Can you really block the flow of PageRank with robots.txt on a middle page?

Why block PageRank using robots.txt rather than nofollow?

Nofollow is no longer an absolute directive since March 2020. Google now treats it as a hint, meaning it can choose whether to follow the link or transfer PageRank. This ambiguity makes nofollow unreliable for completely blocking the flow of SEO juice.

In contrast, robots.txt blocks the crawl of the page itself. If Googlebot cannot access the middle page, it will never discover the outgoing links it contains. As a result, no PageRank is transferred to the final URL, since the engine ignores the existence of the link altogether. It’s a solid technical barrier, not just a suggestion.

How does this middle page actually work?

The principle is based on a two-layer architecture. You create an intermediate URL (for example /go/partner-xyz/) that you block in robots.txt with Disallow: /go/. This page contains a server-side redirect (either 301 or 302) to the final URL of the partner.

From the user’s perspective, the process is smooth: click on the link, a quick pass through the middle page, and arrival at the destination. From Googlebot’s perspective, the journey stops abruptly: the bot reads robots.txt, sees the prohibition, never crawls the middle page, and therefore never discovers the outgoing link to the partner. The PageRank remains confined within your site.

In what situations should you use this technique instead of a simple nofollow?

This method applies to situations where you have no control over the link’s attribute. Typically: a contractual partnership that imposes a dofollow link, a third-party widget that you must integrate with outgoing links, a sponsor that requires visibility without nofollow, or an affiliate platform that refuses restrictive attributes.

If you can simply add rel="nofollow" or rel="sponsored" to the link, do it. It’s simpler and less resource-intensive. But if the dofollow link is imposed and you want to protect your PageRank budget, the blocked middle page becomes the only effective safeguard. It prevents diluting your authority toward destinations that do not strengthen your SEO strategy.

• robots.txt blocks the crawl, so Googlebot never sees the outgoing links from the middle page
• nofollow is just a hint since 2020, not an absolute directive—Google may choose to follow it
• This technique applies to uncontrollable dofollow links (partnerships, widgets, imposed affiliations)
• The middle page must redirect server-side to preserve user experience
• Abusing this method can harm your crawl budget and complicate internal linking

Is this statement consistent with real-world observations?

Yes, it aligns with tests conducted on websites with massive affiliate partnerships. E-commerce sites that deployed blocked middle pages for their thousands of outgoing links to merchants observed better retention of internal PageRank. Strategic pages captured more juice, and rankings improved on targeted queries.

However, it’s important to nuance this: this technique only works if the middle page is fully blocked from being crawled. If you forget an Allow directive in robots.txt, or if Googlebot manages to crawl the page via another path (XML sitemap, external link, cache), the PageRank transfer resumes. The consistency of your robots.txt file is critical.

What risks does this method pose to internal linking?

The main pitfall is to multiply blocked pages without a clear strategy. Each middle page creates an additional URL in your architecture, consumes crawl budget (even blocked, it exists on the server), and adds complexity to analytics tracking. If you have 500 partners, that means 500 intermediate URLs to manage.

Another risk is if you block too broadly in robots.txt (for example, Disallow: /go/ while some pages /go/ should be crawled), you dry up entire sections of your site. The result is orphan pages, sections invisible to Google, and erosion of linking structure. [To be checked] regularly with a robots.txt audit + server logs to avoid accidental blockages.

In what cases does this rule not apply?

If you control the link’s attribute, don’t go through a blocked middle page. Simply use rel="nofollow", rel="sponsored" or rel="ugc" according to the context. It’s cleaner, lighter, and avoids polluting your architecture.

Another case: links to internal resources or strategic partners whose authority you precisely want to enhance by association. If an outgoing link to a reference actor boosts your expertise (for example, a link to a university study or a recognized tool), transferring PageRank is not an issue, it’s a quality signal. Only block what dilutes your juice without a positive SEO return.

How can you effectively set up a blocked middle page?

Create a dedicated URL for each outgoing link you want to neutralize. For example: yoursite.com/go/partner-abc/. On this URL, configure a server-side redirect 302 (Apache, Nginx, or via your CMS) to the final URL of the partner. The redirect should be instantaneous, without any HTML display or JavaScript.

Next, add the directive Disallow: /go/ into your robots.txt file to block the entire /go/ directory from crawling. Verify with Google Search Console's robots.txt testing tool that Googlebot cannot access your middle pages. Finally, replace all direct links to partners in your HTML code with links to your intermediate /go/ URLs.

What mistakes should you absolutely avoid?

Never block a middle page with a meta robots noindex tag or X-Robots-Tag. These directives require Googlebot to crawl the page to read them, meaning the bot will see the outgoing links before adhering to the noindex. The PageRank would have already been transferred. Only robots.txt blocks the crawl upstream.

Another common mistake is using a JavaScript redirect instead of a server redirect. Googlebot executes JavaScript variably depending on available resources. If the middle page is blocked by robots.txt, the bot will never crawl JavaScript anyway. However, if a human reaches the page, a JS redirect may fail or be blocked by an ad blocker. Stick to classic 301/302 HTTP redirects.

How can you check that the setup is functioning correctly?

Test your intermediate URLs in Google Search Console using the URL Inspection tool. You should see the message "Blocked by robots.txt" and no indexing. If the page is indexed or crawled, your robots.txt directive is misconfigured.

Also, analyze your server logs: Googlebot should never query the /go/ URLs. If you see hits from Googlebot on these pages, there is a leak (sitemap, external link, overly broad Allow directive). Lastly, ensure that your users encounter no delays or 404 errors when clicking these links. The redirect must be transparent and instantaneous.

• Create a dedicated intermediate URL for each outgoing link to neutralize (e.g.: /go/partner/)
• Set up a server-side redirect 302 or 301 from the intermediate URL to the final URL
• Add Disallow: /go/ in robots.txt to block crawling of the intermediate directory
• Test with Google Search Console's robots.txt tool to confirm Googlebot cannot access /go/ pages
• Replace all direct links to partners with links to your intermediate URLs
• Verify in server logs that Googlebot never crawls the middle pages