Ecommerce Site Architecture

TL;DR — Ecommerce site architecture is how your store’s pages — categories, subcategories, and products — are organized and linked together. The goal is a shape like a pyramid: your homepage at the top, broad categories under it, then narrower subcategories, then products. The point is that a shopper and Google can get from your homepage to any product in just a few clicks by following links. The single most useful thing to know: Google figures out which pages matter by looking at how you link between them, not by reading your URLs.

What “site architecture” actually means

When people say site architecture (or site structure, or information architecture), they mean two related things:

1. The hierarchy — how your pages nest. A store usually goes homepage → category (e.g. “Shoes”) → subcategory (“Running Shoes”) → product (“Brand X Trail Runner”).
2. The links — which pages link to which. Your navigation menu, your category pages, your breadcrumbs, your “related products” — all of those are links that tie the hierarchy together.

Good architecture means a shopper can find what they want fast, and a search engine can crawl from your homepage to every product without getting lost or stuck.

The pyramid shape

Picture a pyramid:

• Top: your homepage.
• Next: a handful of broad categories (linked from your main menu).
• Below that: subcategories.
• Bottom: individual product pages.

This shape works because it tells Google how your pages relate. “Running Shoes” is clearly part of “Shoes.” “Brand X Trail Runner” is clearly a Running Shoe. Those relationships help Google understand your store.

A lot of guides tell you to make everything “flat” — every product reachable in the fewest possible clicks. That sounds good, but it’s a myth that flatter is always better. Google’s own John Mueller said outright that a “super flat structure” is not better than a “reasonable pyramid structure.” If everything is one click from the homepage, you’ve told Google that every page is equally important — which isn’t true, and isn’t helpful.

The flip side is also a trap: don’t bury products so deep that it takes ten clicks to reach them. The sweet spot is a pyramid that’s shallow enough that important products are only a few clicks from the homepage.

Links are what matter (not your URLs)

This surprises people. Your URL can read /shoes/running/brand-x-trail-runner or just /products/brand-x-trail-runner — and Google mostly doesn’t care which. What Google looks at is your links. As their ecommerce documentation puts it, Google “doesn’t look at the structure of URLs to work out the structure of a site. Instead, it analyzes the linkages between pages.”

So if you want a product to do well, the best thing you can do is link to it more — from your homepage, from category pages, from blog posts. The more (good) links a page has pointing at it, the more important Google assumes it is.

The four things to get right

1. Use real links. Your menu and navigation need to be actual <a href="..."> links. If your menu only works through JavaScript clicks with no real link underneath, Google may not be able to follow it.
2. Use breadcrumbs. “Home > Shoes > Running Shoes > Brand X” at the top of a product page helps shoppers and gives Google a clear hierarchy signal — and it can show up in your Google listing.
3. Control your filters. Those “filter by color / size / price” controls on category pages can create thousands of junk URLs. Left unmanaged, they waste the time Google spends crawling you. (This has its own deep dive — see faceted navigation.)
4. Make your mobile site match. Google crawls the mobile version of your site. If your mobile menu hides categories or strips out content, that’s the version Google sees.

The thing most people get wrong

Don’t restructure your URLs just for SEO. Changing /shoes/running/brand-x to /products/brand-x won’t lift your rankings — URL structure barely matters as a ranking factor. But the change itself breaks links and creates redirects that can hurt you temporarily. Pick a sensible URL format early and leave it alone.

Want the deep version — flat vs. pyramid with the actual Mueller quotes, the duplicate-content trap with multi-category products, mega-menu link equity, and the faceted-navigation crawl-budget fix? Switch to the Advanced tab.

TL;DR — Google reads your link graph, not your URL paths, to infer hierarchy and relative importance — “Google generally doesn’t look at the structure of URLs… it analyzes the linkages between pages.” So a reasonable pyramid (homepage → category → subcategory → product) with strong internal linking beats both a flat structure (Mueller: a super-flat structure is not better than a reasonable pyramid) and a too-deep one (you don’t want to “click through a million times”). URL structure is barely a ranking factor; URL stability is the real risk. The commerce-specific failure points are multi-category products (canonical them), mega menus (link-equity dilution + crawlability), faceted navigation (crawl-budget killer — robots.txt > canonical), and mobile-first nav (your mobile nav is your SEO nav).

