Menu Engineering: How to Maximise Your Restaurant's Profit Margins

Your menu probably features between 30 and 50 dishes. Some of them are making you a lot of money. Others are quietly losing you money — service after service. The problem: without a proper analytical method, you have no way of knowing which is which. That's exactly what menu engineering solves — a discipline born in the 1980s in the United States, and still one of the most powerful — and most underused — tools for maximising margins in an independent restaurant.

Menu engineering involves analysing each dish on your menu along two axes: its popularity (how often it's ordered) and its profitability (how much it actually earns you). By cross-referencing these two dimensions, you get a precise map of your offering — and actionable levers to improve your margins without touching your listed prices.

This isn't abstract theory, nor is it a method reserved for restaurant chains. It's an operational tool you can apply this very week with a spreadsheet, your recipe costings and your sales data.

The fundamentals of menu engineering: understanding the matrix

Where the method comes from

Menu engineering was formalised in 1982 by Michael Kasavana and Donald Smith, two researchers at Michigan State University. Their idea was simple: every dish on a menu can be classified according to its contribution to gross margin and its relative popularity compared to other dishes in the same category.

This approach gave rise to the menu's BCG matrix, a four-quadrant classification tool that every restaurateur should know.

The two axes of analysis

Axis 1 — Popularity (menu mix)

This is each dish's share of total orders within a category. A dish is considered "popular" if it exceeds a threshold calculated as follows:

• Take the number of dishes in the category (for example, 8 starters)
• Calculate the theoretical average share: 100% / 8 = 12.5%
• Apply a 70% coefficient: 12.5% × 0.70 = 8.75%
• Any dish ordered more than 8.75% of the time is considered popular

The 70% coefficient (proposed by Kasavana and Smith) avoids being overly strict: it acknowledges that a perfectly even distribution never exists in practice.

Axis 2 — Unit gross margin

This is the difference between the net selling price and the food cost of the dish. A dish is considered "profitable" if its unit gross margin is higher than the weighted average gross margin of the category.

Concrete example: if your "mains" category includes 6 items and the weighted average gross margin (accounting for volumes sold) is £7.50 / $8.50, any dish generating more than that in unit margin is classified as "high profitability."

The four categories of the matrix

By cross-referencing these two axes, each dish falls into one of four categories:

• Stars (high popularity + high margin): your gems. Popular AND profitable dishes. Don't touch them — protect them, showcase them, and make sure quality remains consistent.
• Puzzles (low popularity + high margin): profitable dishes that are rarely ordered. The challenge is making them more visible or appealing without sacrificing margin.
• Plowhorses (high popularity + low margin): dishes everyone orders but that don't earn enough. You need to increase their margin, either by adjusting the price or by optimising the food cost.
• Dogs (low popularity + low margin): neither popular nor profitable. Candidates for removal or a complete overhaul.

How to run your first menu engineering analysis

You don't need specialist software to get started. A spreadsheet will do. Here's the step-by-step method.

Step 1: Gather your data

For each category on your menu (starters, mains, desserts), you need:

• The name of each dish
• Its net selling price (excluding VAT)
• Its food cost (from your recipe costing sheet)
• The number of units sold over a representative period (ideally 4 to 8 weeks)

The reliability of this analysis depends entirely on the accuracy of your recipe costings. If your food costs are approximate, your results will be too. Take the time to calculate the precise food cost of each dish before you begin.

Step 2: Calculate the indicators

For each dish, calculate:

• Unit gross margin = Net selling price – Food cost
• Total gross margin = Unit gross margin × Number of units sold
• Menu mix = Units sold of the dish / Total units sold in the category

Then, for the entire category:

• Weighted average gross margin = Sum of total gross margins / Total units sold
• Popularity threshold = (100% / Number of dishes) × 0.70

Step 3: Classify each dish

Compare each dish against the two thresholds:

• Menu mix above the popularity threshold? → Popular
• Unit gross margin above the weighted average? → Profitable

Then place it in the corresponding quadrant: Star, Puzzle, Plowhorse or Dog.

