Pricing your own panel: the margin that survives after refunds

The most common way a reseller loses money is not being undercut. It is setting a margin that looked reasonable and never checking what came out of it afterwards.

The number people start with

Cost is ৳27 per 1,000. You sell at ৳31. Margin: ৳4, about fifteen percent. Sell a million units a month and that is ৳4,000. Straightforward.

It is also wrong, because four separate things come out of that ৳4 before any of it is yours.

One: partial deliveries

Your customer's order partials. You refund them the undelivered part; your supplier refunds you the undelivered part. Net effect on margin: zero, in principle.

In practice it is not zero, because your refund and the supplier's do not always match exactly, and because a partial delivery generates a support conversation that a completed one does not. Budget for the conversation rather than the money.

Two: orders you eat

This is the real one. A customer pastes a private profile link. Delivery fails. The supplier does not refund you, because they attempted it correctly and the target was the problem.

Your customer does not accept that. They paid, nothing arrived, and from where they sit that is your fault. You refund them to keep the account, and the loss is yours.

On a beginner customer base this is somewhere between two and five percent of orders. Against a fifteen percent margin, that is up to a third of it gone.

Three: payment fees

Mobile money and card gateways take a cut of every top-up, typically between one and two percent. It comes off the money coming in, not off the profit, which means it scales with revenue rather than with margin.

On fifteen percent, a one and a half percent gateway fee is another tenth of your profit.

Four: your time

The one nobody costs. Every order that goes wrong is a message thread. At a hundred orders a month with a five percent problem rate, that is five conversations — an hour, perhaps two, mostly explaining the same three things.

If your monthly margin is ৳4,000 and you spend eight hours a month on support, you are earning ৳500 an hour. That may be fine. It should be a decision rather than a discovery.

The arithmetic that actually holds

Take your headline margin and subtract:

• Three percent of revenue for orders you absorb
• One and a half percent of revenue for gateway fees
• A realistic hourly value for support time

A fifteen percent markup on ৳27 gives ৳4.05 per 1,000. Absorbed orders take about ৳0.93, gateway fees about ৳0.47. You are at ৳2.65 before you have valued an hour of your own time — roughly ten percent, not fifteen.

That is still a business. It is a different business from the one the first calculation described.

What margin to actually set

Below ten percent you are working for the supplier. At ten to fifteen you have a real but thin business that lives or dies on order volume. At twenty to thirty you have room to absorb problems, and you compete on service rather than price — which is the only sustainable position, because there is always someone cheaper and they are usually about to disappear.

The panels that last are rarely the cheapest. They are the ones that answer messages and refund without an argument, and both of those are paid for out of margin.

Where to charge more, and where not to

A flat percentage across the catalogue is simple but leaves money on the table at both ends. On a service costing a fraction of a taka per 1,000, fifteen percent is a fraction of a paisa and does not cover the support that order might generate. On an expensive service, fifteen percent is a real number and your customer is comparing it.

A common answer is a higher percentage on the cheap end and a lower one on the expensive end. It reads as generous on the services people price-check and it covers your costs on the ones they do not.

The two numbers to watch monthly

1. Absorbed refunds as a share of revenue. If it climbs past five percent, the problem is not pricing — it is that customers are ordering the wrong services, and the fix is better guidance at the order form.
2. Support messages per hundred orders. This is the number that decides whether the business scales. Ten is manageable; thirty means you have bought yourself a job.

Both are lowered by the same thing: telling customers what a service will do before they order it, rather than after.

Do not compete on being cheapest

There is always a panel a paisa below you, and it is usually one of two things: someone who has not done this arithmetic yet, or someone who does not intend to refund anybody. Neither is a competitor you need to match, and both tend to disappear.

What customers actually leave over is not price. It is an unanswered message after a failed order. A reseller who replies the same day and refunds without an argument holds customers at twenty percent over the cheapest option, indefinitely, because the alternative has already burned them once.

The margin that pays for itself

The cost of the second and third things on that list — absorbed refunds and support time — falls sharply when customers order the right service in the first place. Every line of guidance you put next to the order form is margin you keep.

Telling a customer that a service says Max 5k so they should not order 5,000, or that a private profile will fail before they pay for it, costs you one sentence and saves you a refund plus a conversation. That is the highest-return work available in this business, and it is almost entirely free.

Review the numbers quarterly, not never

Costs move — the supplier's, the exchange rate, the gateway's. A margin set once and never revisited is a margin that quietly became something else. Once a quarter, take the same two figures: what you absorbed, and what you actually netted per thousand units sold. Ten minutes of arithmetic, and it is the difference between running a business and hoping.

No comments yet — be the first.