If you’ve been in business a long time, you already know the hard part isn’t getting a menu (or product list) together. The hard part is keeping it profitable while costs keep creeping up and customers get pickier.

That’s exactly where menu engineering earns its keep.

Menu engineering is simply using real sales data to figure out which items are:

• making you money and selling well (keep and promote)

• selling well but barely profitable (fix)

• profitable but not selling (push)

• neither profitable nor popular (remove or rebuild)

And if you’re in food service, the pressure is real: the National Restaurant Association reported median food and non-alcohol beverage costs around 32% of sales for full-service operators in 2024 (and higher for smaller operators).

So let’s walk through a practical menu engineering process you can actually use, without turning it into a spreadsheet hobby.

What menu engineering is, in plain English

Menu engineering is a method that compares:

1. how popular an item is (how often it sells)

2. how profitable it is (how much it contributes after direct costs)

A common approach is the “menu engineering matrix,” which groups items into four buckets (Stars, Puzzles, Plowhorses, Dogs).

You don’t need to memorize the names. You just need to know what to do with each bucket.

Step 1: Pull the only numbers you need

To do menu engineering, you need three basic data points per item:

• Units sold (popularity)

• Price (what you charge)

• Cost (what it costs you to sell it)

For restaurants, “cost” usually means food cost (ingredients), and for retail it’s your product cost. If you also want to factor in labor-heavy items later, great—but start simple.

If you’re using a POS like Clover, this is much easier because you can pull item sales reports and see what’s actually moving (instead of guessing from memory). If your buttons and items are set up cleanly, the reporting becomes reliable—if they’re messy, menu engineering becomes a detective show.

Step 2: Sort items into the 4 menu engineering buckets

Here’s the simple version.

1) Stars (high profit, high popularity)

These are your winners. They sell often and they pay you well.

What to do:

• feature them on your menu (top right, boxes, callouts)

• mention them in staff suggestions

• keep quality consistent (don’t mess with what’s working)

• protect supply so you don’t 86 them constantly

Stars are the backbone of menu engineering. They’re what you build everything around.

2) Plowhorses (low profit, high popularity)

Customers love these. Your bank account… not as much.

What to do:

• tighten portioning or packaging

• renegotiate ingredient/vendor costs where possible

• adjust price carefully (small increases, not sticker shock)

• add a profitable add-on that pairs naturally

Plowhorses are where many established businesses leak profit quietly. People order them all day, and you feel busy… but margins don’t match the effort.

3) Puzzles (high profit, low popularity)

These make money when they sell—but they don’t sell enough.

What to do:

• rename it so customers “get it” immediately

• rewrite the description (benefit-driven, not ingredient list)

• move it to a better spot on the menu

• have staff recommend it with one sentence

• add a photo (if you use photos) or make it a “special”

A lot of puzzles fail because they’re hidden or explained poorly. Menu engineering isn’t always “change the food.” Sometimes it’s just better presentation.

4) Dogs (low profit, low popularity)

These are costing you money and complexity.

What to do:

• remove them

• or rebuild them (new cost structure, new portion, new price)

• or move to a limited-time rotation if you keep it for a reason

Most owners keep Dogs because “someone orders it sometimes.” That’s not a business reason. That’s emotional support for an item.

Step 3: Decide what to push, fix, or remove

This is the part that matters.

What to push

Push your Stars and your Puzzles.

• Stars: because they’re already working

• Puzzles: because they’re profitable and just need a spotlight

In menu engineering terms, you’re trying to shift demand toward the items that pay you best.

Practical ways to push without being salesy:

• “Most popular” tag (Stars)

• “House favorite” tag (Stars)

• “New” or “Chef’s pick” tag (Puzzles)

• Train staff to recommend one item per category

What to fix

Fix your Plowhorses.

Your goal isn’t to kill them—it’s to make them worth selling.

Fix levers (pick 1–2, don’t do everything at once):

• portion size (tighten, standardize)

• ingredient substitution (without lowering quality)

• prep method (reduce waste)

• price adjustment (small)

• bundle or add-on (increase ticket)

What to remove

Remove most Dogs.

Every low-selling, low-margin item creates:

• inventory complexity

• spoilage risk

• training confusion

• slower service

• inconsistent quality

Menu engineering is also operations engineering. Fewer “problem items” usually means smoother execution.

Step 4: Use a simple pricing strategy that doesn’t scare your regulars

If you’ve avoided price increases because you don’t want backlash, you’re not alone. But costs have moved.

The National Restaurant Association notes menu prices rose substantially from 2020 through 2025, citing BLS data showing a 31% increase between February 2020 and April 2025.

So here’s the smarter approach in menu engineering:

• raise prices on Stars carefully (people already buy them)

• raise prices on Plowhorses in smaller steps (watch volume)

• don’t waste time raising prices on Dogs (remove them instead)

• protect entry-level options so customers still feel they have a “safe” choice

If you need a rule: don’t do big jumps on your most price-sensitive, high-volume items. Do small, measured changes and watch results.

Step 5: Make your menu layout do some of the work

Menu engineering isn’t just math. It’s also psychology and convenience.

A few layout moves that tend to work:

• Put your Stars where eyes go naturally (top right, first in a category, highlighted)

• Don’t bury your best items under “misc”

• Limit choices in each category so decision fatigue doesn’t kill ordering

• Keep descriptions clear and benefit-driven (“crispy,” “tender,” “house-made,” “served hot”), not confusing

Step 6: Watch out for the biggest “menu engineering lie”

Here’s the lie: “Our menu is set.”

Menus are not set. Costs change. Customer habits change. Vendor pricing changes. Seasonality changes. And if you don’t revisit your menu, your margins will quietly drift downward.

A realistic menu engineering cadence for established operators:

• Monthly: quick review of top sellers, worst sellers, and margin outliers

• Quarterly: deeper cleanup (remove Dogs, rewrite Puzzles, adjust Plowhorses)

• Seasonally: rotate specials and test new items without bloating the core menu

Step 7: Make sure your POS setup doesn’t sabotage your menu engineering

Menu engineering is only as good as the data.

If staff uses:

• “open item” constantly

• inconsistent naming (“Burger” vs “Cheeseburger” vs “Burger w/ cheese”)

• manual overrides instead of correct buttons

…then your reports won’t reflect reality.

If you want accurate menu engineering, your POS items need to be structured properly (clear categories, modifiers, correct pricing, consistent naming). That’s also why POS matters beyond “taking payments.” It’s the system that tells you what’s actually happening in your business.

A quick “do this this week” menu engineering checklist

If you want action without overload:

1. Pull last 30 days of item sales

2. Identify your top 10 by units sold

3. Identify your bottom 10 by units sold

4. Pick 3 Plowhorses to fix (portion/price/add-on)

5. Pick 3 Puzzles to push (name/placement/staff suggestion)

6. Remove or rebuild 2 Dogs

7. Repeat next month

That’s menu engineering that actually moves profit.

Bottom line

Menu engineering isn’t about fancy charts. It’s about controlling profit with facts instead of guesses.

Push what pays you. Fix what’s popular but weak. Remove what wastes time and margin.