Internal Linking Strategy: A Planning Framework

A step-by-step process for planning and building a site's internal-link structure from scratch — topic mapping, hub/spoke assignment, a link matrix, prioritization, phased rollout, and measurement.

First published: Jul 3, 2026 · Last updated: Jul 26, 2026 · Advanced
demand #2 in Internal Links#9 in Website Structure#103 in Technical SEO#136 on the site
AIAuthored byAI v2

An internal linking strategy is a plan you build before you link, not an audit of a site you already have. The process is six steps: map your topics and keywords into content groups, decide which pages are hubs and which are spokes, draw up a link matrix (source page → target page → anchor text → priority), prioritize link placement toward your highest-value pages, phase the rollout on large or existing sites instead of relinking everything at once, then measure with the GSC Links report, click-depth reduction, and Crawl Stats. The one piece competitors skip is phasing — and John Mueller's own advice on big site changes ('split it time-wise… keep things controllable & trackable') is why staging it is the right call, not just a scaling compromise. There is no official 'internal link matrix' from Google; it's industry process built on Google's general link principles.

TL;DR — This is a planning framework, not an audit — the internal linksAn internal link is a hyperlink from one page on a website to another page on the same website. Internal links help search engines discover your pages and pass ranking signals (PageRank and anchor-text context) between them. article already covers fixing an existing site. The build sequence is six steps: (1) map topics/keywords into content groups; (2) assign hub vs. spoke pages; (3) draw a link matrix — source page → target page → anchor textAnchor text is the visible, clickable text of a hyperlink. It tells readers what they'll find on the other end and gives search engines context about the linked page. → priority; (4) prioritize link placement toward your highest business/search-demand pages (the industry “donor → acceptor” idea, grounded in Google’s “every page you care about needs a link” and “link in context”); (5) phase the rollout on large or existing sites — Mueller’s own advice on major changes is to “split it time-wise… keep things controllable & trackable,” which reframes “you can’t relink everything at once” as the correct method, not a compromise; (6) measure with the GSCA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. Links report, click-depth reduction, and Crawl StatsA Google Search Console report (under Settings) that shows how Google has crawled your site over the last 90 days — total requests, download size, and average response time, broken down by response code, file type, Googlebot type, and purpose. It's only available for root-level properties (a Domain property or a URL-prefix property verified at the site's root).. No search engine documents an official “internal linkAn internal link is a hyperlink from one page on a website to another page on the same website. Internal links help search engines discover your pages and pass ranking signals (PageRank and anchor-text context) between them. matrix” — the methodology is industry process on top of Google’s general link principles. Assume the reader already knows PageRankPageRank is Google's original recursive link-graph algorithm: a page's score depends on the scores of the pages linking to it, and in the published model each page's score is split across its outbound links (the simplified version: links are weighted votes). Google says it's evolved since launch but still part of its core ranking systems., click depthCrawl depth usually means click depth — how many clicks it takes to reach a page from the homepage by following internal links. It can also mean a crawler setting that limits how many levels deep a crawl goes before it stops., and hub-and-spoke — this page doesn’t re-derive them.

Strategy vs. audit — the frame for this whole page

Google’s documentation supports crawlable links and meaningful anchor textAnchor text is the visible, clickable text of a hyperlink. It tells readers what they'll find on the other end and gives search engines context about the linked page., while the exact hub, cluster, and prioritization model remains a site-specific strategy choice. Evidence for this claim Google recommends crawlable HTML links and concise, relevant anchor text, including for internal links. Scope: Google Search crawling and link guidance. Confidence: high · Verified: Google: Link best practices The framework here is a practical methodology rather than a Google ranking-system specification. Evidence for this claim Hub, cluster, and prioritization frameworks are site-specific editorial methods rather than a Google-specified ranking architecture. Scope: Editorial strategy boundary; Google's source supports crawlability and anchor practices, not a fixed link-equity formula. Confidence: medium · Verified: Google: Link best practices

The internal links article owns the audit workflow: find orphans, fix broken links, pull deep pages shallower, map link equity on a site you already have. Site architecture owns the models — silo vs. hub-and-spoke vs. topic clustersSite architecture is how a website's pages are organized, categorized, and interlinked. It controls how crawlers discover pages, how link equity flows, and how clearly search engines understand each page's topical context. Silo structure, hub and spoke, and topic clusters are the three common models. (they’re the same thing once you allow cross-links) and the pyramid Google recommends. And the website structureWebsite structure (site architecture) is a site's visible hierarchy, navigation, breadcrumbs, and URL organization — how pages relate and how people and search engines move between them. Internal linking is the primary signal Google reads to understand that structure, not URL folders. hub owns the big idea that your link graph, not your URL folders, is your structure.

