Menu engineering: four boxes that decide your profit

A menu is not a price list. It is the strongest sales tool in your restaurant, and the only one every customer reads before paying.

Menu engineering is a structured way of answering one question: which items have earned their place?

The two axes

Every item has two numbers:

1. Popularity: units sold over a period (say a month) relative to the rest of the menu.
2. Contribution margin: sell price − plate cost, in money, not percent.

The second point is where most people go wrong. An item at 25% cost selling for 40 gives you 30. An item at 45% cost selling for 200 gives you 110. The second is better for your restaurant, despite the “worse” percentage.

The four boxes

⭐ Star — protect it

Sells well and earns well. Do not touch it: do not change the recipe, do not cut the quality, do not move it into a neglected corner of the menu. Give it the clearest position and make sure its ingredients never run out.

You can raise its price, but carefully and in small steps — this is the item whose price customers remember.

❓ Puzzle — promote it

Great margin, few takers. The problem is presentation, not the dish:

• Move it to the top of its section, or into a highlight box.
• Rename it with a description that explains the value.
• Train staff to suggest it by name.
• Give it one good photograph.

Try that for two months. If it still does not move, it was never a puzzle — it is a dog in a nice suit.

🐴 Plowhorse — fix the economics

Everyone orders it and it earns nothing. Four options, easiest first:

1. Reduce the cost (supplier, specification, better yield).
2. Control the portion with a fixed measuring tool.
3. Raise the price in small steps and watch demand.
4. Attach profitable add-ons (a drink, a sauce, a side).

Plowhorses are dangerous because they give you the feeling of being busy without the profit.

🐕 Dog — remove it

It neither sells nor earns, and it still costs you: menu space, ingredients that expire, one more thing to train.

Before deleting, ask whether it brings in a specific customer (vegetarian, kids, dietary). If yes, keep it deliberately. If no, removing it saves more than you expect.

The layout itself sells

Once items are sorted, arrange the menu to match:

• Seven items per section, maximum. Longer lists push people to the familiar choice — usually not the profitable one.
• Drop the currency symbol. “120” reads lighter than “EGP 120”.
• Avoid a straight column of prices. Aligned prices invite comparison and the cheapest wins. Put the price right after the description.
• Descriptions sell. Ingredient, method, one distinguishing detail. A single sentence is enough.
• Only one highlight. Boxing one item per section works; boxing five cancels the effect.

Where the numbers come from

This analysis needs two datasets together: how much of each item sold, and what each item cost. Restaurants that keep sales in the POS and costs in a separate file spend days reconciling by hand, and postpone the analysis until everyone forgets it.

When both live in one place — as in GoSufra, where recipes connect to inventory and to menu performance reporting — the classification becomes a report you open rather than a project you start.

A quarterly cycle

1. Pull 90 days of sales and the cost of each item.
2. Calculate cash contribution margin per item.
3. Place every item in its box.
4. Take one action per box.
5. Review after 60 days.

Most restaurants discover in the first pass that 20% of items produce 80% of profit — and that a third of the menu could disappear without a single question.