. They are just discovering it in the specific month it happens, as a shock, instead of carrying it as a known, averaged-out line item the rest of the year.
There is a banking parallel here that is worth sitting with. A lender that only looks at interest income and ignores expected credit losses is not being conservative, it is mispricing risk. A landlord who only looks at rent minus bond is doing the same thing: treating a real, predictable cost (vacancy, maintenance, management) as though it will not happen, rather than provisioning for it the way it actually behaves, which is rare in any single month and near-certain over a few years.
What Usika's Financial Reports feature is for
This is the exact gap the Financial Reports feature in your Usika dashboard was built to close. It does not work off assumed averages. It pulls the rent you actually collected and the costs you have actually logged against each property, and turns that into a tax-year report you can export as a CSV or PDF instead of reconstructing it from memory and bank statements every February.
The point is not a prettier spreadsheet. It is replacing the mental shortcut of rent minus bond with the number that was there all along, sitting in your own transaction history, waiting to be added up properly. Once that number is in front of you, you stop guessing whether the R11,000 example above describes your property or falls short of it. You know, the way a lender knows before it prices a loan, not the way a landlord hopes after checking a bank balance.
Sources: PayProp Rental Index, Q3 2025. TPN vacancy data. South African Reserve Bank Monetary Policy Committee, May 2026. Cost line items in the worked example are illustrative, built from standard South African buy-to-let cost ratios, and not a quoted figure from a single named source.