What a website actually costs in Nepal
Website development cost in Nepal runs from roughly NPR 20,000–50,000 for a basic or portfolio site, NPR 50,000–150,000 for a business website, and NPR 130,000–500,000 for an e-commerce store, with custom web applications starting around NPR 150,000 and rising with complexity. Domain and hosting add a few thousand rupees a year on top. Those are the bands quoted prices actually cluster into, drawn from rate cards published by Nepali web development companies as of mid-2026.
The full published range is far wider — roughly NPR 15,000 at the bottom to NPR 1,000,000 at the top. That spread is not evidence that the market is dishonest. It is evidence that “a website” describes at least four different products.
| What you are buying | Typical quoted range | Who it suits |
|---|---|---|
| Static one-page or small brochure site | NPR 20,000–30,000 | A single service, an event, a landing presence |
| Basic / portfolio website | NPR 25,000–50,000 | Individuals, startups, home / about / services / contact |
| Business website | NPR 50,000–150,000 | Custom design, blog, forms, several service pages |
| Industry dynamic site | NPR 60,000–160,000 | Trekking and travel companies, schools, news portals |
| E-commerce store | NPR 130,000–500,000 | Payment gateway, product management, order tracking |
| Custom web application | NPR 150,000–500,000+ | Booking engines, marketplaces, internal integrations |
On top of the build, two recurring costs are unavoidable and small: domain registration at roughly NPR 1,200 to 2,000 a year, and shared hosting at roughly NPR 2,500 to 6,000 a year. A .com.np domain is registered through Mercantile and requires documentation, which is worth starting early because it is the step most likely to delay a launch. Renewal is a separate puzzle — the .np domain renewal and expiry guide lists what to check.
Estimate your build: a quick NPR calculator
The same market bands as the table above, arranged as a calculator. Pick what you are commissioning and it returns the published range you should expect quotes to land in — an observation of the 2026 market, not a quote from me.
Pick a website type and press the button. If the calculator does not load, the band table above contains the same numbers.
What actually drives the number
Two quotes that differ by three times are usually quoting different scopes rather than different quality. Five variables explain most of the gap.
Template or custom design
A purchased theme configured to your brand is dramatically cheaper than a design produced from scratch, and for most Nepali businesses it is entirely sufficient. Custom design earns its cost when your brand is the product — hospitality, high-end travel, anything where the visual impression is doing commercial work.
Configured or built functionality
A booking form assembled from an existing plugin costs a fraction of a booking system written for you. The question to ask is whether your requirement is genuinely unusual. Most are not. Trekking itineraries, room availability, and course enrolment all have well-solved patterns.
Who builds it: freelancer or agency
The same brief priced by a solo freelancer and by an agency will come back with very different numbers, and the gap is mostly overhead rather than quality. An agency quote carries project managers, designers, testers, and an office. A freelancer's carries none of that — what you give up is capacity and continuity, because one person can be ill, busy, or abroad. For a five-page business site, an experienced freelancer is often the better value. For e-commerce with a payment gateway and a hard deadline, an agency's redundancy is worth something. Judge the portfolio and the scope document, not the label.
Who writes the content
This is the line item most often left out of comparison. If the quote assumes you will supply finished text and images for fourteen pages, and you do not have them, your project will stall for months or you will pay for content later at a higher rate. Ask explicitly.
What happens after launch
Hosting, SSL, backups, plugin updates, security patching, and someone to call when something breaks. Maintenance plans published by Nepali developers run from roughly NPR 2,000–5,000 a month for backups and updates to NPR 15,000–40,000 a month when content changes and performance work are bundled. A quote that excludes all of this is not cheaper, it is incomplete.
The expensive mistake: paying twice
Almost every costly piece of SEO remediation I am asked to do exists because a website was designed before anyone asked which searches it needed to appear for. The build was fine. The structure was decided by a designer optimising for a homepage layout rather than for how people search.
