Every restaurant menu has a second menu hiding underneath it, one diners never see. It doesn’t rank dishes by taste or creativity. It ranks them by two numbers: how often people order it, and how much money it actually makes. That grid decides more about what ends up on your table than most people realize, and it didn’t even start in the restaurant industry. It came from a corporate strategy tool built for entirely different products, back in the early 1980s, when two researchers at Michigan State adapted the Boston Consulting Group’s growth-share matrix, originally meant for ranking business units, into something a kitchen could use to rank chicken parmesan against linguine.
The tool is called menu engineering, and once you know it exists, you start noticing its fingerprints on nearly every menu you’ve ever read.
Two Numbers, Four Boxes
The whole system runs on just two measurements. Popularity: what share of orders a dish actually pulls, pulled straight from the point-of-sale system. Profitability: what a dish earns after subtracting what it costs to make, known as contribution margin. Plot every item on a menu against those two numbers and four groups appear on their own. High popularity and high profit. High popularity, low profit. Low popularity, high profit. Low on both. Restaurants gave these four groups names that have stuck around for forty years: Stars, Plowhorses, Puzzles, and Dogs.
Stars Don’t Get Improved. They Get Protected.
A Star is the dish doing everything right already, ordered constantly and earning strong margin every time. The instinct might be to tinker with something that’s already working, add a twist, cut a corner on cost since guests clearly love it regardless. Restaurants that actually understand menu engineering do the opposite. Stars get locked in place: same recipe, same quality, same spot near the top of the menu where the eye naturally lands first. The whole strategy here is protection, not improvement, because guests build loyalty to something staying exactly the way they remember it.
Plowhorses: Everyone Orders Them, Nobody Gets Rich Off Them
A Plowhorse is the opposite problem. High popularity, thin margin. Think of the dish every table seems to order, the reliable burger or the pasta special, the one the kitchen could make blindfolded, that still isn’t pulling much profit per plate. These items usually aren’t going anywhere, guests would notice immediately if they disappeared, so the strategy shifts toward small, careful adjustments instead: trimming portion size slightly, swapping in a marginally cheaper ingredient that doesn’t change the eating experience, nudging the price up by a dollar without anyone really registering it. None of it is dramatic. Compounded across hundreds of orders a week, small margin gains on a high-volume item add up faster than people expect.
Puzzles: High Profit, but Nobody’s Ordering Them
A Puzzle sits in the strange spot where a dish is genuinely profitable, but barely anyone orders it. The problem usually isn’t the food. It’s visibility. Maybe it’s buried at the bottom of the menu where eyes rarely land. Maybe the name doesn’t tell a diner anything appetizing about what’s actually on the plate. Maybe the server never mentions it because they’ve never actually tasted it themselves.
Puzzles get treated as a discoverability problem, not a recipe problem. Solutions tend to be about positioning: moving the item somewhere more visible, giving it a description that actually sells it, training staff to recommend it by name. A Puzzle that gets solved, meaning more people start ordering a dish that was already profitable per plate, is one of the fastest ways a restaurant can grow revenue without changing a single recipe.
Dogs: Not Every One Gets Cut
A Dog is low popularity and low profit, and it’s tempting to assume every Dog should be pulled from the menu immediately. Sometimes that’s exactly right. But experienced menu strategists usually check one thing before cutting anything: why is this item still here. Some Dogs stick around because a small, loyal group of regulars would be genuinely upset to see them go, and losing that goodwill costs more than the plate itself. Others exist purely to round out a menu’s range, the one vegetarian main, the single lighter option, even if it individually underperforms. The honest evaluation matters more than the automatic cut.
Watching It Work on a Real Menu
Picture a simple Italian menu with four pasta dishes. A shrimp linguine sells constantly and earns a healthy margin, that’s a Star, left alone and protected. A chicken parmesan sells just as often but costs more to plate than the kitchen would like, that’s a Plowhorse, quietly nudged on price or portion. A duck ragu earns strong margin on the rare occasion someone orders it, sitting on page three where nobody’s looking, that’s a Puzzle, waiting on a better spot and a better description. And a plain radicchio salad, ordered rarely and barely breaking even, might be a Dog worth cutting, unless it turns out to be the only vegetarian starter on the menu, in which case it earns its spot for a different reason entirely.
None of these decisions get made by gut feeling once a restaurant actually runs the numbers. That’s really the whole point of the exercise.
Why This Actually Matters to the Person Eating There
This ties directly into how restaurants design menus for profit and customer experience: a menu isn’t a neutral list of options, it’s a document shaped by which dishes earn their place, financially, and which ones survive for other reasons entirely. Understanding that doesn’t make dining out feel colder. If anything, it explains why a restaurant’s best dish is usually sitting right where you look first, and why that pasta special never seems to change even as everything else on the menu quietly does.
A Version of This Any Home Cook Can Use
Nobody needs contribution margin math to plan a dinner party, but the underlying question is worth borrowing. Which dishes in your regular rotation get made constantly because they’re easy and well-loved, even if they take more effort than they’re worth? Which ones deserve a better spot in the meal plan because they’re actually great but rarely get made? Running your own kitchen through this lens, even loosely, tends to surface the same thing it does for a restaurant: a few dishes worth protecting exactly as they are, and a few worth either cutting or giving a better chance to shine.