Slugify

What is a slug and what makes a good one?

A slug is the human-readable part of a web address that names the page: in example.com/blog/url-slug-generator, the slug is url-slug-generator. It is written in lowercase, uses hyphens instead of spaces, and contains no accents, punctuation or spaces that a browser would have to escape.

A good one describes the page in a few words and survives being copied into a chat message, an email or a printed page without turning into %C3%A9%20 soup. That is the whole job.

Does a good slug actually help my ranking?

Not directly, and anyone telling you otherwise is overselling it. Google has said URL length is not a ranking factor, and URLs function mainly as identifiers.

What a good slug does help with is real, just less magical. It is visible in search results, so it influences whether someone clicks. It tells a reader what a link leads to before they open it. It survives being pasted anywhere. And it does not break when a system escapes it. Treat slugs as a clarity and click-through decision, not a ranking lever, and you will make better calls about them.

Hyphen or underscore?

Use a hyphen for anything public. This is the one slug decision with a concrete, documented consequence: Google treats a hyphen as a word separator and ignores underscores.

That means seo_slug is read as the single run-on token seoslug, while seo-slug is read as two words. The tool offers underscores because they are genuinely useful for internal keys, file names and database columns, but it warns you when you pick one, because on a public URL it quietly works against you.

How long should a slug be?

Most guides converge on three to five words, or roughly fifty to sixty characters. This page counts both as you type and tells you where you sit.

The length cut here always lands on a word boundary. Set a maximum of forty and the-complete-guide-to-building-a-headless-commerce becomes the-complete-guide-to-building-a, never ...building-a-headles. A slug with a word sliced in half is worse than a slug that is slightly too long, because it stops meaning anything.

Dropping stop words is the other lever. Words like the, and, of and to add characters without adding meaning, so the-best-way-to-learn-about-the-history-of-rome becomes best-way-learn-about-history-rome. The list used here is deliberately conservative, and if removing every word would empty the slug, the original is kept.

How does the accent and alphabet handling work?

In layers. Latin accents are decomposed with Unicode normalization and the marks are removed, so café becomes cafe and Pão becomes pao. Characters that do not decompose get explicit rules: ß to ss, æ to ae, ø to o, ł to l, þ to th.

Cyrillic and Greek are romanised before that, using a character table, so Привет мир becomes privet-mir, Москва becomes moskva and Καλημέρα κόσμε becomes kalimera-kosme.

One honest caveat on that table: it aims at readable ASCII, not at any certified transliteration standard. Several standards exist and they disagree with each other, and Russian and Ukrainian romanise the same Cyrillic letters differently. The table follows the more common Russian convention, so a Ukrainian name may come out spelled differently from its official romanisation. For a slug, readable and stable matters more than certified.

Why are Chinese, Arabic and Japanese left out?

Because they cannot be done with a character table, and doing them badly would be worse than not doing them.

Cyrillic and Greek map letter to letter, which is why a small table handles them cleanly. Chinese, Japanese and Arabic do not work that way: pronunciation depends on context, on word boundaries that are not marked, and on readings that a single character does not determine. Converting them properly takes a dictionary, which would be a large download for every visitor to this page.

So the tool says plainly that it found no URL-safe characters rather than inventing something. If you need slugs in those scripts, use a numeric identifier or a transliteration produced where the content lives.

Can I convert a whole list at once?

Yes. Switch to list mode and paste one title per line, up to three hundred, and every line is converted with the same settings and shown beside its source so you can check the mapping.

Repeated slugs are numbered automatically. Two posts called "Weekly Update" would otherwise produce the same slug and collide as routes, so the second becomes weekly-update-2. It is the kind of thing you only notice after a migration has already broken, which is why it happens here by default.

Everything runs locally in your browser as you type. Nothing is uploaded, logged or stored, and the tool keeps working offline once the page has loaded, so a list of unpublished post titles stays on your machine.