How to price your café menu

Most café owners set their first prices by looking at the café down the road and knocking off a bit. That gets you open. It doesn't tell you whether each sale makes money, and it falls apart the first time milk or beans go up.

This guide covers what to aim for, how to cost a coffee and a dish properly, what GST does to your numbers, which bits of menu psychology are worth your time, and how to put prices up without your regulars walking.

#Food and coffee cost targets for Australian cafés

Your food cost (or beverage cost) is the cost of the ingredients and packaging in an item, as a share of its price before GST. A dish that costs $4.50 to make and sells for $15 before GST has a 30% food cost.

There is no official target. What exists is industry guidance, and the ATO's real-world figures from tax returns.

A few things are worth knowing about those numbers.

The ATO figures are what cafés actually achieve, not what they aim for. They come from tax returns, so they include all the waste, staff meals and over-portioning that recipe costings miss. If your recipes say 28% and your accounts say 38%, the gap is real money leaking somewhere, and it is usually cheaper to find than to price around.

The ATO splits coffee shops by turnover. For businesses turning over $65,000 to $250,000 the range is 34 to 42%; for $250,001 to $600,000 it is 35 to 41%; above $600,000 it is 33 to 38%. In the same data, labour runs 21 to 35% of turnover and rent 6 to 17%, with smaller cafés paying a bigger share in rent.

Cost and labour trade off. Loaded makes the point that cafés run higher food costs than restaurants but usually lower labour. A café that buys in its baking can carry a higher food cost because it isn't paying a pastry cook. A café that makes everything from scratch needs a lower food cost to leave room for the wages.

The menu is what has to hit the target, not every item. Banana bread might run at 12% and a big breakfast at 34%. That's fine if the mix lands where you need it.

#How to cost a coffee

Coffee is your highest-volume item, so a few cents a cup adds up quickly: 5c on 200 cups a day, six days a week, is over $3,000 a year. Cost each size separately, and cost takeaway and dine-in separately, because the cup and lid are a real line.

Here is a regular (12 oz) takeaway flat white, using numbers you can check against your own invoices:

Where those inputs come from: one wholesale roaster put most cafés at $32 to $38 a kilo in January 2025, ranging from $22 to $42 or more, and green bean prices hit records in early 2025, so $40 is a fair 2026 middle. Coles home-brand milk went to $3.55 for 2 litres in April 2026, about $1.78 a litre, and barista milks through a distributor usually cost more. For comparison, Seven Miles Coffee Roasters put the ingredients and packaging of a 12 oz takeaway flat white at about $1.55 in 2024.

Now the price. Say it's $5.90 on your board.

• Take the GST out: $5.90 divided by 1.1 is $5.36.
• Gross profit: $5.36 minus $1.79 is $3.57 a cup.
• Gross margin: $3.57 divided by $5.36 is 67%, so the coffee cost is 33%.

That is well above the 20 to 25% coffee cost suppliers suggest, and takeaway at 2026 prices makes it hard to hit. Dine-in at the same price drops the 26c cup and lid, and the cost share falls to about 29%. To get the takeaway to a 25% coffee cost, you'd need $1.79 divided by 0.25, which is $7.17 before GST, or $7.89 on the board. That's a hard price for a regular in most suburbs, so look at the cost side too: milk waste, how much milk goes in each size, and what you pay for cups.

Don't forget alternative milk. At around $4 a litre for barista oat milk against $2.20 for dairy, the 270 ml in a regular adds about 54c once waste is counted. A 50c surcharge is only 45c after GST, so it doesn't quite cover it. Australian cafés commonly charge 50c to $1; some build it into every price; some absorb it. Any of those works if you chose it knowing the numbers.

#How to cost a dish

The method is the same for every dish:

1. List every ingredient, including the garnish, the oil in the pan and the butter on the side.
2. Work out the cost per unit from the pack you buy: a $5.20, 650 g loaf is 0.8c a gram.
3. Multiply by what goes on the plate, and allow for trim: crusts, heels, stones, peel.
4. Add it up and divide by the serves the recipe makes.
5. Divide by the menu price before GST to get the food cost.

Here is smashed avo on sourdough, using rounded 2026 supermarket prices (your wholesale prices will differ):

On the menu at $18, that's $16.36 before GST, so the food cost is $4.69 divided by $16.36, or 28.7%, and each plate leaves $11.67 towards wages, rent and profit.

To price for a target, divide the cost by the target and add GST back: $4.69 divided by 0.30 is $15.64, plus GST is $17.20. Round up to a sensible café price, $17.50, so you land just under the target rather than just over.

The avocado is 43% of that plate's cost on its own. Avocado prices swing hard with the season, which is exactly why you re-cost when a big ingredient moves (more on that below).

