The SEO mistakes costing Nepali businesses customers — abandoned sites, GBP policy risks, review gating, wasted speed spend — and what fixing each costs.
Most of the SEO problems we see in Nepal are not exotic. They are not penalties, not algorithm updates, not some technical curse buried in the code. They are ordinary decisions that looked reasonable at the time and turned out to be wrong.
That distinction matters, because the fix is usually cheap once you can name the problem. The expensive part is the two years a business spends not knowing.
What follows is a diagnostic, not a checklist. For each mistake, the symptom you would actually notice, why it keeps happening here specifically, and what it costs to correct.
Mistake one: treating a website as a thing you finish
A pattern shows up in almost every first conversation. The business paid for a site three or four years ago, signed off on it, and has not touched it since. The phone number in the footer belongs to a staff member who left. The price list is pre-pandemic. There is a blog with two posts, both from launch month.
The site is not broken. It is just abandoned, and abandonment has a specific cost.
Search results are competitive by nature. You are not being judged against some fixed standard of quality — you are being judged against whoever else answers the same question. A restaurant page that was excellent in 2022 loses ground not because it got worse, but because three competitors wrote something better since.
Here is the part people underestimate: the damage is invisible from inside. Your site still loads. It still looks fine. Nothing alerts you. You simply stop being found for queries you never knew you were losing, and the drop is gradual enough that it never registers as an event.
The correction is not "post a blog every week." That advice has produced an enormous amount of worthless content in Nepal. It is closer to this: pick the four or five pages that actually earn you money, and make sure a customer reading them today gets accurate, current, useful answers. Then check again in six months.
An uncomfortable trade-off
Publishing often does help, when there is something to say. But a business that forces out weekly posts with nothing behind them ends up with a site full of thin pages that dilute attention and impress nobody. Silence is better than filler. Almost every small business would be better served by six genuinely good pages than sixty forgettable ones.
Mistake two: guessing at what customers search
Businesses describe themselves in the language of their own industry. Customers search in the language of their problem. These overlap less than you would expect.
A clinic writes about "comprehensive diagnostic services." Patients type "blood test near me price." A furniture workshop advertises "bespoke joinery solutions." Buyers search "sofa set price Kathmandu." Neither business is wrong about what it does. Both are invisible to the people looking for exactly what they sell.
This is easy to check and almost nobody does. Ask three recent customers how they would have searched for you before they knew your name. You will get answers that contradict your homepage.
Free tools cover most of what a small business needs — Keyword Planner (free with a Google Ads account, though volumes show as ranges without active spend), Search Console once you have some traffic, and the autocomplete suggestions Google offers as you type. Paid tools like Ahrefs or Ubersuggest add depth, but their Nepal-specific data is thin enough that we would not make them the first purchase for a business spending its own money.
The volume trap
A related error runs the other direction: chasing whatever term has the biggest search volume. "Digital marketing" gets far more searches than "SEO for restaurants in Pokhara." It also brings students, job seekers, and researchers from across the region, roughly none of whom will hire you.
Narrow terms convert better and cost less to win. A page that ranks first for a query fifty people make each month, where most of them want to buy, beats a page that ranks eleventh for a query five thousand people make out of curiosity.
Mistake three: half-finished local visibility
Google's local ranking rests on three factors it names openly: relevance, distance, and prominence. Google also states that complete and accurate business information makes a business more likely to appear in local results, and that both review count and rating feed local ranking.
Most Nepali businesses handle this partway. The Google Business Profile exists. It was created once, verified, and then left with no hours, no photos since 2021, a category that was close enough, and reviews nobody has answered.
Two errors here are worth singling out because they carry real risk rather than just lost opportunity.
The first is stuffing keywords into the business name field. "Sharma Traders - Best Electronics Shop in Kathmandu" is a policy violation, and Google's own documentation says extra keywords, city names, or taglines in the name field could result in the suspension of your Business Profile. Suspension is Google's word, not ours. A short-term ranking bump is not worth losing the profile entirely.
The second is review gating — asking happy customers for reviews while quietly routing unhappy ones to a private feedback form. Businesses assume this is clever reputation management. It is a documented policy violation. Google's review policies prohibit discouraging or prohibiting negative reviews, or selectively soliciting positive ones. Incentivised reviews are also banned, including offering something in exchange for revising or removing a negative one.
You can ask every customer for a review. You cannot ask only the ones you expect to be kind.
Google says replies to reviews matter for customers, but makes no specific ranking claim about replying. We still recommend it, on the plainer logic that a prospective customer reading a defensive or absent response is a prospective customer you lost.
Mistake four: mistaking speed for a scoreboard
Site speed is where good intentions go furthest astray, because there is a number attached and numbers invite obsession.
Google's Core Web Vitals thresholds are specific: Largest Contentful Paint at or under 2.5 seconds, Interaction to Next Paint at or under 200 milliseconds, Cumulative Layout Shift at or under 0.1, all measured at the 75th percentile of real visits. INP replaced First Input Delay in March 2024, and the thresholds have been stable since.
What gets lost is Google's own framing around them. Google states plainly that there is no single signal behind page experience, that good scores do not guarantee rankings, and — in a phrase worth quoting exactly — that chasing "a perfect score just for SEO reasons may not be the best use of your time." That is Google's language, and it contradicts most of the speed-optimisation sales pitch circulating in Kathmandu.
