A shop assistant completing a sale on the sell.ke POS, with a barcode scanner and a receipt printer on the counter

    Choosing well

    The best POS system for a small business is the one you do not outgrow in a year

    Every vendor's page on this query says the same thing about itself, which makes the query almost useless to search unless somebody is willing to write down the actual decision. So here is the actual decision, including where we are the wrong answer.

    The trap that catches most first-time buyers is not price and not features. It is buying a till when what they needed was a system. A till is cheap, quick to set up, and completely adequate — until the day you add a second location, start selling online, or want to know which products actually make money. Then you discover the stock figures were never designed to be trusted, and you are migrating a year in.

    The useful framing is therefore not 'which is best' but 'which of these will still be right in two years given where the business is going'. That collapses a long shortlist very quickly.

    What you get

    What this gives you

    Judge it on the record, not the till

    Every POS rings up a sale competently. What differs is whether stock, costs, online orders and accounts live in the same place — which is what you will actually be living with.

    Count the second-till cost

    Per-terminal licensing turns a busy Saturday into a purchasing decision. Ask what a second and third till cost before you ask what the first one does.

    Model transaction fees at your volume

    A percentage of sales is the cost shape that grows fastest and never plateaus. Multiply it against a realistic year, not against a single transaction, before comparing to a subscription.

    Check what happens offline

    If the counter stops when the connection does, you have bought a dependency on your internet provider's worst day, not on your busiest one.

    Ask how you leave

    Products, customers and transaction history should be exportable without a support ticket. A vendor confident in the product is relaxed about this question.

    Prefer one system to three good ones

    A coherent whole beats better parts that disagree. The reconciliation work between best-in-class tools is invisible in a demo and permanent in practice.

    What 'small business' hides

    The phrase covers wildly different needs. A single-counter café with forty items and no stock depth needs speed at the till and very little else. A hardware shop with eight thousand SKUs, supplier accounts and trade pricing needs inventory and purchasing far more than it needs a fast till. A boutique selling on Instagram and in one shop needs the two channels to share stock. The same software is a good answer to one of those and a poor answer to another.

    So the first question is which of those you are, and specifically which part of the business currently costs you the most unpaid time. If it is queueing, buy for speed. If it is not knowing what you have, buy for inventory. If it is reconciling channels, buy for unification. If it is the accountant's annual list of questions, buy for books.

    sell.ke is built for the second, third and fourth of those. For a business whose only real problem is queue speed at a single counter, a simpler and cheaper till is a perfectly sensible purchase and we would rather say so than sell you something with modules you never open.

    The costs that are not on the pricing page

    Hardware lock-in. Proprietary terminals mean the vendor sets the price of every additional and replacement till. Systems that run on ordinary phones, tablets and laptops with commodity 80mm printers and scanners do not have this cost at all.

    Per-terminal or per-user licensing. This is the fee that turns seasonal trading into an annual negotiation, and it is the reason a system that looked cheaper at one till is more expensive at three.

    Payment processing lock-in. Some systems require their own processor, which removes your ability to shop for rates later. Ask whether you may bring your own, and what changes if you do.

    Setup and data entry. The genuinely expensive part of any migration is building the catalogue. Ask what import formats are supported, whether photos and variants come across, and whether anyone helps — this is usually the difference between a week and a weekend.

    The integration tax. If the answer to 'does it do X' is 'yes, via an app', that app has a subscription, an owner and a failure mode. Two or three of those and the cheap system is not cheap.

    A shortlist process that takes an afternoon

    Write down the five things you do most often at the counter and the three questions you most want the system to answer at the end of the month. That page is your test.

    Then use the trial rather than the demo. A demo is the vendor driving on a clean dataset; a trial is you, on your own products, at your own speed. Load fifty real products — not the whole catalogue — and put through a normal hour of trading, including a return, a discount and an awkward split payment, because those are where systems differ.

    Finally, run the reports. Ask the system the three month-end questions you wrote down. If getting an answer requires an export, a spreadsheet and ten minutes, that is what every month will look like.