None of that is re-derived here. This page assumes you know it and answers a different question: how do you actually plan and build the link structure — page by page — before or while you create the content? Everything below is forward-looking.

One honest caveat before the steps: there is no Google or Bing document that names an “internal link matrix” or “keyword mapping before linking” as an official process. Those are industry-developed frameworks built on top of Google’s general principles — that every page you care about should have a link, that you should link to related resources in context, and that structure should stay recognizable. Treat the matrix and the rollout order as good practice grounded in official principles, not as a Google-endorsed methodology. (And keep it in proportion: Gary Illyes has said people “overestimate the importance of links… it hasn’t been [top three] for some time,” so don’t spend a month on a spreadsheet at the expense of the content itself.)

Step 1 — Map your topics and keywords into content groups

Start with the map, not the links. List the topics and target keywords you want the site (or the new section) to cover, and cluster them into groups — each group is a set of pages that belong together because they serve related intent.

Two ways to cut the groups, and you usually blend them:

  • By topic. Everything about “email automation” in one group, everything about “landing pages” in another. This is the natural unit for a content hub.
  • By search intent. Sometimes two keywords sit in the same topic but serve different intents (a “what is” explainer vs. a “how to set up” tutorial vs. a commercial “best tools” page). Those often want to be separate pages that link to each other, not one page trying to do all three.

This is exactly the workflow I’ve described in my own writing on internal links: I plan links out before publishing, and I look for pages already on the site that mention topics I want a new page to rank for. My one-line principle from The Most Important Thing SEOs Overlook: Internal Links still governs the whole thing — “add internal links to related content where it makes sense.” The map is how you make “where it makes sense” a decision you make on purpose instead of one you improvise per article.

Where the plan lives: in a spreadsheet, set up before Step 3. One row per page you plan to create or already have, tagged with its group. That same sheet becomes your matrix in Step 3 and your rollout tracker in Step 5 — build it once.

Step 2 — Identify your hub/pillar pages and cluster/spoke pages

Within each group, decide which page is the hub (the broad overview) and which are the spokes (the specific subtopics). I’m not going to re-explain the hub-and-spoke model — site architectureSite architecture is how a website's pages are organized, categorized, and interlinked. It controls how crawlers discover pages, how link equity flows, and how clearly search engines understand each page's topical context. Silo structure, hub and spoke, and topic clusters are the three common models. covers that, and covers why it’s the same thing as “topic clusters.” Here the question is narrower: how do you tell a hub candidate from a spoke candidate?

Two signals:

  • Breadth of topic. A hub covers the whole subject at a summary level and could plausibly link out to a dozen sub-pages. A spoke covers one slice deeply. If a page’s natural job is to route readers onward, it’s a hub.
  • Search demand. The broad head term usually maps to the hub; the long-tail variations map to spokes. If you already have Search ConsoleGoogle's free tool for monitoring crawling, indexing, and search performance. data, the page (or future page) tied to the highest-demand head term is your hub anchor for that group.

The output of Step 2 is simple: every planned page is now labelled hub or spoke, inside a named group. That labelling is what makes the matrix in Step 3 fall out almost mechanically.

Step 3 — Build the internal-link matrix

The internal-link matrix is the core deliverable of an internal linkingLinks between pages on the same site. strategy: a table that says, for each intended link, which page links to which, with what anchor text, and how important that link is. It turns “we should interlink these” into a concrete, assignable list.

A workable matrix has these columns:

  • Source page — the page the link lives on.
  • Target page — the page the link points to.
  • Suggested anchor text — descriptive, and varied across instances (more below).
  • Link type / placement — hub → spoke, spoke → hub, spoke → spoke, contextual body link, navigational, etc.
  • Priority — high / medium / low, driven by Step 4.
  • Status and date — planned / live, and when it went live (this is what makes the sheet double as your rollout tracker for Step 5 and your before/after baseline for Step 6).

The default link pattern that comes out of a hub-and-spoke map is: hub → each spoke, each spoke → back to its hub, and spoke → spoke where two subtopics genuinely relate. That’s most of your matrix before you’ve thought hard at all. The template lens has a starter table you can copy.

