> ## Content Index > Fetch the complete content index at: https://www.growyoursmallbusiness.org/llms.txt > Use this file to discover other available public pages before exploring further. # Which Dishes on Your Menu Are Actually Making You Money? - URL: https://www.growyoursmallbusiness.org/which-dishes-on-your-menu-are-actually-making-you-money/ - Published: 2026-08-20T15:14:43.000Z - Updated: 2026-08-20T15:14:43.000Z - Author: Mason Martins - Tags: Restaurant, Problem Article **The dish everyone orders and the dish that actually makes you money aren't always the same one — and most menus have never separated the two.** Every restaurant owner has a mental list of which dishes are "doing well" and which ones aren't. Usually it's built from the same two things: what sells, and what the food cost percentage says. It feels like enough information to go on — and most weeks, nobody has time to ask whether it actually is. This is worth ten minutes of that time, because the two numbers most owners lean on to judge a menu were never built to answer the one question that actually matters. **Main Ideas In This Article:** 1. Food cost percentage and sales volume both feel like they measure a dish's success, but neither one tells you what it actually earns — which means most menus have dishes quietly misjudged in both directions. 2. In a business already running on a thin profit margin, a few small adjustments to the dishes you're already selling can greatly improve your bottom line. 3. A real fix has to use actual profit dollars instead of a ratio, weigh that against how often a dish sells, run on data you already have, and end in a specific action. 4. The Menu Matrix is a framework that sorts every dish into one of four groups based on margin and popularity together, and each group comes with one clear, specific move. 5. Start with a fast, same-day ranking to surface a surprise or two, then work through one menu category at a time to get real numbers and a real action for every dish — recheck the results after a few weeks to see what improved. **Key Terms:** 1. Food Cost Percentage — the percentage of a restaurant’s sales revenue that is spent on the food ingredients used to make the dishes sold. 1. Food Cost Percentage = (Cost of Food Used ÷ Food Sales) × 100 2. Profit Margin — the percentage of a business’s revenue that remains as profit after expenses are paid. 1. Profit margin = (Net Profit / Total Revenue) × 100 2. For example, if a restaurant brings in $100,000 in revenue and has $97,000 in total expenses, it has $3,000 left as profit: $3,000 ÷ $100,000 = 3% profit margin 3. POS (point of sale) — the system used to process orders and payments and record what customers buy ### **Popular Isn't the Same as Profitable** Picture a normal Friday night. The kitchen is slammed, and two dishes keep coming back through the window all night. One is the dish everyone orders — the pasta special, the burger, whatever your place is known for. The other moves slower: a steak or seafood dish that goes out four or five times a shift. At the end of the month, you check your food cost percentage, see it's about where it usually sits, glance at what sold the most, and figure you have a decent read on how the menu's doing. That's the moment worth slowing down on. It's usually where the misdiagnosis happens. Most owners judge a dish on two numbers:food cost percentage and how often it sells. Both feel like they should tell you what's working. Neither one actually does. Food cost percentage is a ratio — it tells you what share of a dish's price went to ingredients, not how many dollars it put in your pocket. That distinction matters more than it sounds like it should. Say your pasta special costs $4 to plate and sells for $16 — a 25% food cost, on the high side. It goes out 40 times a night. Your steak costs $9 to plate and sells for $34 — a food cost percentage you'd frame and hang on the wall. It goes out 4 times a shift. Multiply price minus cost by how many actually sell: the pasta clears $480 a night. The steak clears $100\. The dish with the "worse" ratio is carrying nearly five times the profit. Popularity has the same blind spot in reverse. A dish that sells well looks successful by definition. But volume alone doesn't tell you whether that volume is making you money or just keeping the kitchen busy. This doesn’t mean that popularity or food cost percentage don’t correlate with dish success, it means that the numbers owners already look at can hide the weaknesses and opportunities of their menus. Before you can decide what to raise, promote, rethink, or remove, you need to know which dishes are actually pulling their weight — and that starts with looking beyond percentages and popularity to the profit each dish contributes. ### **A Thin-Margin Business Can't Afford Invisible Problems** A typical full-service independent restaurant keeps somewhere around 3 to 5 cents of every sales dollar as profit, after food costs alone eat roughly 32 cents of it. That's not a sign anyone's doing something wrong — it's just what this business looks like. Thin margins are the industry, not a symptom of mismanagement. That's exactly what makes an invisible problem expensive here. In a business with room to spare, one underperforming dish might not matter much. In a business running on 3 to 5 cents per dollar, a couple of cents recovered from dishes you're already selling can turn a 3% profit margininto a 5-6% profit margin. That's close to doubling your actual profit, from money that's already coming through the door. The cost isn't only in the margin, either. Every dish your restaurant offers is using real capacity — prep