Slug Generator
Convert any title or phrase into a clean URL slug instantly. SEO-friendly, with stop word removal, accent transliteration, and bulk mode. Free, no signup.
Slug Generator Tool
Read the slug before you use it. The tool keeps a-z, 0-9 and your separator and
discards everything else, so the output is made entirely of RFC 3986 unreserved characters and
never needs percent-encoding. The cost of that is silent loss: transliteration is a table of 36
Western European accented letters, not Unicode normalisation, so "Hà Nội" becomes
ha-n-i, "Łódź" becomes od and "naïve" becomes na-ve even
with the option on, while Chinese, Japanese, Thai, Khmer and Korean produce an empty slug.
Romanise first. Max length also cuts on character count, not at a word boundary.
SEO Preview
How to Use the Slug Generator
Type or paste your page title
Enter any title, heading, or phrase — the slug generates instantly as you type. Works with article titles, product names, category labels, or any text you need to turn into a URL path.
Choose separator and options
Hyphen is recommended by Google for SEO; enable stop word removal for cleaner slugs. Turn on accent transliteration if your title contains characters like ä, ñ, or é that need to be romanised.
Check the SEO preview
The preview box shows exactly how the URL will look in a browser address bar — so you can confirm it looks clean, readable, and professional before using it in your CMS.
Enable Bulk mode for multiple titles
Toggle Bulk mode on and paste a list of titles — one per line. All slugs are generated instantly and appear in the output field, ready to copy and paste into your spreadsheet or CMS.
URL Slugs and SEO — Why Your URL Structure Still Matters in 2026
What Makes a Good URL Slug for SEO
Google's documentation does not describe the slug as a ranking factor, and it promises nothing about click-through. What its URL-structure guidance asks for is legibility: URLs "constructed logically and in a manner that is most intelligible to humans", using "readable words rather than long ID numbers". A URL like recatools.com/slug-generator/ tells a reader what is behind it; recatools.com/?p=4281 tells the user nothing.
Google's URL-structure guidance is short and worth reading literally. It recommends hyphens over underscores "as it helps users and search engines better identify concepts in the URL", asks for readable words rather than long ID numbers, and says characters outside ASCII "should be percent encoded". It gives no length figure at all — the 60-character habit is a convention about what looks manageable in a browser bar and a search result, not a rule anyone publishes. Beyond that: put the distinctive word early, stay lowercase, and avoid dates unless the content really is dated, because a date in a path commits you to a redirect the day you refresh the piece. Pick one trailing-slash convention and keep it.
One nuance worth knowing: stop words are generally safe to remove from slugs because Google ignores them in keyword matching anyway. However, there are exceptions — "how-to" should keep its structure, brand names containing articles should be preserved, and sometimes a stop word genuinely changes meaning (contrast "a-frame-cabin" vs "frame-cabin"). This tool gives you full control over stop word removal so you can make the call for each slug.
Google publishes no maximum slug length. Every character count you have been given — including the ones on this page — is a convention about readability, not a limit.
Stop Words in URLs: When to Remove, When to Keep
Stop words are the small, common function words of a language — "the", "a", "an", "in", "of", "on", "at", "for", "with", "by" and so on. Search engines filter these out when processing keyword queries, which is why including them in a URL adds length without adding ranking value. The slug "best-restaurants-singapore" and "the-best-restaurants-in-singapore" communicate the same meaning to Google's indexer, but the second version is 20 characters longer and looks less clean in a browser bar.
That said, context matters. The phrase "how-to" is technically two stop words (how + to) but removing them would break the meaning of a how-to guide URL entirely. Brand names that contain articles — "A-Team", "A-Frame" — must be preserved. And occasionally the stop word itself is the keyword: a page specifically about the article "A" in English grammar needs to keep it. Use your judgment and this tool's instant preview to test before committing to a slug.
Multilingual Slugs: Handling ASEAN Languages in URLs
Be precise about what RFC 3986 actually permits, because "Unicode slugs are valid" is a half-truth. The RFC defines an unreserved set — letters, digits, hyphen, period, underscore and tilde — and says anything outside the allowed set is represented as a percent-encoded octet: "%" followed by two hexadecimal digits for each byte. A lowercase ASCII slug joined by hyphens or underscores is therefore made entirely of unreserved characters and never needs encoding anywhere. A Chinese title does: 美食 becomes %E7%BE%8E%E9%A3%9F, three octets per character, which is unreadable in a browser bar and fragile when pasted through chat apps and email clients that re-encode. Google's own guidance agrees — non-ASCII characters "should be percent encoded" — which is a reason to romanise before generating the slug, not after.
Bahasa Indonesia and Bahasa Melayu are particularly well-suited to slug creation because they use the standard Roman alphabet with consistent phonetics — a title like "Cara Membuat Nasi Lemak" becomes "cara-membuat-nasi-lemak" without any transliteration step. Where a site publishes the same page in several languages, a single romanised slug per page keeps every language version addressable by one shareable string — which is a consistency decision, not an SEO one.
