The core problem
Ecommerce SEO is dominated by two issues that are technical before they are editorial: category pages that do not say anything, and faceted navigation generating thousands of near-duplicate URLs that split their own signals.
Category pages are the asset
The category page — not the product page — usually carries the commercial query. Yet most category pages are a grid of products and a sentence of boilerplate. A category page that explains how to choose within the category, what the meaningful differences are, and which option suits which buyer, is competing on completely different terms.
- A real introduction that answers "how do I choose one of these?"
- The buying criteria that actually differentiate products in this category.
- A comparison table where the category supports one.
- An FAQ covering the questions buyers ask before purchasing.
- Internal links to the guides and subcategories that support it.
Faceted navigation, handled deliberately
Filters multiply URLs. Colour, size, price and sort order combine into a crawl space far larger than your catalogue, most of it near-duplicate. Left alone, it dilutes signals, wastes crawl budget and produces indexing that nobody chose.
| Facet type | Treatment |
|---|---|
| Facets with real search demand | Indexable landing page with its own copy |
| Facets with no demand | Canonical to the parent category |
| Sort order and pagination parameters | Excluded from indexing |
| Combined multi-facet URLs | Excluded unless demand is proven |
Product pages that are not the manufacturer description
When fifty retailers publish the same supplier copy, none of them has a reason to rank and none is worth citing. Original specifics — how it is used, what it does not suit, real answers to real customer questions — are what separate a product page from a duplicate.
GEO for ecommerce
Product research is increasingly conversational: "best X for Y", "is X worth it", "what should I look for in X". Those answers are assembled from buying guides and category explainers far more often than from product listings, which is exactly the layer most stores never build. See AI search optimization.