How to Start a Business in Nepal: The Practical Guide

Everything most guides skip: validation, real costs, suppliers, pricing, and the first 90 days of actually trading.

Most guides on starting a business in Nepal end where the hard part begins.

They walk you through the ward office, the Company Registrar, the PAN certificate. Then they stop, as though the certificate were the finish line. It isn’t. The certificate is the easy part. It costs a few thousand rupees and takes a week.

What comes after is where businesses are actually won or lost. Kavi Raj Joshi, co-founder of Startups Nepal, estimates that 90% fail, well above the global average. Almost none of them fail because they filled in a form incorrectly. They fail because the owner never knew, month to month, whether the business was actually making money.

This guide covers the whole thing: how to test the idea before you spend, how to cost it honestly, how to register (briefly, with links to the official sources), how to find suppliers and price your goods, what to record from day one, and a week-by-week plan for your first 90 days of trading.

1. Before anything: prove that people will pay

The most expensive mistake in Nepali small business is building first and asking later. As Joshi puts it, most entrepreneurs “work on assumptions rather than data”. They know the product they want to sell, and they assume a customer exists.

Before you spend a rupee on rent or stock, spend two weeks answering three questions.

Who exactly is the customer, and where are they?

Not “people in Kathmandu.” A specific person, in a specific area, with a specific problem. If you’re opening a kirana pasal, the answer is the households within a five-minute walk. Go and count them. If you’re selling clothes on Instagram, it’s a defined age group with defined spending power.

What are they paying for this today, and to whom?

There is almost always an existing solution, even a bad one. Find it. Go and buy from your future competitor. Note their prices, their busiest hours, what they’re out of stock on, and how they treat customers. An afternoon in a competitor’s shop teaches you more than a month of planning.

Will they actually pay you?

This is the only question that matters, and it has one honest answer: a sale. Sell ten units before you sign a lease. From home, through Facebook Marketplace, out of a bag, to strangers rather than relatives. Family buying from you is politeness, not demand.

Start tracking from that very first test sale. Not in your head. Written down. Karobar lets you record sales and expenses before you’ve registered anything, so those ten test transactions become your first real data: what sold, at what price, to whom, and what it cost you. If you go on to open properly, you already have a baseline. If you don’t, you’ve learned it for the price of nothing.

2. Cost it honestly, including the expenses first-timers forget

Almost every first-time owner in Nepal underestimates their startup cost, usually by a third or more. The visible costs are easy to add up. The ones that sink people are the invisible ones.

Here is a realistic checklist. Figures are illustrative. Your numbers will differ by location and sector, and you should replace every one of them with a quote you have actually obtained.

Two of these deserve special attention.

Working capital is not the same as opening stock

If you spend your entire capital on stock, you have nothing left to restock with when half of it sells. Businesses regularly die with full shelves and no cash. As a rule of thumb, keep at least one month’s purchasing budget aside, untouched.

Your own living costs are a business cost

If the shop cannot pay you a salary, the shop is not profitable. It is being subsidised by your savings, and you should know precisely how long that can last.

Do not count on borrowing your way out of a shortfall. Access to finance is a major or severe constraint for 40.5% of small firms in Nepal (ADB research), average SME interest sits around 12.51% plus service charges, and the average collateral-to-loan ratio for small firms is over 400%, meaning banks want more than four rupees of security for every rupee lent. Plan to survive on what you have.

Set your business up in Karobar before you open and record every one of these as an expense with its category as you pay it. By opening day you will have an exact figure for what the business cost to start. That is a number most owners can only guess at a year later, and the number you need before you can honestly say whether you’re ahead.

3. Registration, briefly

This section is deliberately short, because the official sources are better than any blog and the rules change. Use this as orientation, then verify.

A sole proprietorship (private firm)

This is what most small shops register as. You will typically need a ward recommendation letter from your local municipality, your citizenship certificate, proof of business address (with a landlord’s no-objection letter if you rent), photographs, and a PAN certificate. Registration commonly runs to around NPR 1,000 as an application fee plus a registration fee in the NPR 1,000 to 5,000 range depending on declared capital, with PAN registration at roughly NPR 200. Expect three to seven working days.