Architecture serves two masters

Ecommerce site architecture is the hierarchical organization of a store’s pages — categories, subcategories, products, and supporting content — plus the navigation, internal links, and URL patterns that connect them. It has to satisfy two audiences at once: shoppers who want products in as few clicks as possible, and search engines that need to discover every page, understand how pages relate, and assign relative importance.

The whole topic comes down to one principle that most architecture guides bury or miss: Google reads links, not URLs.

Google reads links, not URLs

This is the load-bearing fact. From Google’s ecommerce documentation:

“Google generally doesn’t look at the structure of URLs to work out the structure of a site. Instead, it analyzes the linkages between pages to gain insights about the relative importance of different pages on a site.”

And the mechanism for “relative importance”:

“the more links a page has to it within a site, the higher the relative importance of the page to other pages.”

So your internal link graph does the heavy lifting. Whether the URL reads /shoes/running/brand-x or /products/brand-x is secondary. The product that’s linked from your homepage, your top category, and three blog posts will get more crawl attention and more PageRank than a product two clicks deep with nothing pointing at it — even though the second one has the “shallower” URL.

The practical chain Google recommends:

“add links from menus to category pages, from category pages to sub-category pages, and finally from sub-category pages to all product pages.”

And the warning for large catalogs:

“If category pages don’t include direct links to all products in a category, Googlebot might not find all of your products by crawling alone.”

When direct linking to every product isn’t feasible at scale, fall back to XML sitemaps and (for products) a Merchant Center feed as secondary discovery paths — but those are a backstop, not a substitute for a crawlable link graph.

Flat vs. pyramid — the myth, busted

Nearly every competitor guide repeats some version of “flat is better” or “keep everything within three clicks.” That advice is half-right and badly framed. The nuance comes straight from John Mueller (Google, Feb 2, 2021 office-hours):

“On the other hand, kind of more the top down approach or pyramid structure helps us a lot more to understand the context of individual pages within the site.”

“So in particular, if we know this category is associated with these other subcategories then that’s a clear connection that we have between those parts.”

“So from my point of view, I think for a lot of sites it makes sense to have more of a pyramid structure.”

And the line that kills the myth outright:

“But it’s not the case that a super flat structure is going to be better than a kind of reasonable pyramid structure.”

Why a totally flat structure hurts: if every page is one or two links from the homepage, you’ve flattened the importance signal too. You’re effectively telling Google “everything here is equally important,” which strips away the context it uses to understand which categories own which subcategories own which products.

But don’t overcorrect into a deep tree, because Mueller capped that too:

“But at the same time you don’t want it to be such that it’s like you have to click through a million times to actually get to the actual content.”

So neither extreme wins. The target is a pyramid that’s shallow enough: broad categories at the top, narrowing to products, with click depth kept low for anything that matters. As a practical rule of thumb, aim for most products within ~3–4 clicks of the homepage on a large store, 2–3 on a small one — but treat that as a guideline, not a Google directive (see the click-depth note below).

The “3-click rule” is consensus, not a Google rule

You’ll see “every page must be within 3 clicks of the homepage” everywhere. It’s useful as a heuristic, but Google has not stated a specific click number. The real risk on a large catalog isn’t exceeding a click count — it’s orphaning pages (nothing links to them) or burying them so deep that crawlers rarely reach them. A long-tail product four or five clicks deep can rank fine if it sits in a logical hierarchy and appears in your sitemap. Click depth is a crawlability and PageRank-flow concern, not a direct ranking factor.

This is also why internal linking beats URL depth. A product five clicks deep in the hierarchy but linked directly from the homepage gets more crawl attention than a product two clicks deep with no inbound internal links. If you’ve got an important product sitting deep in the tree, the fix isn’t to re-architect the whole URL scheme — it’s to link to it from higher-authority pages.

