SEO Strategies for Ecommerce Sites: A Site-Architecture Playbook

By Christoph Olivier, Founder, CO Consulting

Last reviewed: July 2026

The SEO strategies that move an ecommerce site are structural, not cosmetic. Before you touch a single product description, you decide which pages exist, how they nest, which filter combinations Google is allowed to index, and how link equity flows from the homepage down to a product buried on page nine of a category. Get the architecture right and every other tactic works harder. Get it wrong and you fund a crawl of millions of near-duplicate URLs while your money pages stay invisible. This guide stays at the site level on purpose: taxonomy, category-versus-product hierarchy, faceted navigation, internal linking at scale, and crawl budget. On-page product copy and external link building are separate jobs, covered elsewhere.

Why site architecture is the first ecommerce SEO strategy

Site architecture is the ecommerce SEO strategy that everything else depends on. It is the hierarchy of your pages expressed through navigation and internal links, and it decides which URLs get discovered, crawled, and ranked. On a 20-product store it barely matters. Past a few thousand SKUs, architecture is the single biggest lever you have, because it controls crawl budget, index composition, and how PageRank reaches deep pages.

The reason architecture outranks copy in priority is simple: Google cannot rank a page it never crawls, and it will not crawl a page it cannot reach in a reasonable number of clicks. A large catalog can generate millions of crawlable URLs once filters, sorting, pagination, internal search, and variants go live. Your job is to make sure the crawler spends its budget on the couple of thousand pages that earn revenue, not the noise.

Think of it as three decisions stacked on top of each other. First, which pages should exist at all. Second, which of those should be indexable. Third, how strongly each indexable page is linked. Most stores skip straight to writing product descriptions and never make the first two decisions deliberately. For the strategy layer above this, our complete guide to Google SEO in 2026 covers the ranking system this architecture feeds into.

Category pages vs product pages: which to prioritize

Prioritize category pages. They target broad, high-intent transactional terms like “trail running shoes,” they catch shoppers earlier in the decision, and they scale traffic in a way individual product pages rarely match. Well-optimized category pages often generate several times more organic revenue than product pages, yet they stay under-optimized on most sites because teams pour effort into product copy instead.

Both page types matter, but they do different jobs. The table below shows how to think about each.

The practical rule: give your top 20 to 50 category and subcategory pages real, unique intro copy and deliberate internal links, and template the product pages so they are consistently good rather than individually perfect. A category page that ranks passes traffic and authority to every product nested under it. That relationship is why the collection layer is where architecture-focused SEO strategies pay off first.

Design a flat taxonomy every product reaches in three clicks

Build a taxonomy where any product is reachable from the homepage in three clicks or fewer: Home > Category > Subcategory > Product. Pages deeper than three clicks get crawled less often and are at higher risk of being dropped from the index entirely. A flat, logical hierarchy keeps your revenue pages within the crawler’s reach and gives shoppers a clean path.

Depth is the enemy, not breadth. It is better to have a category with 12 subcategories all one click away than a neat five-level tree that pushes products six clicks deep. Map your catalog before you build: list the head terms shoppers actually search, group them into categories, and let real search demand define your subcategories rather than your internal warehouse structure.

1. List your top head and mid-tail keywords by volume and commercial intent.
2. Cluster them into a shallow tree: broad category, then one subcategory layer.
3. Give each node a clean, static URL: /category/subcategory/, lowercase, hyphenated, no session IDs or parameters.
4. Confirm every product template links up to its parent subcategory and category.
5. Crawl the site and check click depth; anything past three clicks gets promoted or cross-linked.

Add BreadcrumbList schema on every category and product page. Breadcrumbs do double duty: they help shoppers orient, and they create consistent internal links from every product back up through its parents to the homepage, reinforcing the hierarchy you designed.

Faceted navigation: the biggest ecommerce SEO trap

Faceted navigation is the most useful UX feature and the most dangerous SEO trap in ecommerce. Filters for size, color, price, and brand let shoppers narrow choices, but every combination can spin up a unique URL. A store with 10,000 products and 50 filter options can generate over 100 million URL combinations, most of them thin, near-duplicate pages that drain crawl budget and dilute link equity.

The fix is a decision, not a plugin: decide which facet combinations earn indexing and block the rest. Do not index by default and do not noindex everything. Use search demand as the arbiter. If people genuinely search “waterproof hiking boots size 10,” that combination deserves an indexable, crawlable page. If nobody searches “blue shoes sorted by price ascending, page 3,” it does not.

Here is the working framework I use with clients, from most permissive to most restrictive.

The whitelist approach is the one that holds up: only allow a facet combination into the index if it earns it. That keeps a tight set of demand-backed landing pages indexable while the long tail of combinations stays out of the crawler’s way. The technical mechanics of robots directives and parameter handling sit in our technical SEO checklist for founders.

Internal linking at catalog scale

Internal linking at scale means linking sideways and upward, not just downward. Most stores only link homepage to collections and collections to products. The gains come from the links teams skip: collection to related collection, product to related product, and body-content links from category pages to their subcategories with descriptive anchor text. These distribute PageRank across the architecture instead of letting it pool at the top.

A category page that only appears in the header nav is under-linked. Add a paragraph of intro copy on your “Women’s Clothing” page that links to “Tops,” “Dresses,” and “Outerwear” in the body, using the anchor text a shopper would search. That single move pushes category-level authority down to the subcategories that actually rank for mid-tail terms.

