The short answer
Site architecture is how your pages link together. Good architecture keeps your important pages close to the homepage, gives every page at least one internal link, and groups related content so both users and Google can see how it fits together.
Breadcrumbs are the visible trail of that structure (Home › Category › Page). Add BreadcrumbList structured data and Google can show the trail in search results and use it to categorise the page. The markup just has to match the breadcrumbs people actually see.
- Keep important pages a few clicks from the homepage so crawlers find them fast.
- No orphans: every page needs at least one internal link from a related page.
- Breadcrumb schema must mirror the visible trail, in the same order.
Click Depth and Flat Architecture
Click depth is the number of links Google has to follow from your homepage to reach a given page. The deeper a page sits, the less often Google tends to discover and recrawl it, which is why a buried page can lag in search even when the content is strong. A flat architecture keeps the hops short: your key pages are reachable in a few clicks rather than buried behind layers of pagination.
This is the honest version of the "3-click rule". Google does not enforce a magic number of clicks. What it does care about is reachability, and its own SEO starter guide asks you to organise the site logically and make it easy to move from broad pages to specific ones. Treat three clicks as a useful target, not a law.
Aim to keep the pages you most want to rank within about three clicks of the homepage. The main navigation, category and hub pages, and contextual links inside your content are the tools that pull deep pages up toward the surface, where crawlers reach them first. Our guide to crawl budget explains why discovery speed matters on larger sites.
Internal Links: The Wiring
Architecture is built out of internal links. They are how authority moves between pages and how Google learns which pages relate to which. One rule from Google is easy to miss: a link only counts if it is a real <a href="..."> tag pointing at a resolvable URL. Links that fire through a script or a button click, with no href, are not reliably crawlable, so build navigation and breadcrumbs from genuine links.
Topic Clusters: Hub and Spoke
The hub (pillar page)
A broad, comprehensive guide on a core topic, for example "Email Marketing". It introduces every subtopic and links out to the deeper pages that cover each one.
The spokes (cluster pages)
Specific articles on long-tail questions, for example "email subject line tips". They link back to the hub and across to related spokes, so the whole topic is connected.
A grouping, not a wall
Clustering helps users and Google see which pages belong to a topic. But it is a way to keep related content connected, not a rule that forbids linking between topics. There is no Google penalty for a link that crosses from one cluster to another when it genuinely helps the reader.
Link where it is useful. The point of a cluster is that related pages are densely connected, not that unrelated ones are walled off. See topic clusters and pillar pages for how to plan one.
Breadcrumb Schema
Visible breadcrumbs solve a navigation problem for readers. To make them work in search too, mark them up with BreadcrumbList JSON-LD. Google states that it uses this markup to categorize the page's information in search results, and the trail can then appear in the result in place of a long URL.
A raw URL in the result:
The same page with BreadcrumbList markup:
The markup needs the required BreadcrumbList, ListItem, position, name and item fields, and the positions and names must match the breadcrumbs a visitor actually sees. A mismatch between the visible trail and the structured data is the most common reason the rich result does not show. If you add other structured data too, keep the same discipline: the markup describes what is on the page, nothing more.
A Worked Example
Illustrative scenario, not client data
A blog has one genuinely excellent guide, but it sits five clicks deep: homepage, blog index, page four of the archive, a category, then the post. No other article links to it. In search it barely shows, and it is a classic orphan-adjacent page. Here is the fix.
- Pull it toward the surface: link to it from a relevant hub page and from the main navigation or a resources menu, so it is reachable in two clicks, not five.
- Wire in related posts: add contextual links to it from the other articles on the same topic, using descriptive anchor text, so it stops being an orphan.
- Add a breadcrumb trail: show Home › Topic › Guide, and mark it up with BreadcrumbList that matches the visible trail.
- Confirm discovery: use the URL Inspection tool in Search Console to check Google has found the page and, once the links are live, request indexing to prompt a recrawl.
Nothing about the content changed. The page simply went from buried and unlinked to shallow and well-connected, which is most of what site architecture is for. The same thinking underlies our guide to internal linking best practices.
