Skip to main content
Guide12 min12 November 2026

Hardware Shop POS in Kenya and Nigeria: Cement, Nails by the Kilo and Fundi Credit

How to run the till in a hardware or building materials shop in Kenya and Nigeria: selling in bags, pieces, kilos and metres, cost prices that move weekly, fundi credit and stock counts.

By Chidi Okeke

Retail & restaurant tech, Lagos, Nigeria

Illustration: what a hardware shop POS must handle in Kenya and Nigeria

The cement price moved twice this quarter and your price list did not

In January 2026 a 50kg bag of cement in Nigeria was going for roughly N9,300 to N9,700. By July, dealers in some states were quoting N15,000, the FCCPC opened a probe into possible price manipulation, and manufacturers pointed at the naira, energy and haulage. In Kenya the swings are gentler but constant: Simba 32.5R around KSh 735, Bamburi Fundi near KSh 780, factory-gate prices lower, upcountry delivery adding roughly KSh 30 to KSh 80 a bag on top.

Now think about what that does to a hardware shop. You bought 200 bags at one price. The next lorry comes in at another. The handwritten price list taped to the counter still shows last month's figure, and your shop assistant sells from it all Saturday because nobody told him otherwise. You did not lose money on theft that day. You lost it on arithmetic.

That is the real job of a till in a hardware shop, and it is not the job most POS software is built for. A supermarket scans a barcode and sells one unit. A hardware shop sells the same nails as a 25kg carton, as a kilo, and as a handful, to a contractor who will pay at the end of the month. This guide is about the till that survives that.

Why a normal shop till breaks behind a hardware counter

Most cheap POS apps carry one hidden assumption: a product has one price and one unit. Scan it, sell it, done. A hardware shop breaks that assumption on the first sale of the morning.

You stock items with no barcode at all, or with barcodes nobody ever scans because the item lives in a sack outside. You sell by weight, by length, by piece and by bundle, from the same stock. Your cost price changes between deliveries, sometimes between weeks. A large share of your sales go out on credit to people you know by face and by phone number. And a good part of your inventory sits in the yard, under a tarpaulin, in the rain.

None of that is exotic. It is the ordinary day of a hardware shop in Kariobangi, in Nakuru, in Onitsha or Mushin. But a till that cannot handle it will quietly produce numbers you cannot trust, and a shop that cannot trust its numbers cannot tell a bad month from a theft problem.

Selling the same stock in bags, pieces, kilos and metres

This is the first feature to test, and it is the one most demos skip. Take three real examples from your own shop.

Nails arrive as a 25kg carton. You sell the full carton to a contractor, a kilo to a homeowner fixing a gate, and sometimes half a kilo. One product, three selling units, one stock figure that has to go down correctly in all three cases. Electrical cable arrives as a 100m roll and leaves by the metre. Paint arrives in 20 litre drums and leaves in 4 litre tins, or poured into a customer's own container. Iron sheets are priced by gauge and by length, and the same mabati profile comes in several lengths.

What you need is a till where one item carries its purchase unit and its selling units with a conversion between them, so that selling 3kg of nails takes 3kg off a carton that the system knows holds 25. Ask the vendor to set that up in front of you with your own nails, not with their demo product. If they have to create three separate products called "Nails carton", "Nails kg" and "Nails half kg", walk away: your stock figure will be wrong by lunchtime, and every stock count after that will be a guess.

Your cost price moves faster than your retail price

In a shop selling biscuits, cost prices drift once or twice a year. In a hardware shop they move with the exchange rate, the fuel price and the cement factory's mood. That changes what you need from the till.

The number that matters is not what you paid for the first lorry. It is what the stock you are holding right now actually cost you on average, because that is what your margin is measured against. A till that records the buying price on every delivery and keeps a running average cost gives you a true margin per item. One that stores a single "cost price" field you typed in once will flatter you, and flattering numbers in a low-margin trade are expensive.

ItemBuying priceSelling priceGross marginWhat kills it
Cement, 50kg bagKSh 700 to 750KSh 780 to 820roughly 8 to 10%Price rise between deliveries
Nails, per kg from 25kg cartonKSh 150 to 170KSh 200 to 230roughly 25 to 30%Short weighing, spillage
Cable, per metre from 100m rollvaries by gauge30 to 40% uproughly 25 to 30%Offcuts nobody records
Paint, 4L tinvaries by brand20 to 30% uproughly 18 to 25%Dead stock in dead colours

Treat those figures as orders of magnitude from the markets, not as your prices. The point is the shape: cement is a volume product on a thin margin, and the small items you sell by the kilo or the metre are where your profit actually lives. A till that shows you margin per item tells you that within a week. A till that only shows total sales never will. It is the same margin discipline a small supermarket lives by, set out in choosing a POS for a small supermarket, applied to a trade where the cost side moves more violently.

