Menu Engineering: Using POS Data to Push High-Margin Items

Every restaurant already sits on the data it needs to make more money per ticket — it's just trapped inside the point-of-sale system. Menu engineering is the discipline of pulling that data out, ranking every item by how profitable it is and how often it sells, then deliberately redesigning the menu to steer guests toward the items that make you the most money. Done well, it lifts margin without raising a single price or cutting a single dish.

This guide walks through how to run menu engineering off your POS data — the exact numbers to pull, how to classify each item, and where to actually push your winners once you know what they are.

What menu engineering actually is

Menu engineering is a two-variable analysis. For every item you sell, you need to know two things: how popular it is (units sold over a period) and how profitable it is (its contribution margin — menu price minus the cost of the ingredients to make it).

Plot those two variables against each other and every item on your menu lands in one of four quadrants. This is the classic framework, and it tells you exactly what to do with each dish:

• Stars (high popularity, high margin): protect and promote — feature them everywhere.
• Plowhorses (high popularity, low margin): re-engineer the cost or nudge the price, and sell add-ons.
• Puzzles (low popularity, high margin): reposition, rename, or upsell into them.
• Dogs (low popularity, low margin): rework or cut.

The whole point is that most menus over-promote low-margin plowhorses and bury high-margin puzzles. Menu engineering flips that: it gets your high-margin items in front of more guests, more often.

The two numbers to pull from your POS

Your POS is the source of truth for popularity — it counts every unit sold, by daypart, by channel, down to the modifier. What it rarely knows on its own is your true food cost per item, so you'll pair sales counts with your recipe costing. Here's what you need:

• Units sold per item over a clean, representative period (a full month, or a quarter to smooth out weekly swings).
• Menu price and any real average discount applied at the register.
• Plate cost — the ingredient cost to produce one order, including obvious modifiers.
• Contribution margin — price minus plate cost. This, not food-cost percentage alone, is what actually pays your rent.

A common trap: chasing the lowest food-cost percentage. An item with a 25% food cost that sells three times a week makes you less real money than one with a 38% food cost that sells two hundred times a week. Menu engineering optimizes for total contribution margin, not a vanity ratio. As a rule of thumb, full-service food costs typically run in the 28–35% range, but treat that as context, not a target — the dollars per plate times volume is what matters.

How to run the analysis, step by step

1. Export item-level sales from your POS for the period. Filter to menu items; strip out voids and comps.
2. Attach a plate cost to each item from your recipe cards, and calculate contribution margin.
3. Find your two averages — average popularity (total units ÷ number of items) and average margin across the menu.
4. Sort every item into a quadrant using those averages as the dividing lines.
5. Act on the quadrant, then re-measure the next period to see whether the moves worked.

That last step is where most operators stall. The analysis is only worth doing if you actually change what guests see — and then check the POS again to prove the change moved margin.

Where to push your high-margin items

Knowing your Stars and Puzzles is half the job. The other half is engineering the moments where guests decide. A few reliably work:

• Anchor high-margin items in prime menu real estate — top-right and top-left of a printed menu, first screen of a digital one.
• Attach modifiers and add-ons to plowhorses so a high-volume, low-margin dish still drives incremental profit.
• Rename and describe puzzles so a high-margin item that nobody notices becomes something people crave.
• Use every ordering surface — the register isn't the only place guests choose. Kiosks, online ordering, and menu boards can all merchandise your winners automatically.

That last point is where a connected stack pulls ahead of spreadsheets. Menu engineering breaks down when your insight lives in one system and your ordering lives in five others that never got the memo.

The connected-stack advantage

Here's the honest constraint of doing this by hand: you run the analysis in a spreadsheet, then have to manually update the printed menu, the online ordering page, the kiosk, and the menu board — and they drift out of sync within a week.

This is the gap INFI is built to close. Your POS runs the register; INFI runs your growth on top of the POS you already have — Square, Toast, Clover, Lightspeed, or HungerRush — without replacing it. That means the high-margin items you identify can be featured consistently across every guest-facing surface at once.

• Self-ordering kiosks merchandise upsells and combos automatically, and typically lift average order value by roughly 15–30% because a screen never forgets to suggest the add-on.
• Commission-free online ordering gives your Stars the same prime placement off-premise that they get on the counter.
• Digital menu boards and customer displays let you re-feature a Puzzle instantly instead of reprinting.

The results show up in the ticket. On INFI, PJ's Coffee lifted average order value by 45%, Juice Press by 15% per order, and Oak View Group venues doubled basket size — the same mechanism menu engineering aims for, applied consistently at the point of decision.

Common mistakes to avoid

• Analyzing once and stopping. Menu engineering is a quarterly rhythm, not a one-time project — costs and tastes move.
• Cutting a Dog that's a loss-leader on purpose. Some low performers exist to complete an occasion; check before you delete.
• Ignoring channel. Your best in-store item may be a poor online seller. Segment your POS data by channel before you decide.
• Optimizing percentages over dollars. Total contribution margin pays the bills.

Bringing it together with INFI

INFI is the Restaurant Growth Platform — kiosks, online ordering, loyalty, displays, and marketing that layer on top of the POS you already run, used by 1,800+ restaurants and founded by former restaurant operator Lucas Liu, Ph.D. It's built to take the high-margin items your menu engineering surfaces and put them in front of guests everywhere they order. Book a demo to see how your POS data can start pushing your best items automatically.