Section 1
The artifact: the contribution-margin-per-cover worksheet
Menu engineering in its classic form, from Michael Kasavana and Donald Smith's 1982 framework, sorts dishes into four boxes by two axes: popularity and contribution margin. Contribution margin there is menu price minus food cost. The re-engineering move for a labor shock is to widen "cost" so it includes the labor a dish consumes, then re-run the same sort. A dish that looked healthy on food margin can fall apart once you count the minutes of skilled hands it eats. Build one row per dish. The columns: Your fully-loaded labor cost per minute is the number the food-cost view hides. Take a line cook's total hourly cost to you, wage plus employer on-costs plus holiday and cover, divide by 60. If a shock just raised your employer payroll contributions, this is the exact number that moved, so recompute it first and let it flow through every row. Then sort the rows by contribution margin and by units sold, and drop each dish into the matrix.
Section 2
The artifact: the contribution-margin-per-cover worksheet (continued)
The classic matrix used food-margin CM. Yours uses labor-inclusive CM, which is why dishes move boxes the moment you run it. The plate that was a Star on food cost can drop to Plowhorse once its fifteen minutes of skilled hands are priced in. That reclassification is the whole point.
Section 3
The decision rule: cut, re-plate, or re-price by the hands
Each box gets a different move, and the move depends on why the dish lands where it does. Stars stay untouched. They sell well and clear a real margin even after labor. Do not re-price them for the sake of it. Their job is to anchor the menu and carry the room. The only change worth making to a Star is guarding it: make sure a prep shortcut or a supplier switch has not quietly degraded the thing customers come back for. Plowhorses are the dishes a payroll shock hurts most, because they sell in volume and the labor cost is now multiplied across every one of those covers. You have three levers, in order of preference. • Re-plate to cut labor minutes. If the dish is popular but labor-heavy, attack the minutes before the price. Can a component be batch-prepped in a quiet window instead of fired to order. Can a garnish that adds ninety seconds per plate and nothing to the guest's experience come off. Every minute you remove is multiplied by the same volume that made the dish a problem, so re-plating a Plowhorse pays back faster than re-plating anything else on the menu. • Re-price with an anchor. If the minutes cannot come out without hurting the dish, raise the price, but do it where the guest has a reference that makes the rise read as fair. A modest rise on a signature that guests already perceive as premium lands better than the same rise on a plain staple. • Reposition on the menu. Sometimes the fix is to stop pushing volume onto your most labor-hungry line. Move it off the eye-line, stop featuring it, and let a higher-CM neighbor take the covers it was absorbing. Puzzles clear a good labor-inclusive margin but do not sell. Here the move is promotion, not surgery: better menu placement, a server mention, a description that sells the value. If you can lift a Puzzle's volume, you are adding high-margin covers without adding a labor-heavy dish to the pass. Dogs cost labor, return little, and nobody misses them. Cut them, or if a Dog exists for a reason (it completes the menu, it serves a regular), re-plate it down to the lowest labor form that still honors the dish. A menu carrying labor-hungry Dogs is paying skilled minutes to lose money on plates few people order.
Section 4
Worked shape: how a dish changes boxes
Take a kitchen with two dishes that look equally healthy on the food-cost sheet. Dish A is a roast plate: expensive protein, high food cost, but assembled in three minutes at the pass because the components hold. Dish B is a plate of hand-finished pasta: cheap ingredients, low food cost, but twelve minutes of skilled hands per cover because it is finished to order. On a food-cost view, Dish B looks like the winner. Cheap ingredients, fat food margin. Run the labor-inclusive worksheet after a payroll shock and the picture inverts. Dish A's three minutes barely move its contribution margin. Dish B's twelve minutes, repriced at the new fully-loaded labor rate and multiplied across a busy service, eat most of the margin the cheap ingredients created. Dish B is the Plowhorse to re-engineer. The fix is not to cut it, because it sells. The fix is to find the minutes: prep the base in a quiet window, simplify the finish, and if the minutes will not come out, price the dish for the hands it actually demands. The food-cost sheet would have told you to leave Dish B alone and cut Dish A. It would have been exactly wrong.
Section 5
The models under the worksheet
Two disciplines sit under the method. The first is the menu-engineering matrix itself, Kasavana and Smith's popularity-versus-margin sort. Its power is that it refuses to treat all dishes the same: a low-margin high-volume dish and a low-margin low-volume dish need opposite responses, and the matrix separates them. The re-engineering step keeps the matrix and swaps the margin definition to include labor, which is the input the shock actually moved. The second is comparative statics: change one input, hold the rest still, and read what breaks. Here the moved input is labor cost per minute. By recomputing only that number and letting it flow through every row, you isolate the shock's real effect and stop yourself from re-pricing dishes the shock never touched. A blanket price rise across the whole menu fails exactly because it ignores this. It over-charges the fast-to-plate dishes that the shock barely affected and under-recovers on the labor-heavy ones that caused the damage.
Section 6
What the worksheet cannot see
It cannot price your guests' tolerance. The worksheet tells you which dishes need a rise; it does not tell you how much your particular room will absorb before covers fall. Watch the mix after any change, because a labor-inclusive re-price can shift demand toward your simpler dishes in ways the sheet did not predict, and that shift can be good or bad depending on which dishes carry your reputation. It also assumes you can measure labor minutes honestly. A kitchen that has never timed its plates has to do that first, standing at the pass with a timer across a real service, because a labor cost you guessed is just a food-cost sheet with extra decimals. And it cannot fix a menu whose problem is structural: if the concept itself is built on hand-finished, fire-to-order plates and the payroll shock has made that model unviable, no amount of re-plating saves it. That is a format decision, not a menu one.
Section 7
The fitness test
You have re-engineered on labor, not just food cost, if you can point to a fully-loaded labor cost per minute that already reflects the shock, a contribution margin per dish that subtracts that labor, and a specific move assigned to each dish: protect the Stars, find the minutes in the Plowhorses, push the Puzzles, cut or strip the Dogs. You have not, if your response to a payroll shock was a flat percentage across the whole menu or a hunt for the worst food-cost dishes. That leaves your labor-heavy sellers under-priced, penalizes the dishes the shock never touched, and misses the one lever that actually recovers a labor cost: the minutes of skilled hands standing between the ticket and the pass.