Should you really avoid using nofollow on your internal links?

Why does Google advise against using nofollow internally?

The main reason relates to PageRank flow and content discovery. When you apply a nofollow to an internal link, you cut off a channel of authority transmission between your pages. PageRank remains stuck on the source page instead of irrigating your entire site.

Googlebot also uses internal links as crawl paths. A nofollow signals to the bot that it doesn't need to follow that link, which can delay or prevent the discovery of certain pages. On a medium-sized site, this practice creates blind spots in indexing.

Does this recommendation change the approach to PageRank sculpting?

Yes, and it marks a clear shift. About a decade ago, some SEOs used nofollow to channel PageRank to strategic pages by blocking flows to pages deemed less important (legal notices, contact forms, etc.). Google has gradually made this technique obsolete by changing how nofollow is treated.

Today, a nofollow on an internal link does not redistribute PageRank to other links on the page. It dilutes it into nothing. In other words, you lose authority instead of concentrating it. This statement definitively puts to rest the idea of PageRank sculpting with nofollow as a viable strategy.

What are legitimate cases for using internal nofollow?

Google's recommendation does not mean that internal nofollow is completely banned. Certain situations justify its use, particularly for user-generated content links in comments or internal forums, even if they are technically internal links within the domain.

Links to low-value pages like cookie consent popups, product filter modals, or session URLs can also receive a nofollow to avoid wasting crawl budget. However, these cases remain marginal and require case-by-case analysis rather than systematic application.

• The internal nofollow blocks PageRank flow and prevents effective redistribution in your architecture
• Googlebot uses internal links for crawling: a nofollow can delay or block page discovery
• PageRank sculpting with nofollow no longer works since Google changed how this attribute is handled
• Legitimate cases only concern internal UGC or technical URLs without SEO value
• A clean link architecture is better than multiplying nofollow directives to correct structural flaws

Is this guideline consistent with what we see on the ground?

Absolutely. Audits of sites that have cleaned up their unnecessary internal nofollow attributes consistently show improved indexing and better PageRank distribution. Often, orphaned or poorly crawled pages regain visibility after removing unnecessary nofollow barriers.

What’s trickier is that some sites have built their entire architecture on historical PageRank sculpting. Abruptly removing all internal nofollow links without rethinking the structure can create temporary imbalances. The transition requires a methodical approach, not just a simple find-and-replace.

What nuances should we consider with this general rule?

Google says "generally recommended," which allows for interpretation. In practice, some e-commerce sites with thousands of filtering facets generate parameter URLs that dilute crawl budget. Applying a nofollow to these links can be tactically justified, even though other solutions (canonical, robots.txt, Search Console parameters) are preferable.

The real nuance is that nofollow should never be used as a patch on a flawed architecture. If you need nofollow to prevent Google from crawling half your site, it’s your structure that’s problematic, not the lack of nofollow. [To be verified]: Google does not communicate a specific threshold beyond which internal nofollow becomes counterproductive, but practical experience suggests that beyond 5% of internal links being nofollow, you likely create more problems than you solve.

In what contexts could this recommendation be misapplied?

The main risk is falling into the opposite extreme: removing all nofollow attributes indiscriminately. On a community site with UGC, removing nofollow from user profile links or signatures can open the door to link spam. Context takes precedence over the general rule.

Another pitfall: confusing nofollow with other directives. Some developers apply nofollow where a simple HTML structure adjustment (moving a link to the footer, reducing click depth) would be more relevant. Nofollow is not an editorial prioritization tool, it's a technical instruction with collateral effects.

What should you do concretely on your site?

The first step is to audit your internal links to identify all present rel="nofollow" attributes. A crawl using Screaming Frog, Sitebulb, or Oncrawl will provide you with a complete mapping. Export the list and categorize these links by type: navigation, editorial content, footer, sidebar, UGC.

Next, create a decision matrix. For each nofollow link, ask yourself three questions: does this link have SEO value? Should Googlebot crawl the target page? Is this link likely to unnecessarily dilute crawl budget? If the answers are yes/yes/no, remove the nofollow. Otherwise, look for an alternative (canonical, robots.txt, deindexing).

What mistakes should be avoided when cleaning up internal nofollow?

The classic mistake is treating all nofollow links the same way. A nofollow on a comment link does not serve the same purpose as a nofollow on a menu link. Segment your analysis by template and context rather than doing a blind global cleanup.

Another trap is removing nofollow without monitoring the impact. Set up indexing tracking (Search Console, server logs) to check that the removal of nofollow actually improves crawling and indexing. If you notice a surge in crawling on pages without value, it means your architecture needs more in-depth work than just removing attributes.

How can you verify that your internal linking is optimized?

Use Search Console coverage reports to identify discovered but unindexed pages or crawled but unindexed pages. A high rate may indicate that your internal links do not transmit enough authority or that some nofollow attributes are blocking PageRank flow.

Also analyze the click depth of your strategic pages. Any important page should be accessible within a maximum of 3 clicks from the homepage. If not, your link architecture needs to be rethought, not compensated for by nofollow. A good internal linking structure makes internal nofollow attributes nearly unnecessary.

• Crawl the entire site to identify all links with rel="nofollow"
• Segment nofollow attributes by context (navigation, content, footer, UGC) before making any decisions
• Remove nofollow from editorial and navigation links to SEO-value pages
• Retain nofollow only on UGC links or technical URLs without SEO interest
• Monitor indexing and crawling in the 4 weeks following changes
• Ensure that the click depth of strategic pages remains under 3-4 levels