No official Australian beverage cost benchmark exists: the ATO bundles food, beverage and all stock into one cost of sales line, which runs 31% to 39% of turnover for restaurants and 31% to 46% for pubs, taverns and bars.
Skai Solutions builds phone agents for Australian and New Zealand venues, which means we read a lot of official cost data on behalf of operators. This page gives you the ATO ranges by venue type parsed from the source file, the reason there is no beverage line in them, and the method for calculating your own.
The honest answer starts with a finding most pages skip: Australia has no official beverage cost benchmark, for any venue type, in any year. We went looking for one properly and it does not exist.
The Australian Taxation Office publishes small business benchmarks drawn from lodged tax returns, and they are the only official figures of this kind in the country. They carry a line called cost of sales. That line bundles food, beverage and every other item of stock sold into one number. There is no separate beverage row for restaurants, coffee shops, takeaway, catering or pubs. So the beverage cost percentage you read online is either imported from another market or somebody's own calculation, and almost never labelled as either.
What the Australian data can tell you is more interesting than a single target, because it lets you compare venue types whose beverage share differs enormously.
We downloaded and parsed the machine readable 2023-24 small business benchmarks from data.gov.au on 7 October 2026. These are the cost of sales ranges, as a share of turnover, by business type and turnover band.
| Business type | Low turnover band | Medium | High |
|---|---|---|---|
| Pubs, taverns and bars | 34% to 46% | 33% to 41% | 31% to 42% |
| Restaurants | 32% to 39% | 32% to 38% | 31% to 36% |
| Coffee shops | 34% to 42% | 35% to 41% | 33% to 38% |
| Takeaway food services | 34% to 42% | 36% to 43% | 35% to 41% |
| Catering services | 31% to 43% | 32% to 41% | 31% to 41% |
| Bakeries and hot bread shops | 31% to 38% | 34% to 39% | 29% to 36% |
The turnover bands differ by type. Restaurants are split at A$65,000 to A$500,000, then to A$2 million, then above. Pubs, taverns and bars are split at A$100,000 to A$750,000, then to A$2.5 million, then above. Coffee shops are A$65,000 to A$250,000, then to A$600,000, then above. Read your own band, not the first column.
Now the comparison that matters. A pub is the most beverage weighted hospitality business the ATO benchmarks. If beverage carried a dramatically better cost percentage than food, pubs should sit well below restaurants. They do not. In the top turnover band a pub reports 31% to 42% and a restaurant reports 31% to 36%: the pub range is wider and its ceiling is six points higher.
Be careful what you conclude from that, because we are being careful too. A pub's rent, mix, gaming revenue and wastage all differ from a restaurant's, so this is not proof that beverage runs worse than food. The defensible conclusion is narrower and more useful: the Australian official data cannot separate the two, and the spread between venue types is small enough that no imported beverage target is doing you any good.
Four numbers, one period, and the period has to be a real stocktake to a real stocktake.
A worked example. Opening stock A$9,400. Purchases A$21,800. Closing stock A$10,100. Beverage cost of goods is 9,400 plus 21,800 minus 10,100, which is A$21,100. Beverage sales ex GST of A$76,000 gives 21,100 divided by 76,000, or 27.8%.
Three traps that make the number meaningless if you skip them. Count stock on the same day of the week each time, because a Friday count and a Monday count are different businesses. Take GST out of the sales figure or the percentage flatters you by about a tenth. And keep staff drinks, spoilage and promotional pours in the cost rather than quietly out of it, because that is exactly where the gap between a good number and a true one lives.
The same method applied to food is on our restaurant food cost percentage page, and the two together plus labour give you the figure that actually predicts survival, which is on our restaurant prime cost page.
The formula is one line. Beverage cost percentage equals beverage cost of goods sold divided by beverage sales excluding GST, multiplied by 100. Beverage cost of goods sold is opening stock at cost plus purchases at cost minus closing stock at cost. On the numbers above that is 21,100 divided by 76,000 times 100, which is 27.8%.
Note what the formula does not contain. There is no allowance for what you poured and did not sell, so staff drinks, spoilage and promotional pours sit inside the cost figure whether you put them there deliberately or not. It is also only as honest as the two stocktakes at either end, which is why a monthly number counted properly beats a weekly one estimated.
This is usually what an operator means by a standard or ideal beverage cost, and it is not a benchmark somebody else publishes. It is your own second number.
Your ACTUAL beverage cost is the formula above: what the stocktake says you spent. Your IDEAL beverage cost is what your own recipes say you should have spent on exactly the drinks the till recorded. You build it by costing each drink once at its poured quantity, then multiplying by the units sold in the period. A 30 ml nip costed at A$1.35 that sold 400 times contributes A$540 of ideal cost, and nothing about that figure depends on counting a shelf.
The gap between the two is the only number on this page that points at a cause. If ideal cost is 24% and actual is 27.8%, that 3.8 points is not a pricing problem and it is not the market. On A$76,000 of beverage sales it is about A$2,900 a period going out as over pouring, wastage, unrung drinks or stock walking. If the two numbers are close and the percentage is still high, the problem is in your buying or your pricing instead, and nothing you do behind the bar will fix it. Running the pair is how you tell those two situations apart, and it is also why a single percentage against an imported band tells you almost nothing on its own.