On anchor text in the matrix: don’t fill every cell with the same exact-match keyword. Google’s guidance is to “write as naturally as possible, and resist the urge to cram every keyword that’s related to the page that you’re linking to.” The internal links article already covers why forcing exact-match anchors has no visible effect — the practical upshot for the matrix is to model descriptive, varied anchors per instance. One more matrix-density rule from Google worth building in: “don’t chain up links next to each other” — links stacked back-to-back are harder to read and each one loses its surrounding context, so spread them through the body.

Step 4 — Prioritize link placement toward your highest-value pages

Not every page deserves equal linking. The whole point of a strategy is to concentrate attention, so Step 4 decides where the strongest, most numerous links point.

Reconcile two lists:

  • Business priority — the pages that convert, sell, or capture leads, regardless of current traffic.
  • Search demand — the pages targeting terms with real volume, per your keyword map and Search Console.

Where those overlap, you have your top tier. Where they don’t, you make a judgment call — usually business priority wins for commercial pages, search demand wins for top-of-funnel content.

A practical prioritization order I’d suggest (this is synthesized best practice, not a single source):

  1. Pages already getting search impressions but few internal links. Fastest ROI — the demand and authority are already there in your GSC data, the page just isn’t being fed internally. Add links and you compound something that’s already working.
  2. Priority / commercial pages, regardless of current traffic — the pages the business needs to rank even if they’re new.
  3. The long tail — everything else, in due course.

The industry “donor → acceptor” idea. Large-site guides (the model is often attributed to JetOctopus, and is laid out well in Digital Applied’s large-site guide) frame this as routing links from donor pages — ones with high crawl activity and impressions — to acceptor pages that are weak and need authority transferred to them, rather than linking at random. It’s a useful lens for Step 4. Label it as an industry framework, though — it’s not Google’s language.

The Google ground rule under all of this: “every page you care about should have a link from at least one other page on your site,” and “think about what other resources on your site could help your readers understand a given page on your site, and link to those pages in context.” That’s the official floor — every priority page linked, in context. The prioritization above is just deciding which of those pages gets fed first and most.

TIP Turn the topic map into reviewable internal-link candidates

Similarity and low-link evidence can prioritize a review queue; they do not prove that every suggested link belongs in the copy.

Score supplied pages with my free Cluster Visualizer Free

  1. Supply the relevant page set with titles, URLs, and current internal links.
  2. Review high-scoring pairs where the target has few or no supplied links.
  3. Add links only where the source passage naturally supports the target, then measure crawl and performance effects.
The score creates a queue; the visible reasons make the editorial decision auditable.

The row scores a proposed link at 76.6 from Internal linking explained to Topic cluster strategy. Reasons include shared terms architecture, internal, links, and SEO; the same URL segment blog; and the target having no supplied links or low PageRank or few supplied links.

Step 5 — Phase the rollout (especially on large or existing sites)

This is the step most guides skip, and it’s the one with real risk. On a large or established site you cannot relink everything at once — and more importantly, you shouldn’t, even if you could.

Why staging is the right call, not a compromise. When you change a lot of a site at once and rankings move, you can’t tell which change did it. John Mueller’s advice on making major site changes captures exactly this:

“If you need to do both, I’d try to split it time-wise so that you can recognize any negative effects in each part, and take action to improve them. If you do everything at once, you’ll never know what to fix, and even if things end up ‘same as before’, you won’t know if one part went down, and was compensated by an improvement on the other part. Keep things controllable & trackable.”

He was talking about combining a domain change with a redesign, but the logic transfers directly to internal linking: a phased rollout isn’t a scaling limitation, it’s how you keep the effect attributable. Roll out a batch, watch it, roll out the next.

A practical rollout order — highest impact, lowest effort first:

  1. The impressions-but-underlinked pages from Step 4 (batch one) — fastest, safest wins.
  2. Priority/commercial pages next.
  3. The long tail, in ongoing batches.

Don’t forget the backlog. A rollout isn’t only “new content forward.” Some of your best opportunities are older, already-published pages that should link to newer related content and don’t. I’ve long recommended going back into existing articles to add links to new related pages — that retrofit is part of the rollout, not a separate project. (Note this is a forward-looking retrofit driven by your matrix, distinct from the reactive orphan/broken-link cleanup the internal links audit covers.)

