Chinese URL Slugs: Pinyin, Chinese Characters or English?

Google Search Central's URL structure guide recommends using words in your audience's language in the URL, and transliterated words where applicable. It even gives a Chinese example. In links, use the percent-encoded form https://example.com/%E6%9D%82%E8%B4%A7/%E8%96%84%E8%8D%B7; the raw form https://example.com/杂货/薄荷 ("groceries/mint") is listed as not recommended. In other words, Chinese URLs are fine, but the href in your HTML should be encoded.

The same page recommends hyphens (-) rather than underscores (_) between words, and warns that URLs are case-sensitive: Google treats /APPLE and /apple as different URLs.

Google's guide to multi-regional and multilingual sites adds three points. Use a different URL for each language version. Localized words in URLs are fine, but use UTF-8 and escape URLs properly when you link to them. And don't redirect users automatically to another language version based on what you guess their language is.

RFC 3986 defines the "unreserved" characters that never need encoding: ASCII letters, digits and the four symbols -, ., _ and ~. Everything else, including every Chinese character, is first converted to UTF-8 bytes and then written as %XX. Hex case does not matter (%E9 and %e9 are the same byte), but the standard says producers should use uppercase. RFC 3987 defines IRIs (Internationalized Resource Identifiers), which may show Chinese characters directly and map back to ordinary URIs with the same rules. That is why the address bar shows Chinese while a copied link is encoded.

For the title 重庆火锅攻略 ("Chongqing hot pot guide"):

Chinese-character slugs are instantly clear to Chinese readers and match the title exactly. They are also the longest once encoded, look like garbage wherever IRIs are not displayed, and are hard to type. Some platforms cap the encoded length: WordPress keeps 200 characters, which is at most 22 Chinese characters, and silently cuts the rest. See our WordPress permalink guide.

Pinyin slugs are pure ASCII and medium length, so every system keeps them as they are, and anyone who reads pinyin understands them. The downsides: Chinese has many homophones and pinyin slugs carry no tones, so a few are ambiguous, and automatic conversion stumbles on polyphonic characters.

English slugs are the shortest and readable internationally. But they need a human translation, they don't match the Chinese title, they are less obvious to Chinese readers at a glance, and they can't be fully automated in bulk.

A common Chinese character takes 3 bytes in UTF-8, and each encoded byte is 3 characters, so one Chinese character = 9 characters. Pinyin needs roughly 3–5 letters per character, plus a hyphen between syllables:

For the nine-character title, the encoded Chinese slug is 81 characters and the pinyin slug 32. Length does not directly affect ranking, but long URLs are harder to read in search results, share cards and logs; see our URL slug length guide. If the pinyin is too long, cut it at a word boundary: with the maximum length set to 24, the generator turns 如何选择笔记本电脑 into ru-he-xuan-ze-bi-ji-ben and never splits a syllable.

One trap is easy to miss: if a system keeps only Latin letters and lacks a pinyin step, a pure-Chinese title gets an empty slug, and 2026年春节放假安排 keeps only 2026. Test your real titles before launch.

Pinyin in URLs usually follows these conventions, and our generator applies them:

1. No tone marks. Tone marks are non-ASCII and get encoded; numbered tones such as zhong4 are hard to read.
2. Write ü as v. 绿茶 becomes lv-cha and 女装 ("women's clothing") becomes nv-zhuang. This is the convention of pinyin keyboards, and it keeps lu and lü apart.
3. Put a hyphen between syllables. Run together, syllables can be ambiguous: 西安 (Xi'an) written as xian looks like the single syllable xian; xi-an is clear.
4. Keep digits and English as they are. 重庆火锅攻略2026 becomes chong-qing-huo-guo-gong-lve-2026.
5. Use lowercase only. URLs are case-sensitive, so one case avoids duplicate URLs.

Should you join syllables into words instead, as in chongqing-huoguo? That is closer to the official pinyin orthography and reads more like words, but it needs accurate word segmentation, which is harder to automate. If you write each slug by hand, joining by word is fine; for bulk generation, one hyphen per syllable is more consistent and easier to proofread.

Many Chinese characters have more than one reading; 行, 长, 重 and 还 are common examples. Our engine is based on pinyin-pro, which picks readings from whole words, and gets most common words right:

The last two rows are real engine output, shown as they are: without context, any automatic converter can misread. In practice:

• Check the pinyin of important pages by hand, especially when the title contains 行, 长, 重, 还, 乐, 发, 调, 着 or 了.
• Fix wrong readings in the slug field. Don't keep a wrong reading just to stay "automatic".
• Don't change a slug lightly once it is published. If you must, redirect the old URL with the 301 redirect generator.

When the same content exists in Chinese and English, a common pattern is a language subdirectory: the English page at /guide/chinese-url-slug and the Chinese page at /zh/guide/chinese-url-slug. Using the same slug in both versions means:

• you can see at a glance that the two pages are translations of each other;
• hreflang annotations and sitemaps are easier to generate and harder to get wrong;
• switching language is just adding or removing /zh.

This site works exactly that way: every page links to its other language version with hreflang in the <head>, and nothing redirects by browser language, as Google recommends. A Chinese-only site does not need English slugs; pinyin or Chinese characters are both fine. For other languages, see how to transliterate slugs.

[ ] One slug style for the whole site: pinyin / Chinese / English
[ ] Pinyin without tones, ü written as v, hyphens between syllables
[ ] Chinese slugs kept short (9 encoded characters per character)
[ ] Non-ASCII characters percent-encoded in HTML links
[ ] No empty or number-only slugs
[ ] Pinyin checked by hand for titles with polyphonic characters
[ ] Each language version has its own URL, with hreflang
[ ] No automatic redirect by browser language
[ ] 301 redirects in place when a published slug changes

Paste a Chinese title into the URL slug generator and turn on Chinese → pinyin (it is on by default on the Chinese pages); you can also set a maximum length and remove stop words. To keep the characters, turn off Smart transliteration (Keep Unicode). To see what a Chinese slug looks like encoded, use the URL encoder/decoder. Convert a whole list of titles with the bulk CSV slug generator, and check length and encoding before you publish with the slug checker.

• Google Search Central: URL structure best practices for Google Search
• Google Search Central: Managing multi-regional and multilingual sites
• IETF: RFC 3986, Uniform Resource Identifier (URI): Generic Syntax, sections 2.1 (percent-encoding) and 2.3 (unreserved characters)
• IETF: RFC 3987, Internationalized Resource Identifiers (IRIs), section 3.1 (mapping IRIs to URIs)
• pinyin-pro: project page and documentation (MIT License)