Home/Blog/Apparel

How to choose a POS for a clothing store

Size runs, colour matrix, returns against the original invoice, inter-store transfers and season-end markdown. A practical test list you can run in an afternoon.

Most POS comparisons are done on feature lists, which is the least useful way to choose one. Every system will claim inventory, reporting and multi-store. The differences show up when you put your own products through it.

Here are eight tests you can run in an afternoon, in a trial environment, that predict whether the system will still be working for you in year three.

1. Enter one style across a full size run

Take a real shirt: four colours, sizes XS to XXL. Create it. Count how many clicks it takes and whether the system understands that these are one product with a grid, or 44 unrelated products.

If it is the latter, every future report will be an aggregation you build by hand, and your size curve will never be visible. This single test tells you more than any feature list.

2. Sell a medium, then look at the stock report

Does the report show you the size-level quantity, or just a style total? Can you see, at a glance, that you have plenty of smalls and no mediums?

The test is whether replenishment decisions can be made from the report. If you still need to walk to the stockroom to know what to reorder, the report is decoration.

3. Return one item and follow it through

Process a return. Then check three things: did the stock come back, did the margin reverse correctly, and is the transaction linked to the original sale?

Systems that fail this test will produce a stock count you cannot trust within six months. It is the single most common gap we find when replacing an older POS.

4. Transfer stock between two stores

If you have more than one location, do a transfer in the trial. Watch what happens to the quantity at each end between dispatch and receipt.

If the stock disappears from store A immediately and appears at store B immediately, you have no in-transit state — which means every transfer is a chance for the numbers to diverge silently.

5. Print a receipt in the second language

Set one user to English and one to Chinese (or whichever pair you need). Ring a sale on each. Print both receipts.

Look for: is the layout broken, are the labels actually translated, does the tax line make sense in both? A translated interface that produces a mangled receipt is worse than no translation, because customers see it.

6. Change a price and check history

Change a product's price, then look at an old transaction. The old transaction must keep the price it was actually sold at.

This sounds trivial and is surprisingly often broken. It matters most when your tax rate changes, because you need historical invoices to reflect the rate in force at the time.

7. Ask what happens when the internet drops

Unplug the network during a sale. Does the till keep working? When it reconnects, does the sale appear once, or twice?

Double-booking revenue is worse than losing a sale, because it corrupts reports you rely on. Ask specifically about duplicate protection on replay, not just "offline mode".

8. Export everything, then open the file

Run a full export and open it in a spreadsheet. Are the fields readable, or is it a database dump with numeric codes?

If the export is unusable, you are locked in regardless of what the contract says. Test this before you sign, not when you want to leave.

A ninth test, if you have time: ask them to load 50 of your real products — with your actual sizes, colours and prices — before the demo. Vendors who will not do this are telling you something about implementation.

What to ignore

  • Feature counts. Ninety features you will never configure are worth less than five you use daily.
  • Screenshots. Every system looks good in a screenshot with three products in it.
  • "AI-powered" anything. In retail POS today this usually means a dashboard nobody opens.
  • Customer logos. Ask for a reference in your country, in your industry. If there is none, the logo wall was not for you.

The decision that matters most

Almost every system on the market will run a clothing shop adequately on day one. The differences appear at month six, when the stock count drifts, the season changes and someone asks a question the reports cannot answer.

Choose the system whose reports you would actually use to make a buying decision. That is the real test, and it is the one most buyers never run.

FAQ

Questions we get on every first call

How many products should we load for a fair trial?

Fifty to two hundred covering your main categories, including your most awkward ones — the products with the most variants and the most unit conversions.

Should we switch during our peak season?

No. Switch in your quietest month, ideally with two weeks of parallel running before you stop using the old system.

What if we only have one store?

Tests 1, 3, 6 and 8 still apply. Multi-store tests can be skipped, but a system that fails the return test will still fail you at one location.

Want this applied to your numbers?

Send a sample of your stock report. We will tell you which of these is happening, on your data.

Request a free trial WhatsApp us sales@gdyun.online  ·  Reply within one business day  ·  Support 09:00–21:00 GMT+8