Subdomain vs subfolder SEO is the decision about whether a section of your site lives at blog.example.com or at example.com/blog. Unlike most technical SEO choices, this one is architectural — it is cheap before launch, expensive after, and almost nobody revisits it once the CMS is wired up.
Google’s position has been consistent since 2012, when Matt Cutts said subdomains and subdirectories are roughly equivalent and advised picking whichever is easier to configure. John Mueller restated it in a 2017 Google office-hours session: “From our point of view, when we talk with the search quality team, they say subdomains and subdirectories are essentially equivalent.” Moz’s long-standing counter-position is that subfolders are the safer default, reserving subdomains mainly for language variants.
So both statements are true at once: Google’s ranking systems do not penalise subdomains, and teams really do see traffic jump when they move a blog from a subdomain to a subfolder. That contradiction is the whole reason this topic has been argued about for a decade, and it is why every article written on it hedges.
The contradiction resolves once you stop looking at the URL and start looking at what else changed.
What actually differs between a subdomain and a subfolder
A subfolder is a path on the same host. example.com/blog is served by the same DNS record, usually the same server, and — critically — usually the same CMS and the same template as the rest of the site.
A subdomain is a separate host. blog.example.com is its own DNS entry. It can point at a different IP, a different server, a different platform entirely. That flexibility is precisely why teams reach for it, and precisely why it causes problems.
Google has to work out, per site, whether a subdomain is part of the same property or an unrelated one. For blog.example.com sitting on a domain Google already understands, that determination is trivial and fast. For customers.example.com hosting user-generated content on a third-party platform, it is not — and Google may reasonably decide to treat the two differently.
That is the honest technical summary. Everything else in this debate is a proxy argument for something operational.
Why subfolders usually win, and why the usual explanation is wrong
The standard explanation is that link equity “leaks” or “splits” across a subdomain boundary and gets consolidated on a subfolder. It is repeated in nearly every article on the SERP. It is also the weakest part of the argument, because Google has said the opposite for over a decade and has never given a mechanism for the leak.
Here is what I actually see when I audit these setups. In my experience auditing sites where a blog sits on a subdomain, the same four things are true almost every time:
- The subdomain runs a different CMS. The marketing site is on Webflow or a custom stack; the blog is on WordPress. Two systems, two templates, two release cycles.
- It is not in the primary navigation. Or it is, but the link is one-way — the main site links out to the blog, and the blog’s own header is a stripped-down version that doesn’t link back to the commercial pages.
- Internal linking stops at the boundary. Blog posts link to other blog posts. They rarely link to the pricing page, the service pages, or the category pages, because the person writing them is working in a different tool and doesn’t have those URLs to hand.
- Analytics and Search Console are split. Two properties, two dashboards, and nobody looking at the combined picture.
None of those are caused by the dot instead of the slash. All of them are caused by the separation the subdomain permits. The subdomain doesn’t hurt you — it removes the friction that would otherwise force you to integrate.
When a team migrates to a subfolder, they fix all four at once, because a subfolder makes it hard not to. Then they attribute the resulting traffic gain to the URL change. That misattribution is the single most common error in this entire topic.
This matters because it tells you what to do if you’re stuck on a subdomain for reasons you can’t change: fix the four items above and you recover most of the difference. It also tells you that migrating to a subfolder while keeping the blog on a separate template with no internal linking will do very little. I have seen exactly that outcome, and the post-mortem is always uncomfortable.
The four cases where a subdomain is the right call
Defaulting to subfolders is correct advice. But “default” is not “always”, and the exceptions are real:
A genuinely separate product or audience. If the content serves a different set of people with different intent and would never be internally linked to your commercial pages anyway, the integration benefit doesn’t exist. A developer documentation site for an API is the classic example.
A third-party hosted platform. Help centres on Zendesk or Intercom, communities on Discourse, status pages on Statuspage — these vendors serve from their own infrastructure. You can usually CNAME a subdomain to them. Getting them onto a subfolder path requires a reverse proxy, which may or may not be worth the operational cost.
Technical isolation you actually need. A single-page app with a different security posture, a different deployment cadence, or a different CDN configuration. Forcing it onto the main host to satisfy an SEO preference is a bad trade when the SEO upside is modest and the engineering risk is not.
Regulatory or infrastructure separation. Data residency requirements, a distinct legal entity, or a subsidiary brand that will eventually be sold. Structural decisions that outlive the SEO argument.
Notice what is not on this list: “our blog is on WordPress and the main site isn’t.” That is the most common reason teams give, and it is a convenience argument, not an architecture argument. It is exactly the case a reverse proxy solves.
The subdomain vs subfolder decision framework
Four questions, in order. Stop at the first “yes”.
- Does a regulatory, legal, or data-residency requirement force separation? If yes, use a subdomain. The SEO conversation is over.
- Is the content hosted by a third party you cannot serve through a proxy? If yes, use a subdomain — and then treat it as a separate property you deliberately link to and from.
- Would this content ever be internally linked to and from your commercial pages? If no, a subdomain is fine. It gains nothing from integration because you were never going to integrate it.
- Is there a hard engineering constraint that a reverse proxy cannot resolve? If no, use a subfolder.
Most teams reach question four and discover the answer is no. That is the point of running the questions in order — it separates the genuine constraints from the convenience preferences that have been dressed up as constraints.
If you’re making this call as part of a broader build, it belongs in the same conversation as your site architecture and URL hierarchy, not as a separate technical ticket.