Fundi and contractor credit is the biggest number in the shop

Ask a hardware shop owner what they are owed and you will often get a long pause. The credit book is a cheap exercise book, sometimes two, and the figures in it are a mix of amounts, part payments and a name like "fundi wa Mama Njeri" with no phone number.

Credit is not the problem. Credit is how the trade works, because a contractor on a site cannot pay cash for every bag as the job goes on. The problem is credit without a limit, without a date and without a record that both sides accept. Three things fix most of it. Each customer gets an account in the till, so every unpaid sale lands in one place and the balance is a number, not a reconstruction. Each account gets a ceiling, in money, and the till warns the person at the counter before it is crossed, because that assistant will not say no to a well known contractor on his own authority. And every credit sale produces a slip the customer takes away, so that an argument three weeks later is about a printed total, not about memory.

There is a harder rule worth adopting: no new credit while an old balance sits past 30 days. It feels rude the first time you apply it. It is cheaper than the alternative, which is funding somebody else's project out of your working capital for free. Unpaid credit belongs in the same column as the other losses that never look like theft: nobody stole anything, and the money is gone all the same.

Quotations and delivery notes, not just receipts

A supermarket till prints one document. A hardware shop needs three, and the gap between them is where money goes missing.

First the quotation. A contractor walks in with a list, you price it, he takes the paper and goes to compare. If your quote is handwritten and unnumbered, you have no idea how many you issued this week or how many turned into sales. If it comes out of the till with a number and a validity date, you can see your conversion rate and, more importantly, you can honour the price without arguing when he comes back on Thursday and cement has moved.

Then the delivery note. Materials leave for the site, often on a borrowed pickup, often before anything is paid. Whatever leaves the yard has to be on a document somebody signs, because the single commonest loss in this trade is goods that went out and were never invoiced. Finally the receipt, with your tax details on it. If you are VAT registered in Kenya your till has to feed eTIMS directly, rather than leave you retyping totals into something else at month end, which is where the errors breed. In Nigeria, keep an eye on the phased e-invoicing rollout: thresholds and dates have shifted more than once, so confirm your own obligation with the revenue service rather than with a vendor's brochure.

Stock counts when the stock is heavy, dusty and partly outdoors

Nobody enjoys counting a hardware shop. The cement is stacked six high, the sand is a heap, the iron sheets are bundled by gauge and the small fittings live in forty tins behind the counter. So most shops count once a year, badly, or never.

The practical answer is cycle counting: pick a category a week and count only that. Cement and sheets on Monday, plumbing on Wednesday, electrical the week after. Twenty minutes at a time, with the till showing you what it thinks you have before you count. You find the gap while it is still small and while the person who created it is still working for you. The method is written out in full in how to do a stock take without closing your shop.

Watch the categories that bleed quietly. Offcuts of cable and pipe, half bags that split in the yard, paint tinted for a customer who never came back. None of these are theft and all of them are loss, and they only show up if somebody records them as wastage instead of letting them vanish into a counting error.

M-Pesa, bank transfer and the cash that never balances

In Kenya a good share of your takings arrive on a Lipa na M-Pesa till. In Nigeria they arrive as transfers and card payments, and your phone buzzes with alerts all day. Either way, the money reaching the phone or the bank is not the same thing as a recorded sale, and the difference is where your evening goes wrong.

A till that does its job records how each sale was paid: cash, M-Pesa, transfer, card, credit. At close, you compare the till's cash figure against the drawer, its M-Pesa figure against the statement, and its transfer figure against the alerts. A gap found tonight has a cause you can name. The same gap found during a stocktake in March has no cause at all, only suspects.

This is also the honest answer to a question many owners ask: is my M-Pesa till not enough on its own? It takes money well, and it tells you nothing about what you sold, at what margin, to whom, on credit or not. The longer version is in is an M-Pesa till enough, or do you need a POS.

What it actually costs to put a real till in a hardware shop

Count three lines. Hardware: an Android phone or tablet, often one you already own, plus a thermal printer for slips and a weighing scale if you sell by the kilo. A decent scale and printer are a modest one off expense, and the full picture of what to buy is in what POS hardware a small shop really needs.

Software: this is where the range is wide. Full ERP style packages sold to hardware distributors in Nairobi and Lagos run to real monthly fees, and they are built for depots with warehouses and sales reps. Offline first mobile apps like digabloPos start free and run on the device you have. In between sit cloud tills that are cheap but assume constant internet, which is a poor assumption in a shop with a metal roof and a generator.

The third line is the one nobody budgets: your time, and your staff's. A till nobody is trained on produces wrong numbers, and wrong numbers are worse than no numbers, because you act on them. Budget two afternoons for setup and one week of checking the close every single evening. For a view of where POS prices actually sit in these markets, how much a POS system costs in Kenya and Nigeria has the ranges.

