Site architecture
📖 4 min readUpdated 2026-04-18
Site architecture is how your pages relate to each other, through navigation, internal links, and URL hierarchy. Good architecture makes every page discoverable in a few clicks, concentrates authority on the pages that matter, and signals topic relationships clearly.
The 3-click rule
Every important page should be reachable from the homepage in ≤3 clicks. This isn't about usability alone, deep pages get less authority flowed to them and crawl less often.
Flat vs deep architectures
Flat
homepage
├── page1
├── page2
├── page3
...
└── page500
Every page one click from homepage. Works for small sites (<100 pages). Doesn't scale.
Deep
homepage
└── category1
└── subcategory1-a
└── subcategory1-a-i
└── product
Discoverable but takes many clicks. Crawl depth penalizes leaf pages.
The sweet spot: broad + shallow
homepage
├── category1
│ ├── subcategory1a (hub)
│ │ ├── post1
│ │ ├── post2
│ │ └── post3
│ └── subcategory1b (hub)
├── category2
│ └── ...
└── about / contact / etc.
Most sites should aim for 2-3 levels deep. Categories (level 2) serve as topic hubs. Every leaf page gets to the homepage in ≤3 clicks.
The hub-and-spoke model
For content-heavy sites (blogs, documentation, e-commerce categories):
- Hub page (e.g., "SEO Guide"), broad topic overview, links to spokes
- Spoke pages (e.g., "Keyword Research," "On-Page SEO"), narrower subtopics, each linking back to hub + to related spokes
This model concentrates authority on the hub, which then passes to spokes, which then interlink with each other. Google sees a coherent topic cluster.
Internal linking patterns that support architecture
- Primary nav, links to top-level categories. Every page.
- Footer, secondary links, auxiliary pages (privacy, terms). Every page.
- Breadcrumbs, visible path from homepage. Pages 3+ levels deep.
- Contextual links, in-content links. Critical for passing authority + relevance signals.
- Related content. "You might also like" / "Read next", keeps users on site, spreads link equity.
URL structure reflects architecture
URLs should map to the hierarchy:
- Homepage:
/
- Category:
/seo-guide/
- Spoke:
/seo-guide/keyword-research/
- Article:
/blog/how-to-keyword-research/
Category-folder URLs signal hierarchy to both users and Google.
Common architectural mistakes
- Homepage linking everywhere. When every page gets a homepage link, none of them get the focus authority-concentration benefits.
- Orphan pages. Critical pages with no internal links in. Get pulled from indexing.
- Tag archives as dupes. WordPress blogs often create tag archives that duplicate category archives. Noindex them.
- Pagination without clear hierarchy. Paginated category pages can balloon into thousands of thin duplicate URLs. Handle with rel="next"/"prev" or (increasingly) just let Google figure it out, noindex deep paginations.
- Navigation that depends on JavaScript. If bots can't follow, it doesn't count.
Tools
- Screaming Frog → visualize crawl depth, find orphans
- Sitebulb → best for architectural visualization
- Ahrefs Site Audit → link distribution analysis