Ecommerce SEO Guide: Categories, Products, Filters and Measurement

Ecommerce SEO is the work that makes the categories, products and filters of an online store easy to find, understand and choose in Google results. Unlike a brochure site, a store has thousands of URLs, stock that changes and search intent with direct commercial value, so mistakes show up in sales.

This guide is a panorama: it walks through each area in the order you would fix it and points to detailed articles. If you would rather have a specialist set the architecture and priorities for your store, there is our ecommerce SEO service. The examples use an imaginary school-supplies shop, so we do not rely on invented data about any real client. Romania-specific points (RON prices, local marketplaces) are explained for foreign readers.

How is ecommerce SEO different from SEO for a brochure site?

The differences are in scale and risk, not in principles.

• A large number of pages. Every product, category, variant and filter combination becomes a candidate URL for crawling.
• Repeated content. Supplier descriptions, near-identical variants, filter pages.
• Changing stock and prices. Pages that appear, disappear or change state.
• Commercial intent. Someone searching “ergonomic backpack grade 1” wants to compare and buy, not read a history of backpacks.
• Heavy competition. On generic terms you meet marketplaces and chains (in Romania, large local marketplaces and international platforms); on niches you can win through specialization.

The practical upshot: in a store, half the SEO work is control, deciding which pages you want indexed and which not, and the other half is quality, making the pages that matter useful.

How do you build the category architecture?

Google's ecommerce documentation explains that it understands site structure through links: from menus to categories, from categories to subcategories, and then to all products. If a product page is not linked from any category, Google may miss it, because a crawler does not type searches into your store's search box.

Practical rules:

1. Place every product under a logical category, without overly deep structures.
2. One primary keyword per category, chosen by real demand, with no overlaps.
3. Direct links from category to products, using <a href> elements, not just JavaScript buttons.
4. Links from the home page or from articles to strategic products and categories; a page that receives more internal links looks more important.
5. Breadcrumbs on every page, with BreadcrumbList markup.

The category is usually the store's most valuable SEO page. How to build one element by element is covered in category page optimization. For choosing phrases, see keyword research, and for distributing links, our guide to internal linking and site architecture.

What makes a product page good for SEO?

A product page has two jobs: be found for specific searches and persuade. The basics:

Images have their own chapter: image optimization for SEO. For variants (color, size), Google's documentation asks that each variant be identifiable by its own URL, for example /backpack/blue or /backpack?color=blue. When you use optional query parameters, the URL without the parameter is the canonical one. More on canonicals in our article on canonical tags and duplicate content.

How do you use Product structured data?

Structured data describes the product to machines: name, image, price, currency, availability, ratings. Google distinguishes between product snippets (pages where you cannot buy directly) and merchant listings (pages where you can buy). Its documentation recommends providing as much rich product information as you can, not a strict list of fields.

A minimal example in JSON-LD:

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Ergonomic school backpack 18 L",
  "image": ["https://example.com/img/backpack-18l.webp"],
  "description": "Backpack with an ergonomic back panel, for grades 1-4.",
  "sku": "GH-18L-WHT",
  "brand": { "@type": "Brand", "name": "ExampleBrand" },
  "offers": {
    "@type": "Offer",
    "url": "https://example.com/backpacks/ergonomic-backpack-18l",
    "priceCurrency": "RON",
    "price": "249.00",
    "availability": "https://schema.org/InStock"
  }
}

Rules that keep you out of trouble:

• The data in the markup must be the same as on the visible page, especially price and stock.
• Add ratings only if you have genuine product reviews.
• Update the markup whenever the price changes; some platforms generate it automatically, others need checking.
• Google does not guarantee that stars or price will show, even when the markup is valid.

Implementation details are in our JSON-LD structured data guide, and you can draft the structure with the schema generator.

How do you control filters and faceted navigation?

Filters (color × size × price × brand) can generate a huge number of URLs, often with the same content. Google's faceted navigation documentation says the most effective way to prevent their crawling is robots.txt, while rel=canonical and nofollow are secondary signals that are less effective over time. One more detail: for filter combinations that return no products, it recommends a 404 status.

Example rules for parameters with no search demand:

User-agent: *
Disallow: /*?*sort=
Disallow: /*?*price_min=
Disallow: /*?*price_max=

A decision table per filter:

Note: a page blocked in robots.txt cannot be read by Google, so any canonical or noindex on it is never seen. Pick one method per URL type. You can draft the rules with the robots.txt generator.

How do you handle pagination?

Google recommends a unique URL for each page in a list (for example ?page=2), each with its own canonical rather than a canonical pointing to page one. Links to the next page should be <a href> elements. Google no longer uses rel=next and rel=prev. “Load more” buttons or infinite scroll can hide products, because the crawler does not click buttons; in that case, add classic links or a sitemap.

What do you do with stock, discontinued products and seasonal pages?

The rules depend on the situation. The table summarizes a sound, widely used policy; Google's documentation explicitly covers empty categories, while the rest is common practice.

The logic of redirects is explained in our guide to 301 and 302 redirects. Mass-redirecting products to the home page is a common mistake: the user does not find what they wanted.

Speed and the mobile experience

Stores often load heavy themes, many scripts and large images. On a phone, that means slow pages at exactly the moment of decision. Core Web Vitals are experience metrics that Google measures; how to read and fix them is in our article on Core Web Vitals. For a first check, run the site speed test.

Content that helps the shopper decide

Not every search around a store is a direct purchase. Many people search “how to choose...” or “difference between...”. Buying guides, comparisons and answers to questions capture these searches and lead readers to categories. Each article should link to the category or products it recommends, and search intent must be respected: see search intent.

How to measure SEO in an online store

A useful report does not stop at traffic: it ties traffic to orders. The structure of a monthly report is in our article on SEO KPIs.

A staged plan

In large stores you cannot fix everything at once, so order by impact: first what affects the whole site (indexing, filters, speed), then high-margin categories, then products with demand. Check each step in Search Console before the next.

1. Clean up indexing: block filters with no demand, fix canonicals and redirects.
2. Organize categories: a clear purpose per category, original content.
3. Prioritize products: those with high margin and demand get the first original descriptions.
4. Add Product structured data and verify it matches the page.
5. Improve mobile speed.
6. Publish guides tied to categories and build authority; see link building.

Common mistakes

• Supplier descriptions copied onto every product.
• Filters indexed with no control and no demand.
• Redirecting every discontinued product to the home page.
• Product markup with prices or stock that do not match the page.
• A canonical pointing to the first page of a paginated list.
• Categories with no text, or identical text across categories.
• No internal links to new products.

When SEO is not the main channel

If you sell unique products at unbeatable prices but with no organic search demand (for example a new, unknown brand), SEO will bring less at first and you need other channels. If the niche is dominated by large marketplaces on every relevant term, the strategy should target long-tail searches and guides rather than generic terms. And if margin or stock cannot support extra orders, growing traffic is not the priority.

Conclusion and next steps

Ecommerce SEO is won through order: you control what gets indexed, make key categories and products useful, mark up data correctly and measure sales, not just visits. Next step: pick the five categories with the highest margin and check each against the list in our category page guide. If you want a plan built for your store, see our ecommerce SEO service.