URL structure: stability matters, depth doesn’t

Google’s URL guidance for ecommerce is mostly about hygiene, not hierarchy:

“use long-term, persistent URLs. Avoid internally linking to temporary parameters, such as session-IDs, tracking codes, user-relative values (location=nearby, time=last-week), and the current time.”

“Minimize the number of alternative URLs that return the same content to avoid Google making more requests to your site than needed.”

“Use ?key=value URL parameters rather than ?value, where possible. URL parameters allow Google Search to understand your site’s structure and crawl and index more efficiently.”

Notice what’s not there: any mandate about nesting depth. Google’s URL docs do not say /category/subcategory/product is better or worse than /product/slug. Depth is a site-structure decision, not a URL prescription.

Mueller has rated URL structure low (roughly 1 out of 7) as a direct ranking factor, and said making URLs artificially flat provides no SEO benefit. The takeaway for ecommerce: URL structure is barely a ranking factor, but URL stability is a real risk factor. Restructuring /category/subcategory/product into /product later causes redirect chains and temporary ranking disruption even if the end state is neutral. Pick a consistent, descriptive format early (/product/black-t-shirt beats /product/3243) and don’t churn it.

Multi-category products and the duplicate-content trap

The one place URL structure does bite you is products that live in multiple categories. If your URL encodes the category path, the same product can be reached at /shoes/running/brand-x and /sale/brand-x and /brand-x-page — three URLs, one product, duplicate content. Two fixes:

• Flat product URLs (/products/brand-x) — one canonical URL regardless of which category path the shopper took to get there. This is the Ahrefs ecommerce guide’s recommended default, precisely to sidestep this problem.
• Category-nested URLs with a canonical — pick one primary category, build the product URL under it, and canonicalize every other path to that primary.

Either way, you resolve the ambiguity with a single canonical. Then you restore the hierarchy signal — which the flat URL throws away — with breadcrumb schema (below).

Internal linking is the actual mechanism

PageRank and crawl attention flow through <a href> links, so your internal linking strategy is your architecture in practice. A few patterns that matter:

• Link best-sellers and priority products from the homepage and high-authority pages. Google explicitly notes you can signal importance by linking popular products from the homepage, blog posts, or newsletters.
• Category pages are link-equity distributors. They take authority from the homepage/nav and pass it down to products. A category page that doesn’t link to all its products leaves some undiscoverable by crawling alone.
• Editorial → product links. Blog content that links to relevant products both passes equity and creates topical context.
• Related products / “customers also bought” add cross-links across the tree so crawlers (and shoppers) don’t hit dead ends.
• Breadcrumbs are internal links too, not just UX.

I’ve made the internal-links case for years. As I wrote in The Most Important Thing SEOs Overlook: Internal Links: “By adding internal links, you are directing the flow of your website authority to the best content, which in turn signals to search engines which pages you consider the most important.” And the part teams forget: “you may be frequently producing new pieces of content in which you add internal links, but are you remembering to go back to your older pieces” — on a store, that means linking new arrivals from existing category and editorial pages, not just publishing them and walking away. The reason this is worth the effort: “Links on your site are a lot easier to get than links from other sites, and they can be just as effective.”

For the full treatment, see internal links.

Breadcrumbs: hierarchy signal + internal links + rich result

Breadcrumbs earn their keep three ways: they’re internal links, they give Google an explicit hierarchy signal, and they can render as a breadcrumb trail in the SERP. From Google’s structured-data docs:

“A breadcrumb trail on a page indicates the page’s position in the site hierarchy.”

Crucially, breadcrumbs let you communicate hierarchy even when your URLs are flat. A product at /products/black-t-shirt can still carry Home > Clothing > T-Shirts > Black T-Shirt via BreadcrumbList markup. And Google’s recommendation is to model the user path, not the URL:

“We recommend providing breadcrumbs that represent a typical user path to a page, instead of mirroring the URL structure.”

