Menu engineering is a methodology developed in the 1980s by two professors at Michigan State University. The core idea: a menu is not just a list of dishes - it's a sales tool, and like any sales tool, it can be designed well or badly.
Restaurants that apply menu engineering principles consistently see higher average check sizes and better margins - it's one of the most effective ways to increase your average check, without changing a single dish. Here's how it works.
The foundation: know your margin and popularity for every dish
Before you can engineer your menu, you need two numbers for every dish:
Contribution margin: Selling price minus food cost. This is the actual profit the dish generates per order - not the percentage, but the dollar amount. A $25 dish with a $8 food cost generates $17. A $12 dish with a $3 food cost generates $9. The $25 dish contributes more even if the percentages look similar.
Popularity: How often the dish is ordered relative to other dishes in the same category. Usually measured as a percentage of orders within its category.
Once you have both numbers for every dish, you can classify each dish into one of four categories.
The four categories
Stars: High contribution margin, high popularity. Your best dishes - guests love them and they make you good money. Protect these. Don't remove them to be creative. Feature them prominently. Consider adding photos.
Plowhorses: High popularity, lower contribution margin. Guests choose these often but they're not your most profitable. Options: raise the price slightly (test whether popularity drops), reduce the food cost (different presentation or smaller portion), or feature the Stars nearby to pull some orders in their direction.
Puzzles: High contribution margin, low popularity. These would make you good money if more people ordered them - but guests aren't choosing them. Usually a presentation or placement problem, not a quality one. Experiment with better descriptions, more appealing names, photos, or positioning on the menu.
Dogs: Low margin, low popularity. These are candidates for removal. They take up space, they require ingredients your kitchen needs to stock, and they don't contribute meaningfully to revenue. Unless there's a strong reason to keep them (a signature dish with heritage value, an item that a loyal regular segment depends on), cut them.
Where to place dishes on the menu
Eyes move in predictable patterns when scanning a menu - a phenomenon rooted in menu psychology. Research consistently shows a few high-attention zones:
Top-right of each category section. When a guest looks at a section (Starters, Mains, etc.), their eye tends to land in the upper-right area first. This is prime real estate. Put your Stars and Puzzles (high margin items you want more orders of) here.
Top of the list. The first item in any list gets disproportionate attention. Don't bury a high-margin dish in position 7 of 10.
Avoid last position. The last item in a section gets less attention than the first. If a dish is performing poorly, consider whether its position is part of the reason.
Following restaurant menu categories best practices, these principles still apply to the order items appear when guests scroll on a digital menu. The first dish in a category gets seen first. Put your best performers there.
Descriptions that sell
The language around a dish affects whether guests choose it. Research by Cornell University found that descriptive menu labels ("pan-seared Cornish salmon with lemon beurre blanc") increased sales by 27% compared to plain labels ("salmon") and led guests to rate the food as better tasting - even when the dish was identical.
What makes a description effective on a menu:
Sensory language: Words that evoke texture, temperature, or taste. "Crispy," "slow-roasted," "hand-rolled," "wood-fired."
Origin references: Ingredients with specific origins feel premium. "Scottish beef," "Cornish crab," "local goat's cheese" all outperform generic equivalents.
Method references: How something is cooked is interesting and communicates care. "24-hour braised," "stone-baked," "cured in-house."
Accuracy: Don't describe a dish as "delicate" if it's hearty, or "spicy" if it isn't. Guests remember when the food doesn't match the description.
Keep descriptions concise - one sentence of genuine information outperforms two sentences of marketing language.
Price presentation
How prices look on a menu affects how much guests spend. A few well-tested principles:
Remove currency symbols. Understanding how to price restaurant menu items goes beyond the numbers themselves. "$18" anchors guests to the concept of spending money. "18" is just a number. Studies show removing currency symbols increases average spend. Worth testing.
Don't right-align prices. A column of right-aligned prices makes it easy for guests to scan just the prices and choose the cheapest option. Embed prices at the end of the description instead: "Pan-roasted chicken, seasonal vegetables, herb jus - 18."
Avoid price anchoring traps. Placing an expensive item at the top of a section makes everything else look cheap by comparison - guests may trade up. Placing the cheapest item at the top pulls them down. Be intentional about which direction you want guests to move.
Applying this to a digital menu
Menu engineering was developed for printed menus, but the principles apply directly to digital ones - with some advantages:
You can test and iterate. On a printed menu, a change means a reprint. On a digital menu, you can reorder dishes, rewrite descriptions, or add photos in minutes. Run a dish in position 1 for two weeks, then position 3. See if order frequency changes. This kind of testing is impossible on print, easy on digital.
Analytics show you what's working. Some digital menu platforms show view counts per dish. Combined with your actual sales data, this tells you whether a dish that isn't selling is simply not being seen (placement problem) or is being seen and not chosen (description or price problem).
Photos are easy to add. A photo next to a dish consistently increases its order rate. On a printed menu, adding photos to a dish is a design and print project. On a digital menu, you upload an image in 30 seconds.
Where to start
If you haven't done a menu engineering analysis before, start with this:
- Export 90 days of sales data from your POS
- Calculate the food cost and contribution margin for your top 20 dishes
- Classify each as Star, Plowhorse, Puzzle, or Dog
- Make three changes: promote one Puzzle to a better position, rewrite the description of one underperforming dish, and seriously consider removing your worst Dog
- Wait 30 days and check whether order patterns have shifted
You don't need to overhaul everything at once. Menu engineering is an ongoing practice, not a one-time project. Small, informed changes compound over time into significantly better margins.
Ready to create your digital menu? Try ArrivaMenu free - no credit card, no time limit.
Create your QR menu today
Free during Early Access. No credit card required.