Operators type liquor cost percentage expecting a tighter band than the one for all drinks, and the honest answer is that no Australian official source publishes one. As the ATO data above shows, cost of sales is reported as a single line that bundles food, beverage and every other item of stock sold, so it cannot be split by sub category at all, let alone into spirits, beer, wine and non alcoholic. Any liquor, beer or wine percentage you read is either imported from the United States or someone's own figure.
You can still produce yours, and it is the more useful version. Map your POS sales categories to your stock categories once so spirits, beer, wine and non alcoholic each have a sales line and a stock line, then run the formula four times over the same period. The spread is usually wider than expected, which is the point: a blended 27.8% can hide a wine line that is fine and a spirits line that is not. Same method as our restaurant food cost percentage page, and the cost per unit side is worked through on what it costs to make a cup of coffee.
Every page on the first results page for this query quotes a target in the high teens or low twenties. It comes from the United States, where the published benchmarks are built on a cost base Australia does not have. Baker Tilly, an accounting firm, describes about 60% of food and beverage sales as the prime cost level a sustainable restaurant runs at. The ATO's own Australian averages put a typical restaurant past that, for reasons that are structural rather than managerial: award rates under MA000119, weekend and evening penalty rates and 12% superannuation have no American equivalent.
On the beverage side specifically, three Australian cost drivers push the other way from the imported figure. Alcohol excise here is indexed and applied per litre of alcohol rather than per sale. Freight to a venue outside a capital city is a real line on a beverage invoice in a country this size. And the merchant fee on a card transaction is charged on the drink as well as the meal, which our restaurant merchant fees page quantifies across a year of volume. None of those three appear in a US pour cost benchmark.
So use the imported number as what it is: a number from a market with a different cost base. Measure your own, and compare it to your own last quarter.
Four moves, cheapest first, and the first two cost nothing.
If none of those four work, the problem is usually mix rather than cost: you are selling the low margin items and not the high margin ones. That is a menu and a floor conversation, not a purchasing one, and the figures on our average spend per head page are where it shows up first.
One caution the ATO states plainly and we repeat. A business outside a benchmark range is not automatically doing anything wrong, only that there may be room to improve. These are lodged tax returns rather than audited accounts, the data year is 2023-24, and the ranges are not targets the ATO sets.
No, and nothing close to one. There is no Inland Revenue equivalent of the ATO small business benchmarks published at this level of detail. The only sector wide New Zealand figure available free is the Restaurant Association of New Zealand's 2025 Hospitality Report, which puts wage costs at about 40% of hospitality revenue. That is a different line on a different definition across a wider sector, so it is not a beverage comparison and we will not present it as one. A New Zealand operator should calculate their own with the method above and benchmark against their own history. Our New Zealand hospitality wages page is the nearest thing to a local cost benchmark that does exist.
This percentage measures what you spend. The other half of it is what you sell, and the cheapest sale in a venue is the booking that was already trying to reach you.
Skai Solutions answers the phone on the first ring, during service and after close, takes the booking and sends it through. The connection to this page is direct: a beverage programme's percentage improves fastest through mix, and mix improves when the room is full enough for the floor to sell, which starts with the Friday enquiry nobody could pick up at 7pm. Setup takes five days.
If your phone mostly carries suppliers and deliveries, a basic message taking plan is the right buy and it is not us. If it carries bookings during service, read what an AI receptionist for a restaurant actually does.
Beverage cost is a venue level share. The item level version of the same question, for the drink most cafes sell more of than anything else, is on what a cup of coffee costs to make, with the published Australian wholesale bean bands and the A$4.69 average cup to check them against.
What is a good beverage cost percentage in Australia?
There is no official Australian figure. The ATO small business benchmarks, the only official data of this kind, publish a single cost of sales line that bundles food, beverage and all other stock together, with no separate beverage row for any venue type. Restaurants report cost of sales at 31% to 39% depending on turnover band and pubs, taverns and bars report 31% to 46%.
How do you calculate beverage cost percentage?
Opening beverage stock at cost, plus purchases at cost, minus closing stock at cost, divided by beverage sales excluding GST, multiplied by 100. Both stocktakes must be real counts taken on the same day of the week, and staff drinks, spoilage and promotional pours belong in the cost.
Why is the usual 20% pour cost target not useful in Australia?
Because it comes from a market with a different cost base. Alcohol excise applied per litre of alcohol, freight across a country this size, and merchant fees charged on the drink as well as the meal all sit on an Australian beverage invoice and none of them appear in a United States pour cost benchmark.
Do pubs have a better beverage cost percentage than restaurants?
The Australian data does not show that. In the top turnover band the ATO reports pubs, taverns and bars at 31% to 42% of turnover for cost of sales, against 31% to 36% for restaurants, so the pub range is wider with a ceiling six points higher. Mix, rent and other revenue differ between the two, so this is not evidence that beverage runs worse than food, only that the official data cannot separate them.
Is there a New Zealand beverage cost benchmark?
No. There is no Inland Revenue equivalent of the ATO benchmarks at this level of detail. The Restaurant Association of New Zealand's 2025 Hospitality Report puts wage costs at about 40% of hospitality revenue, which is a different measure across a wider sector and not a beverage comparison.
Does being outside the ATO range mean I am doing something wrong?
No. The ATO says a business outside a benchmark range is not automatically wrong, only that there may be room to improve. The figures come from lodged tax returns rather than audited accounts, and the data year is 2023-24.