I Crawled 4 More Dev Sites. One Had 9 Orphan Pages Out Of 10

Following up on last week's four-site crawl, I ran the same free crawler against four more sites developers use daily — nodejs.org, vuejs.org, digitalocean.com, and webpack.js.org — 10 pages each, starting from the homepage. Three came back with healthy internal-link structure. One, digitalocean.com, had 9 of its 10 sampled pages come back as orphans.

What I ran

No account needed, so this is reproducible:

• POST https://rankforge.cc/api/analyzer with {"seed_url": "..."}, then poll GET /api/analyzer/<id> until status is completed.
• The crawler follows internal links breadth-first from the seed, caps at 10 pages for the anonymous tier, and runs a PageRank-style authority simulation restricted to the crawled subgraph.
• Raw JSON for all four runs: content/analyzer_20260920_<host>.json.

The healthy three

nodejs.org came back with an average internal authority of 100 across the sample — every page the crawler reached was tightly cross-linked from the others, with /learn topping the authority table. 611 internal links found across 10 pages, only 1 hitting the weakest-20% bucket, 1 redirect.

webpack.js.org was close behind: average authority 79.32, 627 links, two pages (/concepts and /guides/getting-started) tied at the top of the authority table. Dense docs cross-linking, nothing orphaned.

vuejs.org landed in the middle: average authority 46.13, 873 links found, homepage naturally at 100 and /api/application.html next at 28.17. Normal shape for a docs site — a strong homepage feeding an internal hierarchy, no orphans, no redirect chains.

digitalocean.com: 9 of 10 crawled pages were orphans

This is the one worth a closer look. Starting from the DigitalOcean homepage, the crawler reached 10 pages, and 9 of them had zero internal links pointing to them from anywhere else in the crawled set — they were only reachable because something earlier happened to link to them once. All 9 were the same kind of page: old community tutorial-collection landing pages —

/community/tutorial-collections/how-to-install-magento
/community/tutorial-collections/how-to-install-the-mean-stack
/community/tutorial-collections/how-to-install-and-use-memcache
/community/tutorial-collections/how-to-install-wordpress-with-nginx
/community/tutorial-collections/how-to-install-mysql
/community/tutorial-collections/how-to-install-lamp
/community/tutorial-collections/how-to-set-up-ssh-keys
/community/tutorial-collections/how-to-install-apache-tomcat
/community/tutorial-collections/how-to-set-up-a-node-js-application-for-production

DigitalOcean's community section is enormous — this is a 10-page sample, not a site-wide audit, so it doesn't mean the whole tutorial library is 90% orphaned. What it does show is that whatever navigation path the crawler followed from the homepage lands in a cluster of tutorial-collections landing pages that don't link to each other or back into the main site nav. The average internal authority for the sample came back at 20.41, dragged down by that cluster, while the homepage itself held the maximum score of 100 and the strongest tutorial page (how-to-install-and-use-memcache) only reached 11.57.

Tutorial-collection or topic-hub pages like this are usually generated once from a template, linked from a single index or search result, and then never cross-linked to sibling collections again. They're not broken — a user who lands on one from Google gets useful content — but they're structurally isolated from the site's main internal link graph, which is exactly the kind of thing that caps how much authority flows into that whole content cluster.

Why this keeps showing up

Across eight sites now (four last week, four this week), the pattern repeats: large, well-maintained sites with real engineering teams still accumulate pockets of internally orphaned content — usually old campaign pages, supporter walls, or auto-generated topic-collection pages that were linked once and forgotten. None of it is catastrophic. All of it is invisible from a normal site walkthrough, and only shows up once you crawl the actual link graph and check what points to what.

Try it on your own site

Same checks (orphan pages, near-orphans, internal authority distribution, redirect chains) run in one pass, no signup: rankforge.cc/audit.