Menu Engineering and Food Cost

Menu engineering ranks dishes by how often they sell and by how much cash they leave after food. Food cost percentage ranks them by a ratio. The two lists are not the same, and managing only the ratio will push you to promote a cheap plate that does not pay the kitchen and to apologise for a dish that does. AKLI keeps the plate cost current so the second number is not a quarterly guess.

Two different measurements

Contribution margin is money per plate. Food cost percentage is a share of the selling price excluding VAT. A dish at SAR 18 with a SAR 7 plate cost is 38.9% food cost and SAR 11 of contribution. A dish at SAR 95 with a SAR 50 plate cost is 52.6% food cost and SAR 45 of contribution. The percentage says the expensive dish is worse. The cash says it pays for more of the hour. Neither figure includes labor. Neither is actual food cost for the branch. That still comes from inventory. See how to calculate food cost percentage.

The four classes, using the menu’s own average

A standard way to sort a menu, associated with Kasavana and Smith, splits dishes into four classes. Popularity is the item’s share of units sold. The cut-off used here is the average share: with four dishes, that is 25%. Margin uses the average contribution of the dishes on the list, not a target you copied from another restaurant. High means above that average. Low means below it.

• –

  Star: high popularity, high contribution. Keep it. Do not casually cheapen the recipe.

• –

  Plowhorse: high popularity, low contribution. Guests already order it. Look at portion, yield, and price before you print more of it.

• –

  Puzzle: low popularity, high contribution. The cash per plate is good and the mix is not. Describe it, place it, or pair it. Do not assume the food cost percentage is the reason it sits.

• –

  Dog: low popularity, low contribution. A candidate to change or remove, after you check it is not a required side or a loss-leader you chose on purpose.

A worked illustration

These four dishes are an illustration so the arithmetic can be checked. They are not an AKLI customer, and they are not a recommended food-cost target.

• –

  Chicken burger: 400 sold, menu price SAR 28, recipe cost SAR 8, contribution SAR 20, food cost 28.6%.

• –

  Ribeye: 40 sold, menu price SAR 95, recipe cost SAR 50, contribution SAR 45, food cost 52.6%.

• –

  Green salad: 220 sold, menu price SAR 32, recipe cost SAR 4, contribution SAR 28, food cost 12.5%.

• –

  Lentil soup: 30 sold, menu price SAR 18, recipe cost SAR 7, contribution SAR 11, food cost 38.9%.

Units sold are 400 + 40 + 220 + 30 = 690. Mix is 58.0%, 5.8%, 31.9%, and 4.3%. The popularity cut-off is 25%. Average contribution is (20 + 45 + 28 + 11) ÷ 4 = SAR 26. The salad is above both cut-offs: a star. The burger is popular and under SAR 26: a plowhorse. The ribeye clears SAR 26 and misses 25% of mix: a puzzle. The soup is under both: a dog.

Sort the same list by food cost percentage and the salad wins, the burger looks acceptable, the soup looks poor, and the ribeye looks like the problem. The cash sort says the ribeye is the puzzle to sell more of, the burger is the popular plate to re-cost, and the soup is the one to change. You can hold both views. You should not manage the menu on the percentage alone.

Where the plate cost has to come from

The classes are only as current as the recipe cost. If the ribeye price on the card is last quarter’s invoice, the puzzle may already be a dog. Food cost per plate is how that card is built. AKLI costs recipes at weighted average cost: a new delivery at a new price moves the average, including sub-recipes and yield. That is recipe costing. Popularity still needs quantities sold: a POS sales import or a CSV. Without quantities you can still rank contribution. You cannot rank mix.

Contribution margin does not replace actual food cost. If the branch’s inventory COGS is higher than recipe cost times quantity sold, the menu classes are describing a kitchen you do not have yet. Close that gap with theoretical versus actual food cost and the food cost calculator before you reprint the menu on theoretical plates.

Groups and more than one brand

Run the classes per menu, not on a blend of two concepts. A shared commissary does not make the mix the same. Cost the shared batch once, transfer it at cost, and let each brand’s menu price produce its own contribution. That operating pattern is multi-brand groups. AKLI reports food cost per branch. AKLI gives you a live P&L by branch and for the whole group.