Choosing a retail POS in 2026: a practical checklist

Every POS vendor publishes the same feature grid, and every box is ticked. Inventory management: yes. Reporting: yes. Loyalty: yes. Integrations: yes. The grid is not lying, and it is also not useful, because it compares the existence of features rather than the quality of them — and in retail software the gap between "has inventory management" and "has inventory management that works" is the entire decision.

So here are five questions the grids never answer. They are deliberately awkward, and how a vendor responds to them tells you at least as much as the answers themselves.

1. Does inventory update itself?

Live inventory — decremented at the moment of sale, incremented on receipt, with low-stock alerts and reorder suggestions — is the single biggest operational upgrade a store can buy. It is also the feature most commonly claimed and least commonly delivered properly.

The test is simple: after this system is installed, does anyone still need a clipboard? If stock counts remain a weekend activity, you have bought a faster cash register with a report attached. Ask specifically what happens on a return, on a partial delivery, and on an item sold as part of a bundle — those three cases are where weak implementations reveal themselves.

Why this one outranks the others

Because retail is an inventory business. What is on the shelf, what it cost, how fast it moves, and how much of it walks out unpaid are the questions that determine whether a store makes money. Checkout speed matters, loyalty matters, but they are second-order next to knowing your stock position in real time.

2. Will it use the hardware you own?

Scanners, receipt printers, and cash drawers are commodities and have been for years. A modern browser-based POS should drive the ones sitting on your counter today.

Card payment terminals are the legitimate exception, since they are tied to a processor and carry certification requirements. Expect a supported list there. Expect nothing of the sort for scanners.

3. How long is onboarding, really?

Ask the vendor for the exact number of minutes from data import to first sale. Not a range. Not "it varies with your setup". A number, for a straightforward catalog. Pacific POS answers with about ten minutes, and publishes the four steps.

Vendors who answer in weeks are telling you about their architecture — on-premise installation, manual data entry, a technician's calendar. They may still be the right choice for you, but the answer is information, and their comfort with the question is more information.

4. Does it tell you anything?

Reports that list transactions are bookkeeping. You had that before. Insight is knowing your margin by category, which SKUs have not moved since spring, and which hour of which day actually carries your week — without exporting anything to a spreadsheet.

• Margin by category, not just revenue — revenue rankings routinely disagree with profit rankings, and only one of them should drive your buying.
• Dead stock surfaced before it becomes markdown, while there is still a decision to make.
• Sell-through velocity behind every reorder suggestion, so the suggestion is arguable rather than mysterious.
• Hour-by-hour trading patterns, which is what staffing decisions should be made against.

The data for all of this is already in your sales history. The question is whether the system assembles it for you or leaves it as an export.

5. What does year two cost?

Add the subscription, the payment processing margin, any hardware you would be locked into replacing, and the support tier that actually answers the phone. Then run that number over three years.

Payment processing is where the real money usually is, and it is the line most commonly left out of comparisons because it is expressed as a percentage rather than a monthly figure. A fraction of a percent on your card volume can comfortably exceed the entire software subscription. Convert it to dollars per month before you compare anything.

A note on what not to optimise for

Two things get weighted far too heavily in retail POS decisions. The first is the number of integrations, which sounds like optionality but is usually a list of things you will never connect. Ask instead about the two systems you actually use — your accounting package and your e-commerce platform, if you have one.

The second is the interface demo. Every POS looks good when a salesperson who has used it for two years drives it through a scripted flow. If you want a meaningful test, ask to enter your five most awkward products yourself, at speed, without help.