Short answer
Menu engineering sorts each dish by contribution margin (price minus plate cost) and by popularity (its share of items sold). High-margin, popular dishes are Stars; popular but low-margin dishes are Plowhorses; high-margin dishes that rarely sell are Puzzles; and dishes low on both are Dogs. Each group gets a different fix.
Key takeaways
- Use contribution margin in dollars, not food cost percentage, to judge what a dish earns.
- A dish counts as popular when its share of sales reaches 70% of an even share, the rule from the original method.
- Keep Stars visible and consistent, nudge Plowhorse prices or plate costs, reposition Puzzles, and rework or drop Dogs.
- Run the analysis on one menu category at a time with at least a month of sales.
- Change a few things, then measure again before changing more.
Most menus grow by addition. A special sells well and stays. A supplier offers a deal and a new dish appears. Nobody removes anything, and five years later the menu has 40 items, a crowded kitchen and no clear idea which dishes pay the rent. Menu engineering is a simple way to see that clearly. The method was set out by Michael Kasavana and Donald Smith in 1982, and it still works with nothing more than a spreadsheet.
What do you need before you start?
You need each dish's menu price, its plate cost and how many sold over the same period, usually a month or a quarter.
Pull the item sales report from your POS. Use a period without big one-off events, and at least four weeks of normal trading. For plate cost, use a current costed recipe, not a guess. If you haven't costed your dishes recently, the food cost calculator walks through yields and a Q factor for the small stuff.
Work on one category at a time: starters, mains, desserts. Comparing a dessert with a steak tells you nothing useful.
How do you calculate contribution margin?
Contribution margin is the menu price minus the plate cost, in dollars, for one portion.
| Dish | Price | Plate cost | Margin | Sold |
|---|---|---|---|---|
| Burger | $16 | $5.20 | $10.80 | 420 |
| Salmon | $26 | $9.90 | $16.10 | 180 |
| Chicken pasta | $19 | $4.40 | $14.60 | 310 |
| Lamb shank | $29 | $12.50 | $16.50 | 60 |
| Veggie curry | $17 | $4.10 | $12.90 | 90 |
These figures are an illustration, not benchmarks. Then work out the category's average margin, weighted by sales: multiply each dish's margin by the number sold, add the results, and divide by total items sold. In this example that is $14,111 divided by 1,060, or about $13.31. Dishes above it are high margin. Dishes below it are low margin.
How do you decide which dishes are popular?
A dish is popular when its share of the category's sales is at least 70% of an equal share.
With five mains, an equal share would be 20%. Seventy percent of that is 14%. Any main with 14% or more of the 1,060 mains sold, about 148 portions, counts as popular. That 70% rule is part of the original method. It keeps a dish from being labeled unpopular just because one bestseller dominates.
In the example, the burger (40%), the pasta (29%) and the salmon (17%) are popular. The lamb (6%) and the curry (8%) are not.
What are the four menu engineering quadrants?
Every dish falls into one of four groups: Stars, Plowhorses, Puzzles and Dogs.
| High margin | Low margin | |
|---|---|---|
| Popular | Star | Plowhorse |
| Not popular | Puzzle | Dog |
From the example: the salmon and the pasta are Stars, the burger is a Plowhorse, the lamb shank is a Puzzle and the veggie curry is a Dog.
What should you do with each group?
Protect Stars, make Plowhorses earn more, help Puzzles sell, and fix or remove Dogs.
- Stars: keep the recipe, portion and presentation consistent. Give them the best position on the menu and make sure servers can describe them well. Don't discount them.
- Plowhorses: guests love them, so change them gently. Raise the price a little, trim the plate cost with a cheaper side or a smaller garnish, or pair them with a high-margin add-on. A burger with a suggested side upgrade often earns more without anyone noticing a change.
- Puzzles: they earn well but don't sell. Move them to a better spot on the menu, rename them, rewrite the description, feature them as a special or have servers taste them. If nothing works after a few months, consider a lower price.
- Dogs: ask why the dish is there. If it serves a real purpose, make it cheaper to produce. If not, take it off and free up prep time and inventory.
How do you know if your changes worked?
Run the same analysis again after four to eight weeks and compare the average margin and the total margin for the category.
Change only a few things at once, or you won't know what caused the result. Track two numbers: the category's weighted average margin and its total margin in dollars. If the total margin rises while guest counts hold steady, the changes are working. Keep an eye on the bigger picture too. Menu changes affect labor and prep time, which the prime cost calculator brings into the same view.
Common mistakes
- Using stale plate costs. One supplier price rise can move a Star into Plowhorse territory.
- Mixing categories, so desserts look like Dogs next to mains.
- Removing a dish that regulars come in for without asking anyone on the floor first.
- Making ten changes at once.
Menu engineering doesn't replace judgment about your guests and your kitchen. It shows you where to look, which is usually the hardest part.
Frequently asked questions
How often should a restaurant do menu engineering?
Once a quarter works for most independents, plus any time you change suppliers or prices sharply. Seasonal menus can be analyzed at the end of each season.
Why use contribution margin instead of food cost percentage?
Because you deposit dollars, not percentages. A steak at 38% food cost can leave far more money per plate than a pasta at 22%. Food cost percentage still matters for control, but margin decides what to promote.
Do I need software for menu engineering?
No. A spreadsheet with each dish's price, plate cost and number sold from your POS is enough. Software saves time on large menus but the math is the same.
Should I always remove Dogs?
Not always. Some low-margin, low-selling items serve a purpose, such as the one vegetarian main or a kids' dish that brings in families. Keep those on purpose and make them cheaper to produce.