On-Page SEO Guide
Internal linking is not a quota of links per article. It is a site-level decision system: choose the page that deserves support, find relevant source pages, place descriptive crawlable links where readers need them, then measure whether search visibility and navigation improve.
What is internal linking?
Internal linking is the practice of linking one page on a website to another page on the same website. Internal links help people navigate, help Google discover important URLs, provide context through anchor text, and help a site express which pages and relationships matter. For SEO, the useful question is not “How many links should I add?” It is “Which relevant page should support which destination, and why?”
<a href> link with descriptive text. Do not rely on onclick-only navigation or vague anchors.1. What Google actually confirms about internal links
Internal linking advice gets unreliable when useful heuristics are repeated as if Google published them as ranking rules. Start with the parts Google documents clearly.
| Google-documented point | What it means in practice |
|---|---|
| Google uses links to discover pages and understand relevance. | Important pages should be reachable through normal site links, not isolated from the rest of the site. |
Google recommends crawlable <a href> links. | Buttons, router attributes, or JavaScript events that do not produce a real link can make discovery less reliable. |
| Anchor text should be descriptive, concise, and relevant. | “Internal linking audit” tells users and Google more than “click here.” Avoid stuffing every variation into the anchor. |
| Every page you care about should have a link from at least one other page on your site. | Do not leave valuable URLs dependent only on XML sitemaps or external discovery. |
| Internal links remain relevant for generative Search features. | Google’s AI features use the Search index. Pages still need to be crawlable, indexable, useful, and easy to find. |
Primary guidance: Google Search Central’s link best practices. For the broader link-quality context, see Search Counsel’s guide to how Google evaluates links.
2. Why internal links matter for SEO
Internal links do several jobs at once. That is why they can be high-leverage even though adding the HTML itself is simple.
Discovery and recrawling
A crawler can follow internal links from known URLs to other pages. If a useful page is absent from normal navigation and never linked contextually, it is harder for both users and crawlers to encounter naturally. A sitemap may help discovery, but it does not replace a coherent internal path.
Context and URL ownership
Descriptive links help express the relationship between a source page and its destination. If five different pages link to one guide with naturally descriptive phrases around the same subject, the site is giving Google and users a consistent clue about that guide’s role.
Internal authority and importance
Google’s link systems include PageRank concepts, and internal links are part of the link graph under your control. That does not mean one link guarantees a ranking lift. It means internal architecture can influence which pages receive attention and link-based signals inside a site.
User journeys and commercial routing
A strong internal link should also answer the next question. Informational content can route readers into deeper guides, service pages, product categories, tools, or case studies when the destination genuinely continues the task. Search Counsel’s content strategy approach treats that routing as part of the content system rather than an afterthought.
3. Types of internal links and the job each one does
| Type | Main job | What to watch |
|---|---|---|
| Primary navigation | Expose core sections and high-priority destinations. | Do not turn the main menu into a full sitemap. |
| Contextual body links | Connect related ideas at the moment a reader needs the next resource. | Relevance matters more than adding links for a quota. |
| Breadcrumbs | Express hierarchy and help users move upward through a section. | Breadcrumb structure should match the real information architecture. |
| Related-resource modules | Expose useful next steps beyond the body copy. | Avoid generic widgets that link the same unrelated pages everywhere. |
| Footer links | Provide stable access to essential utility and business pages. | Do not use the footer as a keyword-stuffed authority distribution hack. |
| Image links | Make visual elements navigable. | Use descriptive alt text because linked images can provide anchor context. |
4. The Search Counsel internal-link decision framework
The fastest way to waste time on internal linking is to begin with the source article and ask, “Where can I insert more links?” Start from the page that needs help.
This is an editorial workflow, not a Google ranking formula. Its value is that every link has a reason that can be inspected later.
5. How to choose target pages with Google Search Console
Internal linking is most actionable when you already know which URL you want to strengthen. Search Console can help identify pages where additional support is plausible, especially when Google is already testing the URL for relevant searches.
| Search Console pattern | What it may mean | Internal-link action |
|---|---|---|
| Relevant impressions with positions roughly outside page one | Google may understand the topic but the page has not earned strong visibility. | Review intent, content quality, incoming internal links, source relevance, and authority together. |
| Several related queries, but one destination receives weak internal support | The page has query breadth but may be under-signaled inside the site. | Add contextual links from the strongest genuinely related pages. |
| New page indexed but barely discovered through the site | The URL may exist technically without being integrated editorially. | Add links from the hub, relevant siblings, and any page where the new resource answers the next question. |
| High-value service/product page gets little organic visibility | The page may be buried, weak, or poorly connected to informational demand. | Build useful paths from relevant informational and category pages; do not force unrelated blog links. |
| URL already wins strong top positions | Internal linking may not be the highest-priority constraint. | Protect what works. Focus effort on weaker pages unless the link improves the user journey anyway. |
| No impressions because the URL is not indexed or canonicalized elsewhere | This is a technical eligibility problem first. | Fix indexability/canonical issues before expecting internal links to solve ranking. |
A simple GSC workflow
- Open Performance → Search results.
