What is International SEO?

International SEO is the practice of optimising a website so that search engines serve the correct language or regional version of its content to users in different countries. Without it, Google may show a French user your English pages, or show a US user your UK pricing and delivery terms, because it has no reliable signal about which version is intended for whom.

The need for international SEO arises in two situations: when you have content in genuinely different languages (English and German, for example), and when you have the same language targeting different regions (English UK and English US, or Portuguese for Brazil and Portugal). Both require specific configuration to work correctly.

When do you actually need international SEO?

Not every site with visitors from multiple countries needs a full international SEO implementation. The threshold is whether you have distinct content variants intended for different audiences.

You need international SEO if:

• You have pages translated into one or more languages
• You have the same content adapted for different regional markets (currency, spelling, examples, offers)
• You are seeing Google index the wrong language version for users in certain countries
• You have separate country-specific domains or subdirectories that you want ranked in their respective countries

You do not necessarily need it if you have a single English site and happen to receive visitors from many countries. In that case, Google will determine the target audience from content signals and backlink geography.

What are the three URL structure options?

The first decision in an international SEO implementation is where to put each language or regional version. There are three approaches, each with different trade-offs.

Country-code top-level domains (ccTLDs)

Examples: example.fr, example.de, example.co.uk

A separate domain per country. ccTLDs send the strongest possible geotargeting signal to Google: a .fr domain is inherently associated with France. They also give users a clear trust signal that they are on the correct regional version.

The practical costs are significant. Each ccTLD is a separate domain: separate DNS configuration, separate Google Search Console property, separate link building (backlinks to example.fr do not benefit example.de), and separate hosting costs. Maintaining content parity across multiple domains at scale is operationally complex.

ccTLDs make sense for large organisations with dedicated regional teams and budgets. For most sites, the operational overhead outweighs the geotargeting benefit.

Subdomains

Examples: fr.example.com, de.example.com

A subdomain per language or region on your main domain. Subdomains are treated as distinct entities by Google, with their own Search Console property and their own link equity. They are easier to manage than ccTLDs (one hosting environment, one CMS) but share the same fundamental drawback: link equity does not consolidate onto the root domain.

Subdomains are a reasonable choice when technical constraints make subdirectories difficult (for example, different language versions on different platforms). They require correct canonical and hreflang configuration to function properly.

Subdirectories

Examples: example.com/fr/, example.com/de/

All language and regional versions sit under a single domain in distinct subdirectories. This is the simplest structure to manage and the one Google explicitly recommends for most sites. All link equity consolidates on the root domain: backlinks to example.com/fr/ benefit the authority of example.com as a whole. A single Search Console property covers all versions. Technical SEO is managed in one place.

The geotargeting signal is weaker than a ccTLD, but it is addressed through hreflang tags and Search Console geotargeting settings. For the vast majority of sites, subdirectories are the correct choice.

What are hreflang tags and how do they work?

Hreflang is the technical implementation of international SEO. It is a set of HTML link elements (or HTTP headers, or XML sitemap entries) that tell Google which page to serve to users in which language or region.

A basic hreflang implementation looks like this in the <head> of each page:

<link rel="alternate" hreflang="en-gb" href="https://example.com/en-gb/page/" />
<link rel="alternate" hreflang="en-us" href="https://example.com/en-us/page/" />
<link rel="alternate" hreflang="fr" href="https://example.com/fr/page/" />
<link rel="alternate" hreflang="x-default" href="https://example.com/page/" />

The hreflang attribute takes a language code (ISO 639-1) optionally combined with a region code (ISO 3166-1 Alpha-2). en-gb means English for UK users. en-us means English for US users. fr without a region code means French regardless of country.

The x-default attribute is the fallback. It specifies which URL to show when no other hreflang tag in the set matches the user's language and region. This is typically your default language homepage or a language selection page. Missing x-default is one of the most common hreflang errors and is flagged as a warning in most SEO crawlers.

The reciprocal requirement is the rule most sites get wrong. Every page in a hreflang group must reference every other page in that group, including itself. If your English UK page has a hreflang tag pointing to your French page, the French page must also have a hreflang tag pointing back to the English UK page. Hreflang relationships that are not reciprocal are ignored by Google.

This means that for a site with five language versions and 500 pages per version, every page carries five hreflang link elements. At scale, this is managed in your CMS or sitemap rather than manually.