Worked example: "Mains" category at a bistrot

Imagine a Parisian bistrot with 6 main courses, analysed over 6 weeks (roughly 1,200 covers for this category):

• Entrecôte-frites: 380 sold, food cost €6.20, net price €18.50, unit margin €12.30
• Seasonal risotto: 210 sold, food cost €3.10, net price €14.50, unit margin €11.40
• House burger: 290 sold, food cost €5.80, net price €14.00, unit margin €8.20
• Sea bass fillet: 120 sold, food cost €7.50, net price €21.00, unit margin €13.50
• Beef tartare: 150 sold, food cost €6.00, net price €16.00, unit margin €10.00
• Meal-sized salad: 50 sold, food cost €2.80, net price €12.00, unit margin €9.20

Calculations:

• Total units sold: 1,200
• Weighted average gross margin: (380×12.30 + 210×11.40 + 290×8.20 + 120×13.50 + 150×10.00 + 50×9.20) / 1,200 = €10.76
• Popularity threshold: (100% / 6) × 0.70 = 11.67%, i.e. roughly 140 units out of 1,200

Results:

• Entrecôte-frites → Star (popular: 31.7%; margin €12.30 > €10.76)
• Seasonal risotto → Star (popular: 17.5%; margin €11.40 > €10.76)
• House burger → Plowhorse (popular: 24.2%; but margin €8.20 < €10.76)
• Sea bass fillet → Puzzle (margin €13.50 > €10.76; but only 10% of sales)
• Beef tartare → Star (popular: 12.5%; margin €10.00 < €10.76) — actually, note: €10.00 < €10.76, so it's a Plowhorse
• Meal-sized salad → Dog (4.2% mix; margin €9.20 < €10.76)

This kind of result is typical: the majority of independent restaurant menus contain at least one or two dogs and several plowhorses that are silently dragging down overall profitability.

Action strategies by dish category

Classifying your dishes is useful. Acting on it is what actually changes your margins. Here are the proven strategies for each quadrant.

Stars: protect and capitalise

Your stars are the backbone of your profitability. The golden rule: don't change them on a whim.

• Keep the recipe exactly the same — consistency is what brings customers back for that specific dish
• Place them in the hot spots on your menu (more on this in the menu design section below)
• Ensure availability — a star that's sold out means margin lost on the spot
• Monitor food costs — if your supplier raises the price of a key ingredient, react quickly to protect the margin

Puzzles: stimulate demand

A puzzle is a profitable dish that customers aren't ordering enough. Before removing it, look for the reason it isn't selling:

• Is the name clear and appetising? "Free-range chicken with a rich wild mushroom jus" sells better than "Chicken in mushroom sauce"
• Is its placement on the menu visible? Move it into a high-attention zone
• Does your staff recommend it? Train your team to suggest it — a server who says "I'd recommend the sea bass, it's our best catch right now" can turn a puzzle into a star
• Is the price creating a barrier? If your sea bass fillet is the most expensive item on the menu, some diners will avoid it instinctively, even if they want it. Consider repositioning the price or offering it in a smaller portion

Plowhorses: improve the margin

These are the "trap" dishes: everyone orders them, and you feel like things are going well, but they're not earning enough. Several levers are available:

• Raise the selling price — gradually, in increments of £0.50 to £1. On a dish sold 290 times in 6 weeks, every pound of increase represents £290 in extra margin
• Reduce the food cost without degrading perceived quality — switch cuts of meat, adjust portion sizes (a 160g burger patty instead of 180g often goes unnoticed), negotiate better supplier prices, or replace a costly ingredient with a seasonal alternative
• Offer a paid side or upgrade — turn the house burger into "house burger + fries and salad" and raise the price by £2, which increases margin without changing the perception of the dish
• Reduce its prominence — don't place your plowhorses at the top of a category or in the premium visual zones of your menu

This quadrant is often where the quickest wins are found. One or two adjustments on a high-volume plowhorse can have a significant impact on your monthly bottom line. To dive deeper into this topic, see our complete guide to boosting your restaurant's profitability with practical levers.

