How to Fix Site Architecture: Crawl Depth, Orphan Pages, and Hub Pages

TL;DR

Poor site architecture, pages buried deep in the crawl tree, disconnected orphan pages, and absent hub pages, is one of the most common reasons Google under-indexes an otherwise healthy site. Fixing it requires a full crawl audit, a structured internal-linking plan, and hub pages that organize authority before you publish another word.

Author: Guru Editorial  |  Published: June 9, 2026

Why Site Architecture Is a First-Principles Problem

Site architecture is the skeleton every other SEO tactic sits on. If Google cannot reliably discover and crawl your pages, on-page optimization, content quality, and link building all underperform, because the pages those investments support may never reach the index.

The data bears this out. A JetOctopus case study of a large-scale site found that only 40% of pages were being crawled by Googlebot before an internal-link remediation project. After restructuring internal links and reducing crawl depth, coverage climbed to 70%, a 30-percentage-point improvement with no additional content published. This is the leverage that architecture work delivers.

Beyond crawl coverage, architecture shapes how PageRank flows through a site. Pages that sit three or more clicks from the homepage receive a fraction of the link equity that homepage-adjacent pages do. For most sites, the pages that *need* the most authority, pillar topics, transactional landing pages, are often the ones buried deepest.

Diagnosing Your Architecture: The Three Core Problems

Before you can fix anything, you need to understand which of the three structural failure modes your site is running into. They are distinct, but they often compound one another.

Problem 1: Excessive Crawl Depth

Crawl depth is the number of clicks separating a page from the homepage via the shortest internal-link path. A page at depth 1 is directly linked from the homepage. A page at depth 4 requires four clicks to reach.

Google's guidance has long recommended keeping important pages within three clicks of the home page. In practice, sites with large page counts frequently waste a significant share of their crawl budget on low-value URLs, paginated pages, parameter variants, thin filters, while strategically important pages sit at depth 5 or beyond.

The diagnostic is straightforward: run a full site crawl with Screaming Frog, Sitebulb, or SEOguru's technical audit module and pull the crawl depth report. Sort descending. Any page at depth 4+ that generates traffic or carries business value needs to move up the architecture.

Signs your site has a crawl depth problem:

Key category or service pages are at depth 4 or deeper
Log file analysis shows Googlebot spending cycles on parameter URLs and thin paginated pages while ignoring your primary content
Pages appear in your XML sitemap but not in Google Search Console's coverage report

Problem 2: Orphan Pages

An orphan page has zero internal links pointing to it. It may appear in your sitemap, be reachable via a direct URL, and even have backlinks from external sources, but without internal links, Google has no context for where it fits in your site hierarchy.

Across large sites, a significant share of pages receive zero internal links, Semrush's analysis of over 50,000 domains found that 69% of websites had at least one orphan page, and the problem is more severe than most site owners realize. Orphan pages consume crawl budget while contributing a disproportionately small share of organic traffic, a severe misallocation of both crawl quota and indexation effort. The downstream effect on indexation is significant; for a full breakdown of why some pages never make it into the index, see How to Get New Pages Indexed Fast in 2026.

Orphan pages also undermine topical authority. When Google cannot follow an internal path that connects a page to its parent topic, it cannot confidently assess the page's relevance, regardless of how well the page itself is written.

How orphan pages form:

A CMS migration that dropped breadcrumb or category links
Blog posts or landing pages published without being added to a hub or category
Old campaign pages that remained live after the navigation link was removed
Locale or variant pages created programmatically but never linked from the main content tree

Problem 3: Missing or Weak Hub Pages

Hub pages, also called pillar pages, are the architectural glue between your homepage and your content. A hub page covers a broad topic comprehensively, links out to related cluster pages, and receives links back from all of them. Without hub pages, individual pieces of content compete against one another and receive diluted authority.

Research on B2B SaaS sites consistently shows that pillar-cluster architectures outperform flat content structures on keyword coverage and topical authority. Sites with clear hub-and-spoke link structures tend to compound rankings faster because authority flows more efficiently from well-linked hub pages to cluster content, rather than dispersing across disconnected pages.

The Architecture Audit Workflow

The five-step site architecture audit: crawl → depth map → orphan inventory → hub page build → coverage monitoring.

Step 1: Run a Full Site Crawl

Use Screaming Frog, Sitebulb, or a log-file-aware crawler. Crawl as Googlebot, respect robots.txt, and let it run to completion. The output you need:

All URLs discovered, with crawl depth per URL
Inlink count per URL (to identify orphans)
HTTP status codes
Canonical and noindex tags

Export to a spreadsheet. You will be cross-referencing this against your GSC coverage data. If you have Google Search Console connected, pull the list of indexed URLs and compare against your crawl export, any page that appears in your sitemap but not in GSC's "Valid" coverage bucket is a priority to investigate.

Step 2: Build a Depth Map and Prioritize

Sort your crawl export by depth, descending. Flag every page at depth 4+ that meets any of these conditions:

Has external backlinks (check Ahrefs or Search Console's link report)
Has organic impressions in GSC (even with zero clicks)
Is a revenue-driving or conversion page

These pages need a direct internal link added from a shallower page, ideally depth 1 or 2. Do not simply add a link anywhere; the linking page should be topically related and already well-crawled.