For a complete implementation walkthrough, see the hreflang guide. If you need to generate hreflang tags, the hreflang generator tool produces the correct markup for any combination of languages and regions.

How does geotargeting in Search Console work?

For subdirectories and subdomains, you can set a target country in Google Search Console under Settings > International Targeting. This tells Google the intended audience for that property.

For subdirectories, each subdirectory is not a separate Search Console property, so geotargeting is set at root domain level and hreflang does the work of differentiating between regions. For subdomains, each subdomain is a separate property and can have its own geotargeting setting.

ccTLDs do not need a geotargeting setting: the domain extension itself is the signal.

Geotargeting in Search Console is a supporting signal, not a replacement for hreflang. Use both where applicable.

What is the difference between translation and localisation?

Translation converts words from one language to another. Localisation adapts content so it is genuinely useful and credible to users in a specific region. International SEO requires localisation, not just translation.

The distinction matters for search performance in two ways.

Machine translation alone produces thin content. A page translated by a tool without editorial review often reads unnaturally and lacks the specific idiom, cultural references, and phrasing that users in that market would expect. Google's quality systems are effective at detecting low-quality translated content. Pages produced at scale through machine translation without human review are candidates for ranking suppression. See what is thin content for how Google evaluates content quality.

Keyword research is market-specific. Users in different regions search for the same concepts differently. UK users search for "solicitor"; US users search for "attorney". French users searching for a CRM tool use different phrases than Belgian or Swiss French speakers. Effective international SEO requires keyword research conducted in each target market, not keyword translation from the source market.

Translated URLs also require attention: a URL slug translated word-for-word from English may not reflect how users in the target language actually search. Localise the slug alongside the content.

What are the common international SEO mistakes?

Hreflang on pages that are not actually translated. A common shortcut: applying hreflang tags across an entire site when only some pages have genuine language variants. Google ignores hreflang tags that point to content identical to another variant. Only implement hreflang where distinct language or regional versions genuinely exist.

Missing x-default tags. Every hreflang group should include an x-default entry. Without it, users whose language and region do not match any of the variants get no fallback signal. Google flags this as an error in Search Console. It is one of the most common findings in hreflang audits.

Hreflang pointing to redirecting or broken URLs. If a URL in a hreflang group returns a 301 redirect or a 404, Google cannot use it. The hreflang set becomes invalid for that page. Maintain hreflang URLs alongside your main URL inventory and audit them as part of any site crawl. See what is a canonical tag for how canonicals interact with hreflang.

All regional versions sharing one canonical pointing to the root. A mistaken implementation that appears frequently: each language version has a hreflang pointing to its own URL, but also a canonical tag pointing to the default language homepage. This signals to Google that all language versions are duplicates of the root. Google will ignore the hreflang tags and consolidate everything to the root URL, collapsing your international structure entirely. Each language version must have a self-referencing canonical, not a canonical pointing to another language variant.

Incorrect language codes. en-UK is not a valid hreflang value. The correct form is en-GB. Common errors include using the wrong case (hreflang values are case-insensitive in practice, but the ISO codes are upper case for regions), using country codes where language codes are expected, and using deprecated codes. Google's Search Console International Targeting report flags invalid codes.

No localisation of on-page metadata. Title tags, meta descriptions, and Open Graph tags also need to be localised. A French page with an English title tag sends a mixed signal and will underperform in French search results regardless of the body content.

How does international SEO interact with technical SEO?

The hreflang implementation, canonical configuration, and URL structure decisions in international SEO sit entirely within technical SEO. They are not visible to users but directly determine which version of your content Google indexes and ranks for which audience.

The most common technical failure is the canonical-hreflang conflict described above. The second most common is hreflang tags that are present in source code but broken by JavaScript rendering: if your hreflang tags are injected by JavaScript rather than present in the raw HTML, Google may not see them reliably. Serve hreflang in the <head> of the HTML or in your XML sitemap, not via client-side rendering.

For full hreflang implementation details, including how to structure your sitemap entries and validate your tags, see the hreflang guide.

---

Crawly audits hreflang tags across your entire site, flagging missing reciprocal tags, invalid language codes, and hreflang pointing to non-200 URLs. Run a crawl and you will have a complete picture of your international SEO health in minutes. Download Crawly to start your hreflang audit today.