Most businesses in Kathmandu that have tried Google Ads describe the same experience: money went in, clicks came out, and nobody could say what the clicks turned into. The account gets paused after two months and the conclusion is that Google Ads does not work in Nepal.
It does work. It works well for a specific kind of business, and it fails predictably for the rest, usually for reasons that were visible before the first rupee was spent. This guide covers what is genuinely different about running Google Ads in Nepal: who it suits, how you actually pay for it, what a click costs here, how to set a first budget that can prove or disprove the channel, and how to build an account that a Nepali business can keep running.
Why Google Ads works differently in Nepal
Nepal is a Facebook-first market. DataReportal's October 2025 figures put Facebook's ad reach at 14.8 million people, half the population, against 4.35 million on Instagram and 2.3 million on LinkedIn. Roughly 16.6 million people, or 56 percent of the country, are online. Most marketing budget in Kathmandu follows the audience and goes to Meta.
That is exactly why Google search is underpriced here. Facebook reaches people who were not looking for you. Google search reaches people who typed "MBBS consultancy Kathmandu", "dermatologist near Lazimpat" or "ERP software Nepal" into a box because they want to buy that thing now. The number of people doing that in Nepal is smaller than the number scrolling Facebook, but the intent is not comparable, and because fewer local advertisers compete for those searches, the price of that intent is low by any international standard.
The other structural difference is that many searches that matter to Nepali businesses do not come from inside Nepal at all. Hotels, trekking operators, education consultancies, outsourcing firms and exporters are often selling to people in Australia, the UK, the US, India and the Gulf. Google Ads lets you buy that demand by country without opening an office anywhere, which nothing on the Facebook side does as cleanly.
Who should run Google Ads in Nepal, and who should not
Google search advertising rewards businesses whose customers already know what they are looking for and can be served at a value that survives the cost of a click. In Nepal that reliably includes:
- Education and abroad-study consultancies. High-value, high-intent, and searched constantly.
- Private healthcare. Clinics, diagnostics, dental, dermatology, fertility, pain management. People search symptoms and specialities, not brands.
- Hospitality and tourism selling to foreigners. Trekking, tours, hotels and travel agencies targeting searches made outside Nepal.
- B2B services and software. Accounting firms, ERP and software vendors, IT outsourcing, legal and company-registration services.
- Considered purchases. Vehicles, real estate, solar, furniture, appliances, where the buyer researches before visiting a showroom.
It suits far less well:
- Impulse consumer products under a few thousand rupees where the margin cannot absorb a click cost. These belong on Meta or TikTok.
- Brand-new categories nobody searches for yet. Search ads capture demand; they do not create it.
- Businesses without a working enquiry path. If calls go unanswered, WhatsApp messages sit for a day and the form does not work on mobile, ads simply make that failure more expensive.
Before spending anything, check whether people search for what you sell. Google's Keyword Planner shows Nepal-specific volumes once you have an account, even without a running campaign. Our guide to keyword research in Nepal covers how to read those numbers and what to do when the tool shows zero.
How to pay for Google Ads from Nepal
This is the part that stops more Nepali businesses than anything else, so it comes before strategy.
Google Ads does not bill in Nepali rupees. Accounts opened from Nepal are billed in US dollars, and a standard Nepali debit or credit card cannot settle a foreign-currency charge because Nepal Rastra Bank restricts international card payments. There are two legitimate routes.
Option 1: a prepaid dollar card in your own name
Most Class A commercial banks, including Nabil, Global IME, Himalayan and NMB, issue prepaid USD cards to account holders against a PAN and standard KYC. You load the card in rupees, it settles in dollars, and you add it to your Google Ads billing profile like any other card.
The limits are set by Nepal Rastra Bank and have changed several times. As of the April 2026 amendment, registered IT, communication and broadcasting businesses can load up to USD 3,000 a year for software and online services, extendable to USD 5,000 a year for firms that earn foreign currency, with higher tiers for documented exporters. The historic USD 500 personal ceiling has been relaxed according to several banks, but the exact figure your bank will apply depends on your account type and documents. Ask your branch for the current limit in writing before you plan a budget around it, because a card that stops loading mid-month pauses every campaign at once.
Option 2: an agency that bills you in rupees
Many agencies in Kathmandu, including ours, run client campaigns from an account that is funded internationally and invoice the client in NPR, including the 13 percent VAT. This removes the card problem entirely and is often the only practical option for a business that needs to spend more than the card limits allow.
It comes with one non-negotiable condition: the Google Ads account must be yours. The agency should work inside your account through a manager (MCC) link, so that the campaign history, conversion data and quality scores stay with you if the relationship ends. If an agency will only run ads from an account you cannot log into, you are renting your own data.
What to check on either route
- Set the account currency to USD deliberately at creation. It cannot be changed afterwards.
- Keep a rupee-per-dollar rate in your reporting so the finance team sees spend in NPR.
- Keep the media spend and the management fee on separate lines. A single "Google Ads package" figure hides which one you are paying for.
What a click actually costs in Nepal
International benchmark sets put Nepal's average search cost per click around USD 0.65, well below the global search average of roughly USD 2.70 and among the cheapest markets tracked. At around NPR 135 to 140 to the dollar that is under NPR 90 for an average click, and many local service categories run below that.
Averages hide the useful part. Competitive, high-value categories in Kathmandu, such as abroad-study consultancy, private hospitals, real estate and hotels targeting foreign searchers, sit well above the average because the businesses bidding there know what an enquiry is worth. Niche B2B and local service terms often cost a fraction of it because almost nobody is bidding. The only number that matters for your business is the one you observe in your own account after two weeks, and the first job of a small pilot budget is to find it.
