Site structure: hubs, categories, click depth and URL hierarchy

What site structure is

Every site has a structure, whether anyone planned it or not. It is made of two separate layers:

• Link hierarchy: which page links to which. This is what users click through and what Googlebot follows. If the home page links to a category and the category links to a product, the product is two clicks deep.
• URL hierarchy: the folders in the address, such as /kitchen/kettles/. It describes where a page belongs, but it does not create a single path by itself.

Structure matters for three reasons. First, discovery: “Google primarily finds pages through links from other pages it already crawled.” A page without links pointing to it is hard to find. Second, importance: Google uses the link graph to decide which of your pages matter most. Third, users: a visitor who lands on a product from search should be able to see where it belongs and move to the parent category or to related items without using the back button.

A good structure is not a goal in itself. Google's starter guide says organising a site “can help search engines and users understand how your pages relate to the rest of your site”, and adds that search engines will likely understand your pages as they are now. Restructure when the current layout hides important pages or splits topics, not because a diagram looks untidy.

What Google officially says

The same page gives the practical rules. Each of these is a direct quote:

• “For example, add links from menus to category pages, from category pages to sub-category pages, and finally from sub-category pages to all product pages.”
• “As a general rule, the more links a page has to it within a site, the higher the relative importance of the page to other pages on your site.”
• “For example, if you have a best selling product, consider linking to it from the home page or in other content, such as blog posts or newsletters on your site.”
• “Googlebot generally doesn't try to submit searches into a search box as part of crawling a site.”
• “It's strongly recommended to link to all products that you wish indexed.”
• “If it's not possible to link to all pages, use a sitemap or a Google Merchant Center feed.”

Google's link guidelines add two basics that every structure depends on: “Generally, Google can only crawl your link if it's an <a> HTML element (also known as anchor element) with an href attribute.” And: “Every page you care about should have a link from at least one other page on your site.”

From content map to structure

Structure starts with the keyword work you have already done. The lessons “Clustering and the content map” and “Topical authority” explain how to group queries by intent and how to cover a topic fully. This lesson takes that map and turns it into pages and links.

1. One cluster, one page. Each cluster of queries with the same intent gets exactly one URL. Two pages for one cluster compete with each other.
2. Group clusters into topics. Clusters that share a parent topic become one section. The broad cluster at the top of that group (for example “coffee brewing” or “electric kettles”) becomes the hub or category page.
3. Choose the page type by intent. “best electric kettle” wants a comparison or category; “how to descale a kettle” wants an article; “kettle repair london” wants a service or location page. The page type decides where the page sits in the tree.
4. Draw the links before you write. For every new page, decide which hub links to it, which pages it links back to, and which siblings it links across to.
5. Check the depth on paper. Count the clicks from the home page to each money page. If a key page is five levels down in the plan, it will be five levels down on the live site.

Hubs, categories and subcategories

Hub pages

A hub (often called a pillar page) covers a broad topic and links to every page that covers a part of it. On a content site it is a guide such as “Coffee brewing guide”; on a shop it is usually the top-level category. A good hub has its own useful content: a short overview, a clear list of subtopics with descriptive anchor text, and links to the most important detail pages. A hub that is only a list of links is a thin page.

Category and subcategory pages

Categories group similar items; subcategories narrow them down where there is real demand for the narrower group. “Kettles” → “Electric kettles” → “Glass electric kettles” is reasonable if people search for glass kettles. A subcategory with three products and no search demand adds a level of depth and nothing else.

Detail pages

Products, articles and individual services are where most queries land. Each detail page should link up to its parent (in the content and in breadcrumbs) and across to closely related pages.

The link pattern is the important part: links down from the hub, links up from every child, and links across between siblings that a reader would want next. This keeps every page in the section within one or two clicks of the hub, and the hub collects links from all of its children, which tells Google it is the central page of the topic.

Click depth: keep important pages close to the home page

Click depth is the smallest number of link clicks needed to reach a page from the home page. The home page is depth 0, pages linked from it are depth 1, and so on. It is measured along links only: a page reachable only through internal search, a form or a JavaScript event without an href has no depth at all for a crawler.

Google does not publish a target number, but it does say it can use “the number of links it needs to follow to reach a page” to infer importance, and it suggests linking a best-selling product “from the home page”. In practice SEO teams use a rule of thumb: keep the pages that earn money or traffic within about three clicks of the home page. Treat it as a working target, not a Google rule.

Ways to bring important pages closer:

• Link hubs from the main menu so every hub is at depth 1.
• Link best sellers and key services from the home page, as Google suggests.
• Add related links between detail pages, so a page is reachable from its siblings, not only its parent.
• Use breadcrumbs on every page below the top level; they give every page a path back up.
• Make pagination crawlable with plain links, and link the most important items from the first page or from the hub instead of leaving them on page 9.
• Remove empty levels: a subcategory with a handful of items can be merged into its parent.

Do not overcorrect. Linking hundreds of pages from the home page makes the menu useless for people and dilutes the signal that the home page gives to your real priorities.

Flat, deep and siloed structures

Flat structure

A flat site architecture has few levels: home, hubs, detail pages. Discovery is fast and important pages sit close to the top. The risks are an overloaded menu, very long category pages and weak topical grouping, where unrelated pages share one level. “Flat” does not mean linking everything from the home page; it means few clicks between the home page and any page that matters.

Deep structure

A deep structure has many levels of categories. The taxonomy is tidy and each page has a narrow parent, which suits large catalogues. The risks are important pages buried at depth five and beyond, slow discovery of new pages and many near-empty intermediate pages.