Where digabloPos fits, and where it does not

Plainly: digabloPos is a till and a stock system, not a construction ERP. It runs on Android phones and tablets, keeps selling when the internet or the power goes, handles selling units and conversions so nails can leave by the kilo from a carton, keeps a customer account with a credit limit for your fundis and contractors, separates cash from M-Pesa from transfer at close, and prints slips. It starts free, so a hardware shop can test it on real Saturday traffic without committing anything.

What it is not: a tool for a building materials distributor running several depots, a fleet of delivery trucks, purchase orders against supplier contracts and a sales force on commission. That shop needs a distribution ERP and should buy one. The dividing line is roughly whether your stock lives in one yard you can walk across. If it does, a mobile till is almost certainly enough, and anything heavier will cost you money and goodwill from your staff. If you are running two or three branches, managing multiple shop locations covers the middle ground.

Mistakes that cost hardware shops real money

Selling from a stale price list. The most expensive mistake in the trade, and the easiest to fix: when a delivery comes in at a new cost, the price in the till changes the same hour, before the first sale.

One product per selling unit. Creating "cement bag" and "cement half bag" as separate items feels harmless and destroys your stock figures permanently.

Credit with no ceiling. Not because the contractor is dishonest, but because his cash flow is not yours, and his site delays become your overdraft.

Not recording wastage. The split bag, the offcut, the tinted paint nobody collected. Record it as loss and it becomes a number you can manage. Hide it and it becomes a stocktake mystery that makes you suspect your own staff.

Buying a system built for a depot. Paying monthly for purchase order workflows and warehouse bins when you have one yard and two employees is money straight out the door.

And the quietest of all: not closing the day. Five minutes on cash, M-Pesa and transfer, every evening. Shops that do this find problems in days. Shops that do not find them in years, when it is far too late to do anything useful.

How to get this running in a week

Monday and Tuesday: load your fast movers only. Thirty to fifty items covers most of your turnover in a hardware shop, and cement, sheets, nails, cable and pipe will be most of it. Set each one's purchase unit, its selling units and the conversion. Do not try to load all 900 items, because you will stop halfway and abandon the whole thing.

Wednesday: open accounts for your ten biggest credit customers and give each a ceiling you can actually afford to lose. Thursday: run the shop on both systems, the till and your usual book, side by side, and compare them at close. They will not match on the first day. The gaps tell you what you set up wrong, which is exactly what you want to learn now rather than in month three.

Friday: cut the internet and sell anyway, because a till that stops when the connection does is not usable in this trade. See the full argument in why an offline POS is the only kind worth buying. Then on Saturday, your busiest day, use only the till.

By the next Monday you will know two things: whether the software suits your shop, and what your real margin on cement is. Most owners are surprised by the second one, and not pleasantly. That surprise is the whole point of putting a till in.

Frequently asked questions

Can a POS sell the same item in bags, kilos and metres?

A good one can, through selling units with a conversion to the purchase unit: a 25kg carton of nails can leave as a full carton or as single kilos, and the stock figure drops correctly either way. Test this with your own nails or cable during the demo. If the vendor has to create separate products for each unit, your stock numbers will drift within days and never recover.

How do I control credit to fundis and contractors?

Give every credit customer an account in the till rather than a page in an exercise book, set a ceiling in money that you could survive losing, and let the till warn the counter before that ceiling is crossed. Print a slip on every credit sale so the balance is documented on both sides. A useful rule: no new credit while a balance is more than 30 days old.

Do I need eTIMS or e-invoicing for a hardware shop?

In Kenya, if you are VAT registered your sales have to reach KRA through eTIMS, so your till needs to feed it rather than leave you retyping. In Nigeria the e-invoicing rollout is phased by taxpayer size and the dates and thresholds have been revised more than once, so confirm your own obligation with the revenue service directly instead of relying on a vendor's timeline.

What does a POS for a hardware shop cost in Kenya or Nigeria?

Count hardware (an Android phone or tablet you may already own, a thermal printer, a weighing scale if you sell by weight), software, and training time. Distribution ERPs charge real monthly fees and are built for depots. Offline first mobile apps such as digabloPos start free. Compare total cost over a year, including the days your staff spend learning the system.

How often should a hardware shop count stock?

Weekly, by category, instead of annually for everything. Count cement and sheets one week, plumbing the next, electrical after that, with the till showing expected quantities before you count. Twenty minutes a week finds gaps while they are small. An annual count of a full hardware yard is so painful that most shops skip it, which is how small losses become permanent ones.

Put a till behind your hardware counter this weekend

digabloPos sells in bags, kilos, pieces and metres from the same stock, keeps a credit account with a limit for every fundi, separates cash from M-Pesa and transfers at close, and keeps working when the internet drops. Free to start on the Android phone you already have.

Try for free