How to migrate from a subdomain to a subfolder
If you’ve decided to move, the sequence matters more than the speed. This is a site migration with all the usual risks, and the fact that it happens within one domain does not make it safe.
- Crawl and export the subdomain in full. Every indexable URL, its current traffic, and its inbound links. This is your baseline and your redirect source list.
- Build a one-to-one redirect map. Every old URL gets a specific destination. Do not redirect everything to the new section’s homepage — that is treated as a soft 404 and it discards the ranking signal you are trying to preserve.
- Rebuild the templates on the destination. Same navigation, same footer, same breadcrumbs as the rest of the site. If you skip this, you have moved the URL and kept the problem.
- Wire the internal linking before you cut over. Contextual links from the new section into your commercial pages, and from relevant commercial pages back into the new section. This is where the actual gain lives — see the internal linking strategy guide for how to prioritise those links.
- Ship the 301s and keep them permanently. Not for a year. Permanently.
- Add the destination to Search Console and keep the old property. You need the old property to watch the subdomain’s URLs drain out of the index and to catch redirect errors. Do not delete it.
- Monitor for six to eight weeks. Expect a dip in weeks one to three. Judge the outcome at week eight, not week two.
Sites that follow this sequence generally recover and then exceed the old baseline. Sites that skip steps three and four generally end up exactly where they started, having spent a quarter on it.
The reverse proxy option nobody mentions
There is a third option, and it barely appears in the SERP for this keyword. A reverse proxy lets you serve a separately hosted application on a subfolder path of your main domain. The request hits example.com/blog, your edge layer forwards it to the WordPress instance, and the response comes back on your main host.
Cloudflare documents this pattern explicitly for Workers, and most CDNs and edge platforms support an equivalent. It gets you the architectural benefit of a subfolder without merging the technology stacks — which is the constraint most teams actually have.
It is not free. You take on latency at the proxy layer, a caching configuration to get right, and a new failure mode when the origin platform changes its routing. But when the only reason you’re on a subdomain is “the blog runs on a different CMS”, this is the answer, and it is worth pricing before you commit to either side of the argument. A technical SEO review before you build will tell you whether the proxy cost is justified for your traffic volume.
Does this change for international sites?
Partly. Language and country variants are the one case where the subdomain option has a genuine constituency, and Google supports subdomains, subfolders and ccTLDs equally for geotargeting.
The practical argument still favours subfolders for most businesses: one property to consolidate, one set of templates, one crawl surface. Subdomains make more sense when the country sites are operationally independent — different teams, different catalogues, different release schedules. The hreflang implementation is identical either way, and getting that right matters more than the container you choose. If you’re weighing this, the international SEO guide covers the full comparison including ccTLDs.
What changes for AI search
Less than you would hope. Perplexity, ChatGPT and Google’s AI surfaces cite at the URL level, not the domain level, so a well-linked subdomain page can be cited as readily as a subfolder page.
The indirect effect is the one that matters. AI systems favour content they can reach, parse and contextualise. A subdomain that is orphaned from your main navigation is harder to discover and gives the model less context about who published it and why they are credible. That is the same problem as the search case, arriving through a different door — and the fix is the same one.
The short answer
Use a subfolder unless one of the four exceptions applies. But do not expect the URL change alone to do anything for you. The subdomain vs subfolder SEO decision is worth getting right at build time and rarely worth a standalone migration project — what is worth the project is the navigation, template and internal linking work that a subfolder migration usually smuggles in alongside it. Do that work first, measure it, and then decide whether the URL still needs to move. If you want a second opinion on your specific setup before committing engineering time, that is exactly the kind of question a technical SEO audit should answer.
Frequently Asked Questions
Is a subdomain or subfolder better for SEO?
A subfolder is the better default for most sites, but not because Google ranks it higher. Google has said since 2012 that it treats the two as essentially equivalent. Subfolders win in practice because they force integration — shared templates, shared navigation, and internal linking that connects your content to your commercial pages. If you can achieve that integration on a subdomain, most of the gap closes.
Does Google treat subdomains as separate sites?
Google determines this per site rather than applying a blanket rule. For a subdomain on a domain it already understands, with shared branding and reciprocal internal links, Google generally recognises it as part of the same property. For a subdomain hosting unrelated content on third-party infrastructure, it may treat it separately — which is usually the correct interpretation of what you have built.
Will moving my blog from a subdomain to a subfolder increase traffic?
Often yes, but rarely because of the URL change itself. The gain typically comes from the work that ships alongside the move: rebuilt templates, the section appearing in the primary navigation, and internal links into your commercial pages. If you migrate the URLs and keep the blog isolated on its own template, expect very little. Plan the migration around the integration, not the path.
When should you use a subdomain instead of a subfolder?
Four cases justify a subdomain: a regulatory or data-residency requirement, third-party hosted content you cannot serve through a proxy, a genuinely separate product or audience that would never be internally linked to your commercial pages, and a hard engineering constraint a reverse proxy cannot resolve. “Our blog runs on a different CMS” is not on that list — that is the case a reverse proxy is designed for.
Do subdomains split link equity?
There is no documented mechanism for link equity splitting at a subdomain boundary, and Google has consistently said otherwise. What genuinely splits is the internal link graph, because subdomain content is usually built in a separate system and rarely links back to the main site’s key pages. The effect people attribute to lost link equity is almost always missing internal links, which is a fixable content problem rather than an algorithmic penalty.