For products that belong in multiple categories, you can supply multiple BreadcrumbList markups for the different navigation paths. A BreadcrumbList needs at least two ListItem entries (each with position, name, and item) to be eligible for the rich result; the final crumb’s item is optional. (For what it’s worth, Mueller has said breadcrumb placement on the page doesn’t matter for SEO — so optimize it for users.)

Navigation menus and mega menus

Two separate issues live here: crawlability and link-equity dilution.

Crawlability. Google can only follow links that are real anchors:

“Google can only crawl your link if it’s an <a> HTML element with an href attribute.”

JavaScript is fine as long as the rendered DOM contains real anchors — “Links are also crawlable when you use JavaScript to insert them into a page dynamically as long as it uses the HTML markup shown above.” What breaks is onclick-only handlers, routerLink-style attributes, or javascript: hrefs:

“Google can’t reliably extract URLs from <a> elements that don’t have an href attribute or other tags that perform as links because of script events.”

So a JavaScript mega menu can be crawlable — the myth that “JS menus can’t be crawled” is false — but only if it renders proper <a href> links.

Link-equity dilution. Every link in your global nav siphons a slice of the homepage’s authority. A 200-link mega menu spreads that authority thinly across 200 destinations. Mueller has also flagged that very large, frequently-changing mega menus can be hard for Google to interpret. The pragmatic move: keep the global nav to top-level categories (a manageable count, not the whole taxonomy), and let category pages handle the deeper subcategory and product navigation. That’s the pyramid expressing itself in the nav.

Faceted navigation: the crawl-budget killer

This is the single biggest crawl-budget leak on most ecommerce sites. Filters (price, color, size, sort order) each generate crawlable URLs, and the combinations explode — a modest catalog can hide millions of filtered URLs.

Google’s hierarchy of fixes, in their stated order of effectiveness:

“Use robots.txt to disallow crawling of faceted navigation URLs”

That’s the preferred method. The soft signals — notably rel=canonical — are acknowledged but “generally less effective in the long term than the previously mentioned methods.” (This is the reverse of what a lot of guides recommend, which lead with canonical.) Other levers:

• URL fragments (#) for filter state passively avoid crawling, since “Google Search generally doesn’t support URL fragments in crawling and indexing.”
• Return 404 for empty combinations: “If there are no green fish in the site’s inventory, users as well as crawlers should receive a ‘not found’ error.”

The strategy decision comes before the technical one: identify which filter combinations have genuine search demand (and deserve to be indexable landing pages) versus pure-UX filters (which should be blocked or built so they never mint a URL). This topic has its own deep dive — see faceted navigation — and it ties directly to crawl budget.

Mobile-first: your mobile nav is your SEO nav

Google indexes the mobile version of your site:

“Google uses the mobile version of a site’s content, crawled with the smartphone agent, for indexing and ranking.”

That has structural consequences. Content parity is required —

“Make sure that your mobile site contains the same content as your desktop site.” — and so is structured-data parity — “Make sure that your mobile and desktop sites have the same structured data.” And critically:

“Google won’t load content that requires user interactions (for example, swiping, clicking, or typing) to load.”

So if your mobile nav hides categories behind a hamburger that only renders links on tap (via JS click events rather than real <a href> links in the DOM), those category and product links may receive no crawl signal through navigation. Whatever hierarchy you built on desktop has to survive on mobile, as real anchors, with the same content and schema.

Sitelinks: influenced, not controlled

Those clusters of links that sometimes appear under your homepage result are sitelinks, and good architecture makes them more likely — but you can’t pick them. Google: “Our systems analyze the link structure of your site to find shortcuts.” And: “If the structure of your site doesn’t allow our algorithms to find good sitelinks… we won’t show them.” You influence quality through a logical structure, descriptive page titles and headings, and concise, relevant anchor text; you can’t specify which pages appear. If a bad one shows up, the lever is to noindex or remove that page.

