Subdomain or subfolder for your blog: what actually happens when you pick wrong
The question that keeps coming back
Every few months someone in a forum or a client call asks the same thing: should the blog live at blog.example.com or example.com/blog. It sounds like a five minute decision. It isn’t, and I say that as someone who has set this up wrong on a site I actually cared about and spent the better part of a quarter cleaning it up.
I run a portfolio of content sites. Some are on subfolders, a couple are on subdomains because of platform constraints I didn’t choose. I’ve watched both structures get crawled, indexed, and ranked over years, not weeks. This isn’t a theoretical post. It’s what I’d tell a friend before they commit to a structure they’ll be stuck with.
How search engines actually treat the two
Here’s the part that trips people up. A subfolder (example.com/blog) is unambiguously part of the same site in every technical sense: same domain, same cookies, same server in most setups, same root the browser bar shows. A subdomain (blog.example.com) is treated by most crawlers and by browsers as a distinct host. It can have its own robots.txt, its own hosting, its own SSL cert, and historically it’s been common for it to get its own separate entry in tools like Google Search Console unless you set up a domain property.
The practical consequence: authority and trust signals that build up on your main domain don’t move to a subdomain automatically the way they do within a subfolder. A subfolder blog inherits the same domain’s existing backlink profile, its existing crawl history, its existing trust. A subdomain starts closer to its own thing. It can still benefit from association with the root domain, but the connection is looser and less automatic. Search engines have gotten better over the years at treating well linked subdomains as part of the same entity, but “better” isn’t “identical,” and I wouldn’t build a strategy on an assumption I can’t verify.
I don’t know the exact internal logic Google runs to make that call, and anyone who tells you they do is guessing or selling something. What I do know, from watching my own sites’ crawl stats and index coverage over time, is that subfolder content tends to get picked up faster after publishing than content on a freshly stood up subdomain. That’s consistent with subfolders sharing an already-established crawl pattern instead of needing search engines to discover and build trust in a new host from scratch.
What a subfolder buys you
A subfolder blog rides on the coattails of everything your main site has already earned. If your homepage and product pages have backlinks, a reasonable crawl frequency, and some domain history, your blog posts inherit proximity to all of that the moment you publish. New posts tend to get crawled sooner. Internal linking from blog to product pages and back stays inside one site, which is simpler for both users and crawlers to follow.
It’s also just less to maintain. One SSL cert, one hosting environment in most cases, one analytics property if you set it up that way, one Search Console property. When something breaks, you’re debugging one site, not coordinating between two.
The tradeoff is that your blog and your core site now share the same technical foundation. If your CMS is slow, your blog is slow. If your main site has a redirect chain problem or a bloated theme, your blog inherits it too. And if you ever want to run the blog on different infrastructure, maybe a static site generator instead of whatever your ecommerce platform forces on you, a subfolder makes that harder because everything has to live under the same roof, often the same server or the same reverse proxy rules.
What a subdomain is actually for
Subdomains earn their keep when there’s a real technical or organizational reason to separate things, not because someone read that subfolders “always win.” I’ve used a subdomain when the main site runs on a platform that flatly can’t host a blog, think a locked down ecommerce SaaS that won’t let you touch the URL structure below the root. In that case blog.example.com isn’t a choice between two equal options, it’s the only way to have a blog at all without moving off the platform entirely.
It also makes sense when the content genuinely serves a different audience with different infrastructure needs: a support knowledge base that needs its own search index and permissions, a community forum running different software, a docs site built with a tool that only outputs to its own subdomain. In those cases the separation is honest. You’re not hiding a weaker asset next to a stronger one, you’re reflecting that these are actually different things technically.
Where it stops making sense is when someone picks a subdomain because “it looks cleaner” or because a developer set it up that way by default and nobody questioned it. That’s the version I’ve had to fix, and it’s never a quick fix.
The migration tax nobody budgets for
Moving a blog from a subdomain to a subfolder, or the other way, is a full domain migration in every way that matters. You need 301 redirects on every single URL, old to new, mapped one to one, not blanket redirected to a homepage. You need to update every internal link across your site that points to the old location. You need new XML sitemaps submitted, old ones retired properly. You need to watch Search Console for weeks afterward for crawl errors, soft 404s, and orphaned URLs that the redirect map missed.
Even done carefully, there’s usually a visibility dip while search engines reprocess the change. I’m not going to hand you a number of days or a percentage, because I don’t have a controlled experiment to point to and neither does anyone claiming otherwise. What I can tell you is that the dip is real often enough that I plan for it, and that the size of it seems to track with how much content is moving and how clean the redirect map is. A blog with forty posts and a tidy sitemap is a different job than one with four thousand posts, inconsistent URL patterns, and years of external links pointing at the old structure that you don’t fully control.
If you’re choosing a structure for a blog that doesn’t exist yet, this section is the whole argument for getting it right the first time. Migrations are doable. They’re also weeks of unglamorous redirect auditing that I’d rather spend writing content.
Cases where subdomain is the honest answer
I want to be fair to subdomains because the internet is full of posts that treat them as a mistake in every case, and that’s not true. If your blog is going to be run by a completely separate team on separate infrastructure, if it needs its own login system unrelated to your main app, or if your main domain is genuinely maxed out on what its platform allows, a subdomain is the honest structural choice. Pretending otherwise and forcing a subfolder onto infrastructure that can’t cleanly support it usually creates worse problems than the ones you were trying to avoid.
The mistake isn’t using a subdomain. The mistake is using one by default, or using one because a template shipped that way, without anyone asking whether a subfolder was actually available and just wasn’t considered.
What I’d tell you if you asked me directly
If you’re setting up a blog today on a domain you fully control, and there’s no technical wall stopping you, put it in a subfolder. It’s simpler to run, it shares in whatever trust your main domain has already built, and it avoids a migration you’ll eventually have to pay for in redirects and patience. That’s not a guarantee it’ll outrank anything. It’s a bet on fewer moving parts and faster crawling of new content, based on how I’ve seen my own sites get picked up.
If your platform won’t let you, or there’s a real organizational reason for separation, use the subdomain and don’t apologize for it. Just build internal links between the two properties deliberately instead of assuming search engines will connect the dots on their own.
A note on mixed signals
One thing I’d avoid regardless of structure: don’t run half your content strategy on a subfolder and half on a subdomain hoping to “diversify.” I’ve seen that pitched as a way to spread risk or grab more real estate in results. What it actually does is split your internal linking, split your crawl history, and make it harder to tell which half of your content is actually doing the work. Pick one structure for a given piece of content and be consistent about it.
If you’re stuck between the two for your own blog and want a second opinion before you build anything, The SEO Desk is where I write up exactly this kind of decision as it comes up in real sites.