Setting a first budget you can learn from
A pilot budget has one purpose: to produce enough data to tell whether the channel can work for you. Too small and you learn nothing; too large and you fund mistakes. A workable starting point for a Kathmandu service business is NPR 40,000 to NPR 80,000 a month in media spend for two to three months, separate from any management fee.
Work the maths backwards before you start:
- What is a new customer worth to you, on average, in profit? Say NPR 30,000.
- What share of enquiries become customers? Say one in five. An enquiry is therefore worth NPR 6,000.
- If one in thirty clicks becomes an enquiry, you can afford NPR 200 per click and still make money. If your observed CPC is NPR 60, the channel has room. If it is NPR 400, something upstream has to improve first.
If you cannot fill in the first two lines with real numbers, that is the work to do before opening an ad account. Our SEO cost guide uses the same approach for organic budgets and explains why "what does it cost" is the wrong first question.
An account structure that survives a Nepali business
The accounts we inherit in Kathmandu tend to share the same shape: one campaign, one ad group, forty broad-match keywords, no negatives and location targeting set to "Nepal" for a business that only serves Kathmandu valley. That structure spends money on searches from Biratnagar for a clinic in Baluwatar.
Keep it simple and keep it separable:
- One campaign per budget you want to control. If you serve both Nepali and Australian customers, that is two campaigns with different bids, languages and landing pages, not one campaign targeting both countries.
- Ad groups by intent, not by product list. "Dental implant cost" and "emergency dentist" are different people and need different ads.
- Location targeting tight enough to match delivery. Kathmandu, Lalitpur and Bhaktapur for a valley business, with the "presence" setting rather than "presence or interest" so you are not paying for people abroad who are merely interested in Nepal.
- Ad schedule matched to when you can answer. If nobody picks up after 6 pm, do not buy clicks after 6 pm.
The full reasoning, including how to handle Performance Max and brand terms, is in our article on Google Ads account structure that survives contact with reality.
Keywords, match types and the negative list
Nepali search behaviour is mixed-language and mixed-script. People search "best hospital in kathmandu", "kathmandu ma ramro hospital" and occasionally in Devanagari, often with spelling variants. Broad match will happily spend your budget on all of it, including job searches, "salary", "course" and "free" queries that no ad should serve.
Start with phrase and exact match on a small, tight list. Add a starter negative list on day one: jobs, vacancy, salary, course, training, free, download, PDF, meaning, wikipedia. Then review the search terms report weekly and move budget toward the queries that produce enquiries. The routine is set out in search term mining: the weekly habit that protects your ad budget. In most Nepali accounts it recovers 20 to 40 percent of spend within the first month.
Where the click lands
Sending paid traffic to your homepage is the most common way to waste it. The person who searched "invisalign price kathmandu" wants a page about Invisalign, the price or how to get it, and a way to ask a question, in that order, on a phone, in under three seconds.
Build one landing page per intent group. Match the headline to the search, put the primary action above the fold, and make the enquiry options the ones Nepali customers actually use: a tap-to-call number, a WhatsApp or Viber button, and a short form as a fallback rather than the main event. We explain message match and its effect on your Quality Score, and therefore your click price, in landing pages for paid search, and cover the wider conversion work under conversion rate optimisation.
Tracking the conversions that actually happen in Nepal
Most Nepali businesses close enquiries on the phone or in a chat app, not on a thank-you page. If your account only counts form submissions, Google's bidding will optimise toward the wrong people and your reports will say the campaign is failing while the front desk is busy.
At minimum, track four things as conversions: phone calls from ads, tap-to-call clicks on the landing page, WhatsApp or Viber button clicks, and form submissions. Set up Google Ads conversion tracking and GA4 key events on day one, and record which leads became customers in a simple sheet so that after three months you can compare cost per customer, not cost per click. That comparison is the only report worth having.
Google Ads versus SEO in Nepal
They answer different questions. Google Ads buys visibility immediately and stops the day you stop paying. SEO takes months to start and then compounds. A business that has neither should usually run ads on its highest-intent terms now, while the organic work matures, and then reduce paid spend on the terms it starts to rank for.
The mistake is treating them as alternatives. In Nepal, where organic competition is still weak in many categories, a well-run ads account also produces the best keyword research you will ever get: it tells you, with money attached, which searches turn into customers. That list is the brief for your SEO content. We set out how the channels fit together in how digital marketing channels work together in Nepal.
The mistakes we see most in Nepali accounts
- Targeting "Nepal" for a valley business. Half the budget serves people you cannot deliver to.
- Accepting Google's recommendations wholesale. Auto-applied recommendations broaden match types and raise budgets. Turn them off.
- Display and Search in one campaign. The Display Network box is ticked by default and quietly consumes budget on low-intent placements.
- No conversion tracking. Which means no learning, and no way to justify continuing.
- Judging the channel in the first two weeks. Smart Bidding needs conversion data before it is useful. Give a pilot eight to twelve weeks and a weekly search-term review.
- Running ads to a site that cannot convert. Ads reveal a weak website faster than anything else.
- Not owning the account. Covered above. It is the one you cannot undo later.
Working with an agency on Google Ads
You do not need an agency to start. A small, tight account with phone and WhatsApp tracking is achievable for an owner willing to spend a few hours a week on it. An agency earns its fee when the account gets large enough that weekly search-term work, landing page testing, conversion tracking and cross-country campaigns become a real job, or when the card limits make in-house payment impractical.
Whoever runs it, insist on four things: the account is in your name, media spend and fee are itemised, conversions are defined as calls and chats as well as forms, and you receive the search terms report, not just a dashboard. Our paid search service is scoped that way, and our pricing page explains how engagements are sized. If you want a second opinion on an existing account, send us access and we will tell you honestly whether it is worth continuing.