And keep the structure legible as you go. Mueller’s point on internal-link count is that there’s no ideal number, but the structure needs to stay recognizable during a crawl — as reported by iloveseo.com’s write-up of his March 2022 hangout, he described wanting a crawl where “you can still recognize there’s a structure,” and warned it gets “harder to recognize the structure if every page was linked with every other page.” Relayed via a third-party transcription of a Search Central hangout, not a primary transcript — treat as close paraphrase. A rollout that quietly links everything to everything defeats the plan.

Step 6 — Measure whether it worked

You built a baseline in your matrix (Step 3’s status/date columns). Now compare before and after.

Google Search ConsoleA free Google service that reports how a site performs in Google Search and surfaces problems with how Google crawls, indexes, and serves it. It's first-party data straight from Google — but you don't need it to appear in results. — the Links report. This is the primary check. The report’s Top internally-linked pages table shows, in Google’s words, which of your pages “is linked the most from within your own site” — so you can confirm your priority pages actually climbed the list after the rollout. Two of Google’s own recommended uses map straight onto Steps 4 and 6: “Confirm that the core site pages (home page, contact page) are properly linked within your site” and “clean up the traffic flow within your site by confirming that users can get from page A to page B with the fewest clicks.” That second one is your click-depth check in Google’s own words.

Click-depth reduction. The internal links article defines click depth as a metric; here it’s an outcome to verify — did your priority pages move closer to the homepage after the rollout? Re-crawl (Ahrefs Site Audit, Screaming Frog) and compare depth before/after for your top-tier pages.

Crawl Stats — a secondary signal. GSC’s Crawl Stats reportA Google Search Console report (under Settings) that shows how Google has crawled your site over the last 90 days — total requests, download size, and average response time, broken down by response code, file type, Googlebot type, and purpose. It's only available for root-level properties (a Domain property or a URL-prefix property verified at the site's root). “shows you statistics about Google’s crawlingCrawling is how search engines use automated bots (like Googlebot and Bingbot) to discover URLs and download pages. A page has to be crawlable to be indexed, but crawling on its own isn't a ranking factor. history on your website… how many requests were made and when, what your server response was.” A better-linked structure can shift how Google crawls — Google notes that “if a site has information that changes less frequently, or isn’t very high quality, we might not crawl it as frequently.” Be careful here: Google does not frame Crawl Stats as a before/after tool for internal-linking changes — that application is my inference, not official guidance. Read any crawl shift as suggestive, not proof.

Set a cadence. A link matrix isn’t a one-time deliverable. The large-site guides converge on re-measuring and iterating (quarterly is a common cadence), and my own advice to keep linking new content back into older pages is the same point: this is an ongoing process, not a document you file away.

Common mistakes to avoid

  • Treating the plan as a licence to over-link. A matrix that links every page to every other page destroys the hierarchy signal it was supposed to create.
  • Assuming Google endorses “the internal link matrix.” It doesn’t — this is industry process on top of Google’s general principles. Say so; don’t imply an official framework.
  • Big-bang relinking. Contradicts Mueller’s stage-it-and-track-it guidance and leaves you unable to attribute any ranking change.
  • Exact-match anchors in every cell. Model varied, descriptive anchors instead.
  • Filing the matrix and walking away. It’s a living document; re-measure and add links to older content as you publish.

Where this sits in the cluster

This is the execution playbook that sits on top of the internal links fundamentals, the site architecture models, and the website structureWebsite structure (site architecture) is a site's visible hierarchy, navigation, breadcrumbs, and URL organization — how pages relate and how people and search engines move between them. Internal linking is the primary signal Google reads to understand that structure, not URL folders. hub — all in this same cluster, and all auto-linked once published. The neighbouring topics on anchor text, orphan pagesAn orphan page is a page on your site that no other page links to internally. Because crawlers discover pages by following links, an orphan page is effectively invisible to search engines unless it's in an XML sitemap or linked from an external site., and automated internal linkingAutomated internal linking is the use of software — CMS plugins, custom scripts, or build-time tooling — to insert hyperlinks between related pages on a website automatically, based on keyword or topic matching, instead of an editor adding each link by hand. plug into the same plan: anchor text fills your matrix’s third column, orphan pages are what a good plan prevents in the first place, and automation is how you scale the rollout on a very large site.

Add an expert note

Pin an expert quote

New person? Create their unclaimed profile at /admin/experts/ → Pin a quote first.