Find the pages that are hard to reach.
Murkuz connects Google Search Console to Claude or ChatGPT, read-only. Ask Search_Analytics which pages get impressions and which get none: a page with real content but zero impressions is a candidate for being buried or orphaned. Then ask Inspect_URL whether Google has even discovered a specific deep page. You read the data and decide what to link.
A Quick Self-Check
Your most important pages sit within a few clicks of the homepage.
Every page has at least one internal link from a related page: no orphans.
The breadcrumbs users see match the BreadcrumbList markup exactly, in the same order.
Internal links are real <a href> links with descriptive anchor text, not script-only clicks.
Hub pages and the main navigation expose your deep, high-value pages.
Further reading
The clearest material on site structure and breadcrumbs, starting with Google's own documentation. Each one was read in full before it was added here.
Official docsGoogle Search Central
Breadcrumb (BreadcrumbList) structured dataThe primary source for breadcrumb markup. Lists the required BreadcrumbList, ListItem, position, name and item fields, and states that Google uses this markup to categorize the page in search results.
Official docsGoogle Search Central
Make your links crawlableGoogle can only follow a link that is a standard <a> tag with an href pointing at a real URL. The reason your navigation and breadcrumbs have to be real links, not script-driven clicks.
Official docsGoogle Search Central
SEO Starter Guide: the basicsGoogle's own advice to organise a site logically, group similar pages, and make it easy to navigate from broad to specific. The foundation under everything on this page.
GuideAhrefs
Website structure: how to build an SEO-friendly architectureA thorough, worked walk-through of flat architecture, topic clusters and planning a hierarchy before you build. Good for seeing the whole shape of a site, not just one page.
GuideAhrefs
Internal links for SEO: an actionable guideThe hands-on companion to the video. Covers anchor text, finding internal-link opportunities and fixing orphan pages, which is how architecture actually gets built in practice.
Frequently Asked Questions
What is a breadcrumb in SEO?
A breadcrumb is a row of links near the top of a page showing where that page sits in the site hierarchy, usually Home > Category > Page. It lets a visitor jump back up a level, and it gives Google a clear signal of the page's place in your structure. When you add BreadcrumbList structured data, Google can display that trail in the search result in place of the raw URL, which reads more clearly to searchers.
Does breadcrumb schema actually help SEO?
Not as a direct ranking boost, but it has real effects worth having. Google says it uses breadcrumb markup to categorize the information from a page in search results, and the trail can appear in the result in place of a long URL. The one firm rule is that the markup must mirror the breadcrumbs people actually see on the page: the positions and names in the JSON-LD have to match the visible trail.
What is the 3-click rule, and does Google enforce it?
The 3-click rule is a usability guideline: a visitor should be able to reach any page in roughly three clicks. Google does not enforce a specific number. What matters underneath the rule is reachability: the fewer links Google has to follow to find a page, the more readily it discovers and recrawls it. Treat three as a rough target that keeps important pages close to the homepage, not a hard limit.
What is an internal link silo, and will cross-linking between silos hurt me?
A silo, or hub-and-spoke model, groups related pages together: a pillar page links out to its cluster pages and they link back to it and to each other. It helps both users and Google see which pages belong to a topic. There is no penalty for linking between topics when the link genuinely helps the reader. Siloing is a way to keep related content connected, not a rule that walls topics off from one another.
What is an orphan page?
An orphan page is a page that no other page on your site links to. People browsing your site cannot reach it, and Google has a harder time finding it and judging how important it is. Google may still index an orphan page if it appears in your XML sitemap, but a contextual internal link from a related page is the clearest way to signal that the page matters and to pass it some authority.
How deep should my most important pages be?
Keep the pages you most want to rank within a few clicks of the homepage. Pages buried many levels down are discovered and recrawled less often, so they can lag in search. A flat, well-linked structure, supported by a clear main navigation, hub pages and breadcrumb trails, keeps your key pages close to the top where crawlers reach them first and where users find them without digging.