Accepting konbini payments in Japan

Konbini payment lets a Japanese customer order online and pay in cash at a convenience store. The customer receives a payment code, walks to a konbini and settles at the counter. It remains a meaningful share of Japanese e-commerce, and it behaves nothing like a card.

Sequence

The order and the payment are separate events

This is the operational point that matters most. A konbini order is unpaid at checkout and may stay unpaid until it expires. Treating the order as complete means over-reporting revenue, reserving stock unnecessarily and recording abandonment that is actually non-payment. Any Japanese operation using konbini needs an explicit expiry window, a release process and reporting that separates orders placed from orders paid.

Why Japanese consumers still use it

Cash preference, reluctance to enter card details online and the sheer density of convenience stores all sustain it. It also serves customers who hold no card. Removing it to simplify your checkout removes a real segment, so the decision should follow from your own paid-order data rather than from an assumption about how Japanese commerce ought to work.

Fees

Where the cost sits

Typically a fixed fee per issued code or per paid order rather than a percentage, which makes it comparatively cheap on high basket values. The genuine cost is the working capital tied up between order and payment, and the operational overhead of managing expiries. Whether that is worth paying depends on your unpaid rate, and that number is measurable.

Expiry

The expiry window is a commercial setting, not a technical one

Every konbini integration has an expiry window, and in most businesses it is whatever the provider default was on the day of the integration. It is doing two opposite jobs at once. Too short and you cancel orders that would have been paid. Too long and you hold stock against orders that never will be.

The right length is not a default, it is a distribution. Take twelve months of konbini orders that were paid and count how many were paid on day one, day two, day three and so on. Almost every merchant finds the curve collapses after a small number of days and then produces a long, thin tail. Set the window a little past the collapse and both ends of the problem are solved with one number.

Then the setting next to it, which is usually nobody’s decision at all: whether stock is reserved when the order is placed or when the payment arrives. That is a merchandising choice being made by a payments configuration, and on a fast-moving product in a peak week it is the more expensive of the two.

Which payment provider supports konbini in Japan?

GMO Payment Gateway, SB Payment Service and KOMOJU in Japan, alongside dLocal, all document konbini, among others, as do the other Japanese payment providers. The useful questions are the cost per issued code versus per paid order, your unpaid rate, how long stock is held, and whether your reporting distinguishes ordered from paid.

Segment

The customer with no other route

The page above says removing konbini removes a real segment. That is true and, as stated, unmeasurable, because the question is not what share of revenue konbini carries. It is what share of konbini buyers have ever paid you any other way.

Produce it like this. Take twelve months of konbini-paid orders, match them on customer identity, and split into two groups: customers who have also paid you by card or another method at some point, and customers who never have. The first group would probably have bought regardless. The second group is the segment, and it is the only part of the number that answers the question.

Then it decides itself. Take 800,000 euros of konbini-paid revenue where 45 per cent comes from customers who have never paid you another way, so 360,000 euros. At a 40 per cent gross margin that is 144,000 euros of gross profit with no alternative route, against a fixed fee of roughly one euro per paid order, which on 4,000 orders is 4,000 euros. That is not a close decision. If instead the second group is 10 per cent, the method is mostly a convenience for customers who would have paid another way, the fee is still trivial, and the expiry window and the stock reservation above deserve far more of your attention than the method itself.

Reviewing what this costs you

What you pay is set in your agreement, not by the scheme. Start by establishing whether you are overpaying your PSP, or look at how a structured payment RFP settles market coverage before you sign rather than after.

Relevant markets: Japan