Where this fits

This is one piece of the site-structure cluster under Ecommerce SEO. It connects directly to faceted navigation (the crawl side), category page SEO and product page SEO (the page types), and internal links (the mechanism). For how crawlers traverse all this, see crawling.

AI summary

A condensed take on the Advanced version:

• Google reads links, not URLs. “Google generally doesn’t look at the structure of URLs… it analyzes the linkages between pages.” Your internal link graph determines hierarchy and relative importance — “the more links a page has to it within a site, the higher the relative importance.”
• Pyramid beats flat — and beats too-deep. Mueller: a “super flat structure” is not better than a “reasonable pyramid”; but don’t make users “click through a million times.” Aim for a shallow pyramid.
• The “3-click rule” is industry consensus, not a Google rule. The real risk is orphaning pages, not exceeding a click count. A deep-but-linked product beats a shallow-but-orphaned one.
• URL structure is barely a ranking factor; URL stability is the risk. Pick a consistent, descriptive format early; don’t restructure (it breaks links/causes redirect chains).
• Multi-category products create duplicate URLs — fix with a single canonical (flat product URL or a primary category), then restore hierarchy with BreadcrumbList schema (model the user path, not the URL).
• Mega menus: crawlable only if they render real <a href> links; huge menus dilute link equity — keep global nav to top-level categories.
• Faceted navigation is the top crawl-budget leak. Google prefers robots.txt disallow; rel=canonical is “less effective in the long term.” 404 empty filter combinations.
• Mobile-first: Google indexes the mobile version. Your mobile nav must use real anchors with the same content and structured data — “Google won’t load content that requires user interactions… to load.”
• Sitelinks are influenced by structure, titles, and anchor text — not directly controlled.

Official documentation

Primary-source documentation, mostly Google’s ecommerce specialty docs.

Google — ecommerce structure & URLs

• Help Google understand your ecommerce site structure — the menu → category → subcategory → product linking pattern, and “Google analyzes the linkages between pages.”
• Designing a URL structure for ecommerce sites — persistent URLs, ?key=value parameters, minimizing alternative URLs.
• Ecommerce SEO overview — the specialty-docs hub.
• Pagination and incremental page loading — why traditional pagination is more crawlable than “load more” / infinite scroll.

Google — links, breadcrumbs, sitelinks

• SEO link best practices (crawlable links) — the <a href> requirement and what breaks crawlability.
• Breadcrumb (BreadcrumbList) structured data — markup spec; “represent a typical user path… instead of mirroring the URL structure.”
• Learn about sitelinks — how Google generates them from link structure; what you can and can’t control.

Google — crawl, facets, mobile, international

• Managing crawling of faceted navigation URLs — robots.txt as the preferred fix; 404 empty combinations.
• Optimize your crawl budget — capacity + demand; eliminating duplicates.
• Mobile-first indexing best practices — content/structured-data parity; interaction-loaded content.
• Managing multi-regional and multilingual sites — the ccTLD / subdomain / subdirectory tradeoffs; don’t use IP detection.

Bing / Microsoft

• Keeping content discoverable with sitemaps in AI-powered search (Canel & Madhavan, Jul 2025) — sitemap scale limits and the lastmod freshness signal for large catalogs.

Quotes from the source

On-the-record statements from Google and Bing. Each link deep-links to the quoted passage where the source allows it.

Google — how Google reads structure

• “Google generally doesn’t look at the structure of URLs to work out the structure of a site. Instead, it analyzes the linkages between pages to gain insights about the relative importance of different pages on a site.” Jump to quote
• “the more links a page has to it within a site, the higher the relative importance of the page to other pages.” Jump to quote
• “add links from menus to category pages, from category pages to sub-category pages, and finally from sub-category pages to all product pages.” Jump to quote
• “If category pages don’t include direct links to all products in a category, Googlebot might not find all of your products by crawling alone.” Jump to quote

