Almost every Indian ecommerce site I audit has the same wound, and almost none of them know it is there. The category pages look fine. The product pages look fine. But Search Console shows 40,000 URLs discovered and 6,000 indexed, and the category terms that should carry the business are stuck on page two.
The cause is nearly always faceted navigation.
What is actually happening
Filters for size, colour, brand, price and sort order append parameters to the category URL. Each combination is a new URL. A category with six filters and four sort options can generate tens of thousands of crawlable addresses, almost all of them showing a subset of the same products.
Google crawls them. Crawl budget that should go to new products and updated category copy goes instead to /sarees/?colour=red&sort=price_asc&page=3. Link equity spreads across near-duplicates. And the page you actually want to rank, /sarees/, competes with its own variants.
Diagnosing it in twenty minutes
- In Search Console, open Pages and compare "Crawled - currently not indexed" and "Duplicate without user-selected canonical". If either is in the thousands on a site with a few hundred products, facets are the first suspect.
- Run a crawl and group URLs by the presence of a
?. Anything above roughly 20 percent parameter URLs deserves attention. - Check the log files if you have them. The ratio of Googlebot hits on parameter URLs versus product URLs tells you exactly where crawl budget is going.
The fix, in order
Decide which facets have search demand. Colour and brand often do: people search "red cotton saree" and "Samsung washing machine". Sort order and price range almost never do. That single distinction drives everything else.
Make demand facets real pages. If "red cotton sarees" has volume, give it a static, indexable, self-canonical URL with its own H1, its own intro copy and its own title tag. It becomes an asset, not a duplicate.
Block the rest at the crawl layer. Sort, pagination-beyond-reason, price sliders and multi-select combinations should not be crawlable. Use robots.txt disallow patterns for the parameters, and make sure the internal links to them are not plain crawlable anchors. A canonical tag alone is a hint, not a rule, and Google routinely ignores it at scale.
Never combine two facets into an indexable URL unless you have checked the demand. "Red cotton saree under 2000 in Mumbai" is a thin page waiting to be flagged.
Keep pagination honest. Page two of a category should be indexable, self-canonical, and should not canonicalise back to page one. Canonicalising pagination to page one is the single most common way Indian catalogues hide deep products from Google entirely.
What changes after the fix
The realistic outcome is not an overnight ranking jump. It is that crawling concentrates: new products get indexed in days instead of weeks, category pages stop competing with their own variants, and the content and link work you do afterwards actually lands somewhere. On catalogues I have worked on, indexed-page counts typically fall by half while impressions rise, which looks alarming on a chart and is exactly what should happen.
The order matters
Doing content and link building before this is fixed is the expensive mistake. You are pushing authority into a structure that dilutes it. Fix the structure, then the same content budget produces two or three times the result.
If you are scaling a catalogue this year, audit the facets before you commission another word. I cover this sequencing in more depth in my work as an SEO consultant in India, and it pairs directly with the technical checklist in this piece on JavaScript-heavy sites.