Translate URLs and Slugs for SEO: Should You?

Translating URL slugs is optional: Google says localized words in URLs are fine and recommends using words in your audience's language, but it does not describe translated slugs as a ranking requirement. The practical benefit is readability and trust for local users, and the practical cost is more mapping work. Whichever policy you choose, apply it consistently, encode non-ASCII characters properly, and redirect any URL you change.

What does Google say about URLs in other languages?

Two Google documents cover this. The page on URL structure says to use words in the language of your audience, transliterated where applicable, so German-language searchers see German words in the URL and Japanese-language searchers see Japanese. It also recommends readable words over long ID numbers and hyphens over underscores, because hyphens clearly signal a word boundary.

The page on managing multilingual sites adds that it is fine to use localized words in the URL, and that you should use proper UTF-8 and escape URLs correctly when linking to them. Google also recommends a different URL for each language version.

Neither page claims that a translated slug improves rankings. Treat any such claim elsewhere as unverified unless it cites a source.

Should I translate slugs or keep them the same?

Common practice, not an official rule: pick the approach that your CMS and team can keep consistent. For small sites, identical slugs are a perfectly reasonable choice. For content-heavy sites where each language has its own audience, translated slugs usually feel more polished. The keyword research behind them matters more than the slug itself; see international keyword research.

How do I write good translated slugs?

• Use the words people in that market would use for the topic, not a literal conversion of the English slug.
• Keep them short, lowercase where the script has case, and separate words with hyphens.
• Avoid dates, IDs and parameters unless you need them for uniqueness.
• Do not stuff keywords. A slug should describe the page.
• Keep the language folder in the URL as well, so the pattern is /de/preise/, not /preise/. See subdirectories vs subdomains vs ccTLDs.
• Remember that Google treats URLs as case sensitive, so choose one casing convention and enforce it with redirects.

How are non-Latin characters in URLs encoded?

URLs formally allow a limited set of ASCII characters. Anything else is percent-encoded: each byte is written as % followed by two hexadecimal digits. MDN's glossary describes percent-encoding as a mechanism for encoding characters that have a special meaning or are outside the allowed set. For text characters, the bytes used are UTF-8 (the UTF-8 detail comes from how browsers and Google's examples behave rather than from the MDN glossary page).

Google's URL structure page shows the same idea: non-ASCII characters in links should be percent-encoded, and it contrasts gemüse with gem%C3%BCse, recommending the encoded form when linking.

Readable:  https://example.com/de/gemüse/
Encoded:   https://example.com/de/gem%C3%BCse/

What this means in practice:

• In the address bar, browsers typically show the readable form. When you copy the link into some apps, it can turn into the long encoded form. That is expected, not a bug.
• In HTML and sitemaps, write the encoded form in href, <loc> and hreflang values. This avoids ambiguity and matches Google's guidance.
• On the server, make sure your router decodes the path as UTF-8, and that it treats gem%C3%BCse and any other accepted form as the same resource. Pick one canonical form and redirect the rest.
• For domain names, Google says internationalized domain names are fine. Domains and paths are encoded differently (domains use a different ASCII form), so do not assume the same rules apply to both.
• Normalization. The same visible text can be stored as different Unicode sequences, such as a composed or decomposed accent. Whether this causes duplicate URLs depends on your stack; test it if your slugs contain accents.

Always test a sample of non-Latin URLs end to end: load them, copy and paste them into chat and email, and inspect them in Search Console.

What do I do when a slug changes?

Slugs change because of a rename, a better keyword choice or a retranslation. Handle each change as a URL migration:

1. Redirect old to new. Use a permanent server-side redirect, 301 or 308. Google's redirect documentation says these signal that the target should be canonical, and that the old URL becomes an alternate name that fades away over time.
2. Redirect directly to the final URL. Google's site move guidance advises pointing redirects directly at the destination rather than chaining them, and not sending many old URLs to one irrelevant page.
3. Keep the redirect in place. The same guidance says to keep redirects for as long as possible, generally at least one year.
4. Update everything that names the URL. Internal links, canonical tags, the language switcher, XML sitemaps and every hreflang set. The hreflang guide explains why one stale address breaks the reciprocal pair.
5. Redirect each language independently. A change to the German slug should not alter the French URL, but the French page's hreflang set must now point to the new German address.

Sample redirect rule for nginx (adapt to your server or CMS):

location = /de/preisliste/ {
  return 301 /de/preise/;
}

What mistakes should I avoid?

• Redirecting a renamed page to the homepage instead of the new equivalent page.
• Leaving old slugs in hreflang annotations or sitemaps.
• Letting the CMS auto-regenerate slugs from the title every time the title is edited.
• Translating slugs for some pages and not others without a rule.
• Serving the same content at both the encoded and decoded paths without a canonical.

Add these to your recurring checks from the multilingual SEO checklist.

Key takeaways

• Google says localized words in URLs are fine; there is no documented ranking requirement to translate them.
• Pick one slug policy (same, translated or transliterated) and apply it consistently.
• Percent-encode non-ASCII characters in links, sitemaps and hreflang, and test routing end to end.
• When a slug changes, use a direct permanent redirect, keep it long term (Google suggests at least a year) and update hreflang, canonicals and sitemaps.
• Hyphens, readable words and one casing convention keep URLs clean.

Sources

• URL structure best practices – language of URL words, percent-encoding, hyphens, case sensitivity
• Managing multi-regional and multilingual sites – localized words in URLs, separate URLs per language
• Redirects and Google Search – 301 and 308 as canonical signals
• Site moves with URL changes – redirect maps, direct redirects, duration
• Percent-encoding (MDN glossary) – definition of percent-encoding