time, cooler space, a cook's attention during a rush. A dish that's break-even is eating that capacity, night after night, without anyone assessing its place on the menu. It runs the other way too: a genuinely profitable dish can sit under-ordered simply because nothing is pointing guests toward it. That's profit going unclaimed for no reason other than nobody noticed. To be clear about what this isn't: no restaurant closes because of an unoptimized menu. This is a slow leak, not a crisis. But it's gotten more expensive to ignore lately. Food costs have moved up meaningfully over the past few years, and unevenly — some ingredients have tightened hard, others have barely moved. Due to these fluctuations in costs, menus priced well two years ago likely have multiple dishes that have drifted from profitability. The usual fix, raising every price on the menu, is starting to run out of road — guests have absorbed a lot of it already. As a result, knowing which specific dishes actually need attention is quickly becoming the more reliable option. ### **The Fix: Sort Every Dish by What It Actually Earns** The solution has to work in dollars, not percentages, and weigh that figure against how often a dish actually sells. It also has to run on data you already have right now. Finally, it has to end in a specific action for each dish, with a way to check it again down the road. Held up against that, the tools most owners already reach for — watching food cost percentage, a best-sellers list, an across-the-board price hike, even recipe-costing software — each solve a piece of it, but not the whole thing. Here's the solution: instead of judging a dish by one number, judge it by two — what it actually earns in dollars, and how often people actually order it — and look at both at once. Plot every dish in a category on those two measures, and four natural groups show up. Some dishes make good money and sell well, some sell constantly but on thin margins, some are quietly profitable but barely ordered, and some do neither. Each group gets one clear move, and once you know which group a dish is in, the move is obvious. This isn't a new invention. It's called the Menu Matrix, and it comes from a piece of restaurant research from 1982 that's still the standard way this problem gets solved — not because nothing better has come along, but because it is the most direct way to solve this problem. The remainder of this article will walk you through implementing this well-established solution completely, without stopping short and pointing you toward a paid tool to finish the job. ![](https://storage.ghost.io/c/46/4d/464d972f-f4a8-4c19-95e1-e5c8084d7648/content/images/2026/08/data-src-image-10b51f2d-f776-4e97-be49-2d2f7c900d61.png) The Menu Matrix meets everything a real fix would need to do. It uses a dollar figure — contribution margin, price minus true plate cost — never a percentage. It weighs profit and popularity together by design; a dish only lands in a group once both numbers are known. It runs on a sales report and your existing invoices so you already have all of the data you need. And it ends in a specific action that you can implement and track the results of for each dish. ### **Where to Start, and How to Know It Worked** There are two ways into this: a fast version that gets you a same-day gut check, and the full version that actually tells you what to do about each dish. Do the fast one first — it's what usually convinces people the full version is worth the time. **The Fast Look** A rough, same-day pass across your entire menu that ranks every dish by how many actual profit dollars it contributed over the last few months — margin times volume, all at once. The full process below takes real time, and it's easy to put off; this step exists to surface a surprise or two fast enough to make that time feel worth spending. 1. Pull total units sold per dish from your POSacross the last 90 days. 2. For each dish, estimate a rough plate cost — the main ingredients, roughly priced, rounded to the nearest dollar. This doesn't need to be precise. You're doing the same math as the full process, just faster and looser. 3. Subtract that rough cost from the menu price, multiply by units sold, and rank every dish on the menu — not just one category — from highest total profit contributed to lowest. This ranking will surprise you with at least one dish’s spot. That's the point. But notice what it can't tell you: it can flag that a dish is underperforming but it doesn’t tell you what the problem is. It could be the margin, the volume, or both. That's what the full process is for. **The Full Process** This is where you get real numbers and a specific action for each dish. Start with one category, not the whole menu — trying to do everything at once is the most common reason this gets abandoned halfway through. 1. **Pull a clean sales report.** Export an item-level sales report from your POS— often labeled "menu mix" or "product mix" — covering at least 90 days. Pick a stretch that doesn't contain a holiday spike or an unusually slow month. Your POSmay log modifiers, combos, or size variants as separate line items. Consolidate those to each actual menu item, or your popularity numbers will come out wrong before you've even gotten to the cost side. 2. **Select one category first and identify the components.** Pick the category with the widest price spread — usually entrées — and stay there for now. For each dish, list every component as it's actually plated: protein, sides, sauce, garnish, other ingredients, etcetera. 3. **Identify the cost per usable ounce of each component.