The short answer
Ecommerce SEO works when you stop optimizing category and product pages the same way. They do two different jobs. Category pages are your commercial keyword engine: they should rank for head terms like "women's trail running shoes." Product pages are your long-tail conversion engine: they win specific, high-intent searches like "Salomon Speedcross 6 women's size 6." Give each page type its own content depth, internal linking, and schema, then solve the one problem that quietly wrecks large catalogs: faceted navigation.
Get those three things right (page-type strategy, faceted navigation, structured data) and you can outrank stores with bigger budgets and more SKUs. Here is how, with the trade-offs a practitioner actually hits.
Category pages vs product pages: split the strategy
Treat your catalog as two ranking systems.
Category and subcategory pages map to how people search when they are still comparing. Someone typing "men's waterproof hiking boots" wants a selection, not a single SKU. That query has volume and commercial intent, and it is where a category page should win. So a category page needs more than a bare product grid: a short intro that uses the target term naturally, buying guidance, and a few hundred words of genuinely useful copy. Not keyword stuffing, actual help ("waterproof vs water-resistant," "how the sizing runs").
Product pages target the specific, lower-volume, higher-conversion tail. There are far more of these queries than head terms, and they convert better because the searcher already knows what they want. This is the core of how to rank product pages on Google: unique copy, real specs, and reviews.
Do product pages need unique descriptions to rank? In practice, yes. Manufacturer-supplied descriptions are duplicated across every retailer selling that item, so you are competing on identical text. Rewrite descriptions for the products you actually want to rank, prioritizing best sellers and highest-margin items first, since rewriting 10,000 SKUs by hand is rarely worth it. User reviews help here too: they add unique, long-tail-rich text you did not have to write.
If you run SEO across other verticals, the head-vs-tail split shows up everywhere, but the mechanics differ. Compare how it plays out in SEO strategy for SaaS companies and B2B SEO for long sales cycles, both covered in the industry SEO strategies that separate winners from laggards pillar.
Faceted navigation is the whole game at scale
Most guides treat this as a technical footnote. It is not. It is the entire game once your catalog gets large.
Faceted navigation is the set of filters shoppers use: color, size, brand, price, rating. Each filter combination usually generates a unique URL. Do the math: a 10,000-product catalog with many filter values can spawn far more crawlable URL permutations than you have products, and Google documents exactly this problem because it is so common. The result is a flood of near-duplicate pages ("blue shoes size 6 under 5,000 sorted by price") that add no value and dilute the pages you care about.
Why it matters: Google explains that on very large sites, crawl budget becomes a real constraint. Googlebot has a finite crawl capacity for your domain. If it spends that capacity fetching millions of filter permutations, it crawls your new products and updated category pages slower. You are effectively paying, in crawl budget, for Google to index junk instead of your money pages.
So the goal is not "index everything." It is to make crawl and index effort land on pages that earn traffic.
The counter-move: promote demand facets, disallow the rest
Most advice says "noindex your facets." That is blunt, and it throws away real opportunity. The better move, and the one Google's own faceted-navigation guidance supports, is selective.
Some filter combinations match genuine search demand. "Women's trail running shoes size 6" is a query people type. "Nike shoes size 6 blue rated 4 stars price low to high" is not. The strategy:
- Find the facet combinations with demand. Use keyword research to see which filters people search as phrases, and pull Search Console query data to find searches where you already get impressions but land on a weak filtered URL. This is where keyword research tooling earns its keep: you are hunting for filter combos with real, recurring volume.
- Promote 20-40 of them into proper landing pages. Turn the winning combinations into clean, static, indexable URLs (for example /womens/trail-running-shoes/size-6) with their own intro copy and a normal internal link from the parent category. These behave like curated subcategories, not throwaway filters.
- Block the rest at the source. Per Google's guidance, use robots.txt to disallow the filter-parameter URL patterns you do not want crawled (sort order, price sliders, low-value combinations). This stops Googlebot from wasting crawl budget on them in the first place.
A quick reference for handling each facet type:
| Facet / URL pattern | Example | Handling |
|---|---|---|
| High-demand combo | women's trail shoes, size 6 | Static indexable landing page, internally linked |
| Low-demand combo | brand + color + rating + sort | robots.txt disallow the parameter pattern |
| Sort / display order | sort=price_asc, view=grid | Disallow in robots.txt (no unique content) |
| Pure duplicate of canonical | session IDs, tracking params | Disallow and/or canonical to the clean URL |
Why prefer robots.txt disallow over noindex for the junk: noindex still requires Google to crawl the page to see the tag, which spends the crawl budget you are trying to protect. Disallow stops the crawl before it happens. Reserve noindex for pages Google must access for other reasons but you do not want indexed.
Run a technical site audit to see how many parameter URLs Google is actually crawling before and after. It is the fastest way to prove the fix worked: watch the crawl of parameter URLs drop and the crawl of product pages rise.
Structured data: Product and Offer schema
What schema markup do ecommerce sites need? At minimum, Product structured data with a nested Offer, on every product page. Google documents the Product markup that makes listings eligible for rich results: price, availability, and review star ratings shown directly in search.
Practical notes from shipping this:
- Mark up price, currency, and availability accurately. Mismatches between your schema and your visible page can cause Google to distrust or drop the rich result.
- Include aggregateRating and review only when the reviews are real and on the page. Fabricated review markup is a manual-action risk under Google's structured data guidelines.
- Add BreadcrumbList markup so your category hierarchy shows in the SERP, which reinforces the category-to-product structure you built above.
Rich results do not directly boost rankings, but the star ratings and price in the SERP lift click-through, and higher CTR on a stable ranking is free traffic. Validate every template with Google's Rich Results Test before shipping.
Internal linking and site structure
Google's ecommerce best practices come down to a crawlable, logical hierarchy: home to category to subcategory to product, with real HTML links, not JavaScript-only navigation Google cannot easily follow.
The internal linking moves that matter most:
- Link your promoted facet landing pages from their parent category so they are discoverable and pass authority.
- Cross-link related products and "customers also viewed" with descriptive anchors, not "click here."
- Keep important products within a few clicks of the homepage. Deeply buried SKUs get crawled rarely and rank poorly.
You can monitor which pages Google indexes and how often it crawls them through Search Console reporting, and track whether your category and product pages are gaining position with a rank tracker built for large keyword sets. Structure is invisible until you measure it.
Where ecommerce SEO differs from other verticals
The head-plus-tail catalog architecture is ecommerce-specific, but the discipline generalizes. Local and service businesses face the same crawl-and-relevance questions with different tools; see local SEO for restaurants and HVAC lead-generation SEO. Regulated sectors lean harder on E-E-A-T, as in healthcare SEO and law firm SEO. If you are comparing platforms to run all of this, our Ahrefs alternative comparison breaks down the trade-offs, including crawl-audit depth and price. DeployFlare handles the full workflow (audits, keyword research, rank tracking, Search Console) from 499 rupees a month with UPI and GST billing, plus city-level and vernacular SERP tracking if your catalog sells across Indian markets.
A practical order of operations
Starting from a plateaued store, do it in this sequence: audit how much crawl budget faceted URLs are eating, disallow the junk patterns, promote the 20-40 demand facets into landing pages, rewrite descriptions for your top-selling products, add Product and Offer schema, then tighten internal linking. That order fixes the biggest leaks first, and it is usually the difference between a catalog that grows and one that stalls.