Here is the honest version. A site that takes eight seconds on a mid-range Android phone on mobile data is losing customers, full stop, regardless of rankings. That is worth fixing urgently. A site already loading in 2.8 seconds, where an agency is quoting real money to reach 2.4, is a business being sold a decimal point.
The causes are usually the same handful: uncompressed images straight from a phone camera, cheap oversubscribed shared hosting, a WordPress install with twenty plugins where four would do, and a page-builder theme carrying features nobody uses. None of these need an expert to identify. Most need one to fix properly.
Test on what your customers actually hold
Almost every Nepali audience we look at is mobile-first, and mobile-first indexing is now universal — Google completed the transition in 2023 and moved the last remaining sites to its smartphone crawler in July 2024. Whatever exists only on your desktop version effectively does not exist.
Yet sites still get signed off on a designer's laptop over office fibre. Load your own site on a three-year-old phone on mobile data, in daylight, and try to complete an order. The gap between that experience and the desktop mockup is where most conversions die.
Mistake five: publishing content nobody needed
Two failures live here and they need separating, because the standard advice muddles them.
The first is thin content — pages that exist to hold a keyword and say nothing. Google folded its helpful content system into core ranking in March 2024, so this is no longer a separate filter you might dodge. It is part of how pages are assessed.
The second is copied content, and here a persistent myth deserves correcting. There is no general "duplicate content penalty." Google has said so directly, and its position is that duplicate content is not grounds for action unless the intent appears deceptive and manipulative. What actually happens with duplicates is an indexing decision: Google picks one version and filters the rest.
That is not permission to copy. Scraping a competitor's service page is separately covered by spam policy, and beyond policy, it guarantees you can never be the better answer — you have volunteered to be the second copy of someone else's page. But it does mean the fear driving some businesses to avoid republishing their own material across their own channels is misplaced.
What genuinely works is narrower than "write helpful content." Write down the answers you give on the phone all day. The questions customers ask before buying, the ones they ask after, the objections, the comparisons. Those pages are hard for a competitor to fake because they come from doing the work.
Keyword stuffing has not gone away
It still turns up, usually in older sites. Google's spam policies define it as filling a page with keywords to manipulate rankings, and specifically call out blocks of text listing cities and regions a page hopes to rank for. That district-name paragraph at the bottom of a service page is a named example in Google's documentation.
Meanwhile, Google actively encourages putting keywords where visitors will see them — titles, headings, alt text, link text. The line is not frequency, it is whether the text was written for a reader.
Mistake six: buying links because someone offered
Directory packages and "100 backlinks" offers still circulate widely here, and they are cheap enough to seem harmless.
Google's spam policies list buying or selling links for ranking purposes as link spam, including exchanging money, goods, or services for links. Paid placement is legitimate when the link is marked with rel="sponsored" or rel="nofollow" — but a link marked that way is not meant to pass ranking credit, which is precisely why the packages do not mark them.
Google's own position is that its systems nullify unnatural links at scale rather than punishing every site that has them. That is still money spent on nothing, by a business that could rarely afford it, often instead of the content work that would have helped.
Mistake seven: no measurement, or measuring the wrong thing
Businesses that never install Google Analytics or Search Console are working blind. That is common and easy to fix in an afternoon.
The subtler failure is installing them and watching the wrong number. Traffic is the seductive metric because it always goes up if you publish enough. Inquiries are the one that pays salaries.
We have seen sites grow their traffic substantially while inquiries stayed flat, because the growth came from informational articles read by people who were never going to buy. That is not a success with a lag. That is a signal to redirect effort toward pages closer to a purchase decision.
Search Console is the more useful of the two free tools for most owners, and the least used. It shows the actual queries bringing people to you, which is often the fastest correction to a wrong assumption about your own market. Pair it with something dull but honest: ask every caller how they found you, and write it down.
A note on meta descriptions, since they belong to this same family of misplaced effort. Google uses them for snippets, not ranking, and frequently rewrites them anyway based on the query. Write clear ones because they affect whether someone clicks. Do not expect them to move positions.
What actually causes these mistakes
Look across all seven and a common cause appears. Almost none of these are knowledge failures. Nobody thinks abandoned content is good or that buying links is a sound plan.
They happen because SEO has no deadline. Nothing breaks. No customer complains. There is always something more urgent, so the work slides — quarter after quarter — until a competitor who did not let it slide is taking calls that used to come to you.
The businesses that get this right are rarely the ones with the biggest budgets. They are the ones who treated search as a standing responsibility with a name attached, even at two hours a month, rather than a project that finished when the site went live.
Where outside help earns its cost is in the diagnosis and the first correction — figuring out which of these is actually holding you back, and fixing that one properly instead of doing a little of everything. That is the shape most of our engagements at Webpal take, and honestly, the ones that end with the client handling their own upkeep afterward are the ones that worked.
Start here: open your site on your phone, on mobile data, and try to buy from yourself. Then open your Google Business Profile and read it as a stranger would. Most businesses find their most expensive mistake within ten minutes, and it is almost never the one they were worried about.
If you would like a second opinion on what those ten minutes turn up, Webpal runs site audits for businesses across Nepal — and we will tell you plainly if the answer is that nothing needs fixing.