Weigh, don't guess. The single most useful hour you can spend on pricing is in the kitchen with scales: weigh what actually goes on five plates of your three biggest sellers. Portions drift, and a cook who adds an extra 20 g of feta to every plate has quietly raised your food cost without anyone deciding to.

#GST and your menu prices

GST changes the maths in three ways.

What you sell is almost all taxable. The ATO lists all food and drink consumed on your premises, and all hot takeaway food and drink, as taxable. Cakes, pastries and biscuits are taxable wherever they're eaten. Some cold takeaway food can be GST-free, so check specific items with the ATO's food and beverage search tool.

Much of what you buy is GST-free. Bread, unflavoured milk, cream, cheese, eggs, fresh fruit and vegetables, cooking ingredients like flour and sugar, and coffee beans (as a preparation for making coffee, not ready to drink) are generally GST-free. Cups, lids, cleaning products and equipment carry GST, which a registered café claims back. So cost your food from the shelf or invoice price, and your packaging from the price before GST.

Your menu must show the full price. The ACCC requires businesses to show a single total price that includes GST and any other unavoidable charges. Menus have one exemption: a weekend or public holiday surcharge doesn't have to be built into each price, but the menu must say the surcharge applies, its percentage and on which days, at least as prominently as the most prominent price on the menu. A surcharge that applies every day has to be in the price.

The trap is working out margins on the GST-inclusive price. The smashed avo above looks like a 26.1% food cost if you divide $4.69 by $18. It's really 28.7%. Do that across a whole menu and you'll think you're doing better than you are. You must register for GST once your turnover reaches $75,000 a year, and a café selling 50 coffees a day at $5.50, six days a week, passes that on coffee alone.

Card fees are now in your price too. Since 1 October 2026, cafés can't add a surcharge for paying by eftpos, Visa, Mastercard or American Express, following the Reserve Bank's review. The cost of taking cards has to come out of your margin or go into your prices. Our post on the card surcharge ban works through what that means for a typical café.

#Setting the price: cost, market and position

Cost tells you the floor. It doesn't tell you the price. Three things set it together.

Your costs. Below your cost-based price, every sale is working against you. That's the number the calculators give you.

What the street charges. Customers judge your prices against the cafés they already know. Point-of-sale data from Square put the average Sydney flat white at $5.04 in 2025, up from $4.61 in 2023, and CommBank reported flat whites at around $6.50 in Sydney and Perth by early 2026, up about 30% since Covid. Seven Miles notes that CBD cafés often charge less because of volume and competition, while regional cafés can charge more. Walk your area, buy a coffee at each competitor and write their prices down.

Where you sit. Price tells customers what kind of place you are. A café with single origins, proper crockery and table service can charge more than a grab-and-go window, and probably should.

If your cost-based price is above what the street charges, you can trim the cost, accept a thinner margin on that item and make it up elsewhere, or make it noticeably better than the one down the road. What doesn't work is ignoring the gap.

#Pricing psychology that works in a café

A lot of menu psychology is overstated. These are the bits with evidence or plain arithmetic behind them.

Drop the dollar sign. A Cornell study at a restaurant run by the Culinary Institute of America found diners given menus with plain numbers ("18") spent more than those given prices with a dollar sign ("$18"). It was one restaurant and one study, but it costs nothing to try on a printed menu. Your board still has to show the full price including GST.

Make the step up look worth it. If a regular is $5.90 and a large is $6.40, the large feels like good value for 50c and some people will take it. Price sizes so the step up is small in dollars and larger in coffee. Check the large's margin in the coffee calculator first, because it uses more milk and a bigger cup.

Pair things to remove the arithmetic. A coffee and a slice, named and priced as one thing, is a single decision. We covered bundles, and the "ask" that moves average spend more than any menu change, in how to increase average spend in your café.

Pick a price style and stick with it. There's no strong evidence for cafés that $5.90 beats $6, or the reverse. Whole dollars and halves for food and 10c steps for coffee look tidy and are easy to change later.

Don't stack surcharges. Seven Miles makes the point well: a customer who ends up paying $12 for a large soy extra-shot latte, built from three separate add-ons, may not come back. If your add-ons are adding up, fold some into the base price.

#Menu engineering: stars, plowhorses, puzzles and dogs

Menu engineering was set out by Michigan State University researchers Michael Kasavana and Donald Smith in 1982. It sorts every item in a section of your menu by two questions: does it sell? and does it make good money per sale? Money here means gross profit in dollars (price before GST minus food cost), not the percentage.

Take one week of sales from your POS for your breakfast mains, and the gross profit of each from your costings. Here's an illustration:

The simple version of the test: five dishes sold 350 between them, an average of 70 each, so anything over 70 is popular. The average profit per dish sold was $11.05, so anything above that is profitable.