** Obtain the cost of each component from your most recent vendor invoice, and convert to cost per usable ounce, not cost per whole item. For example, if a 10-ounce onion costs 60 cents, and trimming leaves you 8 usable ounces — that's 7.5 cents per usable ounce, not 6\. Cost every component that way. 4. **Identify the ounces of each component in each dish.** Once you have each component’s cost per usable ounce, figure out the ounces used of each component in each dish. Keep a list of each dish and the amount of each component that goes into each dish in ounces. 5. **Calculate plate cost.** Now you should have the cost per usable ounce for each component and the amount of each component in each dish figured out. In order to calculate the plate cost for each dish, all you have to do now is multiply the cost per usable ounce of each component by the total ounces of each component in each dish. For example, if a dish uses 3 ounces of onion and the cost per usable ounce for an onion is 7.5 cents, the onion cost for that dish would be 22.5 cents. Do this for every component under every dish and then add up each component to get the total cost of the components in a dish. Then, add 3% to 8% on top for waste and spoilage to get your plate cost. 6. **Calculate contribution margin.** For each dish calculate contribution margin with this equation: contribution margin = menu price minus plate cost. Average it across the category; a simple average is fine, it doesn't need to be weighted by sales. 7. **Calculate each dish's popularity share.** Divide units sold of each dish by total units sold in its category over the same period, and convert to a percentage. Then calculate your popularity threshold with this equation: 70 / the number of items in the category. Eight entrées, for example, gives you a threshold of 8.75%. Any dish with above 8.75% would count as popular. This formula is the standard equation used in the menu matrix so please use it as given. 8. **Sort every dish into one of four groups.** A spreadsheet with two extra columns — above or below the category's average margin, above or below the popularity threshold — is all you need. - Above both: **Star** - Above popularity, below margin: **Plowhorse** - Above margin, below popularity: **Puzzle** - Below both: **Dog** 9. **Apply the specific fix for each group.** - **Star** — leave it alone. Don't touch the recipe, portion, or price. Give it a prominent spot — top of the category, a boxed or highlighted position. - **Plowhorse** — improve the margin without losing the volume. A modest price increase, usually $0.50 to $1.50, is the most common lever — make the change, then watch volume for several weeks. A slight portion trim, a cheaper ingredient swap that doesn't change the eating experience, or pairing it with a higher-margin add-on can also work. - **Puzzle** — the dish is fine; almost nobody's seeing it. Move it to a more visible spot, rewrite the description with more specific, sensory language, add a photo if your menu is digital, and ask staff to actively recommend it. Also worth checking whether it's priced above what it feels worth — sometimes a small price cut unlocks volume faster than better placement does. - **Dog** — cut it, unless it serves a real purpose beyond the numbers: a vegetarian or gluten-free option, a kids' item, something you keep on purpose. If you're keeping one, you should be able to say why in a single sentence. If you can't, that's your answer. Watch for the instinct to defend a Dog because "people love it" — the popularity threshold already accounts for that. If it's still landing here, that feeling is usually coming from a small, vocal group of regulars, not actual sales share. 10. **Change one thing at a time, then move to the next category.** Make your changes, then leave the rest of the menu alone while a full measurement window passes. Changing several dishes at once means you won't know which change caused which result. Once this category's done, repeat these steps for the next one. Run these at whatever pace is realistic — there's no requirement to do the whole menu in one go. **How You'll Know It Worked** For anything you changed, wait at least 4 to 6 weeks before checking it — ideally a stretch that doesn't overlap another menu change or an unusual event like a holiday or a weather closure. Then, re-pull the same two numbers, contribution margin and popularity share, and compare them to what you recorded before the change. You're looking for movement in a specific direction. A Plowhorse you adjusted should drift toward Star — margin up, popularity holding. A Puzzle you repositioned should also drift toward Star — popularity up, margin holding. If a dish doesn't move at all, that's still useful information: it usually means you pulled the wrong lever. A Plowhorse that took a price increase without losing volume means you probably had more room than you thought. A Puzzle that got better placement and still didn't sell more likely wasn't a visibility problem to begin with — it's genuinely not in demand. Beyond the individual dish, check whether the category's average contribution margin moved up, and if your POSreports cost of goods sold, keep an eye on average contribution margin per ticket across the whole restaurant — that's the number that should trend upward over a full quarter if the work is landing. This isn't a once-and-done project. Rerun the full process every quarter, or sooner if a major ingredient cost shifts. Update your plate costs whenever vendor prices move meaningfully in between, rather than waiting for the quarterly cycle — that habit is what keeps this from going stale the same way the food-cost-percentage approach did in the first place.