Subfolders, Subdomains, or Country Domains: Which Should You Pick?
Subfolders, Subdomains, or Country Domains: Which Should You Pick?
For most B2B companies, subdirectories on your main domain. They are the cheapest to set up, the easiest to maintain, and they keep everything on one host. Country-specific domains give a stronger geotargeting signal, but they cost more and each one can only target a single country.
That is the short answer. The longer answer depends on whether you are selling in different languages, different countries, or both, and those are not the same problem.
We get this question from founders who have just closed their first customer in another market and suddenly need a German page, or a UK pricing page, or both. The decision is architectural, it is hard to reverse, and most of the advice online skips the part where you decide what you are actually targeting.
What Is the Difference Between Multilingual and Multi-Regional?
Google draws the line clearly. A multilingual site "offers content in more than one language." A multi-regional site "explicitly targets users in different countries." You can be one, the other, or both, and the answer changes your architecture.
Google's own examples are useful. A Canadian business with English and French versions is multilingual. A manufacturer shipping to both Canada and the United States is multi-regional. A site with separate USA and Canada versions, plus French and English Canadian content, is both.
Most early-stage B2B SaaS companies think they have a language problem when they have a currency and compliance problem. If your product is sold in English to Dutch buyers, you may need a Netherlands page with local pricing and a local case study, not a Dutch translation.
What Are the Real Options, According to Google?
Google publishes a direct comparison of URL structures in its multi-regional documentation, with the pros and cons for each. Here it is, using Google's own points.
| Structure | Example | Google's pros | Google's cons |
|---|---|---|---|
| Country-specific domain | example.de | Clear geotargeting, server location irrelevant, easy separation of sites | Expensive and can have limited availability, requires more infrastructure, strict ccTLD requirements sometimes, can only target a single country |
| Subdomain with gTLD | de.example.com | Easy to set up, allows different server locations, easy separation of sites | Users might not recognise geotargeting from the URL alone |
| Subdirectory with gTLD | example.com/de/ | Easy to set up, low maintenance on the same host | Users might not recognise geotargeting from the URL alone, single server location, separation of sites harder |
| URL parameters | site.com?loc=de | None listed | Not recommended: segmentation difficult, users might not recognise geotargeting |
Note what is missing from that table: any claim that one option ranks better than another. Google does not say subdirectories inherit authority or that ccTLDs rank worse. The trade-offs listed are about targeting clarity, cost and operational complexity.
Why Do People Say Subfolders Beat Country Domains?
The popular argument is that a subdirectory inherits your main domain's authority while a new country domain starts from zero. That is a reasonable intuition, and Google has never published it as a ranking rule. What Google does publish is that subdirectories are "easy to set up" with "low maintenance" and that country domains are "expensive" and "require more infrastructure."
So the strongest honest case for subdirectories is not a ranking claim. It is that they are cheaper, faster, and simpler to run, and that for a company with one marketing team those three things decide whether the localised pages ever get updated at all. A German subfolder that gets maintained beats a German ccTLD that goes stale in eight months.
Our own view, from building these: the operational argument wins on its own. You do not need to believe anything about link equity to pick subdirectories for a 60 page B2B site.
When Is a Country Domain Actually Worth It?
When the country is a strategic market, not an experiment. Google says ccTLDs are "tied to a specific country" and "provide a strong signal to both users and search engines that your site is explicitly intended for a certain country." That is real value, and it is the only option in the table Google describes as giving clear geotargeting.
There is a trust dimension too, which is not an SEO argument at all. In some markets buyers check the domain. A local domain can read as a local company with local support, which matters in procurement.
The catches are documented. Some countries restrict who may register their ccTLD, so you may need a local entity. And Google notes it treats some vanity ccTLDs, such as .tv and .me, as generic top-level domains rather than country-targeted, so a clever domain hack gives you no geotargeting at all.
How Does Google Decide Which Country a Page Is For?
It combines signals rather than reading one. Google lists country-code top-level domains, hreflang statements in tags, headers or sitemaps, server location via the server's IP address, and other signals including local addresses and phone numbers, local language and currency, links from other local sites, and Business Profile signals where available.
Server location is explicitly described as not definitive, because sites use content delivery networks or host where the infrastructure is better. So if you are on a global CDN, which almost every modern site is, that signal is close to noise.
The practical takeaway is that your URL structure is one input among several. Local currency, a local phone number, a local address and local case studies on the page do real work. We covered the annotation side of this in our hreflang guide.
What Does Google Say You Should Not Do?
Three things, each stated plainly. Do not use IP analysis to adapt your content: Google says "IP location analysis is difficult and generally not reliable" and that it may then fail to crawl your variations properly. Do not rely on cookies or browser settings to switch the content language, because Googlebot "usually originates from the USA" and "sends HTTP requests without setting Accept-Language in the request header."
And do not automatically redirect users between language versions. Google's guidance is to avoid redirecting based on what you think the user's language may be, and to let the user switch instead.
That last one is the most commonly broken rule we see. A site detects a German IP address and force-redirects to the German page, so a German-speaking buyer who wants the English documentation cannot reach it, and Googlebot, crawling from the US, never sees the German page exists.
Do You Need Separate Pages at All?
Only if the content genuinely differs. Google says localized versions "are only considered duplicates if the main content of the page remains untranslated." So an English page for the US and a near-identical English page for Ireland is fine if you annotate them, and pointless if nothing on it changes.
Google also recommends a catchall. If you have specific URLs for English speakers in Ireland, Canada and Australia, provide a generic English page for searchers everywhere else. That generic page can be one of the specific ones.
Before you build five variants, write down what actually differs on each: price, currency, compliance claims, case studies, contact details, legal terms. If the list is short, you may need one page with a currency switcher, not five pages. We wrote about the design side of this in designing for five languages.
What Breaks Most Often in Practice?
Incomplete annotation. Google's rule is that each language version "must list itself as well as all other language versions," alternate URLs "must be fully-qualified, including the transport method," and if two pages do not both point to each other, the tags are ignored. That last clause exists so that another site cannot claim to be your alternate version.
The other common failure is choosing three methods at once. Google supports hreflang in HTML, in HTTP headers, and in sitemaps, calls the three "equivalent," and says explicitly that there is no benefit in Search to using all three, and that managing three implementations is much harder than picking one.
Pick one. On a CMS-driven marketing site, sitemap-level annotation is usually the one that survives contact with a marketing team, because nobody has to remember to edit a template.
So What Should You Actually Do First?
Start with the smallest change that can be validated. Put the new market on a subdirectory, annotate it properly with one method, give it real local content rather than a machine translation of your homepage, and watch whether it earns impressions in that country. If it does, and the market becomes core, you can make the case for a country domain later with evidence instead of a guess.
The mistake we see most is the reverse order: buying four country domains during a funding round, building thin versions of the same site on each, and then maintaining none of them. Google's documentation describes country domains as expensive and infrastructure-heavy for a reason. That cost lands on the team that has to keep them current. The strategy question behind all of this is covered in taking a product into a new geography.
If you are weighing this up for a real site and want a second opinion before you commit, we are happy to walk through it with you at phoenix.studio.
Want a site that performs like this?
Tell us about your project. We will come back with a clear next step, no pressure.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Have a project like this?
Tell us where you want to go. We'll tell you how we'd get you there.