A private limited company

This registers online through the Office of the Company Registrar’s CAMIS portal. You reserve a name (approval holds for 35 days), upload your Memorandum and Articles of Association along with shareholder documents, and pay a fee scaled to authorised capital: roughly NPR 1,000 up to 1 lakh, NPR 4,500 for 1 to 5 lakh, NPR 9,500 for 5 to 25 lakh, and NPR 16,000 for 25 lakh to 1 crore. One shareholder is enough, minimum paid-up capital is generally NPR 100,000, and a clean file takes 7 to 14 days.

Choosing between them

Most single-owner retail shops in Nepal start as sole proprietorships and convert later if they grow. If you are unsure, register simple and revisit the question when your turnover forces it.

4. Finding suppliers and negotiating your first terms

New buyers get worse prices than established ones. That is normal, and it is temporary, but only if you handle the first six months deliberately.

Find more than one

Depending on your sector that means the wholesale markets: Ason and the New Road area in Kathmandu for general goods, Kalimati for fruit and vegetables, Birgunj for imported goods coming through the border. Or manufacturers’ distributors who will come to you. Get quotes from at least three. The spread between the first and third quote is often 10 to 15%, which on a retail margin is the difference between a good year and a bad one.

Ask about the things that aren’t the price

The headline rate matters less than: minimum order quantity, delivery frequency and who pays for it, whether unsold or damaged goods can be returned, and how quickly they can resupply a fast-moving item. A supplier who is 2% more expensive but delivers twice a week is usually cheaper in practice, because you can hold less stock.

Credit terms are the real negotiation

Most suppliers will start you on cash and move you to 15 or 30 days once you’ve proven reliable. Getting to 30-day terms is worth more to a small shop than a 3% discount, because it means your stock is partly financed by the supplier rather than entirely by you. The way to earn it is boringly simple: pay early, every time, for the first few months, and make sure they notice.

Set up every supplier as a party in Karobar and record purchases against them. Two things follow. First, you can see exactly what you owe each supplier and when it falls due, so you never miss the payment that would have earned you better terms. Second, after a few months you can see what you’ve actually bought from each, which turns your next negotiation from “can you give me a better rate?” into “I’ve bought this much from you this year, what can you do?” That is a completely different conversation. Photograph each purchase bill into the app as you go, so a disputed delivery is settled by evidence rather than memory.

5. Pricing: the mistake that kills margin in month one

Here is the most common pricing error in Nepali retail, and it is almost universal.

You buy an item for NPR 80. You sell it for NPR 100. You believe you are making 20%.

You are not. Work it through on a shop turning over NPR 500,000 a month at that markup:

Illustrative figures. Your ratios will differ. The point is the structure, not the numbers.

That 20% margin is really about 9% before your own time, and closer to 3% after it. This is why so many shops feel busy and stay poor. Three things quietly eat the difference:

• Dead stock. Goods that sit unsold are cash you have already spent, sitting on a shelf. A shop with NPR 100,000 of slow-moving stock has NPR 100,000 less to trade with.

• Uncollected credit. Udhaaro extended and never recovered is not a bad debt at the end of the year. It was a discount you gave at the moment of sale without deciding to.

• Your own unpaid labour. If you are working 70 hours a week for less than you’d earn employed, the business is not yet a business.

Price by category, not by habit

Fast-moving staples carry thin margins and should, because customers know those prices exactly. Slower, less-comparable goods carry the real margin. The mistake is applying one blanket markup across everything, which leaves money on the table at one end and prices you out at the other.

Karobar lets you set purchase price, selling price, MRP and a wholesale rate per item, so the margin is calculated for you rather than estimated in your head. More useful still, the product performance report shows which items are actually generating your profit as opposed to merely generating sales. Most owners are surprised. The item you sell most of is very often not the item paying your rent.

