Website Structure

Website structure is the skeleton of your site: how pages are grouped, linked, and nested. It decides which pages Google finds easily and which it buries. To me, a good structure means a user can reach any page in three clicks or fewer, and a crawler can discover every page without hitting a dead end.

I would not waste time on a perfect hierarchy until I have fixed the orphans. An orphan page is a page with zero internal links pointing to it. Google only finds it if you submit a sitemap, and even then it may not trust its importance. I once had a 3000-page e-commerce site where 40% of product pages were orphans. Adding a breadcrumb trail and a category page that linked to each product instantly doubled the number of indexed product pages.

Structure is not a sitemap. A sitemap is a list of URLs you want crawled; structure is the wiring that tells Google which pages are related. I use a simple rule: every page should be reachable from the homepage via a chain of no more than four clicks. If you have a page that takes five clicks, ask yourself if it really needs to exist or if it should be promoted to a higher level.

Navigation menus are part of the structure, but internal linking within content is more important. A menu might only show top-level categories, but a blog post that links to a product page inside a review is the kind of link that carries topical relevance. I prioritise those contextual links over footer links every time.

Structure matters most when you are launching a new site, redesigning an existing one, or noticing crawl coverage dropping in Google Search Console. I have seen a site go from 80% indexed to 95% just by flattening the hierarchy and adding a [technical SEO](/technical-seo/) audit as a routine check.

For a new site, you plan structure before you write a single page. I draw a hierarchy: homepage → category pages → subcategory pages → individual pages. Then I check that each category page links to its subcategory pages and vice versa. That is the minimum viable structure. Without it, Google will still try to crawl, but it will waste budget on low-value pages while missing your best content.

During redesign, structure is where most mistakes happen. Old URLs change, redirects are forgotten, and the new hierarchy buries content that used to rank. I always run a [301 redirect](/301-redirect/) map when moving pages, and I never delete a page without redirecting it to a relevant parent category.

If you have duplicate content issues, structure is often the root cause. Two product pages that differ only by colour might sit in different subcategories, each with a different URL structure. Google sees them as separate pages and dilutes the ranking signal. I use [canonical tags](/canonical-tag/) to point the colour variants to the main product page, and I reorganise the subcategories so that colour is a filter, not a separate path.

Structure also matters for sites with pagination. A blog with ten pages of posts should have a clear hierarchy: the main category page links to page 2, page 2 links to page 3, and so on. Without that, Google may stop crawling after page 1. I implement [pagination](/pagination/) with rel="next" and rel="prev" and ensure each page has a link back to the hub page so the topical signal is not lost.

I used to believe a flat structure was always better. Every page one click from the homepage, minimal hierarchy. I thought that was the fastest way to get every page crawled. I was wrong.

Flat structures work for small sites — under 50 pages — but beyond that, they create a mess. Google cannot tell which pages are the most important because they all have the same link equity. I saw a client with 200 blog posts all linked from the homepage. The homepage had 200 links, most of them pointing to posts that were never read. The crawl budget was wasted on low-value posts, and the best content got less attention than it deserved.

Switching to a siloed hierarchy — grouping related posts under a hub page and linking only the hub from the homepage — improved the site's organic traffic by 40% in three months. The hub page passed authority to the posts inside the silo, and Google started treating each silo as a topic cluster.

I also undervalued the [sitemap](/sitemap/) as a structural tool. I used to treat it as a simple list of URLs. Now I understand that a sitemap can reflect your intended structure, but only if you prioritise the most important pages and exclude thin or duplicate content. I update my sitemap every time I restructure, and I check that the hierarchy in the sitemap matches the hierarchy in my navigation.

Another mistake: ignoring the URL structure. I used to think slugs did not matter much. But a URL like /shop/category/product/ is far more informative to Google than /p?id=123. I now insist on descriptive, keyword-rich slugs that mirror the category tree. It is a small change that makes a big difference in how Google interprets page relationships.