- Filter to the target URL.
- Use a complete comparison period when possible, such as the latest 28 days versus the previous 28 days.
- Review queries, impressions, clicks, CTR, and position together. Do not diagnose from one metric.
- Confirm the queries are relevant to the page’s intended job.
- Then inspect which relevant pages already mention the target topic but do not link to the preferred URL.
If Search Console reporting is the bottleneck, use Search Counsel’s broader on-page SEO and content-strategy resources to separate a relevance problem from an internal-link problem before changing the page.
6. How to find the best source pages
The best source page is not automatically the page with the most backlinks or traffic. Topical relevance comes first. Then consider source strength, destination priority, and reader usefulness.
| Candidate source | Topical fit | Source strength | Reader benefit | Decision |
|---|---|---|---|---|
| High-traffic guide on the same subtopic | High | High | High | Strong candidate |
| Relevant hub or pillar page | High | Often high | High | Strong candidate |
| Older article on an adjacent topic | Medium | Medium | High if the context is natural | Review manually |
| Homepage paragraph unrelated to the target subject | Low | High | Low | Usually skip |
| Random high-authority page selected only for link equity | Low | High | Low | Skip |
Ways to discover candidate sources
- Site search: search your domain for the target topic and review pages that already discuss it.
- Crawler exports: identify pages that mention target concepts but do not link to the preferred URL.
- Content inventory: use your topic map to find hubs, siblings, and related commercial pages.
- GSC: favor sources that already attract relevant visibility when they also fit the topic.
- Backlink/page-strength data: use authority as a secondary tie-breaker after relevance.
Search Counsel’s guide to topical authority makes the same distinction at the cluster level: each URL should own a distinct job while related pages reinforce the broader subject.
7. Anchor text: be descriptive without turning it into a formula
Anchor text should help a reader predict what happens after the click. Google’s current guidance emphasizes text that is descriptive, reasonably concise, and relevant to both the source and destination.
Weak: click here
Better: review our guide to how Google evaluates link signals
Also fine when natural: learn how topical authority and internal relationships fit into a content system
You do not need to force the same exact-match anchor every time, and you do not need an invented percentage threshold for “safe” internal anchors. Vary wording when the sentence calls for it. The destination should remain obvious.
8. Crawlable vs non-crawlable internal links
A link that looks clickable to a user is not automatically an ideal crawl path. Google recommends standard anchor elements with an href pointing to a resolvable URL.
Recommended
<a href="/javascript-seo/">JavaScript SEO guide</a>Riskier as a primary crawl path
<span onclick="goTo('/javascript-seo/')">JavaScript SEO guide</span>Google can process JavaScript and can crawl dynamically inserted links when the rendered markup becomes a normal anchor with an href. The practical test is not “Was JavaScript involved?” It is “What link exists in the rendered HTML, and can a crawler resolve it?” Use URL Inspection or rendered-page testing when the implementation is uncertain.
nofollow as an internal crawl-control system. If a page should not appear in Search, use the appropriate indexing or access control. If a normal internal link is useful to users and points to a page you want Google to understand, there is usually no reason to nofollow it.
9. How PageRank fits into internal linking without bad math
PageRank began as a way to estimate page importance from links. Internal links participate in the link graph, so changing the graph can change how link-based signals move through a site. That is the useful concept.
The mistake is turning that concept into precise modern SEO math. Google does not give SEOs a live “link equity meter,” and modern ranking systems use far more than original PageRank. A page with ten internal links is not automatically better than a page with fifty, and one “power page” link does not guarantee a rank increase.
A practical priority order is:
- Relevance: does the source page genuinely relate to the destination?
- Importance: is the destination a page the site actually wants users and search engines to find?
- Source strength: does the source page earn visibility, links, or prominent placement?
- Placement: does the link appear where the reader benefits from the next step?
For a deeper discussion of PageRank, backlink quality, and what Google has actually documented, see How Google Evaluates Backlinks.
10. How many internal links should a page have?
There is no published Google “ideal number” of internal links per page. Add links when they improve navigation, connect relevant information, or help important content get discovered. Remove or avoid links that are repetitive, irrelevant, templated without purpose, or inserted only to hit a quota.
Usefulness scales with the page. A short definition may need only a few links. A long technical guide may legitimately contain dozens because it references examples, prerequisites, related problems, and sources. The decision should come from the reader task and information architecture, not a fixed “links per 1,000 words” rule.
