Most small business websites are not lost because of ugly design. They are lost because nobody, human or robot, can figure out where anything lives. Site architecture in web design is the part of the job that decides whether a visitor finds your service page in two clicks or gives up after five.
This guide breaks it down in plain language: what site architecture actually is, the difference between flat and deep structures, how URL hierarchy works, how internal linking holds the whole thing together, and a ready-to-copy 5-level map you can apply to a service business today.
What Is Site Architecture in Web Design?
Site architecture is the way the pages of a website are grouped, nested and connected to each other. Think of it as the floor plan of a building. The design is the paint and the furniture. The architecture is where the doors, hallways and staircases go.
A complete site architecture answers three questions:
- Grouping: which pages belong together in the same category?
- Hierarchy: which pages are parents and which are children?
- Connection: how does a user (or a search engine crawler) travel from one page to another?
Site Architecture vs Information Architecture vs Navigation
These three terms get mixed up constantly. Here is the clean separation:
| Term | What it covers | Example |
|---|---|---|
| Information architecture | How content is labelled, grouped and organised so it makes sense | Deciding that “Roof Repair” and “Roof Replacement” both sit under “Roofing” |
| Site architecture | The technical structure: page hierarchy, URLs, click depth, internal links | /services/roofing/roof-repair/ linked from the Roofing hub page |
| Navigation | The visible interface that exposes the structure to users | The header menu, footer links, breadcrumbs, sidebar |
Information architecture is the thinking. Site architecture is the building. Navigation is the signage. Web design as architecture covers this in more depth.
Why Site Architecture Decides Both Usability and Rankings
Structure is one of the rare web design decisions that pays off twice, once for people and once for search engines.
For your visitors
- Faster orientation: a clear hierarchy tells a visitor what you do within seconds of landing anywhere on the site.
- Fewer dead ends: every page offers an obvious next step instead of forcing a back-button retreat.
- Better conversion paths: service pages that sit next to proof (reviews, case studies, pricing) convert far better than isolated pages.
For search engines
- Crawlability: crawlers discover pages by following internal links. A page nobody links to is a page that may never get crawled properly.
- Click depth: pages reachable in three clicks or fewer from the homepage tend to be crawled more often and treated as more important.
- Authority distribution: internal links pass value. A tidy hierarchy funnels authority from your homepage down into the money pages instead of scattering it.
- Topical clarity: grouping related pages under one parent signals to Google that you cover a subject in depth rather than in fragments.
- Sitelinks: well structured sites are far more likely to earn expanded sitelinks under their brand result.
Simple rule: if a human needs a map to find your page, a crawler needs one too.
Flat vs Deep Site Architecture
Every structure sits somewhere on a spectrum between flat and deep.
Flat architecture
Most pages sit one or two clicks from the homepage. Common on small sites with 10 to 40 pages.
- Fast to crawl and easy to navigate
- Authority spreads widely from the homepage
- Becomes chaotic past a certain page count, with menus that turn into a wall of links
Deep architecture
Pages are nested several levels down inside categories and subcategories. Common on large catalogues and multi-location businesses.
- Excellent for organising large volumes of content into logical themes
- Supports clear breadcrumbs and topic clusters
- Risk: important pages buried five or six clicks deep get crawled rarely and rank poorly
Side by side comparison
| Criteria | Flat structure | Deep structure |
|---|---|---|
| Best for | Under roughly 50 pages | Hundreds or thousands of pages |
| Average click depth | 1 to 2 clicks | 3 to 6 clicks |
| Crawl efficiency | High | Depends on internal linking quality |
| Topical grouping | Weak | Strong |
| Menu complexity | Grows messy fast | Stays manageable with dropdowns |
| Main risk | No hierarchy, no topical signals | Orphaned and buried pages |
The answer for most small businesses is neither extreme. You want a shallow hierarchy with real categories: deep enough to group topics, flat enough that nothing sits more than three or four clicks from the homepage. That is exactly what the 5-level map below delivers.
The 5-Level Site Architecture Map for a Service Business
Here is a structure you can copy directly. It works for plumbers, dentists, law firms, agencies, accountants, contractors, clinics, consultants and almost any local service provider.
Level 1: The Homepage
Your single most authoritative page. It should link to every Level 2 hub and to your top three or four money pages. Nothing else.
- States what you do, for whom, and where
- Links to all main categories in the header
- Carries one primary call to action repeated down the page
Level 2: Hub or Category Pages
These are the pillars of your site. Keep them to five or six maximum so the navigation stays readable.
