Card payments in Malaysia and what they cost
Cards are a smaller share of Malaysian online payments than FPX, and a more expensive one. They are also the payment type whose cost varies most, in ways that are invisible on a rate card.
Why the headline rate is a starting point
Your gateway quotes a rate. Your actual blended cost will be higher, and how much higher depends on your customer mix rather than on your negotiation.
Card type matters. Premium and rewards cards typically carry higher interchange than basic ones, because somebody funds the rewards. Your customers' choice of card is not something you control.
Issuing country matters most. A card issued outside Malaysia usually costs materially more than a domestic one. For a store with international customers this is not a rounding difference, and it is the single largest source of variation between a quoted rate and a real one.
Corporate cards cost more than consumer cards.
The consequence: two stores on identical rates can have noticeably different real costs, driven entirely by who shops with them. Your effective card rate is worth measuring from your own settlement data rather than read off an agreement.
Authentication and what it shifts
Malaysian card payments generally involve 3-D Secure — the step where the customer confirms through their bank, often by one-time code.
Two effects, pulling in opposite directions.
Liability moves. With successful authentication, liability for certain fraud-related disputes generally shifts away from the merchant. Genuinely valuable — see chargebacks on your own store.
Conversion drops. Every extra step loses customers. Some abandon at the bank page, some never receive the code, some give up.
That trade is worth being deliberate about rather than accepting as a default, particularly if a meaningful share of your card attempts fail at authentication. That failure rate is measurable and most merchants have never looked at it — see failed payments and the orders they leave behind.
Where card fees hide in your accounts
Blended into one rate. The most common error. If cards, FPX and wallets are averaged, per-order margin is wrong by however far your mix differs from the average, and the mix moves with promotions and traffic sources.
Cross-border premium not separated. International orders can look as profitable as domestic ones while costing more to collect. If you sell abroad at all, separating this is worth doing before deciding whether the international business is worth having.
Currency conversion recorded as a fee. It is not the same thing. Conversion is a rate difference, processing is a service charge, and merging them makes both impossible to manage — see currency conversion at checkout and who pays.
Refund fees uncounted. A refunded card order usually leaves its original fee behind — see what happens to the fee when you refund.
What to actually measure
Four numbers, from your own settlement data, none of which require special tooling once fees are recorded per transaction:
Effective card rate. Total card fees divided by total card revenue. Compare it to your quoted rate; the gap is your mix.
Domestic versus international rate, separately. The difference is often larger than expected.
Authentication failure rate. Card attempts that failed at 3-D Secure as a share of attempts.
Card share of revenue by order value band. Cards tend to dominate at higher values where an FPX flat fee would be cheaper, which is worth knowing before deciding whether to nudge payment method.
The comparison that follows
Once you have a real effective card rate, the comparison with FPX becomes concrete rather than theoretical.
FPX tends toward a flat fee, cards toward a percentage. There is an order value above which FPX is materially cheaper. Below it, cards may be competitive or better.
Whether that is worth acting on depends on how much of your volume sits either side of that line. For a store with a high average order value and meaningful card share, it can be worth a gentle nudge at checkout. For a store selling low-value items where FPX already dominates, it is not worth the effort. Either way the answer comes from your own data, and it cannot be produced from a blended fee rate.
Common questions
Why is my real card processing cost higher than my quoted rate?
Because the quoted rate is a floor. Premium and rewards cards carry higher interchange than basic ones, corporate cards cost more than consumer cards, and cards issued outside Malaysia typically cost materially more than domestic ones. Your effective rate is driven by which customers shop with you, not by your agreement.
Does 3-D Secure help or hurt a Malaysian merchant?
Both. Successful authentication generally shifts liability for certain fraud-related disputes away from the merchant, which is genuinely valuable. It also adds a step that loses conversion — customers abandon at the bank page or never receive the one-time code. The authentication failure rate is measurable and worth looking at before treating the trade-off as settled.
Should international card payments be tracked separately?
Yes. Cards issued outside Malaysia cost noticeably more to process, so international orders can appear as profitable as domestic ones while costing more to collect. Separating them is worth doing before deciding whether international business is worth pursuing, and it is the largest single source of variance between quoted and actual card rates.
How do card costs compare with FPX?
Card fees are typically proportional to order value while FPX tends toward a flat fee per transaction, so there is an order value above which FPX is materially cheaper and below which cards may be competitive. Where that line falls depends on your own fee schedule, and it cannot be identified while fees are recorded as one blended rate.
Related: FPX settlement and what it does to your cash · Malaysian payment gateways and how they actually differ · chargebacks on your own store
Read next
See what you could build
Start a free trial and describe what your business needs in plain language — SmartB Studio builds the module for you.
Start free trial