Subdomain vs Subdirectory

I lean toward subdirectories for most content that supports your core site—blog, news, resource library, product documentation. They keep authority consolidated under one domain, make internal linking straightforward, and simplify analytics because everything lives in one property. For example, a client ran an e-commerce site with a blog under blog.example.com and saw the blog's organic traffic stay flat for months. We moved the content to example.com/blog/, set up 301 redirects, and within three months that blog content contributed 30% of the main domain's new organic clicks. The authority that had been split across two properties merged back into one.

I reserve subdomains for sections that genuinely need independence. If a section runs on a different CMS, requires its own login system, serves a completely different audience, or is a separate web app, a subdomain makes sense. For example, a support portal with its own knowledge base software and separate user accounts works well under support.example.com. A store on a different platform like Shopify also often forces a subdomain. You can still pass some authority through internal links, but it will never be as tight as a subdirectory.

Here is my decision rule: if the content shares the same CMS, the same analytics property, and you want Google to treat it as part of your core site's expertise, use a subdirectory. If the section demands a different team, a different tech stack, or an isolated user experience, use a subdomain. International versions? I prefer subdirectories with [hreflang tags](/hreflang-tags/) unless you need separate hosting or language-specific domains. The one exception I make is for very large programme content that could dilute the main site's focus, but even then I test a subdirectory first.

Some argue subdomains help with crawler budget or let Google treat them as separate sites. In my experience, Google crawls subdirectories just as efficiently if you have a solid [sitemap](/sitemap/) and internal links. I have not seen a crawler budget problem that justified splitting content.

Pitfall one: assuming a subdomain automatically ranks worse. This is the most common myth. Google has stated clearly that neither format has an inherent ranking advantage. I have tested this: a client kept their blog on a subdomain for two years, then moved it to a subdirectory—no ranking change for the blog posts themselves. The difference was only in how authority consolidated across the rest of the site. The subdomain blog's posts performed fine, but the main domain got less link equity from them.

Pitfall two: treating URL structure as the main ranking factor. Content quality, backlinks, and relevance matter far more. I have seen subdomains outrank subdirectory pages simply because the subdomain had better content and stronger external links. Do not agonise over the URL format while ignoring what is on the page.

Pitfall three: moving content without planning redirects, canonicals, and analytics. When you shift content from a subdomain to a subdirectory (or vice versa), you need 301 redirects. Without them, you lose link equity and confuse users. I once saw a site lose 15% of organic traffic because they moved their blog from blog.example.com to example.com/blog/ without redirecting the old URLs—a classic mistake. Also update your [canonical tags](/canonical-tag/) and set up analytics tracking to ensure you are counting the new URLs correctly.

Pitfall four: using a subdomain just for organisational preference. I see teams split content because it feels cleaner. But if the content belongs under the same topical umbrella, a subdomain fragments authority for no benefit. For example, a SaaS company put their knowledge base on a subdomain when it could have lived under /help/. The knowledge base never passed authority to the main site's product pages. After moving it to a subdirectory, internal links between help articles and product pages improved, and the product pages saw a lift in rankings for informational queries.

A less common but real pitfall: forgetting that subdomains are treated as separate sites for Google Search Console and analytics. You need separate properties, which adds overhead for reporting and monitoring. If you work with a team, that extra complexity can slow down [SEO audits](/seo-audits/).

I used to think subdomains were a bad idea for any content that should rank. I had read early advice that said Google treats subdomains as completely separate entities, so splitting authority would always hurt. I recommended subdirectories for everything—blogs, documentation, even a separate e-commerce section that ran on a different platform. That advice was too rigid.

What changed my mind was a client who needed to spin up a job board. The job board ran on dedicated software with its own CMS, databases, and user accounts. Trying to force it into a subdirectory would have required major engineering work to integrate the two systems. The team was small, and they needed to move fast. I recommended a subdomain (jobs.example.com) and helped them set up strong internal links from the main site's career page. The job board ranked well for niche queries and did not harm the main domain's performance. I had prioritised an ideal structure over practical constraints.

I also underestimated the importance of content silos. For a large site with multiple distinct topics, a subdomain can help Google understand that a section is authoritative in its own right. Think of a university: physics.example.edu ranks for physics terms while history.example.edu ranks for history. Trying to cram everything under example.edu/physics/ and example.edu/history/ would confuse the site's overall theme and make internal linking harder because the departments are genuinely separate. I now see subdomains as a valid choice when the content forms a clear topical cluster that deserves its own identity.

Another mistake: I assumed using a subdomain meant you could not pass any authority. That is wrong. Links between a subdomain and the root domain are treated as normal web links, not nofollowed. You can absolutely pass authority through contextual links, even if it is not as clean as a subdirectory. I have seen cases where a subdomain's pages earned backlinks that then indirectly helped the root domain through editorial links. The key is to link between them naturally, not to rely on the URL structure alone.

What I still am not sure about: whether subdomains affect [duplicate content](/duplicate-content/) signals in edge cases. If the same content lives on a subdomain and the main domain, Google might see them as separate sites and not consolidate signals as aggressively. I always use a canonical tag pointing to the preferred URL when duplication happens, but I have not tested a scenario where both versions are deliberately kept live.