Your medium size is not meant to be bought

Three iced white coffees on a kopitiam menu. Look at the steps, not the prices.

Small to medium costs you RM 3.50 for 150ml more. Medium to large costs RM 1.00 for exactly the same 150ml more. Standing at the counter, upgrading to large feels almost free. That is not an accident, and the medium is not there to be sold. The medium exists to make the large look smart.

And the house wins the trade-up. Going small to large adds RM 4.50 of revenue against maybe RM 1.20 of coffee, milk and a bigger cup, so roughly 73 percent of the upsize flows straight to gross profit. The expensive part of serving you, which is rent, staff and the transaction itself, was already paid on cup one.

Once you see the structure you see it everywhere: cinema popcorn, bubble tea, fast-food combos, cloud storage tiers, software plans. The technical name is a decoy or anchoring structure. The practical name is price-pack architecture.

Three rules a ladder should obey

Price the steps, not the packs. The gaps between sizes steer the choice. Make the step toward the size you want to sell feel small and the step away from it feel expensive. Get this wrong and shoppers cluster on your entry size while the premium pack rots on shelf.

Key price points are cliffs, not slopes. RM 9.90 and RM 10.20 are not thirty sen apart, they are in different mental buckets. A smaller pack engineered to sit under the barrier at a higher price per gram is one of the most reliable margin-accretive moves there is. Leave the price point empty and a competitor or a private label takes it.

Never let price per unit invert by accident. Bigger should be cheaper per gram, unless you are deliberately taxing convenience with sachets, on-the-go or single-serve. An accidental inversion gets arbitraged fast: the provision shop buys your family pack, refills the small packs, and you have funded your own competitor.

That is the demand side. Most companies stop here. The expensive mistake lives on the other side.

The side shoppers never see

A manufacturer sells three screen sizes. The shelf shows a tidy price ladder. The factory tells a completely different story.

The 60 inch has the biggest price tag and nearly the worst margin. The 42 inch looks like the affordable entry hero and can barely fund a discount. The 50 inch is the profit sweet spot, and the promotional plan should say so out loud.

• 42 inch. No promotion. It cannot afford one.
• 50 inch. Preferential depth, say 8 percent. Pull volume into the sweet spot.
• 60 inch. Token depth, say 3 percent. Stay in the conversation, protect the halo.

Run gross profit per unit after discount. The 42 inch at list makes RM 200. The 50 inch, even after 8 percent off, makes RM 330. The 60 inch after 3 percent makes RM 240. The discounted SKU is still the most profitable unit you can sell, and every share point you steer into it pays.

Two conditions, or steering does not pay

Depth is a certain cost. Share shift and volume growth are a bet. Steering only pays if both land.

1. Share actually moves toward the sweet spot. If shoppers take the discount on the 50 inch but the mix stays where it was, you have simply reduced the price of your best SKU.
2. Total volume actually grows. In the example above the steering plan adds about 8 percent to total gross profit, but only because volume grew alongside the share shift. Hold volume flat and the identical promotion destroys 17 percent of gross profit.

Which means the honest promotional question is not what depth do I need. It is what share and volume outcome does this depth have to buy in order to pay for itself, and do I believe that outcome?

The two questions for your own portfolio

Demand side: what does each price step tell your shopper to do, and is your medium doing its decoy job or actually stealing sales from the size you want to sell?

Supply side: which SKU is your 50 inch, meaning best real cost position rather than best margin on the standard costing sheet, and does your promotional calendar preferentially feed it, starve it, or, most common by far, not know it exists?

Most promotional calendars are written entirely from the demand side. Depth goes wherever the buyer asked for it, blind to the cost curve underneath. That is not a promotional problem or a pricing problem. It is an architecture problem, and it is fixable in one planning cycle, once someone owns both sides of the ladder.