6. What to record from day one, and what can wait

Nepal has already gone digital on the money side. In a single recent month the country processed 59.26 million QR transactions worth NPR 162.58 billion (Kathmandu Post), with mobile banking at NPR 612 billion. QR volumes are up more than sixtyfold since 2020-21.

Yet most shops taking those QR payments still write the sale in a paper copy at the end of the day, if at all. Money arrives digitally and the record of it is handwritten. That gap is where the confusion lives.

You do not need full accounting on day one. You need four things, and they take about ten minutes a day.

What can wait: formal accounting, balance sheets, depreciation schedules, payroll systems. Those matter later, and mostly they matter to your accountant. Get the four basics right and everything else can be assembled from them.

Karobar covers all four in one place, and a few details matter more than they sound for a Nepali shop. It works offline, so an entry made during a connection drop still saves. It runs in Nepali and switches between Bikram Sambat and AD dates, so the record matches how you actually think about the year. If you already keep an Excel sheet, you can import it rather than starting over. And it backs up to the cloud, which matters more than owners expect, because a lost or water-damaged copy takes the business’s entire memory with it, and phones get lost.

Build the habit in the first week, while the volume is low. It is genuinely hard to start recording in month eight, when there is a year of gaps behind you and no obvious place to begin.

7. Your first 90 days, week by week

A note on the calendar

If you are opening in the second half of the year, festival timing matters more than almost anything else. Dashain begins with Ghatasthapana on 11 October 2026 and peaks at Bijaya Dashami on 21 October; Tihar runs to Bhai Tika on 11 November. Together they are the largest retail event of the Nepali year, worth roughly NPR 200 billion in festival spending (source), with around NPR 137 billion in remittances arriving between mid-September and mid-October.

Stock ordering for that has to happen weeks ahead. And Laxmi Puja, on 8 November 2026, is traditionally when Nepali businesses open new account books for the year, which makes it the natural moment to start keeping records properly if you haven’t already.

The rhythm that makes the 90 days work is small and daily: enter sales and expenses each evening before closing, check the day’s cash position against what’s in the drawer, and once a week look at three reports: what sold, what’s owed to you, and what stock is running low. Fifteen minutes a week of looking at numbers puts you ahead of most shops on your street.

8. When to hire, and what it really costs

The trigger for hiring is not “I’m busy.” It is one of two things: you are consistently turning away customers or closing early because you cannot cover the hours, or you are spending so much time on tasks worth NPR 100 an hour that you have no time for the ones worth NPR 1,000.

Budget for more than the salary

A staff member on NPR 20,000 a month costs meaningfully more than NPR 20,000. The extras are predictable, so plan for them:

Then confront the real question

Once someone else handles sales while you are not in the shop, you need to know what was sold and what came in without having watched it happen. This is the point at which paper records stop working, and it is the most common moment for a small Nepali business to lose money quietly. Not through theft so much as through drift, gaps, and honest mistakes nobody can reconstruct afterwards.

Karobar handles this with staff roles and permissions: your employee can record sales and check stock without seeing your profit margins, your total cash position or your supplier prices. Every entry is attributed. You can check the day’s takings from your phone without standing in the shop, and reconcile the drawer against the record in under a minute.

If you grow into a second outlet, multi-business support keeps each one’s books separate under a single account, and the web and desktop access means you can review both from a laptop rather than a phone screen. See features if you want to check a specific requirement before you commit.

The short version

• Make ten real sales before you sign a lease.

• Add up your costs, then add the six that everyone forgets, especially working capital and your own living expenses.

• Register as a sole proprietorship if you’re small and want it simple; a private limited if you want liability protection and credibility. Use the official sources for both.

• Get three supplier quotes, and negotiate credit terms rather than just price.

• Your margin is smaller than you think. Dead stock, unpaid udhaaro and your own unpaid time are where it goes.

• Record sales, expenses, credit and stock from day one. Ten minutes a day. Everything else can wait.

• Hire when you’re turning away customers, and put controls in place the same week.

The businesses that survive their first year in Nepal are rarely the ones with the best idea. They are the ones whose owners knew their numbers early enough to act on them.

Frequently asked questions