Selling online in Nepal is more viable than it was three years ago and still harder than the platform advertisements suggest. Payments work. Delivery works, mostly. What catches people out is everything between the order and the customer actually having the product in their hands.
This covers the commercial decisions — the ones to make before anyone writes code.
Decide what kind of store you are building
| Model | Works when | Main difficulty |
|---|---|---|
| Own inventory | You control quality and margin | Capital tied up in stock |
| Made to order | Craft, custom, personalised goods | Managing lead times honestly |
| Dropship / supplier-fulfilled | Low capital, testing demand | You own the complaint, not the stock |
| Digital products | No logistics at all | Payment and piracy, not delivery |
Most first-time Nepali stores hold their own inventory, and that is usually right — control over what arrives at the customer’s door is worth a great deal when you are building a reputation from nothing.
Cash on delivery is not optional yet
A meaningful share of Nepali customers still prefer to pay when the product arrives, particularly for a first purchase from a business they do not know. Refusing cash on delivery removes real orders.
But it needs treating as its own operational problem, not just another checkout button:
- Refusal rate. Some COD orders are declined at the door. Budget for it and track it, because it is a real cost in delivery fees and returned stock.
- Cash reconciliation. Your courier collects money and remits it later. That reconciliation is a weekly job someone must own.
- Confirmation calls. Most Nepali stores confirm COD orders by phone before dispatch. It reduces refusals substantially and it is labour.
Treat COD as a payment method with a workflow attached, and it works. Treat it as a checkbox and it silently eats margin.
Digital payments
eSewa and Khalti between them cover the overwhelming majority of online wallet payments here, so supporting both is the practical baseline. Customers are loyal to whichever they have topped up — a missing option is a lost sale rather than a preference.
Card payments matter mainly if you sell abroad. The reconciliation and compliance overhead is not worth it for a domestic-only store.
The engineering side genuinely matters here — signed requests, idempotency, and server-side verification rather than trusting a redirect. Get that wrong and you will either double-charge customers or ship goods that were never paid for. We covered the specifics in integrating eSewa and Khalti.
Delivery is the part that decides your reviews
Inside Kathmandu Valley, same-day and next-day delivery are achievable through established courier services. Outside the valley, timelines stretch and reliability varies considerably by district.
Two rules that prevent most complaints:
Publish honest timeframes by region. “3–5 days inside the valley, 5–9 days outside” beats “fast delivery” and then a week of silence. Under-promising is free.
Give customers tracking, even if it is manual. A status update — confirmed, dispatched, out for delivery — removes most “where is my order” messages. Early on this can be a person updating a field; it does not need to be an integration.
Platform choice
| Option | Suits | Trade-off |
|---|---|---|
| WooCommerce | Most Nepali stores | You own it; you also maintain it |
| Shopify | Fast launch, international selling | Monthly fee in USD; local gateway support is limited |
| Custom build | Unusual workflows, scale | Highest cost; only worth it when you have outgrown the alternatives |
| Social selling only | Testing demand | You are building someone else’s audience, not your own asset |
For most Nepali businesses WooCommerce is the sensible default, largely because local payment gateway support is mature and you keep ownership. Shopify is excellent software whose local payment story is weaker, which matters more here than it does elsewhere.
On social selling: running a store through Instagram or Facebook is a legitimate way to prove demand. It is not a long-term foundation — you cannot rank in search, you cannot own the customer relationship, and a platform change can remove your business overnight.
The costs people underestimate
Product data. Photographing, describing, categorising and pricing a few hundred products is days of work, and it is yours rather than your developer’s. This delays more launches than any technical factor.
Returns. Decide your policy before launch and publish it plainly. An unclear returns policy generates disputes that cost more than the returns.
Ongoing maintenance. An e-commerce site holds customer data and takes payments. It is not a brochure site you can leave alone for a year — see what website maintenance actually covers.
Getting found. A store with no traffic sells nothing. Budget for that from the start rather than discovering it in month three.
A sensible launch sequence
- Apply for merchant accounts immediately. Approval takes longer than development and routinely delays launches.
- Prepare product data while the site is being built. These run in parallel or you wait.
- Agree courier arrangements and confirm real timelines by region.
- Launch with fewer products, done well. Twenty properly photographed products beat two hundred with placeholder text.
- Test the full path yourself — order, pay, receive, return — before telling anyone the store is open.
That last step is skipped constantly and catches things nothing else does.
Frequently asked questions
How much does an e-commerce website cost in Nepal?
A working store generally falls between NPR 150,000 and 500,000 depending on product complexity, integrations and design. The drivers are broken down in what a website costs in Nepal.
Do I need a registered business to sell online?
To accept payments through eSewa or Khalti as a merchant, yes — they require business registration and PAN. You can test demand informally, but a proper storefront needs the paperwork.
Should I use Daraz instead of my own store?
They serve different purposes and many businesses do both. A marketplace brings existing traffic and takes a commission and the customer relationship. Your own store costs more to fill but is an asset you control. Marketplace for volume, own store for margin and brand.
How do I handle returns in Nepal?
Publish a clear policy with a specific window, decide who pays return shipping, and apply it consistently. Most disputes come from ambiguity rather than from the policy itself being strict.
Is cash on delivery risky?
It carries a refusal rate and a reconciliation burden, and it is still worth offering for most Nepali stores because it removes the trust barrier for first-time buyers. Confirmation calls before dispatch reduce refusals substantially.
How long does it take to build an online store?
Eight to fourteen weeks for a straightforward store, with product data preparation usually the critical path rather than development.
Getting started
If you are considering this, do two things this week: apply for your merchant accounts, and start photographing products. Both take longer than you expect and neither depends on the website existing.
When you are ready to build, our e-commerce development work covers the store, the gateway integration and the reconciliation that keeps the numbers honest.