Now the part that matters most before you paste something in. This tool's transliteration is a hand-written table of 36 Western European accented letters, not Unicode normalisation, and anything outside that table is discarded rather than converted — so it does not do the job this paragraph used to recommend it for. Measured: "Hà Nội" yields ha-n-i, not ha-noi; "Phở bò" yields ph-bo; "Łódź" yields od; "Ålesund" yields lesund; "naïve" yields na-ve even with transliteration on, because ï is missing from the table. Chinese, Japanese, Thai, Khmer and Korean produce an empty slug, since every character is stripped. Romanise first — pinyin for Chinese, RTGS for Thai, unaccented quốc ngữ for Vietnamese — and paste the romanised text here. Check the output before you publish it; a silently mangled slug is worse than no slug, because it goes live looking deliberate.
Bahasa Indonesia and Bahasa Melayu are the happy case: standard Roman alphabet, no diacritics, so a title converts cleanly with nothing to transliterate.
Hyphens, underscores, and the cost of changing a live URL
Google's wording on the hyphen is milder than the folklore: it recommends hyphens over underscores "as it helps users and search engines better identify concepts in the URL". Both are legal — RFC 3986 lists hyphen, period, underscore and tilde as unreserved.
Google's URL-structure page gives no maximum length, and no minimum. What it does ask for is readable words rather than long ID numbers — /wiki/Aviation over /index.php?topic=42&area=3a5ebc944f41daa6f849f730f1.
The riskiest use of this tool is regenerating a slug that is already live. A changed path is a new URL: old links break unless you redirect, and the redirect has to be maintained. Generate before publishing, not after.
Stop-word removal is on by default here and it is not always safe. "How to build a shed" becomes build-shed, losing the how-to framing; "A-Frame cabin" loses the A. The tool keeps the unfiltered words if removal would leave nothing, and that is the only guard it has.
Transliteration here is a hand-written table of 36 accented letters, not Unicode normalisation. Letters outside it are dropped, so "naïve" becomes na-ve even with the option switched on — ï is simply not in the table.
Measured Vietnamese: "Phở bò Hà Nội đặc biệt" comes out as ph-bo-ha-n-i-c-bi-t. Polish "Łódź" becomes od. Romanise before pasting, and read the output.
Chinese, Japanese, Thai, Khmer and Korean input produces an empty slug. The character filter keeps a-z and 0-9 only, so every character is removed and nothing is left to join.
RFC 3986 §2.1 encodes each out-of-set byte as "%" plus two hex digits. A UTF-8 Chinese character is three bytes, so one character becomes nine URL characters: 美 is %E7%BE%8E. Google says non-ASCII "should be percent encoded".
Indonesian and Malay are the clean case: standard Roman alphabet, no diacritics to lose, so a title converts here without needing the transliteration option at all.
The max-length option cuts with a plain character count and then trims a trailing separator — it does not stop at a word boundary. "Understanding distributed consensus algorithms" capped at 18 gives understanding-dist.
Frequently Asked Questions
-
A URL slug is the human-readable part of a web address that identifies a specific page. In the URL
recatools.com/slug-generator/, the slug is slug-generator. Good slugs are short, lowercase, use hyphens as separators, and include the page's primary keyword. They appear in browser address bars, search results, and shared links — making them important for both SEO and user experience. -
Hyphens, on Google's own recommendation: "We recommend using hyphens (-) instead of underscores (_) to separate words in your URLs, as it helps users and search engines better identify concepts in the URL." Note how modest that is next to the version you usually hear, which asserts that an underscore fuses two words into one indexable token. Google's published wording claims only that hyphens help identify concepts. Both characters are legal — RFC 3986 lists hyphen, period, underscore and tilde as unreserved, so neither ever needs percent-encoding — and most content systems default to hyphens. Use the hyphen, but use it because it is recommended and conventional, not because an underscore is broken.
-
Less than the effort usually spent on it. Google's documented position is about legibility rather than ranking: use readable words instead of long ID numbers, prefer hyphens to underscores so concepts are identifiable, percent-encode anything outside ASCII. The benefit that follows is a reader who can see what a link is before clicking it, and a URL that survives being pasted into a message intact. That is worth having and costs nothing at publication time — but it is a small effect, and rewriting a slug on a page that is already live costs more than it returns.
-
Stop words are common function words (the, a, an, in, of, on, at, for, with, by…) that Google filters out in keyword matching. Removing them from slugs shortens the URL without losing meaning — "best-restaurants-singapore" is cleaner than "the-best-restaurants-in-singapore". However, keep stop words when they change meaning: "how-to" guides need their structure, and brand names containing articles should be preserved. This tool lets you toggle stop word removal on or off.
-
Malay and Indonesian, yes — standard Roman alphabet, no diacritics, so "Cara Membuat Nasi Lemak" becomes
cara-membuat-nasi-lemakwith nothing to convert. Everything else in the region needs romanising first, and the tool will not tell you it failed. Vietnamese loses its vowels: "Hà Nội" comes outha-n-iand "Phở bò" comes outph-bo, because the transliteration table covers Western European accents only. Chinese, Thai, Khmer, Japanese and Korean produce an empty slug, every character having been stripped. Convert to pinyin, RTGS or unaccented quốc ngữ yourself, paste that, and read the result before you use it. -
Changing a URL causes a temporary ranking drop — even with a 301 redirect in place. Google takes time to discover the redirect, crawl the new URL, and transfer ranking signals. Expect 2–8 weeks for full recovery. Always implement a 301 redirect from the old URL to the new one, update internal links, resubmit the new URL in Google Search Console, and monitor rankings closely. Avoid changing URLs unless you have a strong SEO reason to do so.