Siloed structure

A silo keeps each topic in its own section and links mostly within that section. It makes topical grouping very clear. The risk is rigidity: when a reader on a cold brew article needs a grinder, a strict silo forbids the link that would actually help. Use silos as the default grouping and add cross-links wherever a user would want them.

For most sites the answer is a shallow, logical hierarchy (home → hub or category → subcategory where needed → detail page) plus contextual cross-links. Large catalogues need more levels; compensate with menu links, featured links and related links so that depth in the taxonomy does not become depth in clicks.

URL hierarchy vs link hierarchy

The two hierarchies are independent. Google says it “generally doesn't look at the structure of URLs to work out the structure of a site”; it analyses the links. A URL can look deep and be close to the home page, or look shallow and be unreachable:

URL folders still have uses. “Parts of the URL can be displayed in search results as breadcrumbs, so users can also use the URLs to understand whether a result will be useful for them.” And on larger sites: “If you have more than a few thousand URLs on your site, how you organize your content may have effects on how Google crawls and indexes your site.” Google explains that grouping topics in directories can help it learn how often the URLs in each directory change, for example a rarely changing /policies/ folder next to a frequently changing /promotions/ folder.

Practical URL rules from Google's URL structure guide:

• “When possible, use readable words rather than long ID numbers in your URLs.”
• Separate words with hyphens, not underscores: /summer-clothing/, not /summer_clothing/.
• “Whenever possible, shorten URLs by trimming unnecessary parameters (meaning, parameters that don't change the content).”
• URLs are case sensitive, so /APPLE and /apple are different URLs for Google. Use one case, normally lowercase.

Do not build deep paths just to mirror the menu. When a product sits in several categories, a URL that contains the category path creates one URL per category for the same product. Keep product URLs independent of the category (/product/blue-kettle/) and let breadcrumbs and links express the hierarchy. Choose subfolders or subdomains on practical grounds; Google's advice is: “From a business point of view, do whatever makes sense for your business.”

Structure patterns by site type

Blogs

Chronological blogs bury content by design: each new post pushes older ones down the paginated list. Build topic hubs from your clusters, link every post to its hub, and link related posts to each other. Tags are useful only if each tag page is a real, maintained collection; dozens of tags with one or two posts each add thin pages and no structure.

Online shops

Follow Google's pattern: menu to categories, categories to subcategories, subcategories to all products. Link best sellers from the home page and from relevant content. Create filtered landing pages only where a filter combination has its own demand, and handle the rest as described in the lesson “Pagination and faceted navigation”.

Service and local sites

Give each core service its own page and group services under a few hubs. Case studies, prices and FAQs link back to the service they support. Location pages work when each one has real local content such as the address, team, local projects and reviews; a copy of the same page with a different city name does not.

Restructuring an existing site

Changing the structure of a live site changes links, and often URLs. Done carelessly, it moves or loses pages that already rank. Work in this order:

1. Inventory. Crawl the site and export every URL with its click depth, internal links, status code and canonical. Add clicks from Search Console and external links, so you know which URLs carry value.
2. Design the target tree from the content map, then compare it with the crawl: which pages move, merge, split or disappear.
3. Keep URLs where you can. Many structural fixes (new hubs, menu changes, related links) need no URL change at all. Change URLs only when the old ones are genuinely wrong.
4. Map every changed URL to its new URL, one to one, and set up 301 redirects. Merged pages redirect to the page that absorbs them. The lesson “Redirects and site migrations” covers the redirect side in detail.
5. Update internal links in menus, breadcrumbs, content and product cards so they point to the final URLs, not to redirects. Update canonicals and the XML sitemap the same way.
6. Crawl the staging or new version before release and check depth, broken links and redirect chains.
7. Monitor after launch: Page indexing and Performance in Search Console, server logs if you have them, and a fresh crawl to confirm the new depth.

Measuring click depth with a crawler

Any site crawler that starts at the home page and follows links reports a depth (or “level”) for each URL. Use it like this:

1. Crawl from the home page with link following only, so depth reflects the real link graph. Make sure the crawler renders JavaScript if your menus depend on it.
2. Build the distribution: how many indexable pages sit at depth 0, 1, 2 and so on. A long tail at depth 4+ is the first thing to fix.
3. Filter your priority pages (top categories, best sellers, core services) and check each one's depth and number of internal links.
4. Compare with the sitemap: URLs in the sitemap that the crawl never reached have no link path at all. These are orphans, covered in the next lesson, “Internal linking and orphan pages”.

Common mistakes

• ✕Important pages buried deepBest sellers or core services five or more clicks from the home page.
• ✕Search box as the only pathGooglebot does not type into search boxes, so those products are not found by crawling.
• ✕Navigation without real linksMenus built on click events with no <a href> give crawlers nothing to follow.
• ✕Treating URL folders as structureAdding folders to URLs does not change click depth; links do.
• ✕Category paths in product URLsOne product in three categories becomes three URLs.
• ✕Hubs that are only link listsA hub needs its own useful content, not just a directory of links.
• ✕Empty intermediate levelsSubcategories with no demand and a few items add depth and thin pages.
• ✕Restructuring without redirectsChanged URLs without one-to-one 301s and updated internal links lose their accumulated signals.

FAQ

Sources

• Google Search Central — Help Google understand your ecommerce site structure
• Google Search Central — SEO Starter Guide
• Google Search Central — Link best practices for Google
• Google Search Central — URL structure best practices for Google Search

Quotes checked against the official pages on 30 October 2026.