Loyalty that pays for itself

Most loyalty programs are run as a cost: you spend margin to rent retention, and finance spends every quarter trying to make the spend smaller. That framing quietly caps how ambitious the program is allowed to be. Flip it, and everything downstream changes.

The bet I would make again: build the loyalty program as its own P&L, not as a discount line. If the program earns its own revenue, you can afford to be generous where generosity actually moves behavior, instead of defending a budget.

At Traveloka I led and grew Points, the company's first loyalty program, across Indonesia, Singapore, and Malaysia. The move that mattered was not richer rewards. It was building the program a revenue line of its own: points-exchange and co-brand partnerships with banks, so Points generated income while it drove retention. Once it paid for itself, the question stopped being "how much can we afford to give away" and became "what behavior do we want to buy."

Three things I would tell anyone starting here.

1. Decide whether it is a cost center or a business before you design a single screen. The answer changes the roadmap, the partnerships, and who you hire. Most teams never make the call explicitly, so the program defaults to a discount and stays small.

2. Earn the moments inside the journey, not on a rewards tab. Points only change behavior if they show up where the decision happens, at search and checkout, when someone is weighing price. Hidden in a separate tab, they are a line item nobody reads.

3. Find the revenue partner the program unlocks. A loyalty base is a distribution asset. For us it was banks, points-exchange and card benefits. The partner pays for the generosity, and the generosity drives the retention. That loop is the whole game.

The reason this is worth the trouble: a program that funds itself can keep getting more generous, and generosity is what people actually tell their friends about. A discount line can only ever shrink.

If your loyalty program is a budget the CFO wants to cut, you built a cost. Build a business instead.