The most common message I get starts the same way. We built a website last year. It looks good. Nobody finds it.
The website usually does look good. That is the confusing part. Somebody was paid, something was delivered, everyone signed off. And then it sat there, invisible, while the owner slowly concluded that SEO is a scam or that Google hates small businesses.
Neither is true. In almost every case it is one of four things, and they are distinguishable from each other in about twenty minutes.
First, check whether Google has actually seen it
Before diagnosing anything, do this. Search Google for:
site:yourdomain.com
That asks Google to list every page it has indexed from your domain. The result tells you which conversation you are having:
- No results at all. Google does not have your site. This is a crawling or indexing problem, and it is the most fixable of everything on this page.
- Far fewer pages than you have. Google has part of your site. Something is blocking the rest.
- All your pages are listed. Google has your site and has decided not to rank it highly. This is a completely different problem, and no amount of technical work will fix it.
Most people never run this check and spend months solving the wrong problem. If your pages are indexed and you are still invisible, skip to the last section — the technical sections will waste your time.
Reason one: the site never told Google what it is
When I audited 34 live Nepali websites in August 2026, 41 per cent had no <h1> element at all, and 47 per cent had no structured data. Nine of the 34 had neither.
Think about what that means from Google's side. It arrives at a page. There is no headline element telling it the subject. There is no machine-readable description telling it this is a dental clinic in Sanepa with these opening hours. There is a meta description, maybe, and about 60 per cent of the time there is a canonical tag.
Google will still work something out from the body text. It is good at that. But you have made it guess, and it is competing against a page that did not make it guess.
This is the cheapest thing on this list to fix. One H1 per page that states the subject. A correct schema block. A meta description you have actually read. None of it changes your design.
Reason two: nobody is searching for what the site says
This is the reason people least want to hear, and it is the most common one among sites that are properly indexed.
A trekking company writes a homepage headline like "Your Journey Begins With Us". A clinic writes "Committed to Excellence in Patient Care". These are not search terms. Nobody types them. The page is optimised for a brochure, and it is competing in a medium where the entry ticket is matching what someone typed.
The people searching are typing things like "annapurna base camp trek cost", "dentist open saturday kathmandu", "root canal price nepal". Those are the queries with intent behind them, and they are usually nowhere on the site.
The fix is not keyword stuffing. It is noticing that your customers describe their problem in words you never use, and then using those words where they belong: in headings, in page titles, in the first paragraph. If you sell trekking, you need a page about the cost of the trek, because that is what people are actually asking.
I have written about how to work this out properly in the piece on search intent.
Reason three: everything lives inside JavaScript
This one is invisible from the front end, which is why it survives so long.
Some sites render nothing in their initial HTML. The server sends a near-empty page, JavaScript runs, and the content appears. In a browser this looks perfect. To a crawler it is an empty room. Google can render JavaScript, but rendering is queued and it is not guaranteed, and anything that depends on a third-party script that fails will simply not be there.
You can test this in one command. In a terminal:
curl -s https://yourdomain.com | grep -o "<h1.*</h1>"
If your headline does not come back, it was not in the HTML. Or, without a terminal: open your page, view source, and use Ctrl-F to look for a sentence you can see on the screen. If the sentence is not in the source, a crawler did not get it either.
I ran into a version of this on my own site. Every page loaded a shared JavaScript file, and that file got truncated to zero bytes on the server. The site kept returning HTTP 200. Nothing alerted. But the hero photograph, the book cover and the headline statistics were all hidden behind a CSS rule that only JavaScript could remove, so for three days visitors saw blank space where the content should have been. The fix was to make the HTML render on its own and treat the JavaScript as decoration. It is a better architecture regardless of crawlers.
Reason four: the site is slow where your customers are
Nepal has 16.6 million internet users, and 82.8 per cent of mobile connections run on 3G, 4G or 5G (DataReportal, Digital 2026: Nepal). That is a mobile-first market on connections that vary enormously by time and place.
In the audit, median time to first byte was 817 milliseconds, which is fine. But 13 of 34 sites took over a second, five took over 2.5 seconds, and one took 10.4 seconds. Seven sites had five or more render-blocking scripts in the head; the worst had 21.
Speed rarely makes a site invisible on its own. What it does is compound. A slow site gets crawled less often, so new pages take longer to appear. It loses visitors before the page paints, which feeds back as a poor engagement signal. And on a congested mobile connection in the evening, a 2.5-second server response becomes something much worse.
Also worth checking: 73 per cent of the images in that audit had no width and height attributes, which causes the page to jump around as it loads. That is a Core Web Vitals failure that costs nothing to fix.
If you are indexed and still not ranking
Then none of the above is your problem, and this is the honest part.
Google has your pages. It has understood them. It has decided that for the queries you care about, other pages deserve to be higher. Usually because those pages are more thorough, more specific, more current, or because the sites they sit on have been publishing consistently for years and yours has not.
There is no technical trick for this. The work is producing something genuinely better than what currently ranks, on the specific questions your customers ask, and then continuing to do that for longer than your competitors are willing to.
That is slower than anyone selling SEO wants to admit. It is also the only part that keeps working.
The order I would work in
- Run
site:yourdomain.com. Establish which problem you actually have. - Check your homepage source for an H1, a canonical, a sane meta description, and whether your visible text is in the HTML.
- Write down the ten questions customers ask you before they buy. Check whether any page on your site answers them in their words.
- Add width and height to your images. Add alt text. Both are free.
- Only then start worrying about server speed, scripts and Core Web Vitals.
Most of that is a day of work by someone who knows where to look. If you want a second opinion on which of the four you are dealing with, get in touch — telling you which problem you have is usually enough for you to solve it yourself.