What UniShopping does not do

Every platform has a list like this. Most of them keep it internal, and you find out on a Tuesday afternoon with a customer on the phone.

Here is ours, up front. Each of these is a deliberate choice, not an oversight — we would rather ship a shop whose money and stock are correct than a longer feature list that is approximately right. Where there is a way to get the job done anyway, it is written underneath.

No exchanges

You cannot swap a customer's size 10 for a size 12 in one movement. There is no flow that returns one item, reserves another, and settles the difference.

Do this instead: refund and re-order. Refund the return to store credit and the customer spends it on what they actually wanted, in one visit and without their card leaving their pocket — which covers most of what people mean when they ask for an exchange. Refund to the original card instead and they are credited and charged separately, which is worth telling them before they see two lines on their statement.

Either way the return records the reason and restocks the original, so your stock stays right. If the price differs, the two transactions differ — that is the honest version of an exchange, just with more paperwork.

No gift cards

You cannot sell a gift card, and there is no code for a customer to redeem at checkout or at the counter. This one is not on the way: gift cards are an anonymous, transferable liability with unclaimed-property rules that vary by state, and that is a bigger commitment than it looks from the outside.

Do this instead: a discount code, if what you want is a promotion. It is not the same product — a discount code is not money a customer has already paid you — but it covers most of what small shops reach for gift cards to do.

Store credit is a different thing, and it is here. It is money held against a named customer's account at one shop, usually as the result of a return, spendable online or at the counter as a tender applied after tax. It is not covered by the decision above precisely because it is not transferable: it belongs to a customer you have a record of, which is what keeps it outside the unclaimed-property rules. Every issue, redemption and expiry is a ledger row, so what you owe reconciles to the cent. See Returns and refunds.

Domestic selling only

One shop, one currency, one country. There is no second price list for another market, no duties or import charges calculated at checkout, and no cross-border tax handling. A customer overseas can buy from you, but they will pay in your currency and sort out their own customs.

Do this instead: if you sell meaningfully into a second country, price for it deliberately — a shipping rate that covers the freight, and a note on your shipping policy page saying duties are the buyer's. That is what most shops your size do regardless.

No live carrier rates

Checkout does not ask Australia Post, DHL or FedEx what a parcel costs. Rates come from what you set up yourself: flat rates, rates that vary by order value, and rates that vary by weight.

Do this instead: set weight bands from your own carrier's price list and check them once a quarter. Weight-based rates get most shops within a few percent, and they never leave a customer staring at a spinner because a carrier's API is slow.

This matters most for heavy or oversized goods, where a flat rate quietly loses money on the long shipments and overcharges on the short ones.

No order editing after placement

Once an order exists, its lines are fixed. You cannot add an item, remove one, or change a price on it.

Do this instead: cancel the order and place it again with the correct lines, or — when the customer only wants to add something — take a second order. Cancelling releases the units the order was holding and closes its fulfilments, so nothing is left half-committed; putting stock back on the shelf stays your call, because a cancelled order's goods may already be in a van.

The one thing you should not do is fulfil an order that no longer matches what the customer is paying for. The records are the same records your reports, your tax and your accountant read.

Search is Postgres-based

Storefront search is built on the database, not on a dedicated search index. It handles ordinary searching well: product titles, descriptions, partial words. It is good to roughly 2,000 products.

Past that you will start to notice it — searches get slower, and results get less relevant before they get slower. There is no typo tolerance, no synonyms, and no learning from what customers click.

Do this instead: while your catalog is under a couple of thousand products, lean on categories and collections for browsing rather than expecting search to carry it, and put the words customers actually use into your product titles. If you cross that size, tell us — the fix is a real search index, and the trigger for building one is a shop that needs it.

No app ecosystem

There is no app store, no third-party plugins, and no way for an outside developer to add functionality to your shop. Every integration is one we have built ourselves — Stripe for payments, Resend for email, the printer clients — or it does not exist.

Do this instead: read the data rather than install something. Products, orders, customers and inventory export to CSV; invoices, payments, refunds and payouts export in a shape Xero and MYOB will import; a webhook tells another system when an order is created, paid, fulfilled or refunded; and a read-only API token, scoped to one shop and revocable, lets a tool fetch what it needs on its own. See Integrations.

What none of that is, is a write API — an outside system cannot change your catalog, your stock or your orders.

If you depend on a specific tool, ask before you commit to us rather than after. A first-party integration is a thing we can decide to build; an app ecosystem is not something you can work around at the edges.

Why the list exists

None of the above is on this page because we forgot. Each one buys something: no exchanges kept the returns path simple enough to be correct, no order editing kept payment, stock and fulfilment from multiplying into a state space nobody can test, no app ecosystem kept every integration something we can actually support.

That trade is only fair if you can see it before you commit. If one of these is a deal-breaker for your shop, better to know now — and better still to tell us, because what merchants are actually blocked on is how this list gets shorter.