The five most common and most expensive versions:
- All services on one page. A single “Services” page listing eight offerings cannot rank for eight different searches. Each service that people search for separately needs its own URL. Splitting this after launch means new pages, new content, and internal linking work — several times the cost of doing it at build time.
- Content that cannot be edited. If adding a page or changing a paragraph requires the developer, your site will not change, and a site that does not change does not build search presence. Insist on a CMS you can actually use.
- No URL plan. Auto-generated URLs like
/page-id-47, or a structure that has to be reorganised later. Every URL change needs a redirect, and redirects that are missed lose whatever the old page had earned. - JavaScript-only content. Sites where the text only appears after scripts run can be indexed inconsistently. This is a solved problem when raised at build time and an awkward one afterwards.
- Launching without analytics. Google Analytics and Search Console cost nothing and take twenty minutes. Sites launched without them have no baseline, which means nobody can later prove what any subsequent work achieved.
None of these are exotic. All of them are cheaper to prevent than to repair, and none of them require you to spend more on the build.
What to put in the brief
You do not need technical knowledge to protect yourself here. You need seven sentences in the brief.
- A page list. Name every page the site will have, including one page per service or product category that customers search for separately.
- A URL structure. Readable, lowercase, hyphenated, describing the page —
/services/technical-seo, not/p?id=12. - Content responsibility. State in writing who writes and who supplies photographs, and by when.
- CMS access. You can add pages, edit text, and change images without the developer.
- Analytics at launch. Google Analytics and Search Console installed and verified, under accounts you own.
- Speed on a real phone. Not a desktop demo — a mid-range Android device on mobile data, which is how most of your Nepali visitors will arrive.
- Everything in your name. Domain, hosting, analytics, and CMS admin registered to you, with the developer added as a user.
That last point deserves emphasis. Domains registered under a developer's account and hosting bought on their card are the most common reason a business cannot leave a supplier it has outgrown. It is not usually malicious. It is convenience at setup that becomes leverage later.
If you already have a website
Businesses often assume a rebuild is the answer to poor search performance. Sometimes it is. Frequently it is the most expensive way to fix a problem that has nothing to do with the design.
A rebuild is justified when the site cannot be edited, is not mobile-usable, is built on something unmaintainable, or has a structure that genuinely cannot express what you sell. A rebuild is not justified when the real issue is that the site has four pages and needs twenty, or that nobody has ever done keyword research, or that the Google Business Profile is unverified. Those are content and configuration problems, and rebuilding the site solves none of them while resetting whatever search history the domain has accumulated. If the real question is why the site is not ranking, diagnose that first — it is usually cheaper than a rebuild.
Redesigns also carry a specific risk: sites regularly lose rankings after a rebuild because URLs changed without redirects. If you do rebuild, map every existing URL to its replacement before launch. This is a half-day of work that prevents the single most common post-redesign disaster.
The sequence that costs least
If you are starting from nothing, the order that avoids paying twice is straightforward.
Establish what people actually search for in your category first. This is a few days of keyword research, and it determines your page list — which is the single most consequential structural decision in the entire project. Then brief the build against that page list. Then launch with analytics and a verified Google Business Profile. Then, once there is a working site with a sensible structure, invest in content and search visibility.
Businesses that follow that order spend less overall than businesses that build first and retrofit search later, because the retrofit is nearly always more expensive than the planning would have been.
Frequently Asked Questions
The short version
Expect NPR 25,000 to 50,000 for a small site, NPR 50,000 to 150,000 for a business site, and NPR 130,000 upward for e-commerce, plus a few thousand a year for domain and hosting. Judge quotes on scope rather than headline price, and ask specifically about content, post-launch support, and ownership. Decide your page list before the build rather than after it. And register everything in your own name, because the cheapest website in Nepal is the one you never have to pay for twice.
If you are about to commission a build and want the structure sanity-checked before it starts, send me the brief. It takes me an hour and it is free — it is far cheaper for both of us than fixing it in a year.