- /services/ overview of everything you offer, linking to each service page
- /locations/ if you serve several towns or regions
- /about/ team, story, credentials, certifications
- /blog/ or /resources/ for educational content
- /contact/ quote form, phone, address, opening hours
A hub page is not a placeholder. It should have real content: an explanation of the category, a grid of child pages, and internal links pointing down.
Level 3: Core Service Pages
One page per service you actually sell. These are your commercial pages, the ones that should rank for buying keywords.
- /services/roof-repair/
- /services/roof-replacement/
- /services/gutter-installation/
Each of these should link up to its parent hub, sideways to closely related services, and down to supporting content. siteinspire.com walks through the specifics.
Level 4: Sub-Services, Locations and Segments
This is where you capture long tail and local searches without duplicating your Level 3 pages.
- /services/roof-repair/emergency-leak-repair/
- /locations/manchester/roof-repair/
- /services/accounting/for-restaurants/
Warning: only create Level 4 pages when you have genuinely unique content for them. Twenty near-identical city pages with the town name swapped is thin content, not architecture.
Level 5: Supporting Content
Proof, education and trust material that feeds authority back up the tree.
- Blog articles answering questions your customers actually ask
- Case studies and project galleries
- FAQ pages, pricing guides, comparison posts
- Reviews and testimonials pages
Every Level 5 page should link back to the Level 3 or Level 4 page it supports. That is the whole point of publishing it. A fuller account is out there.
The map in one table
| Level | Page type | Typical URL | Purpose |
|---|---|---|---|
| 1 | Homepage | / | Positioning and routing |
| 2 | Hub page | /services/ | Group and distribute authority |
| 3 | Service page | /services/roof-repair/ | Rank and convert |
| 4 | Sub-service or location | /locations/leeds/roof-repair/ | Capture long tail and local intent |
| 5 | Supporting content | /blog/how-long-does-a-roof-last/ | Build trust and pass links upward |
URL Hierarchy: Making the Structure Visible
Your URLs should read like a breadcrumb trail. Anyone glancing at the address bar should know exactly where they are.
URL rules that never fail
- Use lowercase only and separate words with hyphens, never underscores or spaces.
- Mirror the hierarchy in the folder path: /services/category/page/.
- Keep it short. Three folder levels is plenty for a small business.
- Drop filler words like “the”, “and”, “our”, dates and ID numbers.
- Never change a URL casually. If you must, set a 301 redirect from the old address.
- Pick one format for trailing slashes and stick to it site-wide.
Messy vs clean URLs
| Messy | Clean | Why |
|---|---|---|
| /page-id-4471?ref=menu2 | /services/roof-repair/ | Readable and descriptive |
| /Our_Services/Roof%20Repair | /services/roof-repair/ | No caps, no encoded spaces |
| /blog/2019/03/17/post-1 | /blog/how-long-does-a-roof-last/ | Dateless and topic based |
| /services/roofing/repair/emergency/leeds/north/ | /locations/leeds/roof-repair/ | Fewer levels, clearer intent |
Internal Linking: The Glue of Site Architecture
A hierarchy on paper means nothing if the links do not follow it. Internal linking is what turns a list of pages into a navigable structure.
The four link directions
- Downward: hub pages link to all their children. This is how crawlers discover new pages.
- Upward: children link back to their parent, usually through breadcrumbs.
- Sideways: siblings link to each other when genuinely related, for example roof repair to gutter installation.
- Contextual: blog posts link into the service pages they support, with descriptive anchor text.
Internal linking checklist
- Every page is reachable from the homepage in three clicks or fewer.
- Zero orphan pages: no page exists without at least one internal link pointing to it.
- Anchor text describes the destination. “Read our roof repair guide” beats “click here”.
- Breadcrumbs are enabled site-wide with structured data.
- Your highest value service pages receive the most internal links.
- The footer carries a compact set of key links, not a dump of every URL on the site.
- An XML sitemap is submitted in Google Search Console and updated automatically.
Before and After: A Messy Sitemap Rebuilt
Before: the sprawl
A typical small business site after five years of ad hoc additions:
- /home-new-v2/
- /roofrepair/
- /ROOF-Replacement-Services/
- /page-id-88/ (a gutter page nobody links to)
- /services/ (empty page with only a phone number)
- /leeds-roofers/ and /roofers-in-leeds/ (near duplicates)
- /blog/2018/09/untitled-post/
- /contact-us-page/
Problems: inconsistent naming, duplicate pages competing for the same keyword, an empty hub, orphaned pages, no hierarchy at all, and no breadcrumbs. Crawlers see a pile of loose pages instead of a business with three clear service lines.