Stars sell well and make good money. Keep them exactly as they are, keep the quality up, and give them the best spot on the menu. Don't fiddle with the recipe to save 20c.

Plowhorses sell well but make less per plate. People like them, so don't drop them. Nudge the price up a little, look for a cheaper ingredient that eats the same, or trim a portion that's more generous than it needs to be. Ten per cent more profit on the big breakfast is 94c a plate, or about $89 a week at 95 plates, without selling a single extra meal.

Puzzles make good money but don't sell. Usually it's a visibility problem: move them up the menu, rename them, describe them better, or have staff recommend them. If they still don't move after a few weeks, they're taking up prep and fridge space.

Dogs sell poorly and make little. Take them off, unless there's a reason to keep them, like a vegan option that brings in a group of four.

The lesson that surprises people: percentage isn't profit. The shakshuka might have a higher food cost percentage than the avo and still earn more per plate. You bank dollars, not percentages. Run this once a quarter on each section of the menu, and before any menu reprint.

#Reviewing prices when costs move

Costs don't rise neatly, so pricing needs a habit, not a one-off.

Know which ingredients matter. A handful of ingredients usually drive most of a café's food and drink cost: milk, coffee, eggs, avocado, bread, bacon, cheese. Keep their current invoice prices somewhere you'll see them.

Re-cost when one moves. When Coles took home-brand milk from $3.20 to $3.55 for 2 litres in April 2026, that was an 11% rise. If your milk went up by a similar 18c a litre, a regular coffee with 270 ml and 10% waste costs about 5c more. Across 200 cups a day that's about $10 a day, or over $3,000 a year on six-day weeks, from one line on one invoice.

Decide what you're protecting. There are two ways to pass on a cost rise, and they give different answers.

• Keep the same dollars per cup. The cost went up 5c, so put the price up by 5c plus GST. Your gross profit per cup stays at $3.57, but the margin percentage slips a little.
• Keep the same margin percentage. At a 67% margin, a 5c cost rise needs 5c divided by 0.33, which is about 15c before GST. That keeps pace as everything else goes up too.

A sensible default: protect the dollars on high-volume items like coffee, so the board price moves as little as possible, and protect the percentage on food, where 50c on a $17.50 dish is easier to absorb.

Watch beans, but don't panic about them. Arabica futures reached a record of about US$4.20 a pound in February 2025 and had eased to about US$3.70 by January 2026, according to Rabobank. At a 20 g dose, every $1 a kilo your roaster adds is 2c a cup. Re-cost from your roaster's invoice, not the headlines: Padre Coffee puts raw beans at about two-thirds of the cost of a bag of roasted coffee, so a move in the futures price only partly reaches what you pay.

Review everything at least once a year. Seven Miles recommends a yearly review and points out that cafés which hold prices for years end up making one large, jarring jump, after years of shrinking margins.

#How to raise prices without losing regulars

The arithmetic first, because it takes the fear out of it. Fifty cents on a $5.90 coffee is 45c after GST. On 200 cups a day, six days a week, that's about $91 a day and $28,000 a year. With a cup making $3.57, you'd have to lose about 11% of those cups, 22 or 23 a day, before the rise left you worse off.

That doesn't mean you won't lose anyone. It means the bar for the rise to be a mistake is high. How you do it makes the difference.

Go small and often. Abdullah Ramay of Pablo & Rusty's recommends rises of 15 to 25 cents at a time, regularly, rather than one big jump. Seven Miles notes that $5.80 to $5.90 is less noticeable than $5.90 to $6.00, but at some point you have to cross the round number.

Tell people why, briefly. A small sign or a line on the menu ("Milk and coffee costs have gone up, so our prices have too. Thanks for sticking with us.") does more than silence. Brief your team so nobody is caught saying "I don't know, the boss did it".

Raise where it hurts least. Large sizes, extra shots and food items with low price awareness can carry more of a rise. The small flat white is often the price people remember and compare, so move it last and least.

Don't discount your way back. A permanent "early bird" discount to soften a rise just undoes the rise. If you want to reward people, reward the ones who keep coming.

Look after the regulars on purpose. A regular who feels known is more likely to shrug at 30 cents than a stranger is. Remembering names and orders does most of this; a loyalty card does the rest, because the reward is a small, predictable cost spread over regulars who are already coming. Back Again does it with a digital card customers tap at the counter, but paper cards work too, and our loyalty program calculator shows what any card really costs. For a wider plan, see your first 100 regulars.

Measure it. For two to four weeks after a rise, watch transaction numbers and average spend, not just takings. If takings are up and transactions are steady, it worked. If you want to see how a price change moves your break-even, try the break-even calculator.