Key takeaways
- Check indexation before anything else — a page Google has not indexed cannot rank, and this is the single most common cause.
- Page speed is a harder constraint in Nepal than most guides assume, because real connection speeds outside Kathmandu are slower than the benchmarks.
- Targeting terms nobody searches for looks identical to failing SEO, but the fix is completely different.
- Most non-ranking sites have fixable technical problems rather than fatal ones. A rebuild is rarely the cheapest answer.
A site that does not rank is rarely broken in an exotic way. Across audits of Nepali business sites, the same six causes come up repeatedly — and they are usually cheaper to fix than the rebuild people assume they need.
Here they are in rough order of how often I find them.
1. Google has not indexed the pages
A page that is not in the index cannot rank, and this catches more sites than every other cause combined.
Open Google Search Console, go to Pages, and look at what is excluded and why. Common culprits:
- A
noindextag left on from a staging environment. This happens constantly — the site launches, nobody removes the tag, and the site is invisible for months. robots.txtdisallowing directories that contain real content.- No sitemap submitted, on a site with weak internal linking.
- Canonical tags pointing at the wrong URL, so Google consolidates several pages into one it was not meant to.
If you have never set up Search Console, do that first. Everything else in this list is guesswork until you can see what Google actually sees.
2. The site is too slow for the connections it serves
Google treats 2.5 seconds as the threshold for the largest element to paint. Most guides then benchmark on a fast connection and declare victory.
That does not hold here. On the mobile connections common outside Kathmandu, a page taking six seconds loses a meaningful share of visitors before they see anything — and bounce behaviour of that kind is not something ranking survives indefinitely.
The usual causes are unglamorous: page-builder templates loading scripts nobody uses, uncompressed images at full camera resolution, five webfonts where one would do, and a chat widget plus analytics plus a popup tool all loading before the content.
Run PageSpeed Insights on your own site, on mobile, and read the opportunities list. It is usually accurate.
3. You are targeting terms nobody searches for
This one is disguised, because it looks identical to failing SEO from the inside — effort in, nothing out.
The difference is that the pages might be ranking perfectly well for phrases with no search volume. A page ranking first for a term nobody types produces exactly as much traffic as a page ranking nowhere.
Check what your pages actually receive impressions for in Search Console, and compare that against what you assumed they targeted. The gap is often large. Businesses tend to write around internal vocabulary rather than the words buyers use, and in Nepal there is an additional split between English and Nepali phrasings for the same intent.
4. Thin content across too many pages
Ten short pages covering variations of one topic will lose to one thorough page, every time. They also compete with each other, so none of them accumulates enough authority to rank.
If you have pages titled Web Design, Website Design, Web Design Services, and Best Web Design, you do not have four assets. You have one asset split four ways.
Consolidate, redirect the losers to the winner, and write one page that fully answers the question.
5. No local signals
For any business serving a geography, local SEO is a separate discipline that many sites skip entirely.
The basics: a claimed and completed Google Business Profile, consistent name-address-phone details across every directory and profile, and pages built around how people actually search — which in Nepal is usually a service plus a place.
This is high-leverage and cheap. It is also the fastest-moving of everything in this list; a properly maintained Business Profile can start producing phone calls within two months.
6. Nothing has changed for two years
Search results move because competitors act. A site that was optimised once in 2023 and left alone is not holding position — it is drifting down while others publish.
This does not require industrial content output. Two well-researched pages a month, answering questions buyers genuinely ask, sustains steady growth for most businesses here. Fewer, better pages beat volume.
How to work out which one you have
In order, because each step makes the next one meaningful:
- Set up Search Console. Look at Pages and Performance.
- Run PageSpeed Insights on mobile, on your three most important pages.
- Compare the queries you receive impressions for against the terms you meant to target.
- Search your own main service term and see who ranks. Read their top page and be honest about whether yours is better.
Most of what this surfaces is fixable without touching the design. A rebuild only becomes the cheaper option when the platform itself blocks the fundamentals — no control over titles, unfixable speed, or URLs that change every time someone edits a page.
Diagnose before you spend. The audit costs a fraction of the rebuild, and it frequently shows the rebuild was never necessary.
Questions
How long before a new site ranks?
Indexation usually happens within days to a few weeks once Search Console is set up and a sitemap is submitted. Ranking for competitive commercial terms takes three to six months in Nepal, longer in categories where established players hold years of accumulated links.
Do we need to rebuild the site?
Usually not. A rebuild only becomes the cheaper option when the platform blocks fundamentals — no control over page titles, unfixable page speed, or URLs that change on every edit. An audit tells you which situation you are in before you spend anything.
