Question 1

Why hyphens rather than underscores?

Accepted Answer

Google's documentation on URL structure recommends hyphens, because they are read as word separators while underscores have historically joined words into a single token. Both are legal in a path (RFC 3986 lists hyphen, underscore, period and tilde as unreserved characters), so this is about how the words are read, not whether the URL works. Underscores remain the better choice for filenames and asset paths, which is why the option is here.

Question 2

What happens to accented and non-Latin characters?

Accepted Answer

With transliteration on, accents are decomposed and dropped, so é becomes e and ș becomes s. Letters that have no decomposition are looked up instead: ß becomes ss, ø becomes o, đ becomes d. Cyrillic and Greek are romanised through tables, so Привет becomes privet and Ελλάδα becomes ellada. Scripts with no Latin equivalent at all, such as Chinese or Japanese, and emoji are dropped. Switch the option off to keep the original characters, which are valid in a URL but arrive percent-encoded when someone copies the link.

Question 3

Why is & turned into the word and?

Accepted Answer

Because deleting it changes what the URL says. "R&D Budget" would slug to r-d-budget, which reads as nothing. The same applies to %, +, @, # and the currency symbols, each of which is spelled out. It matters beyond readability for & in particular: it is the reserved character that separates query parameters, so it can never appear raw in a path.

Question 4

Can I convert a whole list of titles at once?

Accepted Answer

Yes. Every non-empty line is slugged on its own and comes back in the same order, so a column of post titles pasted from a spreadsheet returns a matching column. Because two different titles often slug identically ("Part 1: Setup" and "Part 1 - Setup" both give part-1-setup), repeats are numbered part-1-setup-2, part-1-setup-3 and so on, which is what WordPress, Ghost and most other CMSs do with a collision.

Question 5

Should I remove stop words?

Accepted Answer

Only when the slug is genuinely long. Dropping them shortens the path, but it also makes the URL read less like a sentence, and Google has said repeatedly that words in a URL are a very small ranking factor. Question words are never removed even with the option on: taking "how" out of how-to-proof-dough leaves a slug that no longer says what the page is.

Question 6

How long should a slug be?

Accepted Answer

Three to five words is the usual advice, and it holds up for a practical reason rather than a ranking one: URLs get truncated in search results, in chat previews and in analytics reports, and the first few words are what survive. The length limit cuts back to the last whole word rather than stopping mid-word, unless the first word on its own is already over the limit.

tooldura

URL Slug Generator

About URL Slug Generator

How to use URL Slug Generator

Frequently Asked Questions