Google — URLs, links, breadcrumbs

• “use long-term, persistent URLs. Avoid internally linking to temporary parameters, such as session-IDs, tracking codes, user-relative values (location=nearby, time=last-week), and the current time.” Jump to quote
• “Minimize the number of alternative URLs that return the same content to avoid Google making more requests to your site than needed.” Jump to quote
• “Google can only crawl your link if it’s an <a> HTML element with an href attribute.” Jump to quote
• “We recommend providing breadcrumbs that represent a typical user path to a page, instead of mirroring the URL structure.” Jump to quote

Google — facets, sitelinks, mobile

• “Use robots.txt to disallow crawling of faceted navigation URLs.” Jump to quote
• “Our systems analyze the link structure of your site to find shortcuts.” Jump to quote
• “Google uses the mobile version of a site’s content, crawled with the smartphone agent, for indexing and ranking.” Jump to quote

John Mueller, Google (office-hours, Feb 2, 2021 — via Search Engine Journal’s verbatim reproduction)

• “So from my point of view, I think for a lot of sites it makes sense to have more of a pyramid structure.”
• “But it’s not the case that a super flat structure is going to be better than a kind of reasonable pyramid structure.”
• “But at the same time you don’t want it to be such that it’s like you have to click through a million times to actually get to the actual content.” Read the coverage

Architecture audit checklist

A pass to confirm both shoppers and crawlers can navigate your store:

• Your store follows a pyramid: homepage → top-level categories → subcategories → products (not totally flat, not buried deep).
• Important products are within a few clicks of the homepage and linked from high-authority pages (homepage, top categories, editorial).
• No orphan products — every product is linked from at least its category page.
• Category pages link to all their products (or you have sitemaps/feeds as a backstop for very large catalogs).
• Navigation and menu links are real <a href> anchors in the rendered DOM — not onclick-only / javascript: / routerLink handlers.
• Global nav is trimmed to top-level categories; deeper navigation lives on category pages (mega menu isn’t dumping the whole taxonomy into every page).
• Breadcrumbs are present, use BreadcrumbList schema, and model the user path (not the raw URL).
• Multi-category products resolve to a single canonical URL.
• URL format is consistent, descriptive (/product/black-t-shirt, not /product/3243), and stable — no recent or planned restructures purely for SEO.
• Faceted navigation is controlled: robots.txt disallow on filter URLs (not relying on canonical alone); empty filter combinations return 404.
• Mobile nav exposes the same categories/products as desktop, as real anchors, with the same content and structured data.
• XML sitemap lists canonical, indexable URLs with accurate lastmod, submitted in Google Search Console and Bing Webmaster Tools.

The mental models

1. Links are the structure; URLs are just labels. Google infers hierarchy and importance from your link graph, not your URL paths. Before you touch a URL scheme, ask what links to the page and what it links out to. That’s what moves the needle.

2. The pyramid, between two failure modes. Too flat = no importance signal, Google can’t tell what owns what. Too deep = pages starve for crawl attention and PageRank. Aim for a shallow pyramid: homepage → categories → subcategories → products, with important pages kept close to the top via internal links.

3. Depth is about discoverability, not a magic number. The “3-click rule” is a heuristic. The actual risk is orphaning, not exceeding a click count. A deep-but-linked page beats a shallow-but-orphaned one.

4. URL decision rule: stable beats clever. Pick a consistent, descriptive, persistent format once. Don’t restructure for SEO — the change costs you (broken links, redirect chains) more than the new structure gains you (≈nothing, as a ranking factor).

5. The four commerce-specific leaks to plug.

• Multi-category products → one canonical + BreadcrumbList schema.
• Mega menus → real anchors, trimmed to top-level categories.
• Faceted navigation → robots.txt disallow (preferred over canonical); 404 empties.
• Mobile nav → it IS your SEO nav; parity in links, content, and schema.

Ecommerce architecture — cheat sheet

Flat vs. pyramid (what each does)

Click depth guidelines (rules of thumb, not Google rules)

Depth is a crawlability + PageRank-flow concern, not a direct ranking factor. An orphaned shallow page loses to a linked deep page.

