Nearly every online store in Nepal I have looked at makes the same allocation error. Effort goes into individual product pages — descriptions, titles, images — and the category pages are left as whatever the theme generated. The searches, meanwhile, are overwhelmingly for categories.
Think about what people actually type. Not "Samsung Galaxy A15 6GB 128GB Awesome Blue". They type "mobile price in Nepal", "running shoes Kathmandu", "office chair price Nepal". Those are category and comparison queries, and the page that should answer them is usually the weakest page on the site.
The category page is the money page
A category page in most Nepali stores is a grid of products, a paginated list, and nothing else. No text, no heading that means anything, no reason for it to be considered an answer to anything.
What a competing page has is a short, genuinely useful introduction: what is in this category, what the price range actually is, what to consider when choosing, what is currently available. Two hundred words of real orientation, above or below the grid, written for someone deciding rather than someone already convinced.
That is not a trick and it is not padding. Someone searching "office chair price Nepal" wants the range, the tiers, and what changes between them. A grid does not tell them. A paragraph does, and the grid is right underneath it.
Add the price range in text, in NPR. It is the single most searched attribute in this market and it is usually only present inside product tiles that are rendered by script.
The duplicate description problem
Manufacturer descriptions get pasted into product pages, unchanged, by every store selling that product. Twenty Nepali sites carry the identical paragraph for the same phone.
Google has to pick one, and it will not pick yours on the basis of the text, because the text is not yours. So it falls back to other signals — authority, links, engagement — where the largest store wins by default.
You do not need to rewrite a thousand products. Rewrite the ones that matter. Take your top twenty by revenue and add three or four sentences that only you could write: what it is actually like, who it suits, what it does not do, what it compares to at that price in Nepal. That is a day of work and it differentiates the pages that earn.
For the long tail, accept the duplicate description and compete on the category page instead. That is the correct allocation, and it is the opposite of what most stores do.
Out-of-stock products, which is where URLs go to die
This is the most consequential technical decision an online store makes and it is usually made by accident.
When a product goes out of stock, stores typically do one of three things: delete the page, hide it, or leave it up. Deleting is almost always wrong. That URL may have accumulated links, rankings and search history, and deleting it throws all of that away for a product that will very likely come back next month.
What to do instead, depending on the situation:
- Temporarily out of stock: keep the page live, keep it indexable, say clearly that it is out of stock and when it is expected. Offer alternatives on the same page. The page keeps its equity and the visitor gets something.
- Discontinued, with a successor: 301 redirect to the replacement product, not to the homepage. Redirecting to the homepage is treated as a soft 404 and wastes the equity you were trying to preserve.
- Discontinued, no successor: redirect to the parent category.
- Never coming back and never had traffic: let it 404. That is a legitimate answer and a clean one.
A store that deletes out-of-stock products is running a slow migration against itself every month. The mechanics are the same as any URL change, and the same rules apply — make the list before you delete anything.
Faceted navigation, quietly generating thousands of pages
Filters for colour, size, brand and price usually create a new URL for every combination. Left alone, a store with a few hundred products generates tens of thousands of near-identical crawlable URLs.
The effect is that crawlers spend their time on filter permutations instead of your actual products, and your genuinely useful pages get visited less often.
The straightforward handling: canonical tags on filtered views pointing at the unfiltered category, and disallow the filter parameters in robots.txt. Keep indexable only the filter combinations that people actually search — brand within category is usually worth keeping, colour almost never is.
Product schema, and the honest limits of it
Product markup with price, currency and availability can produce price and stock information directly in the search result, which measurably improves click-through on commercial queries. It is worth doing and most Nepali stores have not done it.
Two cautions. Set the currency to NPR — I have seen stores marking Nepali prices as USD, which produces a wrong and alarming number in the results. And keep availability accurate; markup that says in-stock for something you do not have is worse than no markup, because the click is wasted and the visitor is annoyed. The broader argument for which markup earns its place is here.
Speed matters more here than anywhere else
Category pages are the heaviest pages on a store — thirty product images, filter scripts, a carousel — and they are the pages that receive the search traffic. On a mobile connection that combination is punishing, and the visitor leaves before the grid has rendered.
Dimensions on every image, lazy loading below the fold, and deferring the scripts in the head will do more for a Nepali store than most content work. The measured spread across Nepali sites is wide enough that this is a choice rather than a constraint.
Where I would start on Monday
Take your five highest-revenue categories. Write two hundred honest words for each, including the price range in NPR. Check what happens to a product URL when it goes out of stock, and fix that policy if it deletes pages. Then rewrite the descriptions on your top twenty products.
That is a week of work, it addresses the queries with actual commercial intent, and none of it requires a rebuild. Everything else — the thousand long-tail products, the filters, the schema — can follow once the pages that earn are the pages you have worked on.
If you want the category and out-of-stock handling reviewed on a live store, send me the domain, or read more about how I approach this kind of work.