Menu Engineering: How to Design a More Profitable Restaurant Menu

Menu engineering is the practice of analyzing your menu items by profitability and popularity, then designing the menu to sell more of the right things. Developed in the 1980s at Cornell, it remains one of the highest-leverage tools available to restaurant operators -- and one of the most underused.

The core insight: not all menu items are equal. Some are popular and profitable. Some are popular but unprofitable. Some are neither. And some are both profitable and ignored. Menu engineering tells you which is which.

The Four Menu Engineering Categories

Every item falls into one of four quadrants based on margin (profit per plate) and popularity (how often it is ordered):

Stars -- High Margin, High Popularity

Your best performers. They sell well and make money. Keep them consistent, make them prominent, and do not change them without a good reason.

Plowhorses -- Low Margin, High Popularity

Sell a lot but do not make much per plate. Customers love them, but they drag your food cost up. Options: raise the price slightly, reduce portion or ingredient cost, or feature them in combos where other items carry more margin.

Puzzles -- High Margin, Low Popularity

Would be great for your business if people ordered them -- they just do not. The problem is usually visibility or perception. Try better menu placement or a more compelling description.

Dogs -- Low Margin, Low Popularity

Do not sell and do not make money. They add kitchen complexity, tie up inventory, and clutter your menu. Remove them or completely reinvent them.

How to Run a Menu Engineering Analysis

You need two things: plate cost for every dish, and sales mix data for a defined period (last 30-90 days).

Step 1: Calculate contribution margin per item Contribution margin = Menu price minus plate cost. A dish at $18 with a $5.40 plate cost has a contribution margin of $12.60.

Step 2: Calculate popularity index Popularity index = Times sold divided by total covers x 100. An item is popular if its index is above the menu average.

Step 3: Plot the quadrant Put contribution margin on one axis, popularity on the other. Draw lines at the averages. Every item lands in one quadrant.

Real example: A weekend brunch menu with 18 items. After analysis, 5 items turn out to be Plowhorses, accounting for 42% of orders but averaging 38% food cost. The operator raises prices on two by $2 each and reformulates a third to reduce plate cost by $1.10. Food cost on those items drops 3 points over the next quarter.

Menu Layout and Descriptions

Physical placement directly affects sales. Research shows the upper-right quadrant of a menu gets the most attention.

Use placement strategically:

• Move Puzzles to prominent positions before removing them
• Put Stars where they are easy to find
• Do not bury high-margin items in the middle of a long list

Descriptions matter. "Braised short rib with rosemary jus and truffle mashed potato" outsells "beef with potato" even if it is the same dish. Sensory and origin language consistently increases perceived value.

Upsell Architecture

Design natural upsells that add margin -- a premium add-on, a protein upgrade, a specialty sauce option. Beverage pairings and dessert prompts built into the menu are also menu engineering tools.

How Often to Re-Engineer Your Menu

A full menu analysis once or twice a year is the baseline. Track sales mix data continuously so you can spot shifts quickly.

Menu engineering is not a one-time project. It is an ongoing operational discipline.

Frequently Asked Questions

How long does a menu engineering analysis take?

With accurate plate cost data and a sales mix export from your POS, a full analysis takes 2-4 hours the first time and under an hour once your systems are in place.

What if I do not have plate costs for every dish?

Start with your top 10 items by volume. Those account for most of your food spend and will give you the most actionable data quickly.

How much can menu engineering improve profitability?

Operators who run menu engineering analyses and act on them typically see 1-3 percentage point improvements in food cost within one to two menu cycles. On a $1M restaurant, that is $10,000-$30,000 in recovered margin.

What is the best way to improve a Puzzle item?

Try moving it to a more prominent menu position first. If that does not help, rewrite the description with more sensory language. If it is still not moving after one menu cycle, consider a modest price reduction or remove it.

---

Ready to take control of your food costs? Try CostLab free for 14 days