How Do I Compare Venues Fairly Across a Restaurant Group?

A steak-heavy bistro running 30% food cost can be better run than a café at 26%, once each is checked against what its own recipes say it should cost. Fair multi-venue restaurant reporting judges every site on the gap to its own target, uses one short list of numbers cut off on the same day each week, and saves direct comparison for the same dish, where like really is like.

The usual version looks different. Someone builds a Monday spreadsheet from four POS exports and three stocktake sheets. The CBD bistro shows 31% food cost, the beachside café shows 27%, and head office asks the bistro’s head chef what went wrong. The chef points out that the bistro sells steak and the café sells toast, everyone gets defensive, and nothing changes. The numbers were right. The comparison was wrong.

Which numbers should you compare across venues?

Compare a small set of numbers that mean the same thing at every site. Five is enough for a weekly group view:

• Food cost % against that venue’s target. The gap to target is comparable; the raw percentage often isn’t.
• GP dollars. A venue at 33% food cost that does twice the covers can put more money in the bank than a venue at 27%. Percentages don’t pay rent.
• Variance: theoretical vs actual food cost. Theoretical is what the recipes say the week should have cost, given what the POS sold. Actual is what the stocktake says it did cost. The gap is waste, over-portioning, price rises, theft and count error, and it is the most comparable number you have. (For what usually sits inside that gap, see why actual food cost runs higher than your recipes say.)
• Waste, logged in dollars. Weight is fine for the kitchen, but head office needs dollars to rank problems.
• Labour as a share of sales. For most venues this is the biggest line. Restaurant & Catering Australia’s 2024 benchmarking found wages were nearly 28% of turnover, and that almost 30% of businesses made a loss in 2023-24. A group that watches food cost and ignores labour is watching the smaller leak.

Keep the definitions identical. If one venue counts staff meals as waste and another doesn’t, or one stocktakes Sunday night and another Monday morning, the comparison is broken before it starts. Write the rules down once: count day and time, what counts as waste, how staff meals and comps are recorded, which revenue is in (functions, delivery, retail) and which is out.

Why does raw food cost % mislead?

Raw food cost % misleads because it mostly measures menu mix, not how well a kitchen is run. A venue that sells a lot of protein, seafood or premium wine by the glass will run a higher percentage than one selling coffee and smashed avo, however tight the kitchen is. The ATO’s small business benchmarks show how much "normal" varies by format. In 2023-24 tax data, restaurants averaged cost of sales of 34% to 35% of turnover, while pubs, taverns and bars averaged 37% to 40%, and the smallest pubs ranged anywhere from 34% to 46%. Those figures cover all cost of sales, food and drink together, but the point stands: different formats have different normal.

Other things skew the raw number:

• Functions. A Saturday wedding at a set price per head can push a week’s food cost up or down by a few points.
• Size. A small venue spreads fixed waste (the end of a tray of herbs, the last of a sauce batch) over fewer sales.
• Stocktake timing. Count before a big delivery and closing stock looks thin; count after and it looks fat.
• Price points. Two venues on the same recipe but different menu prices will show different percentages for the same plate.

The fix is to compare each venue’s variance and its gap to its own target. A worked example, with hypothetical numbers:

Say the bistro’s recipes and sales mix say food cost should be 29% (theoretical) and the stocktake says it was 30% (actual). The café’s theoretical is 23% and its actual is 26%.

On raw percentages the café looks better. On variance, the bistro is 1 point out and the café is 3 points out, so the café is the one losing money it shouldn’t.

In Excel, with opening stock in B2, purchases in B3, closing stock in B4, food sales in B5 and theoretical food cost % in C5:

Multiply the variance by food sales to get dollars: on $40,000 of weekly food sales, 3 points is about $1,200 a week. Dollars are what head office should rank by.

How do you compare the same dish across venues?

Compare the same dish by plate cost, not percentage: if two venues make the same burger from the same master recipe, their plate cost should be within a few cents. When it isn’t, there are only four places to look:

1. Price paid. Is one venue buying the brioche or the beef from a different supplier, or on an older price? In a group it’s common for one venue to still be paying last quarter’s rate because nobody compared invoices across sites. Pull the latest invoice for the item from each venue and line up the unit prices; our guide to catching supplier price rises shows how to do it every week rather than once.
2. Yield. One kitchen trims a brisket harder, or buys whole and portions itself while the other buys portioned.
3. Portioning. 180 g of chips in one venue, 230 g in the other. Weigh a few plates on a busy night.
4. The recipe itself. Local chefs adjust. Sometimes that’s an improvement worth rolling out to the group; sometimes it’s drift.

Start with your top ten sellers that appear at more than one venue. Line up plate cost, menu price and GP dollars per plate side by side. The dish with the biggest spread in plate cost is your first conversation, and it’s a much easier conversation than "why is your food cost high?". If you need a refresher on building the plate cost in the first place, Open Pantry’s guide on how to calculate food cost percentage walks through it.

What should head office see each week, and what should venue managers see?

Head office should see one page per week with every venue on it, ranked by dollars at risk; venue managers should see their own venue in detail, with the reasons. The two views answer different questions.

Head office view (weekly, same cut-off for every venue):

• GP dollars and food cost % against target, per venue
• Variance in points and dollars, per venue
• Labour as a share of sales, per venue
• The three biggest dollar problems across the group, with the venue named
• Same-dish plate cost spread for the top sellers

Venue manager and head chef view (daily and weekly):

• Which dishes are over target and why (price, waste, yield, count)
• Waste by item, deliveries short, credits outstanding
• Price rises on this week’s invoices

When one venue’s number does move, split it into price, waste, yield and count before anyone is called in; why GP drops in a given week walks through the order to check.

Pick a cadence and hold it. Weekly is right for most groups: monthly is too late to fix anything, and daily food cost % across venues is mostly noise from deliveries and count timing. Hold the meeting on the same day each week, review the ranked list, agree one action per venue, and check last week’s actions first.

Reporting time is worth protecting too. In Lightspeed’s 2026 APAC survey, almost three in five venues said AI-powered tools save them more than an hour a day, with reporting (57%) among the most cited uses. Treat that as vendor-reported, but if your area manager spends Monday morning stitching spreadsheets, that time is better spent in the venues.

How Restaurant AI Manager helps with multi-venue restaurant reporting

Restaurant AI Manager, Open Pantry’s AI restaurant operations software, reads each venue’s orders, invoices, stock and POS sales and shows where margin went. Open Pantry is an Australian ordering and operations platform for food and beverage suppliers and the restaurants they supply, founded in 2018. For groups, every venue’s GP, labour and orders sit side by side, and GP per dish and per venue is live from invoices and POS sales, with the same dish compared across venues. When GP slips, it’s split into price rises, waste, yield and count error, in dollars that add up. Master recipes, shared lists and roles keep definitions consistent, and chefs see food cost, not wages. See the multi-venue management and head office pages, or open the free live demo with no sign-up and pick the group role to see a bistro, a café and a production kitchen side by side.

In summary: fair multi-venue restaurant reporting

Fair multi-venue restaurant reporting compares each venue to its own target, ranks problems in dollars, and uses variance and same-dish plate cost rather than raw food cost %. Set the definitions once, cut off on the same day, review weekly, and send one action back to each venue.