Menu Engineering for Restaurants: Find Your Stars and Fix the Duds

Most restaurant menus grow by accident. A dish gets added because a guest asked for it, another because the chef liked it, and a few more for a festival that ended months ago. A year later you have 80 items, the kitchen needs stock for all of them, and you are not sure which dishes actually make you money.

Menu engineering is a simple way to fix this. You look at every dish on two measures — how much profit it makes per plate and how often it sells — and then decide what to push, what to change and what to drop. You do not need a consultant. You need your sales report, your food cost per dish and an honest hour of thinking.

Step 1: Collect two numbers for every dish

For each item on your menu, you need:

• Number sold in a fixed period, for example the last 30 days. Take this from your billing software's item-wise sales report.
• Contribution margin per plate: selling price (before tax) minus the food cost of that plate.

If you do not yet know the food cost of each dish, start there. Our guide on how to calculate food cost shows the simple method using a recipe card.

Here is an example with made-up numbers, only to show the method:

• Paneer Tikka: price ₹320, food cost ₹110, margin ₹210, sold 240 plates
• Dal Makhani: price ₹260, food cost ₹60, margin ₹200, sold 310 plates
• Mutton Rogan Josh: price ₹420, food cost ₹230, margin ₹190, sold 90 plates
• Veg Manchurian: price ₹240, food cost ₹70, margin ₹170, sold 40 plates

Step 2: Find your averages

Work out the average margin per plate across the menu section (starters, mains, drinks and so on) and the average number sold per item. Do this section by section — comparing a cold drink with a main course does not tell you much.

Every dish is then either above or below average on profit, and above or below average on popularity. That gives you four groups.

Step 3: Sort every dish into four groups

Stars — high profit, high popularity

These are your best dishes. Guests love them and they make good money. Keep the recipe and portion consistent, keep them in the best spot on the menu, and do not discount them without a reason.

Workhorses — low profit, high popularity

Guests order these a lot, but each plate earns less. Do not remove them; people come for them. Instead, look at the cost side: can the portion be standardised, can a garnish be simplified, can you buy a key ingredient better? A small, careful price increase may also be fine because demand is already strong.

Puzzles — high profit, low popularity

These make good money per plate but few people order them. Often the problem is visibility or description. Move them higher on the menu, give them a clearer name, add a short line about what makes them special, or ask staff to recommend them.

Duds — low profit, low popularity

These take up menu space, kitchen time and stock, and give little back. Try fixing one or two (new recipe, new price) but be willing to remove the rest. A shorter menu usually means less wastage and faster service.

Step 4: Use menu layout to guide choices

How a menu looks changes what people order. A few simple ideas:

• Put stars and puzzles at the top of each section and in boxes or highlighted areas.
• Avoid long price columns with rupee signs lined up — guests start comparing prices instead of dishes.
• Use short, honest descriptions. "Slow-cooked overnight" says more than "delicious".
• Keep each section to a manageable number of items so guests do not feel lost.
• On a QR menu, use photos for your stars and puzzles; they often help a dish get noticed.

Step 5: Train staff to recommend the right dishes

Your servers influence a large share of orders. Tell them which two or three dishes to suggest each week — usually puzzles you want to grow and stars you want to protect. Let them taste these dishes so they can describe them honestly.

Step 6: Repeat every month or season

Menu engineering is not a one-time exercise. Ingredient prices change, seasons change, and guest taste changes. Run the same analysis every month, or at least every season, and compare. You will start to see which changes worked.

Keep a simple record: what you changed, when, and what happened to sales and margin afterwards. Over time this becomes your own playbook.

Common mistakes to avoid

• Using food cost percentage alone. A dish with 40% food cost can still earn more rupees per plate than one with 25%. Look at rupee margin too.
• Removing a workhorse too quickly. If it brings guests in, it may support the rest of the bill.
• Ignoring recipe consistency. If portions vary from cook to cook, your margin numbers are guesses.
• Changing too many things at once. Then you cannot tell what worked.

How DUOZENT helps

DUOZENT's POS billing records every item sold, so item-wise sales reports are ready whenever you need them. With recipe-based inventory, each dish has a recipe linked to stock, so you can see the food cost behind every plate. AI insights highlight items that are selling well or slowing down, and the QR table ordering menu makes it easy to move dishes, add photos and update descriptions without reprinting. If you have old sales data in a spreadsheet, the sales-sheet upload brings it in so you can start from real history.

Want to see which of your dishes are stars and which are duds? Book a free demo — 14-day free trial + free leakage report.