Google-confirmed guidance vs SEO folklore
<a href>.11. Internal linking architecture: hubs, clusters, services, and ecommerce
Internal linking works best when every page has a defined role. The goal is not for every page in a cluster to rank for the same broad query. The goal is for each page to own a distinct task while helping users move through the larger subject.
| Site type | Useful internal-link pattern | Common failure |
|---|---|---|
| Publisher / agency | Pillar → focused guides → relevant sibling guides → service path when appropriate. | Dozens of articles with no preferred owner for each intent. |
| SaaS | Feature hub → use cases → documentation → comparison/help content. | Blog earns visibility but never supports feature or solution pages. |
| Ecommerce | Category → subcategory → product; buying guides link into relevant categories/products. | Every product depends on faceted URLs or search boxes for discovery. |
| Local / service business | Service hub → specific services → relevant location pages → supporting FAQs/case studies. | Location pages all cross-link mechanically with no user reason. |
Search Counsel’s own content strategy architecture is organized around nine topic hubs with focused supporting guides. That structure is useful only when the links follow genuine topic relationships rather than a template.
12. Worked example: choosing one internal link from start to finish
Scenario
Target page: /javascript-seo/
Potential source: a technical SEO guide discussing rendering and crawlability.
Weak implementation
“For more information, click here.”
Stronger implementation
“If navigation or content depends on client-side rendering, check whether your JavaScript SEO implementation leaves important links available in rendered HTML.”
| Check | Assessment |
|---|---|
| Source-to-target relevance | Strong: both pages concern crawl/render behavior. |
| Reader benefit | Strong: the destination explains the next technical problem. |
| Anchor clarity | Strong: “JavaScript SEO implementation” predicts the destination. |
| Crawlability | Strong when implemented as a normal <a href>. |
| Forced for SEO only? | No. The link is useful even if search engines did not exist. |
That final check is underrated. If a link makes the article worse for a reader, “link equity” is not a good excuse to keep it.
13. A complete internal-link audit: what to find and what to do
A useful audit is not a list of errors. Each finding needs a decision. Crawl the site, combine that crawl with sitemap and Search Console URLs, then review high-value pages manually.
| Problem | How to detect it | Typical decision |
|---|---|---|
| Orphan / isolated page | Crawl from homepage, then compare against sitemap and GSC URL sets. | Link it, redirect it, noindex it, or retire it depending on purpose. |
| Priority page with few useful incoming links | Inlink count + manual relevance review + business value. | Add contextual links from the strongest relevant pages. |
| Broken internal link | Crawler 4xx report. | Replace, remove, or restore the destination. |
| Redirecting internal link | Crawler 3xx report. | Update the source link to the final preferred URL when appropriate. |
| Wrong destination | Manual review of the surrounding sentence and destination intent. | Point to the page that actually answers the reader’s next question. |
| Generic anchor text | Anchor-text export or manual scan. | Rewrite when a more descriptive phrase genuinely improves clarity. |
| Competing destinations for the same intent | GSC query/page overlap + SERP review. | Clarify URL ownership before adding more links to both pages. |
| Deep important page | Click-depth report. | Add a useful hub, navigation, breadcrumb, or contextual path. |
| Onclick/router-only navigation | Rendered HTML inspection. | Expose the destination with a crawlable anchor when it should be discoverable. |
| Excessive sitewide template links | Compare repeated links across crawl templates. | Simplify when links add noise rather than navigation value. |
| Internal links to non-canonical URLs | Crawl + canonical report. | Link directly to the preferred canonical URL. |
| Useful page excluded from normal site journeys | Manual navigation test + analytics/GSC. | Integrate it where readers logically need it. |
For orphan-page handling specifically, Search Counsel’s indexing-problems guide explains why the correct action depends on whether the page is valuable, replaced, intentionally private/noindex, or genuinely obsolete.
14. How to measure whether internal-link changes worked
Do not declare success because a crawler shows ten new inlinks. That only proves implementation. Search impact requires post-change observation.
Record the baseline before editing
- Target URL and canonical.
- Indexed status.
- Incoming internal-link count and important source URLs.
- Clicks, impressions, CTR, and position for the target URL.
- Relevant query set and top-20/top-10 query counts if available.
- Organic landing sessions and business outcomes when analytics is available.
After implementation
- Recrawl and confirm the new links resolve directly to the preferred URL.
- Inspect rendered HTML for any JavaScript-generated links.
- Allow time for recrawling and reprocessing; do not judge the change from the next day.
- Compare complete periods such as 28 days before versus 28 days after, while noting seasonality and other SEO changes.
- Review whether the target gained relevant query breadth and stronger position buckets, not just whether total impressions moved.
- Check whether pages that already performed well were harmed by broader site changes.
