Starting a Water Delivery Business in Pakistan: What It Actually Takes

A small bottled-water delivery business in Pakistan needs three things before it needs anything else: a reliable water source you can treat, enough bottles to cover two delivery cycles, and a round of customers close enough together to serve profitably. Get those three right and the business works at almost any scale. Get the third one wrong and no amount of plant investment saves it.

This is a practical guide to what the parts cost, what order to buy them in, and where new suppliers actually lose money. It is written from what we see across the businesses using our software, not from a template.

The four parts of the business

1. Treatment and filling

You need to produce water that is safe, consistent and defensible. At the small end that means a filtration and UV or RO setup and a manual or semi-automatic filling line. At the larger end, an automated line with rinse, fill and cap in sequence.

The temptation is to spend here first, because it is the part that feels like the business. Resist it. A modest line that runs reliably and fills a round you already have is worth more than a large one waiting for customers.

2. Bottles

This is the part new entrants consistently underestimate. You do not need one bottle per customer — you need roughly two and a half to three per customer, because at any moment a bottle is either at the customer, in transit, or in the plant waiting to be filled.

A 400-customer round therefore needs somewhere around 1,000–1,200 bottles in circulation. At Pakistani replacement cost that is a serious sum, and it is capital that is permanently tied up. It is also the single largest asset you will fail to keep track of — see how to track returnable bottles, because this is where most of the money quietly goes.

3. Delivery

A loading vehicle, a rider, and fuel. The economics here are all about density: a rider serving 80 stops within a few kilometres is profitable; the same rider serving 40 stops spread across the city is not, at any price you can charge.

This is why the third requirement — a geographically tight round — matters more than the plant. Ten customers in one lane are worth more than thirty scattered across a district.

4. The office

Someone has to take orders, raise bills, chase money and count bottles. At the start this is you. The mistake is assuming it stays free: by the third route it is a full-time job, and if nobody is doing it properly the business will look busy and profitable right up until the year-end count.

Licences and compliance

Requirements vary by province and change, so treat this as a checklist to verify locally rather than a definitive list. In broad terms, expect to deal with:

• Business registration and an NTN.
• Provincial food-authority registration or licensing (Punjab Food Authority, Sindh Food Authority, and their equivalents).
• Water quality testing at defined intervals, with records kept.
• PSQCA marking requirements where they apply to your packaging and claims.
• Local trade licensing and any environmental or groundwater extraction permissions in your area.

Verify each with the relevant authority before you commit capital. A plant that cannot be licensed is an expensive shed.

Working capital: the number nobody plans for

Here is the trap. You deliver all month and bill at month end. Customers pay through the following month, some of them late. Meanwhile you are buying caps, labels, electricity, fuel and wages weekly.

Plan for at least two months of operating costs in working capital, on top of every asset purchase. Businesses that skip this do not fail because they are unprofitable; they fail because they are profitable and illiquid, which looks identical from the inside until the day wages are due.

The three mistakes that close new suppliers

Mistake one: buying customers instead of building a round

A new supplier takes any customer who calls, anywhere. Six months later there is a round with 200 customers spread across an area a rider cannot cover in a day, every stop is a long drive, and the fuel bill eats the margin. It is very hard to unwind, because saying no to existing customers is worse than never having taken them.

Build outward from a core. Turn down distant customers early, however painful, or price them to reflect what they actually cost.

Mistake two: competing on price alone

Water delivery has a real floor cost per bottle — treatment, cap, label, bottle depreciation, fuel, wage, and the amortised cost of the bottles that never come back. Undercutting the established supplier by a few rupees a bottle wins customers who will leave for the next few rupees, and it does it at a margin that cannot fund the bottles you need to buy.

Reliability wins in this market far more than price does. A household that has been let down twice in a heatwave will pay more, permanently, for a supplier who turns up.

Mistake three: no control over the round

The rider has the customer list, holds the stock, collects the cash and reports the day verbally. This is fine while there is one rider and you know every customer. It stops being fine the moment you cannot personally verify what happened, and the transition is invisible — you find out months later, from a count that does not balance.

Two controls fix most of it: a per-rider cash ledger closed at the end of every round, and a bottle count that reconciles daily.

Sensible order of investment

1. Prove a round exists. Twenty committed customers in one area, served from a modest setup, tells you more than any market survey.
2. Buy bottles to depth, not width. Enough bottles for the customers you have, plus the cycle. Running out of empties is how you lose a customer.
3. Add the second rider before the second area. Densify what you have first.
4. Put controls in before the third route, not after the first loss. This is the point at which paper stops working.
5. Then scale the plant to the round you have actually built.

What to measure from day one

• Bottles outstanding, per customer and in total.
• Deliveries per rider per day, and stops per kilometre.
• Collection rate: what you billed against what you received, monthly.
• Bottle losses per month, as a number and as a cost.
• Customers gained and lost, monthly — churn tells you about reliability faster than any complaint does.

Five numbers. If you know these, you are running the business. If you do not, the business is running.

Irsaal was built to produce exactly those five numbers for water suppliers in Pakistan, without anybody keeping a spreadsheet. The trial is fourteen days, in rupees, with no card — and if you are still at twenty customers with a register that balances, our honest advice is to come back at three routes.