Three couriers, one delivery promise

A customer orders from your storefront on Monday and from your Shopee listing on Tuesday. Same customer, same shop, same two items. One arrives by the courier you chose. The other arrives by whichever courier the marketplace assigned. Different tracking number, different app, different delivery window, and if either goes wrong they will ask you about it, not the platform.

That is the shape of fulfilment in a multichannel business. You do not get to pick one courier and be done, because you do not control how every channel ships.

You are running two fulfilment models at once

Marketplace orders mostly ship through the platform's own logistics arrangement. Sell on Shopee and the consignment is likely to move through SPX Express. You get a label, a pickup, and a status feed you do not control and cannot negotiate.

Storefront orders are yours to route. You choose based on destination, weight, cost and how quickly you have promised it: Ninja Van for domestic parcels, DHL when it is going further or needs to be fast.

Both models work. The difficulty is that they produce completely different operational lives. One gives you no choice and full visibility through the platform. The other gives you full choice and visibility only if you go and get it.

And the customer experiences neither of those distinctions. They experience one business that either delivered on time or did not.

The three costs of leaving it fragmented

Somebody becomes the tracking department. When a customer asks where their order is, the answer lives in a courier portal, and which portal depends on which channel the order came from. Multiply by the number of enquiries a day and you have quietly created a job nobody was hired for.

Freight cost stops being knowable. Quoted rates and billed rates are not the same thing. Surcharges, remote-area fees, weight corrections, failed-delivery redeliveries and returns all land after the fact. If the real cost never gets attached back to the order, then your margin per order is an estimate, and your margin per channel is a guess built on that estimate.

Nobody notices what stalled. An order that has not moved for four days is invisible unless something is looking for it. The first person to notice is normally the customer, which is the most expensive way to find out.

What actually needs to join up

Not the couriers themselves. They can keep working exactly as they do. Three things need to come back to the order.

The consignment, raised from the order. The despatch is created from the order record rather than re-keyed into a courier portal, so the address, contents and reference come from the same place the sale did. Re-keying is where wrong addresses come from.

The status, on the order. Not a link to a tracking page, but the delivery state sitting on the order itself, so anyone answering the customer can see it without knowing which courier was used or which channel it came from.

The real freight cost, back on the order. What you were actually billed, against the order it belongs to. This is the one that changes decisions, because it is the difference between knowing what a channel earns and assuming it.

That is the whole scope, and it is worth being precise about what it is not. SmartB does not plan routes, does not optimise a delivery run, and does not track a vehicle on a map. Those are the province of logistics software and it does not pretend otherwise. What it does is make sure the order knows where it got to and what it cost to get there.

Ship-from-store, and why it is a stock problem

Retailers with a shop and an online channel eventually notice they are holding stock in two places and running out in one of them while the other has plenty.

Ship-from-store fixes that, and click-and-collect is the same idea pointed the other way. Both are attractive and both fail for the same reason when attempted early: they need the stock figure to be trustworthy at a location level, not just in total. Promising a customer collection at an outlet that turns out not to have the item is worse than not offering collection at all.

So the order of work is the unromantic one. Get one stock figure across channels first, as covered in running a shop and a marketplace on one system. Get it trustworthy per location second. Only then offer customers a promise that depends on it.

The returns tail

Returns are where multichannel fulfilment gets genuinely awkward, because the return path rarely matches the outbound one.

A marketplace return follows the platform's process and its timetable. A storefront return comes back however you arranged it. The item returns to stock, or it does not because it came back damaged. The refund happens, and the original commission may or may not reverse in proportion.

Each of those has a money consequence that belongs to the original order, which is why returns sit at the join between fulfilment and reconciliation rather than neatly inside either. A return that is recorded as a stock movement but never reconciled against the fee treatment of the original sale is a small, permanent error repeated at volume. Where omnichannel money actually lands covers that side of it.

Common questions

Can I use my own courier for marketplace orders?

Usually only within whatever the platform permits, which varies and changes. In practice most sellers ship marketplace orders through the platform's arrangement and keep their own courier accounts for storefront and direct sales. The aim is not to unify the couriers, it is to unify what you know about the deliveries.

Does SmartB plan delivery routes or track my vehicles?

No. Route planning, delivery-run optimisation and vehicle tracking are outside what SmartB does. It works at the order level: raising the consignment, bringing delivery status back onto the order, and attaching the real freight cost to it.

Why does the billed shipping cost differ from what I quoted?

Surcharges, remote-area fees, weight or dimension corrections, redeliveries after a failed attempt, and return legs. None is unusual and all of them arrive after the order looked finished. They only distort your margin if nobody attaches them back to the order.

Should I offer click-and-collect?

Only once your stock figure is reliable at the location level. Click-and-collect converts a stock accuracy problem into a customer-facing promise, so it rewards businesses that have already unified stock and punishes those that have not.

One business, whatever the label says

The customer sees a delivery from you. They do not see the marketplace's logistics arrangement, your courier account, or which of the three you happened to use.

Making that true operationally does not require running a single courier. It requires the order to know what happened to it, regardless of who carried it.