Floreant POS in Mexico: what free open source really costs
Free software is not the same thing as a free system. Floreant is real and it works, and the bill it sends arrives as your own time instead of a monthly invoice.
Home / Blog / Best POS system for small restaurant: the real cost
Two quotations with the same headline figure can differ by a wide margin once every device is counted. The gap is usually hidden in the words per screen and per terminal.
Anzul Aqeel · Founder & Director, Anunzio International FZC · 2 September 2026
The best pos system for small restaurant use is usually the one whose full cost you can predict a year ahead. Headline pricing rarely tells you that, because the line that grows is not the software line.
This article puts the published numbers next to the ones nobody puts on a quotation, then gives a way to test a system in a week rather than a year.
Two vendors publish their pricing openly, which makes them useful for illustration rather than as recommendations.
Lightspeed lists its restaurant tiers publicly, with Starter at $69/month, Essential at $189/month and Premium at $399/month, and a kitchen display system at $30/screen per month across all tiers.
Square's restaurant pricing works differently, with a free tier and paid tiers, and add-ons charged per device: $30/mo. for the Square KDS app per device and $50/mo. for the Square Kiosk app per device on one tier, with custom pricing for businesses processing over $250,000 annually.
Read the pattern rather than the numbers. The monthly cost you are quoted covers the software per site, while the parts a real kitchen needs are priced per device. A small restaurant with a screen in the kitchen and a screen at the pass is buying the same thing twice.
Hardware. Terminals, printers, cash drawers, stands, cables. It is a capital cost with a life of a few years, and in Kuwait it usually arrives with shipping and customs on top.
Per-device fees. See above. Count every screen you will actually run, not the one in the demonstration.
Card processing. A percentage of every sale, charged forever. Over a year it commonly exceeds the software fee several times over, and it is the number most owners never compare.
Your own time. Setting up a menu, training staff, and the first fortnight of mistakes. Real money, spent quietly.
Leaving. The cost of moving your menu and customer list to something else. If the export is a photograph of a screen, this cost is enormous and nobody warned you.
Cloud based restaurant POS systems removes some of the hardware cost by moving the software to a browser, and used POS systems for restaurants is worth reading before buying anything new at all. An EPOS system for a restaurant covers the same category under the name used in some markets.
When I was reselling this software, the deal was almost always won on the monthly figure and lost, a year later, on everything attached to it. I would rather a customer read the second page of the quotation than the first. — Anzul Aqeel, founder of Get Menu
Day one. Load your ten best-selling dishes, not the whole menu. If that takes more than an hour, note it, because the whole menu will take twenty times longer.
Day two. Change a price at your busiest hour. Time yourself.
Day three. Break something deliberately. Cancel an order, refund a line, send a wrong item. Watch what the system does and what it leaves for a human.
Day four. Ask support a real question by whatever channel is offered, and time the reply.
Day five. Export everything. Open the file. If you cannot read it, stop here.
The weekend. Run a real service on it beside your current method. Nothing else tells you the truth.
Any product that survives that week is a serious candidate. Most do not reach day five.
Get Menu is an ordering system, not a register, and it is priced as one flat monthly fee with 0% taken from your sales. There is nothing per device, because the devices are phones your staff and customers already own.
Your own ordering page on an address belonging to your restaurant, with 12 page layouts, a font picker and a photo style toggle.
A code for every table, generated for you, with a designer for the printed card and a print pack sent ready for the printer. The table is the address, and a group at one table orders into a single bill.
Orders that arrive as messages. A confirmation to the customer, a ticket to the managers group and a rider dispatch with the GPS pin, all within about 2 seconds, plus a log of every message and whether it arrived.
Delivery that fits a city of towers. Building and unit picked from a list, a delivery fee per building, and free delivery above a threshold you set.
Prices that hold. Every line re-priced from your live menu at checkout, options and add-ons with their own prices, required choices, and a minimum order value.
Stock counts per item and per option, falling as orders arrive, flipping to sold out at zero, with a movement history showing where every unit went.
Loyalty in the phone's wallet. Stamps, tiers your regulars grow into, and a Google Wallet card that updates itself. Apple Wallet is built and waiting on a certificate.
Offers that cannot run away. Coupons with caps, exclusions and expiry dates, ten kinds of upsell, and bundles at a fixed price, with the best discount winning rather than two stacking.
Records. Tax invoices carrying your registration number as a PDF, numbered in one sequence, with a sales dashboard and a spreadsheet export whenever you want it.
The trial runs 15 days with no card details, which is exactly long enough for the week above plus a second look.
We do not sell hardware, so there is no shipping and no customs on our account.
There are no online payments, so we charge no processing percentage and hold no card details at all.
We do not run a kitchen display screen, ingredient stock, recipe costing, staff rotas or table bookings, and there is no pre-order or collection order type.
We are not a register. A counter sale still goes through whatever you already own.
Add the yearly software fee, the yearly per-device fees, the hardware divided by three years, and your best estimate of card processing on last year's turnover.
That single figure is what you are actually choosing between. Almost nobody calculates it, which is why almost everybody is surprised in month fourteen.
Less than it usually does, once every device is counted. Build one yearly figure covering software, per-device charges, hardware spread over its life, and card processing, then compare that number rather than the monthly headline.
The software may be. The processing percentage on every sale is not, and it is usually the largest line in the whole arrangement over a year.
Only if paper is failing you. Many small kitchens do fine with orders arriving as messages on a phone the team already watches.
No. Payment is cash or your own card machine when the food arrives, which is why no card details are stored anywhere.
You export your customer list on any day you choose, old order details clear automatically after 120 days, and a customer can be erased without breaking your accounts.
Free software is not the same thing as a free system. Floreant is real and it works, and the bill it sends arrives as your own time instead of a monthly invoice.
Most owners search for one kind of stock software and actually need the other one first. Knowing which is which saves a New Zealand kitchen both money and staff hours.
The useful question is not whether to join a delivery app. It is which parts of your business you are happy for somebody else to own, and which parts you are not.
Fifteen days with everything switched on. No card, no commission, ever.
Questions from other owners are answered in the community.