# Subfolder vs subdomain > The evidence behind blog.domain.com versus domain.com/blog, what migrations actually produced, and when a subdomain is still the right call for docs. Source: https://saas-marketing.net/comparisons/subfolder-vs-subdomain/ Topic: SaaS SEO Type: comparison Published: 2026-09-11 Last updated: 2026-09-11 Publisher: SaaS Marketing (saas-marketing.net) License: CC BY 4.0. Quote or republish with attribution and a link to https://saas-marketing.net/comparisons/subfolder-vs-subdomain/ ## Short answer Google treats subdomains and subfolders as it sees fit, and John Mueller has repeatedly said either can work. In practice subfolders consolidate authority signals more reliably and are simpler to measure, so put the blog at domain.com/blog. A docs subdomain is defensible when your documentation platform cannot be proxied and the engineering cost of a reverse proxy exceeds the traffic at stake, which for most teams under 20,000 monthly docs sessions it does. ## Key takeaways - Put the blog in a subfolder. The evidence is not overwhelming but it is one-directional and the migration cost is low. - Google's public position is that both work. Its practical behaviour is that subfolders inherit site-level signals faster. - Reported migration gains of 30 to 100 percent are confounded by redesigns, content pruning and internal linking done at the same time. - A reverse proxy costs real engineering time forever, not once. Budget ongoing maintenance, not a one-week project. - Status pages and the app belong on subdomains. Nobody should proxy status.domain.com. - If docs already rank well on a subdomain and drive signups, migrating them is a bet with downside and modest upside. --- This argument has been running since 2012 and it still eats a week of engineering time at most Series A SaaS companies. Someone reads a case study about a 100 percent traffic lift, an engineer points out that the docs platform only supports a CNAME, and the meeting ends with nothing decided. Here's what the evidence actually supports, and where it runs out. ## What has Google actually said about subdomains? Google's public line, repeated by John Mueller across many Search Central office hours and social posts, is that Google can work with either and will figure out whether a subdomain belongs to the main site. That has been consistent for a decade. It is also not the same as saying the two are identical. The gap between the statement and the observed behaviour matters. Google saying "we can handle it" describes capability, not speed or reliability. What practitioners consistently see is that a brand new subdomain takes longer to earn crawl frequency and starts from a weaker position than a new folder on an established domain. That's not a contradiction of anything Google said. It's the practical texture underneath it. "Either works" and "they perform identically" are different claims. Google has made the first one repeatedly. It has never made the second one. The consolidation argument is the one worth holding onto. Internal links between a subfolder blog and your pricing page are ordinary internal links. Between a subdomain and the root domain they are cross-host links, and while Google may well treat them as internal, you have introduced uncertainty for no gain. Good [SaaS site architecture](/guides/saas-site-architecture/) removes uncertainty rather than adding it. ## What did real migrations produce, and what confounded them? The reported numbers are genuinely good and genuinely unreliable. Public migration write-ups from SaaS companies cluster around a 30 to 100 percent organic traffic increase within six months. Almost none of them isolate the variable. **30% to 100%** Reported organic growth range in the six months after a subdomain to subfolder move Think about what a team does during a migration. They audit every URL, which surfaces duplicate and thin pages nobody had looked at since 2022. They prune. They rebuild internal linking because they're touching the templates anyway. They fix the redirect chains that had been quietly bleeding equity. Then they attribute the result to the folder. | Change made during a typical migration | Plausible share of the gain | |---|---| | Subfolder consolidation itself | Real but modest | | Content pruning and consolidation | Often the largest single factor | | New internal linking between blog and product | Substantial, especially for commercial pages | | Template and Core Web Vitals improvements | Small to moderate | | Redirect chain cleanup | Small, occasionally large if chains were bad | None of this means don't migrate. It means set your expectations from the consolidation alone, then treat the pruning gains as what they are: available to you whether or not you move the folder. The sequencing advice in [SaaS website migration SEO](/playbooks/saas-website-migration-seo/) exists precisely so you can tell which change did what. Do the pruning and internal linking first, let it settle for eight weeks, then migrate. You will annoy everyone who wants it done in one sprint, and you will be the only team in your category that actually knows the answer. ## The engineering reality nobody puts in the SEO post Most SaaS companies do not choose a subdomain. They inherit one, because the tool they bought only offers one. That is the real constraint and SEO advice that ignores it is useless. Docs platforms, help desks and community tools ship with a CNAME setup by default. Getting them onto domain.com/docs means a reverse proxy: a rule at your CDN or edge layer that catches requests to a path and forwards them to the vendor's origin, rewriting headers and links on the way through. Cloudflare Workers, Vercel rewrites and nginx all do this well enough. **What a reverse proxy actually involves** Step five and six are the honest cost. A proxy is not a project, it's a permanent dependency