Your first week with a new POS system

Most shops that change POS system do it in one weekend, because that feels like the responsible way to do it. Close on Saturday on the old system, open on Monday on the new one, and hope the middle went well.

It usually does not. The catalogue turns out to be half loaded, nobody has practised a refund, and the first genuinely awkward transaction happens with four people waiting.

A staged week works far better, and it costs nothing extra. Here is an order that holds up.

Day one: get the products in, badly

The instinct is to import a perfect catalogue. Resist it. A perfect catalogue is a project that takes three weeks and stops the switch happening at all.

Export what you have from the old system, or from the spreadsheet where it actually lives, and import it with four fields per product: name, barcode, price, and cost. That is enough to sell.

Categories, images, suppliers and descriptions are all improvements you can make while the shop is running, one aisle at a time, and they are much easier to do once you can see the products in the system.

Two things are worth getting right at this stage, because they are expensive to fix later. Barcodes have to be exact, since a wrong one is worse than a missing one. And cost prices have to be real, because every margin figure you ever look at is built on them.

Day two: count the stock you are about to trust

Opening stock figures are the part people skip, and then wonder for six months why nothing reconciles.

You do not need to count the whole shop. Count your top twenty lines properly and enter those. Set everything else to zero and let the figures build from what you receive and sell from here on, or count the rest over the following fortnight in the same rolling way you would run any stock take.

What matters is knowing which figures are real. A shop where twenty products have trustworthy stock and the rest openly do not is in a much better position than one where every number is a guess wearing a uniform.

Day three: set up the people, not just the till

Create a login for every person who works the counter, and give each one only what the job needs.

This is the step that is always postponed, and it is the one that pays. Separate logins mean every sale, discount and refund carries a name. That is not about distrust. It is about being able to answer “what happened on Tuesday afternoon” without interrogating anyone.

Keep refund permission narrower than sell permission. Everyone needs to sell. Not everyone needs to be able to send money back out of the till.

Do this before training rather than after, so people learn on their own login and never share one out of habit.

Day four: rehearse the awkward transactions

Selling is the easy part, and it is the only part most training covers. Then the first difficult moment happens live.

Spend an hour on the ones that actually come up. A part-paid sale. A discount on one line rather than the whole basket. A return with a receipt, and one without. An exchange for a different size. A sale interrupted halfway through because the customer went back for something, which means parking the basket and picking it up again.

Have each person do each of these once, on a real till, with a made-up customer. It is twenty minutes each and it removes almost all of the first-week panic.

Day five: run both systems for a day

On the last day before you switch properly, ring a normal day’s sales through the new system while the old one stays available as a fallback.

You are not testing whether the software works. You are testing whether your shop works on it: whether the barcodes scan, whether the prices are the ones on the shelf, whether the receipt prints the way you want, and whether the person who is quietest about training is actually fine.

Every problem found on this day is a problem found without a queue.

The weekend: switch, and keep the old one readable

Switch on a quiet morning rather than a busy one, and keep the old system readable for at least a quarter. You will need it for a warranty question, a supplier query, or a comparison against last year.

If the old system is a subscription, check what happens to the data when you stop paying before you stop paying. Some go read-only, some export on demand, and some simply close. Export everything you might want while the account is still fully alive, even if you never open the file.

Week two onwards: improve the boring half

Once the shop is running, the catalogue is where the remaining value is. Add suppliers so purchase orders work. Set reorder levels on the lines you cannot afford to be out of. Tidy categories so the till screen matches how you actually think about the shop.

None of it is urgent, all of it compounds, and doing it a little at a time is the only version that ever gets finished.

One thing to check before you start

Whatever system you are moving to, confirm two things on day zero: that you can get your data back out yourself, in a file, without asking anyone; and what happens if the internet drops mid-sale.

StoreWave imports products from a file to get you started and exports every report to one, so the first answer is yours to check rather than ours to promise. On the second, checkout retries by itself and tells you when the connection is back, and the catalogue stays readable through a short drop, so a brief outage is an inconvenience rather than a closed shop.