Running a food stall? Why you need a POS app (and which one is free)

You run a food stall. A kottu cart in Wellawatte. A roast chicken counter at Hyde Park Corner. A short eats stand outside an office complex. A takeaway hatch on Galle Road. The till is a steel cash box. The “POS” is a notebook with the day’s takings in pencil. The product list is in your head.

This has been the model since street food existed. The question is whether it should still be the model in 2026 — and what changes if you replace the cash box with the phone you already own.

This post is the honest answer. We are not going to pretend a POS is mandatory. We are going to tell you what it actually changes, and when those changes are worth the ten minutes it takes to set up.

What you are currently paying for “no POS”

The notebook-and-cash-box system has real costs. Most of them are invisible because you have lived with them for years.

Cashier shortage at end of day. When the takings come up Rs. 1,800 short, you do not know if it is your son who ran the stall over lunch, the cousin who covered while you went to the bank, or just a maths error. No paper trail, no answer. Eat the loss.

No idea what is actually selling. You think the cheese kottu outsells the chicken kottu 2 to 1. Maybe it is the other way around and you are over-prepping chicken every morning. You will never know without writing every sale down by item.

Running out at the worst time. Halfway through Friday rush, you run out of egg roti. Three customers walk away. You did not realise you were down to ten this morning because nobody counted at open.

Cannot price test. You raise short eats by Rs. 5. Do you sell less? You think maybe. You have nothing to compare against.

No proof of takings. When a supplier wants to give you 30-day credit, they ask “what is your monthly revenue?” You have to estimate. They are estimating from your estimate. Trust margin is thin.

Holiday hand-off is impossible. You cannot take a Sunday off because the only person who knows the prices and the right change for Rs. 380 + Rs. 220 is you. The stall is closed when you are.

None of these are dramatic individually. Added up over a year, they are the difference between a stall that quietly grows and one that quietly stalls.

What changes when you add a POS

We are going to be specific. Six things change, and they happen on day one.

1. The end-of-day cash count is decided

Open the assistant at end of day. Say “How much in cash today?” It tells you. You count the cash box. The two numbers match or they do not. If they do not, you have a specific gap to investigate.

If your cousin covered for two hours and was on his own PIN (more on this), you can ask “What did Sunil sell from 1pm to 3pm?” and the answer is exact.

2. You can see what is actually selling

“What were my top 5 items today?” The assistant answers in plain language. After a week, you have a real read on cheese kottu vs chicken kottu, vegetable roti vs egg roti, samosa vs vade. After a month, you stop over-prepping the slow movers and start prepping more of the quick ones. This shows up directly in margin — less waste, fewer stockouts.

3. Stock alerts before the customer notices

If you tell the POS your morning stock — “I have 30 egg roti, 40 vegetable roti, 20 samosa, 25 vade” — every sale decrements those automatically. When egg roti gets to 5 remaining, the assistant tells you. You either start making more or you mark it sold out before a customer walks up and asks.

4. Price test, fast

You raise the price of fish bun by Rs. 10 on a Tuesday. The POS records sales. On the following Tuesday, you compare. “Compare fish bun sales this Tuesday to last Tuesday.” The assistant answers. Now you have data.

This is the first time most small operators actually run a price test instead of guessing.

5. You get a receipt that proves takings

Some customers want a printed receipt — especially anyone billing the company expense account. Even when they do not ask, every sale exists as a record. If you ever go to a bank or a supplier and need to show monthly turnover, the POS gives you a one-line answer instead of a rough estimate.

6. Your cousin can run it without learning anything

You teach your cousin one screen: tap product, tap product, tap pay, tap cash, done. He covers for two hours. Every sale he rings up is tagged to him. You take the afternoon off. You see the numbers when you come back. Nobody had to learn software.

These are not theoretical wins. They show up the same week you set the POS up.

”But my customers want it to be fast”

This is the main objection from food-stall operators and it is a fair one. If a POS slows your line by even three seconds, the line backs up at lunch rush and you lose customers.

