SEO · Guide

Technical SEO for a multilingual website: a practical checklist

Localized pages need more than translated words. Clear language URLs, self-canonicals, equivalent hreflang links and crawlable navigation help people and search engines understand them.

Original diagram connecting equivalent English, Italian, French and German pages through language annotations.
Original diagram connecting equivalent English, Italian, French and German pages through language annotations.Original CherTra News illustration

Give each language a stable URL

A translated page should have a predictable, crawlable URL that identifies its language version. A language subdirectory is one common choice, though a site can use other consistent structures. Avoid relying only on a browser-language redirect or a client-side selector that hides the alternative pages from crawlers.

Every version should show its actual language in the document’s lang attribute and use a translated title and description that accurately describe that page. A translation should read naturally for its audience, not merely repeat the source-language phrase structure.

Use canonicals and hreflang for different jobs

A canonical points to the preferred URL for that page’s content. A translated page with distinct useful content generally needs its own self-referencing canonical; it should not be canonicalized to the English version just because that version is the default.

Hreflang annotations identify equivalent language or regional versions. Each page should link to itself and the true equivalents, and those pages should reciprocate. Do not link pages with materially different content as if they were translations, and do not publish hreflang references to routes that return errors.

Discovery, indexing, and ranking are different stages

A simplified search journey. Crawling makes a page available for processing; indexing is a separate decision, and ranking depends on the query and many signals.

  1. Link discovered

    A crawler finds a URL through links or a sitemap.

  2. Crawl request

    The crawler requests the page and checks access rules.

  3. Page processed

    Content and relevant page signals are evaluated.

  4. Indexing decision

    The search engine decides whether and how to index it.

  5. Search result

    If indexed, the page may be considered for relevant queries.

Localize the whole journey

Navigation labels, forms, validation messages, contact options and important calls to action should match the page language. Keep a clear way to switch language without forcing a user to restart a task. When an exact equivalent does not exist, the switcher should not imply that one does.

Review metadata, structured data language, social-preview text and image descriptions. A page with translated body copy but English-only title and navigation can create a confusing result in both search and the user experience.

Validate the site as a set of language relationships

Generate an inventory of canonical URLs and their alternates. Check that every indexable language page appears in the sitemap, has the expected HTML lang, a unique title and description, and reciprocal hreflang references. Crawl links to find localized routes that accidentally return a 404 or redirect to the default language.

Only translate pages when the content can be kept accurate. For time-sensitive articles or complex legal material, publishing fewer complete localized versions is better than creating empty or machine-placeholder pages that imply a translation exists.

Sources & further reading

Have a factual correction or a source to suggest? Contact the editorial desk.