URL structure fast facts

• Google reads links, not URL paths, to infer hierarchy.
• URL structure ≈ 1/7 as a ranking factor (Mueller) — stability is the real risk.
• Descriptive > numeric: /product/black-t-shirt beats /product/3243.
• Persistent URLs; avoid session IDs / tracking params in internal links.
• ?key=value parameters over bare ?value.
• Multi-category product → one canonical + hierarchy via BreadcrumbList schema.

Navigation & links

• Crawlable = real <a href>. JS menus OK if they render anchors. onclick-only / javascript: = not crawlable.
• Trim global nav to top-level categories; push deeper nav onto category pages.
• Breadcrumbs: BreadcrumbList schema, model the user path, multiple trails allowed for multi-category products.

Faceted navigation (crawl-budget order of preference)

1. robots.txt disallow filter URLs — Google’s preferred fix.
2. URL fragments (#) for filter state — passively un-crawled.
3. rel=canonical — “less effective in the long term.”
4. 404 empty filter combinations.

Mobile-first

• Google indexes the mobile version → mobile nav = SEO nav.
• Same content + same structured data as desktop. Interaction-loaded content may not be seen.

Sitelinks — influenced (structure, titles, anchor text), not controlled. Bad one? noindex/remove the page.

Where should a new ecommerce page live?

Architecture mistakes to avoid

Making every product one click from the homepage

Why it fails: a super-flat graph removes meaningful priority and overwhelms the navigation. Do instead: use a reasonable pyramid and reserve prominent links for categories and products that matter most.

Treating URL folders as the hierarchy

Why it fails: Google derives structure from links, not directory depth. Do instead: build crawlable parent-child links, breadcrumbs, and category paths; keep URLs stable and readable without relying on them to communicate importance.

Exposing every facet as a crawlable link

Why it fails: combinations multiply into a large duplicate URL space. Do instead: decide which filtered demand deserves a curated landing page and constrain the rest.

Designing desktop and mobile navigation separately

Why it fails: important links hidden from the mobile version may be missing from the version Google primarily evaluates. Do instead: preserve essential hierarchy and ordinary anchor links across responsive states.

Tools for auditing ecommerce architecture

• Use the Internal Link Cluster Visualizer with a supplied link graph to find orphans, weakly connected sections, and pages receiving little internal importance.
• Use the Link Analyzer on representative category and product HTML to verify crawlable anchors, breadcrumbs, and destination URLs.
• Use the Faceted Navigation Auditor to classify parameter patterns before promoting or blocking filtered pages.
• Use Scout Site Audit Free for a bounded crawl that can expose broken routes and pages the intended pyramid does not reach.

Standing architecture metrics

Crawlable click depth by template

Metric: distribution of shortest internal-link depth for categories, subcategories, and products. What it tells you: whether important inventory is buried or disconnected. How to pull it: a repeatable crawler export, segmented by template. Benchmark / realistic range: establish a baseline by template and investigate material regressions; there is no universal three-click ranking rule. Cadence: monthly and after navigation or taxonomy releases.

Orphan and weakly linked page count

Metric: indexable sitemap URLs with no crawlable inlinks, plus priority pages with very few inlinks. What it tells you: where the declared catalog and actual link graph disagree. How to pull it: reconcile sitemap URLs with a crawl and analytics or GSC landing pages. Benchmark / realistic range: true indexable orphans should be zero; minimum inlink needs depend on the store hierarchy. Cadence: monthly.

Organic landing-page coverage

Metric: share of intended category and product templates receiving organic impressions. What it tells you: whether architecture makes the important catalog discoverable and eligible to surface. How to pull it: Search Console page exports joined to the canonical URL inventory. Benchmark / realistic range: compare by template and inventory state, using the store’s own trend rather than an invented industry target. Cadence: monthly with quarterly taxonomy review.

