Internal Link Mapper and Internal PageRank Calculator

Paste a Screaming Frog All Inlinks export and get internal PageRank for every page, click depth, true orphans, near-orphans, weak anchor text, links to redirects and a section linking matrix, with navigation links set aside. Simulate new links before you add them. Free, nothing uploaded.

Live v3.0 · October 2026 Free · No loginReads Screaming Frog exportsCSV and Markdown
Internal link analysis PageRank, damping 0.85
Paste a Screaming Frog Bulk Export > Links > All Inlinks CSV, or one link per line as source,target or source,target,anchor. Columns are read from the header. Up to 200,000 links.
Click depth is counted from here.
Used when there is no Link Position column. A target linked from this share of pages is treated as navigation.
Find true orphans and simulate new links
Paste your sitemap URLs or a Search Console pages export. Any URL here that no link points to is a true orphan, which a crawl alone cannot find.
One source,target per line. The What-if view shows how PageRank and depth would change.
Analysed in your browser · nothing is uploaded
Internal link analysis ·

Paste your links and press Analyse.

What this internal link mapper does

Paste the internal links from a crawl, and the mapper turns them into the measures that decide how authority moves around a site: internal PageRank for every page, click depth from the homepage, orphans, near-orphans and dead ends, anchor text quality, links to redirects and errors, nofollow internal links, and a map of which site sections link to which. It reads a Screaming Frog All Inlinks export as it comes, header row and all.

Its main difference is what it leaves out. Navigation, header and footer links point at the same few pages from everywhere, so they make every page look well linked. The mapper sets them aside, using Screaming Frog’s Link Position column when it is there, and computes everything a second time on the links your content actually makes. A page that every menu links to but no article does can look well linked in a raw inlink count. Here it shows up as navigation-only.

Crawler suitesBrowser link toolsThis mapper
NeedsA paid crawler or a site audit subscriptionA live URL, or an uploaded crawl CSV (LinkJuice)A pasted export; nothing is uploaded
Internal PageRankYes, under their own namesLinkJuice, yesYes, on all links and on in-content links only
Template links separatedSomeLinkJuice, yesYes, by Link Position or share of pages
Orphans against your URL listSome, from sitemapsRarelyYes
Anchor text auditScreaming Frog (non-descriptive anchors), SitebulbOne-page checkersEvery target: generic, empty and URL anchors, with image alt text counted
What-if for planned linksNoNoYes
PricePaid licence or subscriptionLinkJuice: from EUR 19 a month after a free trialFree, CSV and Markdown exports

Tools reviewed in October 2026: Screaming Frog, Sitebulb, Semrush and Ahrefs site audits, Netpeak Spider, LinkJuice, Sitechecker, LinkBoss and RankAI.

How internal PageRank is calculated

PageRank treats each link as a vote that passes on a share of the linking page’s own score. The original PageRank paper by Sergey Brin and Larry Page describes a damping factor, usually set to 0.85, which models a reader who mostly follows links but sometimes jumps to a random page instead. This calculator uses that method: every page starts equal, each passes 85% of its score evenly across its outgoing links, the remaining 15% is spread across all pages, and the scores are recalculated until they stop changing. Pages that link nowhere would leak score out of the system, so this calculator uses a common convention and spreads their share evenly across all pages.

Scores are shown from 0 to 100 against the strongest page, twice: once on every followed link, and once on in-content links only. The first shows what your templates do. The second shows what your writing does, and it is the one you can change page by page. Google’s ranking systems use many more signals than this, and its internal version of PageRank is not public, so treat the figure as a map of how links distribute attention on your site, not as Google’s number.

How to analyse internal links from a Screaming Frog export

  1. Crawl the siteAny crawler that exports links works; in Screaming Frog, use Bulk Export, Links, All Inlinks. The export includes anchor text, status codes, follow and link position, all of which the mapper reads.
  2. Paste the export and the homepage URLRows that are not hyperlinks, such as images and scripts, are skipped. Click depth is measured from the homepage you give.
  3. Add your sitemap or Search Console URLsA crawl only finds pages that something links to. Pasting your full URL list is how true orphans, which nothing links to, show up.
  4. Work through the issues listTrue orphans and navigation-only pages first, then links to redirects, near-orphans and deep pages. Each has the URLs and what to do.
  5. Test the fix before making itPaste the links you plan to add into the what-if box and see which pages gain PageRank and move closer to the homepage.

Orphan pages, click depth and anchor text

Google’s link best practices say that every page you care about should have a link from at least one other page on your site, and its SEO Starter Guide notes that Google primarily finds pages through links from pages it has already crawled. An orphan breaks both. The mapper separates three kinds: pages nothing links to at all, which only a URL list reveals; pages linked only from navigation; and near-orphans with one or two in-content links.

Depth is about clicks, not folders. John Mueller said in 2018 that Google does not count slashes in URLs, and that what matters is how many links you have to click through to reach the content. The mapper counts clicks from the homepage over followed links and flags pages four or more clicks away. Google’s guidance also says there is no magical ideal number of links a page should contain, so the issues list flags pages with too few links in, not a target count.

Anchor text tells users and Google what the linked page is about. Google’s guidance says good anchor text is descriptive, reasonably concise and relevant to both pages, and gives “click here” and “read more” as examples of anchors that are too generic. The anchors view lists every target receiving generic, empty or bare-URL anchors from your content.

What this mapper cannot do

It does not crawl. It analyses the links you paste, so its picture is only as complete as your export. It cannot see JavaScript-inserted links your crawler missed, it does not know your rankings or traffic, and it does not suggest new links from page text. For planning which pages should exist and link together before you build them, use the Topical Map Generator, and for the reasons a page cluster competes with itself, see keyword cannibalisation.

Frequently asked questions

A score for each page based on how links within your own site point to it, calculated with the PageRank method from Brin and Page’s 1998 paper with a damping factor of 0.85. It shows where your internal links concentrate attention. It is not Google’s own figure, which is not public.
An orphan has no links pointing to it, so neither users nor crawlers can reach it by clicking. A dead end has no links out, so it passes nothing on. Because a page with no links out never appears as a source in an inlinks export, the mapper finds those through your URL list, and separately lists crawled pages whose only links out are navigation.
Google has not published a number. John Mueller said in 2018 that what matters is how many clicks it takes to reach the content, not the URL structure. The mapper flags pages four or more clicks from the homepage as the ones worth linking to from higher up.
Because they link to the same pages from everywhere, they make every page they point to look well linked and hide pages that no content links to. The mapper keeps them in the “all links” view and click depth, and leaves them out of the editorial view.
Google’s guidance says there is no magical ideal number. What matters is that every important page has links from related pages, with descriptive anchor text.
Yes. Any CSV with source and destination columns works; anchor, status code, follow and link position columns are used when their headers are recognised. Plain source,target lines work too.
No. The analysis runs in your browser and nothing is sent anywhere.