← All posts
19 Sept 2026 · 5 min read

Menu Engineering: How to Find Which Dishes Actually Make You Money

Menu engineering plots every dish by popularity and margin. Learn the four categories — stars, plowhorses, puzzles and dogs — and what to do with each.

Skip the arithmetic: try the free menu price calculator.Open it →

Your best-selling dish is not necessarily your most profitable one. Menu engineering is a simple way of looking at every item on two axes at once — how often it sells and how much it earns — so you can see which dishes are carrying the restaurant and which are quietly costing you money.

The two numbers you need for each dish

  • Popularity — how many of this dish you sold as a share of everything sold, over a decent period (a month or more).
  • Contribution margin — the selling price minus the cost of the ingredients. Note this is rupees, not a percentage: a ₹400 dish with a 40% food cost leaves more money per plate than a ₹100 dish with a 25% food cost.

Both numbers depend on knowing each dish’s true ingredient cost, which means costing recipes properly instead of estimating.

The four categories

Compare each dish to the average popularity and average margin for your menu, and it lands in one of four boxes:

  • Stars — high popularity, high margin. Protect these. Keep quality and portion consistent, feature them prominently, and be careful about changing them.
  • Plowhorses — high popularity, low margin. Guests love them, but each sale earns less than it should. Try a modest price increase, a smaller portion, a cheaper ingredient that guests will not notice, or pairing them with a higher-margin side.
  • Puzzles — low popularity, high margin. They earn well but do not sell. Try repositioning on the menu, renaming, a better description or photo, or having staff mention them. If nothing works, they may not belong.
  • Dogs — low popularity, low margin. Candidates to remove or rework, unless they serve a specific purpose such as a signature item or a dish that brings a particular crowd in.

A quick illustrative example

Suppose butter chicken sells 300 plates a month at a ₹120 margin, while a grilled fish sells 40 plates at a ₹210 margin, and a paneer wrap sells 260 plates at a ₹45 margin. Butter chicken and the paneer wrap are both popular, but only the first is a star; the wrap is a plowhorse worth re-pricing or re-costing. The fish earns well per plate but barely sells, which makes it a puzzle: is it hidden on the menu, or is it simply not wanted here? (These figures are made up to show the method, not benchmarks.)

Some cautions

  • Do not judge a dish on a few days of data. Weekends, festivals and weather distort a small sample.
  • Popularity can be an artifact of placement. A dish at the top of a page or with a photo will out-sell an identical one buried in the middle.
  • Some low-margin items exist to make the rest of the menu work — a tea that pulls guests in for snacks, for instance. Look at the whole basket, not just the single line.
  • Costs change. Re-run the numbers when a major ingredient’s price moves.

How to start

Pull your sales by item for the last month or two. Cost your top twenty dishes from their recipes. Work out the contribution margin for each and sort them into the four boxes. You will usually find two or three clear actions — a re-price, a removal, a repositioning — that pay for the exercise straight away.

With recipe-level costing in EasyBills, cost per plate and margin per dish are calculated for you and stay current as ingredient prices change, so this analysis becomes something you look at regularly instead of a once-a-year project.

Next

Restaurant Billing Software in India: What to Look For Before You Buy

Also worth reading

AI Assistant for Restaurants: What It Can Actually Do (and What It Can’t)

See how EasyBills handles this.

Book a walkthrough