Most menus are priced the same way they’ve always been priced — cost the dish, add a margin, round to a sensible number. But customers don’t read a menu like a spreadsheet. Small, deliberate changes to how a dish is priced and positioned can shift what people order, without changing the food at all.
The framework: 4 types of dish
In the early 1980s, researchers Michael Kasavana and Donald Smith developed the standard tool for this — the menu engineering matrix. It sorts every dish into one of four categories, based on two things: profit and popularity.
- Stars — high profit, high popularity. Your best dishes by both measures.
- Plowhorses — high popularity, lower profit. Everybody orders them, but the margin is thin.
- Puzzles — high profit, lower popularity. Good margins, just not enough orders.
- Dogs — low on both. Quietly costing you shelf space and kitchen focus.
Most menus have a mix of all four, whether anyone’s mapped it out or not.
What to do with each category
Stars — protect them:
- Don’t discount them
- Don’t bury them at the bottom of the page
- Don’t cut corners on the recipe to save a few pence
Plowhorses — fix them:
- Try a small price increase — high order volume means even a modest rise adds up
- Consider a slightly adjusted portion size
- Pair with a higher-margin side or add-on
Puzzles — promote them:
- Move them to a more prominent position on the page
- Give them a more appetising description
- Get the team recommending them directly
Dogs — reconsider them:
- Try one genuine repositioning or repricing attempt first
- If nothing shifts, it’s usually fine to remove them
- Free up the prep time and menu space for something that earns it
Why small changes beat big ones
The instinct with a struggling dish is often to overhaul it entirely — new recipe, new name, big price cut. The research points the other way:
- Customers react to what’s in front of them right now, not to what changed since last month
- A Plowhorse priced 50p higher rarely loses orders
- A Puzzle moved to a better spot on the page, with a slightly better description, often gains them
- Modest, deliberate adjustments tend to shift behaviour more reliably than dramatic ones
Where the data actually comes from
This only works if you know which dishes are popular and profitable — not just one or the other. That’s where back office reporting does the real work:
- Real sales-by-item data, paired with recipe costing
- Turns “I think our carbonara sells well” into an actual answer
- Popularity and profitability are very often two different questions — worth knowing which is which before you touch a price
The takeaway
Menu pricing psychology isn’t about tricking anyone — it’s about recognising that price, description and position all quietly shape what gets ordered, whether you’ve designed for that or not. The menu engineering matrix gives you a simple way to sort your dishes into what needs protecting, fixing, promoting or reconsidering.
Want to see your own menu through this lens? Book a demo with YUMA and see what your sales and cost data actually says about your menu.