Kaspa’s sell screen is engineered for the food-stall use case specifically. A typical sale:

• Tap “Cheese Kottu” — 0.3 seconds.
• Tap “Cash” — 0.3 seconds.
• Tap “Done” — 0.3 seconds.
• Total: under a second.

The screen learns your most-tapped items and surfaces them as big buttons at the top. The cashier does not “search” for anything most of the time — they tap the picture of the kottu and it is added. If a customer orders three items, that is three taps. Pay is one tap. Done is one tap. Five taps total for a three-item sale.

That is faster than counting cash. You will not lose your line. You will probably move more customers through it.

We have shop owners on takeaway counters telling us that Kaspa shaved 30 seconds off their average sale time. Not because of any single feature — because every sub-second of friction was engineered out.

”What about the printer? I do not want to lug hardware around”

You do not need a printer. Kaspa’s sell screen works fine without one — sales complete, totals track, reports work, cash is counted. The customer just does not get a paper receipt.

For most food stalls, that is fine. Customers paying cash do not ask for receipts. Customers paying card need a receipt, but if you are not running card today (most food stalls in Sri Lanka are cash-only), this never comes up.

When you do want a printer — for the occasional office worker who needs an expense receipt, or because you want to stick the order ticket to a clip at the prep station — any 58 mm Bluetooth thermal printer works. Throw it in a Tupperware to keep it weather-safe and you have a fully digital stall.

”Cash doesn’t need a POS”

Half-true. Cash does not need a POS to settle the payment. Cash needs a POS to record what was sold.

If all you cared about was “money in the box at end of day matches money out”, you would not need to know what was sold either. But you do care, even if you do not articulate it that way. You care because:

• You want to know if cheese kottu is worth keeping on the menu.
• You want to know whether to raise prices in November.
• You want to know if your cousin Sunil is reliable.
• You want to know whether the rainy week actually cost you what it feels like it did.

Cash plus a notebook gets you partway. Cash plus a phone POS gets you the rest of the way, for free, with the same physical till.

A practical setup for a food stall

The whole setup, end to end, for a typical kottu / roti / short-eats stall:

1. Use your phone. Any phone running Chrome works. Both your phone and your cousin’s phone will end up being a register. Open pos.trykaspa.com on yours.
2. Sign in with your phone number. SMS code. You are in.
3. Add your menu. Type or speak each item: “Cheese kottu 850”, “Chicken kottu 1100”, “Vegetable roti 220”, and so on. Do this once. Edit as your menu changes.
4. Add a cashier with a PIN for whoever covers you. “Add a cashier named Sunil PIN 1234.”
5. Open the sell screen. Tap an item. Tap “Pay”. Tap “Cash”. Done.

Time to working register: about 8 minutes if you have a 12-item menu. Less if shorter. See the 10-minute setup playbook for a walkthrough.

When the WiFi or mobile signal drops — and it will, multiple times — sales keep working. The phone runs the POS from local storage. See how Kaspa handles patchy connectivity. When the signal comes back, sync happens automatically.

What it costs

$0.

Not “free trial”. Not “free up to 100 transactions”. Not “free for the first device”. $0, forever, unlimited sales, unlimited products, unlimited cashiers, unlimited devices. See pricing for the line items.

A Pro tier is in flight (premium AI, advanced reports) and it will be optional. The core POS — what a food stall actually needs — stays free.

We can offer this because we do not have a sales team in San Francisco and we do not pretend our software is more complicated than it is. Small shop, simple POS, free.

”Why not just keep using the notebook?”

Because you will get tired of writing the notebook in the rain.

Honestly that is the practical reason most stall operators eventually digitise. Notebooks get wet, get torn, get lost. The phone is in your hand anyway. The phone does not get soggy.

The other reasons — knowing what sells, having a cashier shortage that decides itself, being able to take Sunday off — those compound.

Try it this week

Friday evening. After the rush is done. Open pos.trykaspa.com on your phone. Sign in with your number. Spend 10 minutes adding your menu. Saturday morning, open the sell screen. Tap, tap, tap. Run the day on the POS. Saturday night, open the assistant and say “What did I sell today?”

If the answer is not faster than your notebook, you lost 10 minutes. If it is — and we are confident — you keep going.

That is the whole pitch.