Step 3: Inventory and Triage Orphan Pages

Filter your crawl export for pages with zero inlinks. Cross-reference with GSC to segment orphans into three buckets:

Bucket	Condition	Action
Resurrect	Has impressions or backlinks	Add to a hub page and relevant cluster content
Consolidate	Thin content, no links, no impressions	301-redirect to the most relevant live page
Remove	Outdated, no value, no links	Noindex or remove and redirect

The resurrection bucket is where the quick wins live. Adding one internal link from a well-crawled hub page to an orphan with existing impressions often accelerates indexation within weeks, not months. Every change of this type should move through a formal approval record, here is why that matters at scale.

Building Hub Pages That Actually Work

A hub page is not a table of contents. It is a substantive piece of content that covers a broad topic at the level a knowledgeable generalist would expect, typically 1,200-2,500 words, while explicitly linking to the deeper cluster pages that cover subtopics.

The structural requirements of a working hub page:

Covers the parent topic with enough depth to rank independently for head-term queries
Contains contextual links to every cluster page (not a bare list, links embedded in relevant sentences)
Each cluster page links back to the hub page with consistent anchor text
The hub page itself is linked from a high-authority page (the homepage, a top-level nav item, or a primary service page) so it sits at depth 2 or less

Choosing your hub topics:

Start from your existing content inventory, not from a keyword tool. Group pages by semantic theme. Anywhere you have three or more pieces of content on related subtopics without a clear parent page, that is a hub-page gap. The hub pages you need are already implied by the content you have published.

For a more detailed treatment of building the link network inside a hub, see our guide on internal linking at scale.

Crawl Depth vs. Orphan Pages: Which to Fix First?

Both are urgent, but the sequencing matters. Here is how to prioritize based on site size and situation:

At scale, orphan-page remediation and crawl-depth reduction deliver compounding impact, large sites often see the biggest gains from tackling both in the same sprint.

Prioritization guidance:

Small sites (under 500 pages): Orphan pages are the faster win. A handful of internal links can reconnect dozens of orphaned pages within a single content sprint.
Mid-size sites (500-20,000 pages): Start with crawl depth. A flat architecture dramatically improves coverage, which makes orphan page discovery more reliable.
Large sites (20,000+ pages): Both are urgent simultaneously. Treat them as parallel workstreams: one team flattening the architecture, another clearing the orphan backlog. Botify research confirms that unoptimized large sites have only 40% of strategic URLs crawled each month.

Maintaining Architecture as You Publish

Architecture is not a one-time project. Every content sprint you run is an opportunity to introduce new orphan pages or push existing pages deeper in the crawl tree.

The practices that prevent regression:

Link every new piece of content to a hub page before publishing. Make it a pre-publish checklist item.
Run a monthly or quarterly crawl diff to catch net-new orphans and depth increases before they compound.
Track GSC coverage trends alongside your content output. A rising count of "Discovered - currently not indexed" URLs is a leading indicator that architecture is degrading. Review SEOguru's on-page and coverage tracking features for how to automate this monitoring.
Keep redirect chains short. Every hop in a redirect chain costs crawl budget. Chains of three or more hops should be collapsed to a direct 301.

Architecture Fixes vs. Content Fixes: Understanding the Difference

A question that comes up frequently: when a page is underperforming, should you fix its architecture first or rewrite its content? The answer depends on the failure mode.

Symptom	Most Likely Cause	First Fix
Page ranks but gets no clicks	Weak title/meta description	Content (on-page)
Page has impressions but rank is inconsistent	Crawl depth too high, low authority	Architecture
Page not appearing in GSC at all	Orphan page or crawl block	Architecture
Page indexed but no impressions	Keyword targeting mismatch	Content
Rankings dropped after migration	Redirect chain or internal link loss	Architecture
Strong external links but poor rank	PageRank dilution from depth	Architecture

If a page is not being crawled or indexed reliably, no amount of content improvement will help. Architecture is always the prerequisite.

Frequently Asked Questions

What is a good crawl depth for SEO?

Most SEO practitioners recommend keeping all important pages within three clicks of the homepage. Pages at depth 1-2 receive the most crawl attention and internal link equity. Pages at depth 4+ are frequently under-crawled on large sites, particularly if they lack strong inlinks from shallower pages.

How do I find orphan pages on my site?

Run a full site crawl with a tool like Screaming Frog or Sitebulb, then filter for pages with zero inlinks. Cross-reference that list against your XML sitemap and Google Search Console coverage data. Any page in the sitemap with zero internal links is an orphan regardless of whether it appears indexed.

How many internal links does a hub page need to its cluster pages?

There is no fixed number, but every cluster page on a topic should receive at least one contextual inlink from the hub page. On large topic clusters (20+ cluster pages), it is acceptable to group cluster links into a structured section on the hub page rather than forcing all links into body prose.

Will fixing site architecture hurt my current rankings?

It should not, provided you are not removing pages or changing URLs without proper 301 redirects. Adding internal links and creating hub pages is additive work. Consolidating thin orphan pages via 301 redirects may temporarily affect impressions for those individual URLs, but the equity consolidates to the target URL and typically improves overall topic-level rankings.

How long does it take to see results from an architecture fix?

Crawl improvements can show in Google Search Console's coverage data within two to four weeks of Googlebot re-crawling the affected sections. Ranking improvements tied to better PageRank distribution typically take six to twelve weeks to fully manifest, depending on site size and how frequently Googlebot visits the affected pages.

What is the difference between a hub page and a pillar page?

The terms are used interchangeably in most SEO contexts. Both refer to a comprehensive, broadly-targeted page that links to and receives links from a cluster of related, narrower pages. "Hub" is more common in site architecture discussions; "pillar" is more common in content strategy discussions. Functionally they serve the same structural role.