Resources worth your time

My related writing

• The Most Important Thing SEOs Overlook: Internal Links (Search Engine Land) — why internal links direct authority to your best pages, and the part teams forget (going back to link old content).
• The Beginner’s Guide to Technical SEO — where site structure sits in the bigger technical picture.

On this site

• Ecommerce SEO — the pillar hub.
• Faceted navigation — the crawl-budget deep dive.
• Category page SEO and product page SEO — the two page types this architecture connects.
• Internal links — the mechanism that makes architecture work.
• Crawling and crawl budget — how bots traverse your structure.

From others

• Ecommerce SEO: A Beginner’s Guide (Chris Haines, Ahrefs) — covers hierarchy, URL structure, and the flat-product-URL recommendation for multi-category products.
• 11 Ways to Improve E-commerce Category Pages for SEO (Sam Underwood, Ahrefs) — operational detail on pyramid structure, breadcrumbs, and category URL patterns.
• How to Structure Your Website Architecture for SEO (Ahrefs) — the general (non-ecommerce-specific) structure guide.
• John Mueller Recommends Pyramid Site Structure (Roger Montti, Search Engine Journal) — the verbatim source of the pyramid quotes.

From around the industry

• Google Recommends Pyramid Navigation Structure (Barry Schwartz, Search Engine Roundtable) — Schwartz’s coverage of the same Mueller office-hours session; useful secondary confirmation of the pyramid recommendation.
• Google: URL Structure Not So Important For SEO (Search Engine Roundtable) — Mueller rates URL structure ~1/7 as a direct ranking factor; argues stability matters more than depth.
• Ecommerce SEO Guide (Jenn Mathews, Search Engine Land) — broad practitioner guide covering site structure, crawl budget, and category architecture.
• Importance of Link Architecture (Google Search Central Blog, 2008) — the foundational Google post explaining how internal link structure distributes PageRank through a site.
• Managing Crawling of Faceted Navigation URLs (Google) — official primary source for the robots.txt-first approach to faceted nav; the article’s quotes link here.
• Bing Webmaster Guidelines (Microsoft Bing) — Bing’s counterpart to Google’s crawl/indexing guidance; covers link reachability requirements and sitemap submission.

Test yourself: Ecommerce Site Architecture

Five quick questions on ecommerce hierarchy and internal links. Pick an answer for each, then check your work.

Ecommerce Site Architecture

Ecommerce site architecture is how an online store's pages — categories, subcategories, and products — are organized and linked. Google reads the link structure (not the URL path) to work out hierarchy and relative importance, so a logical pyramid plus good internal links matters more than how deep the URLs look.

Related: Faceted Navigation, Crawl Budget, Internal Link, Canonicalization

Ecommerce Site Architecture

Ecommerce site architecture is the hierarchical organization of an online store’s pages — categories, subcategories, products, and supporting content — together with the navigation, internal links, and URL patterns that connect them. It has to serve two audiences at once: shoppers who need to reach products in as few clicks as possible, and search engines that need to discover every page, understand how the pages relate, and assign relative importance.

The single most misunderstood point is how Google reads structure. Google’s ecommerce documentation says it generally infers site structure and relative importance from the links between pages rather than from the URL path alone. So your internal link graph — what links to what, and how many links point at a page — does the heavy lifting, not whether the URL reads /category/subcategory/product or /products/slug.

The design goal is a pyramid: broad categories at the top, narrowing to specific products at the leaf level, with enough cross-linking that neither users nor crawlers hit dead ends. A completely flat site is not inherently better, while excessive depth can hide products from shoppers and bots. Neither extreme wins.

The recurring failure points are commerce-specific: faceted navigation that explodes into hundreds of thousands of near-duplicate URLs and drains crawl budget; products that live in multiple categories and need a canonical to avoid duplication; mega menus that dilute link equity; and mobile navigation that — under mobile-first indexing — is the navigation Google crawls. URL structure itself is barely a ranking factor; URL stability is the real risk, since restructuring breaks links and triggers redirect chains.

Related: Faceted Navigation, Crawl Budget, Internal Link, Canonicalization

Changelog