15. Internal linking for AI Overviews and AI Mode
There is no special internal-link formula that guarantees AI Overview or AI Mode inclusion. Google says the same SEO foundations still apply to generative Search: content must be technically eligible for Search, crawlable, indexable, useful, and easy to discover. Internal links support those foundations by connecting important content and helping Google find it.
Google’s 2026 guidance explains that generative Search can use retrieval-augmented generation and query fan-out, where the system issues related searches to gather information across subtopics. That creates an opportunity for a strong page to be useful across several related questions, but it does not mean you should create a separate thin page for every fan-out query.
What improves this page’s AI-search readiness
- Self-contained answers: important questions begin with a direct, defensible answer before the detail.
- Unique information: frameworks, worked examples, original audits, and measured experiments give retrieval systems something beyond commodity definitions.
- Source clarity: factual claims link to primary Google documentation where possible.
- Internal discoverability: relevant Search Counsel hubs and supporting pages should link to this guide naturally.
- Useful visuals: diagrams and annotated screenshots should explain a real task, not decorate the page.
- Technical eligibility: the canonical page must be indexable and eligible for a normal Search snippet.
Google explicitly says there is no special schema required for AI features. Do not add “AI schema,” unnecessary AI text files, or markup that does not match visible content just to chase citations. See Google’s AI features and your website and its 2026 generative AI optimization guide.
For Search Counsel’s broader treatment of citation-focused content, see AI Overviews: How to Get Your Content Cited. Keep that page focused on AI citation strategy; keep this page focused on internal-link decisions.
16. Common internal linking mistakes
- Adding links by quota. A fixed density encourages irrelevant links and distracts from the reader journey.
- Choosing sources by authority alone. Relevance should come before raw page strength.
- Linking every keyword to a commercial page. Informational intent should not be interrupted by unrelated sales destinations.
- Using vague anchors everywhere. “Click here” wastes context when a descriptive phrase fits naturally.
- Forcing the same exact-match anchor repeatedly. Natural language is clearer and easier to maintain.
- Ignoring URL ownership. If two pages compete for the same intent, more links to both can reinforce the confusion.
- Leaving new content isolated. Publishing is not finished until relevant existing pages point to the new resource.
- Sending links through unnecessary redirects. Update internal links to the preferred destination where practical.
- Using JavaScript interactions without a real crawlable link. Check the rendered HTML.
- Calling implementation a result. New inlinks are not the same thing as new rankings, traffic, or leads.
FAQ: internal linking for SEO
Are internal links a Google ranking factor?
Links are part of Google’s systems for discovery, relevance, and link-based signals such as PageRank. It is safer to say internal links can influence how Google finds and understands pages than to treat one internal link as a guaranteed ranking boost.
How many internal links should I add to a page?
Google does not publish an ideal number. Add links that improve navigation, connect genuinely related information, or help important pages get discovered. A long resource may need many more links than a short page.
What is the best anchor text for an internal link?
Use wording that is descriptive, reasonably concise, and relevant to the destination. The reader should be able to predict what the linked page is about without needing a formula for exact-match percentages.
Do internal links pass PageRank?
Internal links participate in the site’s link graph, so PageRank concepts are relevant. Modern Google ranking is more complex than a simple “link juice” calculation, so use PageRank as a structural principle rather than a precise SEO meter.
What is an orphan page?
An orphan page is a URL with no discoverable internal link path from the crawled site. It may still be discovered through a sitemap or external link, but valuable pages should normally be connected to relevant site journeys.
Should every important page be within three clicks of the homepage?
Keeping important content reasonably easy to reach is a useful architecture principle, but “three clicks” is a heuristic rather than a published Google ranking threshold. Judge the structure by clarity, importance, and real user paths.
Do JavaScript internal links count?
They can. Google says dynamically inserted links are crawlable when the rendered result is a normal anchor element with an href. Onclick-only or non-anchor navigation is less reliable as a crawl path.
Do internal links help AI Overviews?
They can support the same foundations that matter for AI features: crawlability, discovery, clear site structure, and useful content. Google does not publish a special internal-link method or schema that guarantees AI Overview inclusion.
Sources and methodology
This guide separates documented Google guidance from Search Counsel’s editorial recommendations. The decision framework, source-selection matrix, audit process, and measurement workflow are practical SEO methods; they are not presented as Google ranking formulas.
- Google Search Central: Link best practices for Google
- Google Search Central: AI features and your website
- Google Search Central: Optimizing for generative AI features
- Google Search Essentials
- Search Counsel: How Google Evaluates Backlinks
- Search Counsel: Building Topical Authority
- Search Counsel: Content Strategy
- Search Counsel: How to Fix Indexing Problems
If your pages are already useful but the site does not route authority and readers toward the pages that matter, start with the target-page audit above. For a broader system-level review, see Search Counsel’s content strategy services and on-page SEO services.