between your edge config and a vendor's release schedule. Teams that skip the monitoring discover the breakage through a traffic drop, three weeks late. ## A decision table by surface Different surfaces, different answers. Treating your whole domain as one decision is the error. The status page row is the one people forget. If status.domain.com routes through the same edge config as your marketing site, an edge outage takes down the page customers use to find out about the outage. Keep it separate and ideally on a different provider. ## When a docs subdomain is the right call When the proxy cost exceeds the traffic at stake. That is the whole test, and it deserves an actual number rather than a shrug. Estimate the migration upside honestly: if docs currently pull 12,000 organic sessions a month and you think consolidation buys 20 percent, that's 2,400 incremental sessions. At a 1.5 percent docs-to-signup rate that's 36 signups a month. Now price the proxy: two engineering days to build, plus perhaps a day a quarter forever, plus the tail risk of a silent break. For most teams that ratio is fine above 20,000 sessions and marginal below it. A 40-person infrastructure SaaS had docs on a subdomain pulling strong rankings for API and integration queries built up over four years. The proxy was possible but their docs vendor shipped frontend changes weekly. They left it, spent the engineering time on internal linking from docs back to product pages instead, and got most of the benefit with none of the risk. That's the pattern to copy when you decide against migrating: capture the consolidation benefit you can get through linking. Cross-host internal links from docs to pricing, from blog to docs, and a shared navigation still do real work. The clustering approach in [SaaS topic clusters](/guides/saas-topic-clusters/) applies across hosts, it's just less tidy. ## How to migrate without losing a quarter If you are moving, the risk is entirely in execution. The folder choice cannot hurt you. A botched redirect map absolutely can. **Pre-migration checks** The seventh item is the one that saves your job. Without a baseline you cannot distinguish a migration problem from an algorithm update that happened the same week, and something always happens the same week. Pair this with the crawl setup in [technical SEO for SaaS](/guides/technical-seo-for-saas/) and the tooling notes in [technical SEO crawlers](/tools/technical-seo-crawlers/). Expect a dip. Two to six weeks of softness is normal, and the teams who panic and roll back at week three are the ones who end up with the worst outcome of all: two migrations and no gain. If you run [SaaS SEO experiments](/guides/saas-seo-experiments/) you will already have the patience and the measurement discipline for this. ## What to do this week If your blog is on a subdomain, put the migration on the roadmap but do the content pruning and internal linking first. You will get most of the traffic gain before you touch infrastructure, and you will know what the folder move actually contributed. If your docs are on a subdomain, calculate the incremental sessions and put a real engineering cost against them. If the ratio isn't obviously good, leave it and spend the time on [SaaS content refresh](/guides/saas-content-refresh/) work instead. And if you're building something new, start in a subfolder. It costs nothing on day one and saves an argument two years from now. More on the wider programme at [SaaS SEO](/saas-seo/), and the enterprise-scale version of this problem in [enterprise SaaS SEO](/playbooks/enterprise-saas-seo/). ## Frequently asked questions ### Does Google treat subdomains differently from subfolders? Google has said for years that it can handle both and will work out whether a subdomain is part of the main site. In observed practice, subfolders inherit domain-level authority and crawl priority more quickly, and reporting is simpler. Treat the difference as real but smaller than most SEO blog posts claim. ### Should my SaaS blog be on a subdomain or subfolder? Subfolder, at domain.com/blog. It consolidates link equity with your commercial pages, keeps internal linking between blog and product pages trivially simple, and puts everything in one Search Console property. The only good reason to use a subdomain is a CMS that genuinely cannot be proxied and an engineering team that will not build the proxy. ### What did companies gain by moving a blog from subdomain to subfolder? Publicly reported gains cluster between 30 and 100 percent organic traffic growth in the six months after migration. Nearly all of those migrations also involved content pruning, redesigns and new internal linking, so the subfolder change is one variable among several. Expect a gain, but do not budget on the headline numbers. ### Is a reverse proxy worth it for a docs site? It depends on the traffic. If docs drive 50,000 monthly sessions and influence signups, a proxy pays for itself. Under roughly 20,000 sessions, the ongoing maintenance burden, caching bugs, TLS renewal and the on-call risk usually exceed the SEO gain. Be honest about who maintains it in year three. ### Where should the help centre live? Subfolder if your help desk supports custom domains with path-based routing, which Zendesk and Intercom do to varying degrees. If it only offers help.domain.com, leave it there. Help centre content mostly serves existing customers, so its organic contribution to acquisition is usually small enough not to justify infrastructure work. ### Does a subdomain hurt if it has its own strong backlinks? Less than you might think. A subdomain with genuine independent authority, like a widely linked research hub, loses little by staying separate. The cost is opportunity: those links are not reinforcing your commercial pages. Weigh that against migration risk before moving anything that already performs.