The most common starting point is to price from your target food cost percentage. Cost the plate accurately, decide the food cost you want that item to run, and divide:
Menu price = plate cost ÷ target food cost %
Example: a $4.00 plate at a 30% target → 4.00 ÷ 0.30 = $13.33 (round to $13.50 or $13.95).
The same math as a multiplier: 1 ÷ 0.30 = a 3.3× markup. A 33% target is roughly 3×. This gives you a floor, not a final answer — it ignores how popular an item is and what the guest will happily pay.
Menu engineering plots every item on two axes — popularity (how often it sells) and contribution margin (price minus plate cost). The classic four categories:
| Category | Profile | What to do |
|---|---|---|
| Stars | High popularity, high margin | Protect them; feature prominently; hold quality. |
| Plowhorses | High popularity, low margin | Nudge price up, trim plate cost, or re-portion. |
| Puzzles | Low popularity, high margin | Re-name, re-place, or promote — the money is good if it sells. |
| Dogs | Low popularity, low margin | Fix, re-price, or cut. |
This framework, popularized by Kasavana and Smith, is why chasing food-cost percentage alone can mislead: a dish with a “bad” 40% food cost but a high dollar margin may out-earn a “good” 25% item that barely sells.
Food-cost math sets a floor, but two other approaches keep it honest. Competitor / market pricing asks what comparable places charge for a similar dish in your area — you can sit above or below, but you should know where you stand and why. Value (demand-based) pricing asks what the guest perceives the item is worth: a signature dish, a hard-to-copy specialty, or anything with a great story can carry a price well above its food-cost floor, while a commodity item like a soft drink or a side salad is judged against what everyone else charges. The best price is usually the highest of the three lenses that your guests will still happily pay.
Individual prices have to add up to a viable business. After setting them, check your blended food cost, your prime cost, and whether your projected mix clears break-even. Re-price at least once or twice a year as supplier costs move; small, regular increases are absorbed far better than one big jump.