Dogs: make the call

A dog doesn't sell and doesn't earn. Three options:

• Remove the dish — this is often the best decision. A shorter menu is easier to manage, reduces waste and simplifies production. If a dish accounts for only 4% of your sales, the vast majority of your customers won't even notice it's gone
• Overhaul it completely — new name, new presentation, perhaps a signature ingredient. This is a new dish, not an improvement
• Keep it only if it's an "image" dish — a vegan or gluten-free option may be necessary to avoid losing entire tables, even if it doesn't sell much. In that case, at the very least make sure its unit margin is healthy

Menu design: the invisible weapon of menu engineering

Menu engineering doesn't stop at number-crunching. The way your menu is designed directly influences what your customers order.

Hot spots on the menu

Eye-tracking research conducted in the restaurant industry has shown that diners don't scan a menu in a linear fashion. On a two-panel menu (the most common format), the areas that attract the most attention are:

• The upper centre of the right-hand page — this is the premium zone, where the eye lands first
• The first and last item in each category — the primacy and recency effect, well documented in cognitive psychology
• Visually highlighted items — a simple border, a subtly different background or an icon is all it takes

Place your stars and puzzles in these zones. Relegate your plowhorses to less visible positions.

The decoy effect (decoy pricing)

A classic menu engineering technique: include a deliberately expensive dish to make the other options look more attractive by comparison. If your rib-eye steak is £28 and your côte de boeuf to share is £50, the rib-eye suddenly seems very reasonable — even if it's your highest-margin dish.

A word of caution: the decoy must remain a genuine dish — available to order and well executed. The goal isn't to mislead the customer, but to use natural cognitive biases to steer choices towards dishes that earn you more.

The optimal number of dishes per category

A common mistake among independent restaurateurs: offering too much choice. The paradox of choice (studied by psychologist Barry Schwartz) shows that too many options cause hesitation, dissatisfaction and a retreat to "safe" choices — often your plowhorses.

The operational recommendation for maximising margin:

• Starters: 5 to 7 items
• Mains: 6 to 9 items
• Desserts: 4 to 6 items

Beyond that, every additional dish dilutes attention, complicates stock management and increases the risk of waste. If you're deciding between a paper menu and a digital menu, bear in mind that a digital menu makes testing and quick updates far easier — a real advantage for iterating on your menu after each analysis.

Writing dish descriptions

The name and description of a dish directly influence its popularity. A few principles from menu engineering:

• Use sensory adjectives: "melt-in-the-mouth," "crispy," "infused with fresh thyme" speak to the reader's senses
• Mention provenance when relevant: "Lake District lamb," "tomatoes from our local grower" — this justifies the price and builds perceived value
• Avoid overly long descriptions: 15 to 25 words maximum. Beyond that, diners stop reading
• Remove currency symbols where possible: research in hospitality (Cornell University) has suggested that dropping the £ or $ sign reduces price sensitivity. In practice, this works mainly in fine dining; in a bistrot, keep the symbol for clarity

Frequency and tracking: menu engineering as an ongoing process

Running a menu engineering analysis once a year is better than never. But it's not enough. A restaurant's menu is a living organism: food costs fluctuate, customer tastes evolve, and seasonality shifts volumes.

The right cadence

• Full analysis: with every menu change (ideally 3 to 4 times a year, following the seasons)
• Key indicator tracking: monthly — at a minimum, monitor the menu mix and the weighted average gross margin for each category
• Tactical adjustments: ongoing — if a dish drops in popularity or a food cost spikes, don't wait for the next menu overhaul

Indicators to track

Beyond the matrix, keep an eye on:

• Average spend per head: well-executed menu engineering should increase the average spend, not the number of covers
• Overall food cost ratio: it should stay within your target range (typically between 25% and 35%, depending on your positioning)
• Number of dishes per category: resist the temptation to add without removing
• Stars' share of revenue: if your stars account for a growing share of sales, your work is paying off

Training your team

Menu engineering only works if your front-of-house team is on board. Your servers are your number one sales tool:

• Brief them on which dishes to push — identify 2–3 "priority" dishes per service (your stars and puzzles)
• Give them the talking points — not a sales script, but genuine selling points: "the risotto is made with fresh wild mushrooms that came in this morning," "the rib-eye is our best-seller, it's been dry-aged for 30 days"
• Let them taste the dishes — a server who has tasted a dish recommends it naturally, with genuine conviction

Common menu engineering mistakes to avoid

Even with the right method, certain pitfalls keep coming up.

