Part I · Stop the money leaks

Chapter 2: Your bestselling dish might be your worst earner.

Time60–90 minutes

You need4 weeks of item sales, your recipes, recent invoices

You leave withWhat every dish earns and a price test

The leak

Most owners judge a dish by how often it sells or by its food cost percentage. Neither tells you what it earns. A dish with a “bad” 38% food cost can make you more money than a “great” 15% one, because rent and payroll are paid in dollars, not percentages.

What matters is contribution per plate (price minus what’s on the plate) multiplied by how many you sell. And when an ingredient goes up (Chapter 1), every plate it’s on quietly earns less, at the same menu price.

What it costs you

Worked example · Mirela is fictional

Mirela’s menu, one week. Plate costs for the burrata, rigatoni and branzino are worked out from her recipes and latest invoices; the others are costs she already knew.

Dish	Price	Plate cost	Per plate	Sold a week	A week	Food cost
Burrata starter	$18	$6.89	$11.11	240	$2,666	38%
Arancini starter	$14	$3.20	$10.80	110	$1,188	23%
Grilled octopus starter	$21	$6.60	$14.40	60	$864	31%
Rigatoni alla vodka main	$24	$3.42	$20.58	150	$3,087	14%
Short rib main	$38	$14.50	$23.50	120	$2,820	38%
Cacio e pepe main	$23	$3.10	$19.90	70	$1,393	13%
Branzino main	$36	$15.84	$20.16	45	$907	44%
Tiramisu dessert	$13	$2.40	$10.60	100	$1,060	18%

The burrata is the bestseller, and one of the lowest earners per plate. Its main ingredient just went up 75 cents (Chapter 1), so every plate costs $0.75 more to make at the same $18: about $180 a week at today’s sales. Meanwhile the short rib, with the same 38% food cost, earns more than twice as much per plate.

Sort every dish into four groups

Compare dishes within their course, starters with starters and mains with mains, because a starter will never earn what a main does. A dish sells well if it sells at least 70% of what the average dish in its course sells (the standard menu-engineering rule). It earns well if its contribution per plate is above its course’s average, weighted by sales, or within 7.5% of it. A course needs at least three dishes before the comparison means anything.

Group	Sells well?	Earns well?	What to do
Stars	Yes	Yes	Protect them. Keep quality and portions consistent. Best spot on the menu. Staff recommend them.
Workhorses	Yes	No	Guests already love them. If a cost went up, test a small price step; otherwise look at the portion, side or garnish.
Puzzles	No	Yes	Help them sell: better menu position, a clearer description, a staff pick, a special.
Rethinks	No	No	Review them before you cut them. They take prep time and walk-in space.

At Mirela, the starters average 137 plates a week and $11.51 a plate; the mains, 96 plates and $21.32. The rigatoni, short rib and cacio e pepe are Stars, and so are the burrata and arancini, which earn close to their course’s average. The grilled octopus and the branzino earn well and sell less: Puzzles. Tiramisu is her only dessert, so there’s nothing to compare it with yet.

The price decision

Before raising a price, work out how many sales you could lose and still earn the same:

Sales you can lose = 1 − old contribution ÷ new contribution

Burrata from $18 to $19, the price that earns back what the ingredient increase took: the contribution goes from $11.11 to $12.11. So Mirela could sell 8% fewer burratas, 221 a week instead of 240, and still make the same money. Lose fewer than that and she’s ahead. A gentler first step is $18.50. For a bestseller people order on sight, either is a reasonable four-week test.

Raise prices in this order: dishes whose costs went up first, in small steps. Leave your Stars alone unless their costs jumped. Never raise everything at once; regulars notice a whole new menu, not one line. An increase above 10% isn’t forbidden, but check demand, nearby prices and value first.

The fix

Export item sales from your POS for the last 4 weeks. Look for a report called “product mix” or “item sales.” You need dish name, quantity sold and price.
Cost your dishes in the Menu Profit Doctor, right below: from your recipes and invoice prices, or with plate costs you already know. Start with your top sellers; five dishes is a good start, three is the minimum.
Read the redline. The tool compares each dish with the rest of its course, finds cost creep since your last invoices, and gives you price options with the sales each one can afford to lose.
Pick one action per group: protect a Star, test a price, push a Puzzle for two weeks. Prompt 3 rewrites a Puzzle’s description.
Check again in four weeks with fresh sales numbers, judged on contribution, covers and average check.

Tool · About 10 minutes for the first dish

Find what every dish really earns.

Cost your dishes from your invoices, compare them course by course, and leave with a price test and what to push, keep or rethink. There’s a worked example inside.

No sign-up. Your numbers stay on this device.

The prompts

Prompt 1 · Cost a dish

Here is my recipe for [dish], for one portion: [ingredients and amounts]. Here are my current unit prices: [paste from your Price Tracker]. Work out the cost of each ingredient for one portion, then the total plate cost. Add a line for oil, salt and garnish at [$0.25]. Show every calculation so I can check it. If any ingredient has no price, list it as missing instead of guessing.

Prompt 2 · Read the menu

Here is my menu with course, price, plate cost and weekly sales for each dish: [paste]. Within each course, group each dish as a Star, Workhorse, Puzzle or Rethink, using these dividing lines: sells well means at least 70% of what an average dish in that course sells; earns well means a contribution per plate above, or within 7.5% of, the course average weighted by sales. For each dish whose cost went up, tell me the new contribution at a $0.50 and a $1 price increase, and how many sales I could lose before I earn less. Show your working.

Prompt 3 · Help a Puzzle sell

Rewrite the menu description for [dish] in our voice. Under 20 words. Mention one real ingredient or technique: [what’s special about it]. Make no claims we can’t back up, like “best” or “healthy.” Give me three options.

Three decisions this week

i.One price test on a dish whose cost went up, at a price that passes the check above.

ii.One Puzzle gets a new description and a staff push for two weeks.

iii.One Rethink (if you have one) gets a date to be reworked or removed.

Your takeaway · Chapter 02

What you leave with

Run the Menu Profit Doctor with your own numbers and this section fills in: your result, a PDF of it and your plan as text.

Your menu, dish by dish, with a price testWaiting for your numbers.
Numbers from another device?Load a backup file you downloaded there.