Crawls your pages with navigation, header and footer stripped out, so only editorial links count. Then shows which pages have no inbound links at all, and where to add them based on the terms two pages genuinely share.
Most internal link audits count every link on the page, which means the navigation dominates the result. If your menu appears on 500 pages, every page it links to shows 500 inbound links, and the number becomes meaningless.
A page linked only from the menu looks well-linked in most tools. In practice it has no editorial support at all.
A page with no editorial links pointing at it is the most expensive problem on a site: it exists, it costs crawl budget, and nothing supports it.
Two pages covering the same ground with no link between them is either a missing link or a cannibalisation problem. Both are worth finding.
Nav, header and footer are removed before any link is counted, so what remains is links placed inside the content.
Per page, how many editorial links point at it and how many it sends out. Pages sort with the fewest inbound first.
Any crawled page with zero inbound editorial links is listed separately and its link suggestions sort to the top.
The distinctive terms on each page, with common words removed. Two pages sharing several and linking to neither is flagged.
The anchor text of every editorial link is recorded, so you can see how pages currently describe each other.
Word counts alongside link counts, because a thin page with many inbound links is usually the wrong page to be supporting.
Enter any page and we crawl outward from it, up to twelve pages on the same host. Navigation is removed before counting, so only editorial links appear in the result.
Fetches up to twelve pages, so this one takes twenty to sixty seconds.
Pages with zero inbound editorial links. Fix these first — a single contextual link from a related page is usually enough.
Page pairs sharing several distinctive terms with no link between them. Read the shared terms; if they describe the same thing, add the link.
Two pages sharing eight or more terms may not need a link at all. They may need merging into one stronger page.
Suggestions are ordered with orphan targets first, then by how many terms the pair shares.
This is enough to see the shape of a section and find the obvious gaps. It is not a full-site link graph, and it does not judge whether the links it suggests are actually good ideas.
Enough to reveal a pattern. A full site map is something we run as part of an engagement, not in a browser.
Anchors are recorded but not judged. Whether an anchor is useful depends on the sentence around it, which needs a human.
Frequency-based, not semantic. Two pages using the same words may be related, or may simply be in the same industry.
A page reachable only from the sitemap will not be crawled — which is itself worth knowing.
Because a link that appears on every page carries almost no information about which pages matter. Editorial links are placed deliberately, which is what makes them a signal.
There is no number worth quoting. The useful question is whether every page you care about has at least one editorial link pointing at it from a related page.
Probably, though not enough to obsess over. A link inside the body copy is worth more than one in a related-posts strip at the bottom.
No. Dense reciprocal linking between near-identical pages usually signals cannibalisation rather than depth. Link deliberately.
Internal linking is one of the few things fully within your control and one of the least done well. We map it across entire sites, then rebuild the structure around what the data shows.