Mistake 1: Relying on food cost percentage rather than gross margin

Many restaurateurs think in terms of food cost percentage. "This dish runs at 28% food cost — that's fine." But menu engineering thinks in pounds and pence of margin, not percentages — and that's a fundamental difference.

A dish with a 35% food cost selling at £22 net generates £14.30 in margin. A dish with a 25% food cost selling at £10 net generates £7.50 in margin. The second has a better ratio, but the first earns you nearly twice as much per plate. This is one of the classic mistakes that undermine profitability in independent restaurants.

Mistake 2: Ignoring seasonality

An analysis carried out in January won't hold true in July. Ordering patterns change dramatically with the seasons: salads and fish surge in summer, stews and gratins dominate in winter. Recalculate your matrix every season.

Mistake 3: Removing a popular dish without preparation

If you need to take a plowhorse off the menu, don't do it overnight. First introduce an alternative (ideally a future star or puzzle), let it establish itself, then retire the old dish. A customer who can no longer find "their" regular dish may decide not to come back.

Mistake 4: Not keeping recipe costings up to date

Menu engineering relies on accurate food costs. If your recipe costings are six months old and supplier prices have shifted, your analysis will be skewed. Update your costings at least once a quarter — and whenever there's a significant change in supplier pricing.

Mistake 5: Analysing the menu as a whole rather than by category

Comparing the margin on a starter with that of a main course is meaningless: the price levels and customer expectations are entirely different. The analysis must always be carried out category by category: starters against starters, mains against mains, desserts against desserts.

Menu engineering and digital tools

Going digital makes menu engineering considerably easier. A QR code menu connected to your POS system can give you real-time browsing data: which dishes are viewed most, how long customers spend on each section, and which dishes are viewed but not ordered.

This browsing data, cross-referenced with your sales figures, sharpens your analysis. A dish that's frequently viewed but rarely ordered is likely a puzzle with a pricing or description problem. A dish that's rarely viewed AND rarely ordered is an invisible dog — it isn't even being seen.

Platforms like ALaCarte.direct let you update your digital menu in real time, making testing far more agile than with a paper menu: you can reposition a dish, tweak a description and measure the impact within days.

For restaurateurs looking to go further with their customer data, CRM for restaurants opens up exciting possibilities: understanding not just what sells, but who is buying what.

💡 Good to know: create your free QR code digital menu with ALaCarte.Direct — online in minutes, editable at any time.

Conclusion: take action this week

Menu engineering isn't a discipline reserved for large restaurant groups. It's a pragmatic tool, accessible to any independent restaurateur with a spreadsheet and their sales data. Here are the concrete steps to get started this week:

1. Update your recipe costings for the 10 best-selling dishes on your menu — precise food costs, down to the penny
2. Export your sales data from the last 4 to 6 weeks from your POS system
3. Build your matrix for a single category (start with mains — that's where the margin impact is greatest)
4. Identify your dogs and plowhorses — these are your two action priorities
5. Make a decision for each problem dish: removal, price repositioning, food cost optimisation or description overhaul
6. Brief your team on the 2–3 dishes to recommend as a priority at the next service

Menu engineering is an iterative process. Your first analysis won't be perfect, and that's fine. What matters is that you start looking at your menu through the eyes of a business manager, not just a chef. Each analyse-act-measure cycle brings you closer to a more profitable, more readable and — paradoxically — often shorter and more coherent menu.

Your menu is your top salesperson. It's time to give it what it needs to do its job.