Four patterns carry most of the weight:

• Category-to-subcategory body links: descriptive anchors inside intro copy, not just the nav menu.
• Related-collection blocks: a “you might also like” row linking sibling collections, so authority flows laterally.
• Product-to-product links: curated related products, not just algorithmic upsells, to strengthen thin product pages.
• Breadcrumb links: every product links back up its parent chain on every page load.

Handle pagination cleanly while you are in there. Use self-referencing canonicals on paginated pages; do not canonicalize page 2 and beyond back to page 1. Canonicalizing deep pages to page 1 tells Google to ignore that content, so the products only listed on page 4 never get crawled and never get found. Let the paginated series stand on its own with self-canonicals so deep inventory stays discoverable.

Crawl budget and index bloat on large catalogs

Crawl budget becomes a real constraint once an ecommerce site crosses roughly 50,000 to 100,000 crawlable URLs. Past that threshold, Google rations its crawling, and every wasted URL, a sort parameter, a filter combination, an internal search page, is a page it does not spend on your revenue-driving inventory. Index bloat is the visible symptom: thousands of thin, duplicate URLs flooding the index and burying the pages you want to sell.

The strategy is to shrink the crawlable surface to the pages that matter. That means the facet whitelist above, robots.txt blocks on sort and search parameters, a clean XML sitemap that lists only indexable, canonical URLs, and a routine to prune or noindex out-of-stock and discontinued products rather than leaving them to accumulate. A sitemap that mirrors your intended index gives Google a clear signal of which URLs to prioritize.

This matters more in 2026 because AI crawlers now account for a meaningful share of bot traffic, and some crawl tens of thousands of pages per referral they send. A faceted URL mess no longer just wastes Googlebot’s budget; it wastes the budget of the AI systems increasingly deciding which stores get surfaced in answers. The stores that keep their crawlable surface tight get their real inventory seen; the ones that do not fund the crawl of their own noise. If you are triaging where to spend effort, our SEO audit run-order guide sequences these fixes by impact.

A worked example: a 30,000-SKU catalog

Here is how the pieces fit on a real-shaped store. Take a 30,000-SKU home goods retailer with 40 categories, 180 subcategories, and eight filter attributes. Raw, that configuration can expose well over a million crawlable URLs once sorting and pagination multiply out. The architecture job is to collapse that to the roughly 2,500 URLs that should rank.

1. Keep 40 category and 180 subcategory pages indexable, each with unique intro copy and body links to their children: 220 pages.
2. Whitelist about 300 demand-backed facet pages (category + brand, category + one key attribute) with static URLs: 300 pages.
3. Set every other multi-facet combination to noindex, follow, so link equity still flows but the pages leave the index.
4. Block sort, view, and internal search parameters in robots.txt entirely.
5. Keep the roughly 1,800 in-stock product pages indexable; noindex or 301 the discontinued ones.

The result is around 2,300 to 2,500 indexable URLs instead of a million, every one reachable within three clicks, every category linked laterally and downward. That is the entire game: fewer, stronger, well-linked pages that a rationed crawl budget can actually cover. When we run this pattern, the recurring win is category-page rankings climbing within a quarter as crawl equity concentrates. To build the ongoing process around it, see our monthly SEO workflow, and if you want this done with you, book a consultation.

Frequently asked questions

What are the most important SEO strategies for ecommerce sites?

The highest-leverage ecommerce SEO strategies are structural: design a flat taxonomy where every product is within three clicks of the homepage, prioritize category pages over product pages, control faceted navigation so only demand-backed filter combinations get indexed, link collections laterally and downward, and protect crawl budget by keeping the crawlable URL count tight. On-page copy and link building matter, but they sit on top of this foundation.

Should ecommerce SEO focus on category pages or product pages?

Focus on category pages first. They target broad, high-intent transactional terms, capture shoppers earlier, and often drive several times more organic revenue than product pages. Give your top category and subcategory pages unique copy and deliberate internal links, then template product pages to be consistently strong at scale. Category pages rank and pass authority down to the products nested under them.

How does faceted navigation hurt ecommerce SEO?

Faceted navigation generates a unique URL for every filter combination, so a store with thousands of products and dozens of filters can spin up millions of thin, near-duplicate pages. These drain crawl budget and dilute link equity. The fix is a whitelist: index only facet combinations people actually search, set the rest to noindex-follow, and block sort and internal-search parameters from crawling entirely.

How many clicks from the homepage should ecommerce products be?

Every product should be reachable within three clicks of the homepage, following Home > Category > Subcategory > Product. Pages deeper than three clicks get crawled less often and risk being dropped from the index. Keep the hierarchy flat and broad rather than deep, and use breadcrumbs plus category-to-subcategory body links to keep deep inventory within the crawler’s reach.

When does crawl budget become a problem for an ecommerce site?

Crawl budget becomes a real constraint once a site crosses roughly 50,000 to 100,000 crawlable URLs, which faceted navigation and pagination can trigger fast. Past that point Google rations crawling, and wasted URLs starve your revenue pages. Shrink the crawlable surface with a facet whitelist, robots.txt blocks on parameters, a clean sitemap of only canonical URLs, and pruning of out-of-stock products.