Cloud POS vs a traditional till: which suits a small shop

The comparison is usually framed as old against new, which is not useful to anyone actually deciding. Plenty of shops run happily on a till that was installed on a machine in 2019, and plenty of shops on cloud systems have had a bad afternoon because of a router.

The useful framing is this: both kinds of system have a single point of failure, and you are choosing which one. A locally installed till depends on that computer. A cloud POS depends on your internet connection. Everything else in the comparison follows from that one sentence.

The two architectures, briefly

A traditional or on-premise POS is software installed on a machine in the shop. The database is on that machine, or on a small server in the back. It runs whether or not anything outside the building is working.

A cloud POS keeps the database on a server elsewhere and reaches it over the internet. The till is a browser or an app that draws what the server sends. Many of them are web based, meaning there is nothing to install at all.

Where each one costs you

Upfront money. Traditional systems are usually a licence, bought once, often bundled with hardware the vendor supplies. Cloud systems are usually a monthly fee, sometimes with a free tier, and typically run on hardware you already own. For a shop opening its doors with a tight budget, this is not a close contest.

Money over five years. It gets closer. A one-off licence stops costing you anything, in theory. In practice it starts costing you in year three, when the version you own no longer runs on a current operating system, and the upgrade is another licence. A monthly fee is at least predictable. Work out both over five years rather than comparing a licence price to a monthly one, because that comparison always flatters the licence.

Your time. This is where the traditional system quietly loses, and it never appears on a quote. Somebody has to install it, keep the machine patched, run backups, and test that the backups restore. In a small shop that somebody is you, on a Sunday. A cloud system moves all of that to the vendor. You are paying a monthly fee partly to not be a system administrator.

Hardware you cannot reuse. A traditional POS bought as a bundle often only runs on the terminal it came with. When you outgrow it or it breaks, the hardware has no second life and no resale value. Hardware you already own does.

Where each one fails

The traditional system fails locally and totally. The machine dies, is stolen, or fills its disk, and the shop cannot sell. More importantly, if the backups were not real, the sales history is gone. Not degraded. Gone. This is the failure that ends up costing small shops the most, and it is common precisely because backing up is a task with no deadline until the day it has one.

The cloud system fails intermittently and partially. The connection drops and, depending on the system, you either cannot sell, or you can sell in a reduced way, or you barely notice. Your data is not at risk, because it is not in the building. What is at risk is the next twenty minutes of trade.

Notice the shape of the difference. The traditional risk is rare, catastrophic and permanent. The cloud risk is more frequent, milder and temporary. For most small shops the second is the better trade, but that depends entirely on how good your connection actually is, and that question deserves a proper answer rather than a shrug.

Where the cloud is straightforwardly better

Some of these have no counterargument.

• Multiple people, multiple devices. A second till is another browser, not another licence and another install.
• Looking at the shop remotely. Traditional systems can do this with work, a VPN and a static address. Cloud systems do it because that is what they are.
• More than one location. Two shops on two local databases means reconciling two sets of numbers by hand, forever. One shared database is not an improvement, it is a different category of thing.
• Never losing data to a dead machine. Covered above, but it bears repeating, because it is the single most common serious loss in small retail IT.
• Getting features without doing anything. The software improves while you are asleep.

Where the traditional system is straightforwardly better

Fewer situations, but they are real, and a vendor who denies them is selling rather than advising.

• Genuinely unreliable internet. Not “occasionally slow”. If your connection drops for an hour at a time and there is no mobile signal to fall back on, a local system is the correct engineering answer. Do not let anyone talk you out of your own experience of your own building.
• Very high transaction speed. A busy supermarket lane scanning a hundred items needs the database on the same machine as the scanner. Almost no small shop is in this category, but the ones that are know it.
• A regulatory requirement that the data stay on site. Rare, but it exists in some places and some trades. If it applies to you, it decides the question.
• You already own one that works. “Modern” is not a business case. If the installed system does what you need and the backups are real, the correct action is to keep taking money with it.

How to actually decide

Four questions, in order. The first one that gives a clear answer decides it.

1. Is your internet genuinely reliable, and do you have a fallback? Test it honestly over a fortnight, not by how it feels. If the answer is no and there is no mobile tethering option, choose local and stop here.
2. Do you need more than one till, more than one location, or access from outside the shop? If yes, choose cloud and stop here.
3. Is there a real person who will run and test backups every week, indefinitely? If not, choose cloud. An unbacked local database is a slow-motion accident.
4. What is the five year cost of each, including the hardware refresh and the operating system upgrade? Now compare.

If none of those decided it, either choice will serve you, and you should pick on how the software feels to use for eight hours. That matters more than architecture, and it is the one thing you can only judge by trying it.

Where StoreWave sits

StoreWave is a cloud POS, and it makes the trade above deliberately. It runs in a browser on hardware you already own, needs no install and no backups from you, and gives every store its own address you can open from anywhere. Card payments are recorded rather than processed, so your existing terminal and rate are untouched.

It is also honest about the other side of the trade: it keeps selling through a brief connection wobble, and it does not pretend to queue a full afternoon of sales through an outage. If your connection is the kind that goes away for an hour, buy a local system. That is the right answer, and we would rather say so than have you find out on a Saturday.

The free plan covers five team members and twenty products, which is enough to run the four questions above against your own shop before spending anything. See free versus paid for where the ceilings sit.