Crawl depth explained

Crawl depth is the click-distance from the homepage (depth 0) to a page, following the shortest chain of internal links. The homepage's direct links are depth 1, what they link to is depth 2, and so on. Crucially it's about clicks, not URL folders: a page at /a/b/c/d/e is depth 1 if the homepage links straight to it, and a page at /article can be depth 6 if it's only reachable through five pagination hops.

[Home] depth 0
   |
   ├─► /products        depth 1
   │      └─► /products/shoes   depth 2
   │             └─► /products/shoes/running  depth 3
   │
   └─► /blog/post-from-2019      depth 1
          (homepage links to it directly, so
           it's shallow despite the URL)

Depth is a shortest-path calculation over your internal link graph — a breadth-first walk out from the homepage, where each page's depth is the fewest links needed to reach it. Because it is the shortest path, one good link is enough to make a page shallow no matter how buried it is elsewhere.

That also means depth is only as accurate as the link graph you measured, and several things quietly change the answer:

• chevron_rightJavaScript navigation — if the menu is not in the server-rendered HTML, a text-only crawl never sees those edges and reports the whole site as far deeper than it is.
• chevron_rightNofollowed internal links — whether they count depends on the crawler's configuration, so two tools can legitimately disagree.
• chevron_rightThe starting URL — depth from the homepage is the convention, but a site where most entry is from search has a different practical shape.
• chevron_rightCrawl budget — if the crawl stopped early, pages beyond the cut appear infinitely deep rather than deep.
• chevron_rightRedirects — a link through a 301 is still an edge, but chained redirects add hops that some tools count and others collapse.

Two independent mechanisms make depth matter, and they compound:

• chevron_rightCrawl frequency falls with depth. Search engines crawl shallow, well-linked pages more often, because they look more important and are reached sooner in the crawl. A deep page is recrawled rarely, so its updates are indexed late and it can drift stale.
• chevron_rightAuthority falls with depth. Because internal authority decays with every hop, a deep page inherits only a heavily-reduced share — the same idea as authority travel distance. Depth and authority are two readings of one underlying problem.

Most deep tails are not built deliberately — they are produced by pagination, one publication at a time.

A blog index links the ten most recent posts. Post eleven moves to page 2, which is itself depth 2, making that post depth 3. By post ninety it is on page 9 — reachable only after eight pagination clicks, so depth 10 despite nothing about it changing. The page did not get worse; the conveyor belt moved.

  week 1    /blog  ->  post          depth 2\n  month 2   /blog  ->  page/2  ->  post   depth 3\n  month 6   /blog  ->  page/5  ->  post   depth 6\n  year 1    /blog  ->  page/12 ->  post   depth 13\n

• chevron_rightEcommerce — the same effect on category pagination, made worse when out-of-stock products drop out of the default listing entirely.
• chevron_rightDeep filters — a product reachable only by selecting three facets in sequence is effectively very deep, because a crawler will not guess the combination.
• chevron_rightSingle-path sections — a whole subsection hanging off one navigation link inherits that link's depth, and loses everything if the link is removed.
• chevron_rightDocumentation — versioned or deeply nested docs where the useful page sits five levels below the entry point.

The structural answer to all of them is the same: stop relying on sequential paths for discovery, and build hubs that link directly to everything in a topic. That is what content clusters do — they collapse a chain into two hops.

Depth is best read as a distribution — how many pages sit at each click-depth — not a single number. A healthy site has most of its important pages shallow and only low-value pages in the deep tail.

HEALTHY                       UNHEALTHY
depth  pages                  depth  pages
  0    █                        0    █
  1    █████                    1    ██
  2    ████████  <- bulk of     2    ███
  3    ████         valuable    3    █████
  4+   ██  (minor only)         4+   ████████████  <- valuable
                                          pages buried here

Depth is a useful proxy for internal authority, and treating them as identical will mislead you in both directions.

Depth counts hops on the shortest path. It does not care whether that path runs through a page with a hundred outbound links or one with five, or whether the linking page has any authority to pass. Two pages at depth 3 can be in completely different positions.

• chevron_rightShallow but starved — a page linked once from a sitewide footer is depth 1 and receives almost nothing, because the footer divides what it passes across every link on it.
• chevron_rightDeep but well-supported — a page at depth 4 with several contextual links from strong, relevant articles can hold more internal authority than a depth-2 page linked only from a nav.
• chevron_rightShallow and strong — the intended case: a short path through pages that themselves have authority and few competing outbound links.

This is also why "flatten everything" fails as a strategy. Adding every page to the navigation makes them all depth 1 while dividing the homepage's authority across hundreds of links — the arithmetic in link equity. The depth number improves and the pages are no better off.

It is a fair challenge. Search engines are far better at discovery than they were, sitemaps list every URL regardless of linking, and plenty of deep pages rank perfectly well. So why care about click-distance at all?

Because a sitemap solves discovery and nothing else. It tells a search engine a URL exists. It does not pass authority, carries no anchor text, and gives no signal about the page's importance relative to everything else you publish. Those come from links, and links are what depth measures.

• chevron_rightDiscovery — largely solved by sitemaps. A deep page will usually get found.
• chevron_rightCrawl frequency — not solved. Sitemap presence does not make a page crawled often; link prominence does, which is why deep pages have stale cached versions.
• chevron_rightAuthority — not solved at all. A sitemap entry passes none. This is the gap that matters most.
• chevron_rightPerceived importance — not solved. How you link a page is how you tell a search engine what you think of it, and a sitemap treats every URL identically.

The reason to measure it first anyway is cost. Depth needs only a crawl, and it reliably points at the parts of the site where the linking is weakest. Authority modelling gives the better answer; depth gives a decent answer immediately.

Depth isn't a standalone metric — it's downstream of your linking. Deep pages are usually deep because they're only reachable through pagination, deep filters, or a single buried nav link. Fixing depth means adding contextual links from shallow, authoritative pages, strengthening hubs, and wiring pages into clusters that pull whole topics up at once.

RankForge maps your full depth distribution, flags valuable pages stranded in the deep tail, and recommends the specific links that would flatten them — feeding the same dimension of the Structural Health Score. For the hands-on process, read how to improve crawl depth.

• chevron_rightWhat the ideal click depth actually is — why the right target scales with site size, and why average depth is the wrong number to optimise.
• chevron_rightCrawl depth vs crawl budget — two things that get confused constantly, and how to tell which one is your bottleneck.
• chevron_rightThe 3-click rule — where it came from, why it fails at scale, and the version worth keeping.