Internal linking in 2026: hub & spoke, breadcrumbs, and recommended-content blocks

Internal linking has moved well beyond “add a few links and hope for the best”. In 2026, strong results come from a clear structure that helps people orient themselves, helps search engines crawl efficiently, and stays maintainable as the site grows. The most reliable pattern is a hub-and-spoke set-up supported by breadcrumb navigation and well-designed blocks of recommended materials that earn their place through relevance.

Hub & spoke architecture that scales without turning into a link farm

A “hub” is the page that sets the overview and the standards: it defines the topic, covers the core questions, and links out to the most important subtopics. The “spokes” are those subtopic pages, each going deeper and satisfying a specific intent. In practice, this works best when the hub is not a category dump, but an editorial page with a clear structure, a short table of contents, and links that are explained in context (why that spoke matters, and what the reader will get from it).

In 2026, the biggest mistake I still see is confusing hub-and-spoke with aggressive cross-linking. Your spokes should link back to the hub (so the hierarchy is obvious), and they can link to a small number of sibling spokes when there is a natural next step (for example, “measurement” after “implementation”). If a spoke links to ten loosely related pages “just in case”, it stops being helpful and becomes noise for both users and crawlers.

Technical execution matters: links should be crawlable, meaning standard anchor links with an href attribute. If your navigation relies on non-link elements, complex scripts, or click-handlers that don’t produce a real URL, you’re adding friction for discovery and crawling. Keep the basics boring and dependable, then use JavaScript for enhancements rather than for the core route through the content.

Anchor text, link placement, and a practical “link budget” per page

Anchor text should describe what the user will see after clicking. That sounds obvious, but it’s still common to find “click here” or vague anchors repeated across a site. A good rule is: if you read only the linked text, you should still understand the destination. This is as much about usability as it is about helping search engines interpret the relationship between pages.

Placement is just as important as wording. Put the most important internal links where a reader naturally needs them: after a key definition, at the end of a section that raises a follow-up question, or in a short “next step” area after the main content. Links buried in a crowded footer or scattered inside unrelated paragraphs rarely get used, and they dilute the signal of what the page is actually about.

Set an internal guideline for link volume. There’s no magic number, but there is a point where additional links stop improving navigation. A practical approach is to define “must-link” targets (hub, primary conversion page if relevant, one or two highly related resources) and then allow a limited number of contextual links that genuinely help understanding. This keeps editors consistent and stops pages from becoming messy over time.

Breadcrumbs as navigation first, and structured data second

Breadcrumbs work when they reflect a real hierarchy: Home → Section → Subsection → Current page. For users, they reduce backtracking and help confirm “where am I?” For site owners, they create a consistent internal path to key sections and encourage cleaner information architecture instead of endless layers of tags and filters.

In 2026, breadcrumbs are also a place where many sites quietly break things: the breadcrumb trail doesn’t match the actual structure, uses inconsistent naming, or jumps across sections because of convenience. Treat breadcrumbs as part of the editorial and UX system, not as decoration. If a page can’t be placed cleanly in the hierarchy, that’s a signal to revisit taxonomy or the hub strategy.

Where it makes sense, add breadcrumb structured data that matches what users see on the page. The key is alignment: visible breadcrumbs, internal links, and markup should tell the same story. If your structured data describes one path but the page shows another, you’re creating avoidable confusion and potential reporting issues in tools like Search Console.

Common breadcrumb pitfalls: faceted pages, duplicates, and inconsistent trails

Faceted navigation (filters like brand, colour, price) is the classic breadcrumb trap. A filtered list can explode into thousands of URL combinations, and breadcrumbs can become meaningless if they reflect every filter choice. The safer approach is to keep breadcrumbs tied to the core category/hub hierarchy, and handle filters as on-page state rather than as “new levels” in the breadcrumb trail.

Another pitfall is duplicate content created by multiple breadcrumb routes to the same page. If an article can sit under three different categories, you need to pick a primary home for the breadcrumb, then support cross-category discovery with recommended-content blocks (contextual modules) rather than by changing the breadcrumb path depending on where the user came from.

Consistency is the quiet winner: use the same naming conventions across hubs and sections, keep URLs readable, and avoid frequent structural rewrites unless you’re ready to manage redirects properly. Breadcrumbs only become a strength when users see the same logic everywhere they go.

Recommended-content blocks that improve journeys, not just pageviews

Recommended-content blocks (related articles, “recommended materials”, “next reading”) are where internal linking becomes genuinely user-led. The goal is to anticipate the next question a reader will have, not to force traffic into unrelated pages. In 2026, these blocks work best when they are curated by rules that reflect intent: beginner vs advanced, informational vs transactional, and problem → solution sequences.

Keep the module tight: a small set of strong recommendations beats a long list. Give each item a clear title and, where possible, a one-line explanation that sets expectations. This reduces pogo-sticking and makes the block feel like editorial guidance rather than a random selection. It also helps your team maintain quality because they can quickly sanity-check whether each recommended item truly belongs.

Build maintenance into the process. Recommended-content blocks often start well and then decay as new pages are published. A simple monthly review for top landing pages (based on organic entrances) is usually enough: confirm that recommendations still match the page intent, replace outdated items, and ensure you’re linking to the current best resource on that subtopic rather than to an old post that merely shares a keyword.

How to choose recommendations: intent mapping, freshness rules, and testing

Start with intent mapping. For each hub and each spoke, write down the top 3–5 follow-up intents a user might have after reading the page. Then map those intents to your best supporting pages. This avoids the common “same-site keyword matching” approach that produces weak recommendations because it ignores what the reader is trying to do next.

Define freshness rules so the block stays credible. Freshness doesn’t always mean “newest”, but it does mean “still correct”. For topics that change quickly (regulations, pricing, technical standards), your rule might be “review every quarter”. For evergreen topics, it might be “review annually, or when performance drops”. Either way, you’re making upkeep a normal part of publishing, not a panic job after rankings fall.

Finally, test with outcomes that matter. Track click-through rate on the module, but also watch what happens after the click: time on page, scroll depth, and whether users return to the hub or continue deeper. If a recommendation block gets clicks but causes fast exits, it’s a sign that the promise (anchor/title) doesn’t match the destination, or that the recommendation isn’t actually the next best step.