After: the 5-level rebuild
- / (homepage)
- /services/ (real hub with a description and links to all three services)
- /services/roof-repair/
- /services/roof-replacement/
- /services/gutter-installation/
- /locations/
- /locations/leeds/ with /locations/leeds/roof-repair/
- /about/ and /about/reviews/
- /blog/ with topic posts linking into the matching service page
- /contact/
What changed: duplicates merged and 301 redirected, the empty hub filled with real content, orphans linked from their parent, URLs normalised to lowercase, breadcrumbs added on every template. Same business, same services, dramatically clearer signals for users and crawlers alike.
How to Plan Your Site Architecture in 7 Steps
- Inventory every existing page. Export a crawl or list your URLs in a spreadsheet with traffic and conversions for each one.
- Decide the fate of each page: keep, merge, rewrite or redirect. Ruthlessness here saves months later.
- Group pages into five or six categories based on how customers think, not on how your internal departments are organised.
- Map keywords to levels. Broad terms go on hubs, commercial terms on service pages, long tail and questions on supporting content.
- Draw the tree before touching the CMS. A whiteboard, a spreadsheet or a free diagram tool is enough.
- Define URLs and redirects for every page in the new tree, then build the navigation to match exactly.
- Verify after launch: crawl the site, check click depth, hunt for orphans, confirm redirects return 301, resubmit your sitemap.
Seven Site Architecture Mistakes to Avoid
- Menus with 15 top level items. If everything is a priority, nothing is.
- Empty category pages that exist only to hold a dropdown.
- Burying money pages five clicks deep behind decorative landing pages.
- Creating a page per keyword variation instead of one strong page per service.
- Clever menu labels like “Solutions” or “What We Craft” when “Services” works better.
- Orphaned landing pages built for a campaign and never linked afterwards.
- Redesigning without a redirect map, which is the fastest known way to lose organic traffic overnight.
A Quick Self-Audit You Can Run Today
- Open your homepage and try to reach your most profitable service page. Count the clicks. More than three is a problem.
- Read your top ten URLs out loud. If you cannot guess the page content from the URL, rewrite it.
- Check whether every service page links to at least one case study or blog post, and vice versa.
- Search site:yourdomain.com in Google and compare the count with your real page count. A big gap points to crawl or indexing issues.
- Look at your navigation on mobile. If it takes more than two taps to reach Contact, fix it.
Final Word
Good site architecture is invisible when it works. Visitors simply find what they need, and search engines quietly crawl, understand and rank your pages. The 5-level map above is deliberately simple because simple structures survive: they stay clean as you add services, locations and content over the years.
If your current site grew page by page without a plan, restructuring is usually the highest return work you can do before spending anything on ads or new content. At Berardo Modern we plan the tree first, then design on top of it. Get in touch if you want a structural audit of your website.
FAQ: Site Architecture in Web Design
How many levels should a small business website have?
Three to five levels is the sweet spot. Most service businesses need only three: homepage, hub, service page. Levels four and five come into play when you add locations, sub-services or a content library.
Is a flat site architecture better for SEO?
Flatter is generally better for crawling, but only up to a point. Once you exceed roughly 50 pages, a fully flat structure loses all topical grouping. Aim for a shallow hierarchy where important pages stay within three clicks of the homepage while still living in logical categories.
Do URLs really need to match the site hierarchy?
It is not a strict ranking requirement, but hierarchical URLs make the structure obvious to users, help breadcrumbs, simplify analytics segmentation and make future maintenance far easier. There is no good reason to avoid them on a new build.
What is an orphan page?
A page with no internal links pointing to it. Search engines struggle to discover it and users can only reach it via a direct link. Either link it properly from a relevant parent page or remove it.
Should every blog post link to a service page?
Almost always yes, when the connection is natural. Supporting content exists to build trust and channel visitors and link value toward the pages that generate revenue. Use descriptive anchor text and avoid stuffing five identical links into one article.
Can I change my site architecture without losing rankings?
Yes, with preparation. Map every old URL to its new destination, implement 301 redirects, keep your internal links updated, resubmit your XML sitemap and monitor Search Console for a few weeks. Rankings often dip briefly before recovering and improving.
What tools help visualise site architecture?
A spreadsheet is enough to start. Beyond that, a site crawler will show click depth and orphan pages, Google Search Console reveals indexing gaps, and any basic diagram tool works for sketching the tree before you build it.
