Engineering Your “Star” Items (high profit + high sales)

Finally, put it all together with menu engineering — so your best items get seen and ordered. Star items are the winners: dishes that customers already love and that you earn well on. With better placement, naming, descriptions, photos, and smart category structure, you can increase sales of these stars without changing the kitchen or adding new items.

Before you engineer anything, make sure your numbers are real. Stars only exist when your costs and portions are consistent, so start with Food Cost Percentage Explained (and what to aim for) and lock in consistency via Portion Control: The Hidden Key to Better Profit.

Menu engineering is the process of comparing popularity and profitability per item, then adjusting how the menu presents items so guests naturally choose the best outcomes. The classic way to visualize this is the menu engineering matrix (Stars, Puzzles, Plowhorses, Dogs), explained clearly in Toast’s Menu Engineering Matrix guide. Toast POS

Stars sit in the top-right: high popularity + high profitability. Your job is to keep them there and sell more of them.

You only need two metrics for each item:

Popularity: how many units sold in a period (e.g., last 30 days)

Contribution margin: selling price − item cost

Both Lightspeed and DoorDash explain contribution margin in practical terms: it’s the money left after food cost, and it’s what you should optimize when engineering a menu. See Lightspeed’s menu engineering walkthrough and DoorDash’s menu engineering tips. Lightspeed+1

Simple process (works for any restaurant):

Export sales for the last 30 days (or 90 days if you have seasonal swings).

List each item with units sold and its contribution margin.

Mark anything in the top group for both metrics as Star.

If you’re still fixing pricing basics, do that first: How to Price a Dish: Cost + Margin + Psychology.

Using the matrix language from Toast and common menu engineering practice: Toast POS+1

Stars: high profit, high sales → promote and protect

Plowhorses: low profit, high sales → adjust portions, price, or positioning

Puzzles: high profit, low sales → improve name, description, placement

Dogs: low profit, low sales → remove, replace, or reposition carefully

Lightspeed also summarizes the categories and the recommended actions (keep stars front and center; adjust plowhorses; rework puzzles; rethink dogs) in real-world operator examples like this Lightspeed story on using menu data to boost profitability. Lightspeed

To avoid the common traps during cleanup, see The Most Common Pricing Mistakes Restaurants Make.

You don’t need to “sell harder.” You need to remove friction and guide attention.

Practical placement tactics that consistently help stars:

Put stars in the top 1–3 positions of a category (people scan top-down).

Give stars more visual breathing room (shorter descriptions around them, clean layout).

Avoid hiding stars among too many similar options (category overload kills decision speed).

If you run a digital menu or multi-location setup, the same logic applies — but you can apply it faster across locations. A recent industry guide aimed at operators, Deliverect’s menu engineering article for multi-location restaurants, emphasizes starting with each item’s cost and contribution margin, then using engineering to push profitable items. deliverect.com

Star names should do one thing: make the item feel like the obvious choice.

Quick naming rules that lift star performance:

Use specific ingredients guests recognize (“slow-roasted,” “chargrilled,” “crispy”).

Add a distinctive cue (region, signature sauce, house-made element).

Avoid internal kitchen terms (“Chef’s special #3”) unless your brand is built on it.

For items that are profitable but not selling (Puzzles), naming is often the highest ROI fix — and the matrix framework helps you see that clearly. Toast POS+1

A great star description doesn’t explain everything — it answers the silent questions:

What is it?

What does it taste like?

Why should I pick it instead of the safe option?

Use a simple structure:

Flavor cue (smoky / creamy / spicy / fresh)

Key ingredients (what matters)

Signature detail (house sauce, special prep)

When description updates don’t move the needle, the issue may be pricing structure. That’s where How to Price a Dish: Cost + Margin + Psychology ties back into engineering (anchors, good/better/best).

Photos can increase orders — but too many photos can cheapen the feel and overwhelm scanning. The easiest rule:

Use photos for stars, not for everything.

Use consistent lighting/background to keep the menu clean.

Make the photo match the real plate (trust matters more than “beauty”).

If you sell bundles, star photos also help combos convert — just make sure the bundle doesn’t damage margin. Pair this with How to Build a Profitable Combo / Meal Deal so a popular deal doesn’t quietly erase profit.

If your category has 18 items, stars get diluted. Category structure is a selling tool:

Break big categories into smaller “decision-ready” groups (e.g., “Signature Bowls” vs “Bowls”).

Put stars in a category where they’re the clearest option (not buried among lookalikes).

Avoid having multiple stars compete head-to-head in the same tight cluster unless you’re sure it lifts total sales.

This is also where portion control protects you: if a star becomes more popular after engineering, you need stable portions so increased volume doesn’t create margin drift. Portion Control: The Hidden Key to Better Profit is the safety net.

Recompute popularity + contribution margin (30–60 minutes). Lightspeed+1

Confirm your top stars still have stable portions and costs.

Make one improvement per star (placement OR name OR description OR photo).

Track result for 2–4 weeks (don’t change 10 variables at once).

Re-run the matrix and repeat.