Menu engineering for Malaysian restaurants: where to start
Every restaurant has dishes that sell well, dishes that make money, and a few that do neither. The trouble is that they are rarely the same dishes. Menu engineering is a simple way to see which is which, using data your point-of-sale system already holds, and then to reshape the menu so the dishes that sell are also the ones that earn.
What menu engineering measures
Menu engineering looks at each dish on two measures:
- Popularity: how many of that dish you sell compared with the rest of the menu section.
- Contribution margin: the selling price minus the food cost of one portion. This is the ringgit amount each plate contributes towards labour, rent and profit.
Margin in ringgit matters more than food-cost percentage here. A dish with a 40% food cost that sells for RM38 may contribute more per plate than a dish with a 25% food cost that sells for RM14.
The four groups
Plotting every dish against those two measures sorts the menu into four groups. The names come from the classic menu engineering model used in hospitality schools:
| Group | Popularity | Margin | What to do |
|---|---|---|---|
| Stars | High | High | Protect. Keep the recipe and portion consistent and give it the best spot on the menu. |
| Ploughhorses | High | Low | Re-engineer the cost: portion, garnish, ingredients or supplier. Consider a small price rise or a high-margin pairing. |
| Puzzles | Low | High | Help it sell: a clearer name and description, a better position on the menu, or staff recommending it. |
| Dogs | Low | Low | Remove or replace, unless it plays a specific role, such as a dish a loyal group of regulars expects. |
A practical way to start
1. Pull four to eight weeks of item sales
Export quantity sold and selling price for each item from your POS. Use a normal trading period and leave out festive weeks or heavy promotions, which distort popularity. Analyse each menu section separately: mains against mains, drinks against drinks.
2. Cost every dish properly
Use a standard recipe with current supplier prices, including garnishes, sauces and the packaging used for takeaway. Most margin surprises come from dishes that were costed once at opening and never again.
3. Calculate and classify
For each dish, work out its contribution margin and its share of the items sold in that section. A common rule of thumb treats a dish as popular if it sells at least 70% of an even share. In a section of ten dishes, an even share is 10%, so any dish above 7% counts as popular. Compare each dish's margin with the section average to decide whether it is high or low.
4. Act on the groups, one change at a time
Change a few things, then measure for four weeks before changing more. If you re-price, re-describe and move a dish all at once, you will not know which change worked.
Menu design matters as much as the maths
Once you know which dishes to promote, the printed or digital menu has to do the work:
- Give Stars and Puzzles the most visible positions, typically the top of each section and any boxed or highlighted area.
- Write short, specific descriptions for the dishes you want to sell. Name the key ingredient or the cooking method.
- Avoid lining prices up in a column down the edge of the page, which invites diners to choose on price alone.
- Keep sections short. Long menus slow ordering, add prep work and increase waste.
- On QR and online menus, use photos and the order of items deliberately. The first items in a list get the most attention.
Common mistakes
- Cutting a popular dish because its food-cost percentage looks high. If it brings in a strong ringgit margin and draws guests, it may be one of your best dishes.
- Raising prices across the board. Targeted changes to Ploughhorses usually protect margin with less pushback from regulars.
- Ignoring the kitchen. A dish that is profitable on paper but slows the line at peak can cost more in lost covers than it earns.
Where to go from here
Menu engineering is most useful as a routine rather than a one-off project. If your POS and recipe costs are set up well, the analysis can be refreshed every quarter in under an hour, and it gives you an evidence-based answer the next time a supplier raises prices.
Frequently asked questions
How often should a restaurant re-engineer its menu?
Review the numbers at least every quarter, and whenever a major supplier price changes. A full redesign is usually needed less often, but the analysis should be routine.
Do I need special software for menu engineering?
No. Item sales from your POS and a costed recipe for each dish are enough to start, and a spreadsheet will do the calculation. Some POS and inventory systems can automate it once recipes are set up.
