Food cost percentage for South African restaurants: the VAT trap

Nearly every food cost calculator you will find online was built for the United States, where sales tax is added at the till rather than included in the menu price. Use one as a South African operator and it will quietly overstate your margin by about 13%, because it treats the full menu price as if it were all yours.

That gap is the difference between a menu that looks profitable and one that actually is.

The calculation, done properly

Food cost percentage is your ingredient cost divided by the selling price excluding VAT, times 100. In South Africa VAT is 15%, so the first move is always to strip it out: divide the menu price by 1.15.

1. Take the menu price: R120.
2. Divide by 1.15 to remove VAT: R104.35. That is your actual revenue.
3. Take the ingredient cost at yielded weights: R31.30.
4. Divide and multiply by 100: 31.30 ÷ 104.35 × 100 = 30% food cost.
5. Gross profit is what is left: R104.35 − R31.30 = R73.05, or 70%.

Run the same dish through an American calculator and it compares R31.30 against the full R120, returning 26% food cost and a gross profit that does not exist. Four percentage points sounds small until it is applied across an entire menu.

The second mistake: costing at raw weight

A whole rump does not weigh the same once it has been trimmed. Neither does a case of onions once the skins and ends are gone. Costing at purchase weight rather than yielded weight under-costs every plate that uses it, and it compounds quietly across a menu.

Weigh what actually reaches the plate, not what came off the delivery vehicle. For proteins in particular the difference between the two is often 15 to 25%.

What a good number looks like

Most South African kitchens aim for 28 to 35% on food and 18 to 25% on drinks, but the right target depends entirely on your rent, your labour bill and your volume. A high-rent Sandton site and a suburban family restaurant should not be chasing the same number.

Percentages do not pay wages. A dish at 35% food cost returning R95 of gross profit beats one at 25% returning R40 — and a kitchen optimised purely for percentage will quietly stop selling the dishes that actually pay the rent.

Rank the menu by cash, then by percentage

Work out the cash gross profit on every dish and sort by it. Then look at how often each one sells. The four combinations tell you what to do without any further analysis:

• High cash profit, sells often — protect it. Do not change the spec, do not move it on the menu.
• High cash profit, sells rarely — a marketing problem, not a costing one. Move it, describe it better, get staff recommending it.
• Low cash profit, sells often — the dangerous one. It is busy and it is not paying. Re-cost it, re-spec it, or reprice it.
• Low cash profit, sells rarely — take it off. It is costing you prep time, stock lines and menu space.

Load-shedding costs belong in the calculation

Generator diesel and spoilage after outages are real costs of putting food on a plate, and most operators leave them out of costing entirely because they sit in overheads. They do not have to appear in every dish costing, but they must appear somewhere before you judge whether your gross profit is actually covering the business.

Doing this for every dish

The arithmetic is simple. Doing it for eighty dishes, and redoing it every time a supplier moves a price, is what nobody has time for — which is why most menus are costed once and then slowly drift out of date.

FoxEra costs every dish and re-costs it automatically when a supplier price changes, with VAT handled properly and everything reported in your own currency. It is free — no subscription and no contract. Honest limits: we are new, with no South African customers yet, and rand billing is not configured, so while the free tier works there is no way to pay us today even if you wanted to.