    Compare

    Where sell.ke fits, and where it does not

    Written to be useful rather than flattering. The right answer for you may be one of the first two columns.

    What you are comparingA simple tillBest-of-breed stacksell.ke
    Best whenOne counter, few products, no online salesYou have a systems person and specialist needsStock, channels or books are the thing costing you time
    Inventory depthBasic or noneDeep, in a separate productDeep, in the same system
    Selling onlineSeparate platformSeparate platform plus syncIncluded from the Basic plan
    AccountingExport to an accountantA separate ledger productDouble-entry, from the Growth plan
    Where it is weakerStops being enough quicklyPermanent integration upkeepFewer specialist extremes than a dedicated tool
    Extra tillsOften per-terminalOften per-terminalNo per-terminal licence

    Questions

    Best POS for a small business — questions

    What is the best POS system for a small retail business?

    Judge the shortlist on four things rather than on feature counts: whether stock is one figure across every channel you sell on, whether the system knows your cost prices well enough to report margin, what a second till and a second location cost, and whether the counter keeps working offline. Those four predict whether you will still be happy in two years better than anything else on a comparison page.

    How much should a small business pay for a POS system?

    Subscription pricing in this market broadly runs from roughly ten to a hundred dollars a month per location depending on depth, with transaction-fee models charging a percentage instead. sell.ke starts at USD 12 (KES 1,499 in Kenya) a month with no per-terminal licence and no cut of your sales. The figure to compare is annual total including extra tills, extra locations and any transaction percentage — not the headline monthly price.

    Is a free POS system good enough to start with?

    It can be, provided you understand what funds it. Free usually means either payment processing at a percentage, a single location, or no inventory tracking. Any of those may be fine for a first year. The risk is that the cheapest thing to start with is frequently the most expensive thing to leave, because the data you accumulate in it is what you have to migrate.

    Should I buy POS hardware from the software vendor?

    Only if the software requires it. Receipt printers, barcode scanners and cash drawers are commodity hardware; an 80mm thermal printer and a printer-driven drawer will work with anything that speaks the standard. Buying a bundle is convenient and occasionally cheaper, but a system that only works with its own hardware has priced every future till for you.

    How long does it take to switch POS systems?

    The software takes an afternoon; the catalogue takes the time. Budget a day for a small shop importing a clean spreadsheet, and about a week for a multi-location business with thousands of SKUs and opening stock counts to establish. Run both systems in parallel for a fortnight regardless of size — the fourteen-day trial exists to make that free.

    How much does sell.ke cost?

    Plans run from USD 12/month for a single-location till to USD 115/month for unlimited scale, with the online store on your own domain included from USD 23/month. In Kenya the same plans are KES 1,499 to KES 14,999/month, billed in shillings. Every plan starts with a 14-day trial and no card, and there is no per-terminal licence and no commission on your sales. There is no hardware to buy either — sell.ke runs on a phone, tablet or laptop you already own.

    What hardware do I need?

    None to start. sell.ke runs in a browser and as an Android app, so a phone, tablet or laptop you already own is a working till on day one — the phone camera reads barcodes. When volume justifies it, add an 80mm thermal receipt printer (Bluetooth or USB), a handheld scanner and a cash drawer the printer opens. There is no proprietary terminal to buy and no per-terminal licence, so a second till on a busy Saturday costs nothing.

    How long does setup take?

    A single-counter shop is usually selling the same day. The work is the product list, not the software: import a spreadsheet, scan barcodes, or let Amina build products from a supplier invoice or a photo of your price list. A multi-location business with thousands of SKUs and opening stock counts should plan about a week and run the old system alongside for a fortnight. The 14-day trial is long enough to do exactly that.

    What happens to my data if I leave?

    You export it. Products, customers, suppliers, stock movements and transactions are all exportable, and the accounting side produces standard statements rather than a proprietary format. We would rather you stayed because moving is not worth it than because moving is not possible — a business whose records are hostage is a business that resents its software.

    Try it on your own products

    Fourteen days, no card. Import your product list or let Amina build it from a photo of your price list, and run it alongside whatever you use now.