-
Google's URL-structure documentation gives no length guidance at all — not a maximum, not a recommended range. The 60-character figure you will meet everywhere, this tool's own default assumptions included, is a convention about readability: a slug that short stays legible in a search result, survives a paste into a chat message and can be typed from memory. Treat it as a target to aim at, not a line to stay under, and prefer a longer slug that reads properly to a shorter one that has been cut into fragments. If you do use the max-length option, check the output: it cuts on character count and will truncate mid-word.
-
Only include dates when the content is genuinely time-sensitive and meant to expire — news articles, event listings, or annual reports. For evergreen content (how-to guides, tool pages, reference articles), avoid dates entirely. Date-based URLs age poorly: readers see "2021" in a URL and assume the content is outdated, reducing CTR. When you eventually update the content, you'll need a 301 redirect which temporarily hurts rankings. Timeless slugs are almost always the better choice.
-
Yes, with some notes. For Roman-alphabet languages (French, Spanish, Portuguese, German, Malay, Indonesian, Vietnamese with diacritics), enable the "Transliterate accents" option to convert characters like é, ñ, ü, and ç to their Roman equivalents. For non-Roman scripts (Chinese, Thai, Arabic, Korean), romanise or transliterate the text first before pasting it into this tool. The stop word removal list is English-only, so disable it when processing non-English titles.
-
A canonical URL is the definitive, preferred version of a page's URL — the one you tell Google to index when multiple URLs might serve the same content. Slugs are a key part of canonical URL design because an inconsistently named slug (with/without trailing slash, different capitalisation, with/without www) can create duplicate content issues. Always generate your slug once, set the canonical tag to that exact URL, use it consistently in internal links, and never change it without implementing a 301 redirect.
Method & sources
How it computes
Converts a title to a URL path segment in the browser: optionally maps 36 Western European accented letters to ASCII, lowercases, replaces every character outside a-z0-9 with a space, splits on whitespace, optionally drops English stop words (keeping the unfiltered words if that would empty the slug), joins with a hyphen or underscore, collapses repeats and trims the ends. The output is made only of RFC 3986 unreserved characters, so it never needs percent-encoding anywhere in a URL.
What this tool implements
- RFC 3986 (STD 66), §2.3: the unreserved set is ALPHA / DIGIT / '-' / '.' / '_' / '~'. Both offered separators are in it, so neither hyphen nor underscore ever requires encoding — the preference for hyphens is Google's recommendation and convention, not a syntax rule.
- RFC 3986 §2.1: anything outside the allowed set is percent-encoded as '%' plus two hex digits per octet, which is why a three-byte UTF-8 CJK character becomes nine URL characters. Stated on the page with a worked example.
- Google's URL-structure guidance is reported as written: hyphens recommended over underscores 'as it helps users and search engines better identify concepts in the URL'; readable words rather than long ID numbers; non-ASCII characters 'should be percent encoded'. It states no length limit, and the page now says so.
- Transliteration is a hand-written 36-entry table, NOT Unicode NFD normalisation with combining-mark removal. Every character outside the table is discarded, and the page states the measured consequences for Vietnamese, Polish, Turkish, Scandinavian and CJK input rather than leaving a reader to discover a mangled slug after publishing.
- Stop-word removal is skipped when it would leave an empty slug; that is the tool's only guard, and the page names the cases where removal changes meaning (how-to, A-Frame).
- Max length cuts on character count and trims a trailing separator only — it does not stop at a word boundary, and the page says so with a worked example.
Sources
- Berners-Lee T, Fielding R, Masinter L. Uniform Resource Identifier (URI): Generic Syntax. RFC 3986 / STD 66, Internet Engineering Task Force; January 2005 — §2.1 percent-encoding ('a character triplet, consisting of the percent character "%" followed by the two hexadecimal digits'), §2.3 unreserved characters ('unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~"'), §3.3 path segments. https://www.rfc-editor.org/rfc/rfc3986.txt
- Google Search Central. Keep a simple URL structure — 'We recommend using hyphens (-) instead of underscores (_) to separate words in your URLs, as it helps users and search engines better identify concepts in the URL'; 'When possible, use readable words rather than long ID numbers in your URLs'; 'Characters in the non-ASCII range should be percent encoded'. No URL length guidance is given. Last updated 2025-12-10 UTC; fetched 2026-09-13. https://developers.google.com/search/docs/crawling-indexing/url-s…
- The Unicode Consortium. Unicode Standard Annex #15: Unicode Normalization Forms — the NFD decomposition plus combining-mark removal that a general-purpose transliterator uses, and that this tool's 36-entry table substitutes for. https://www.unicode.org/reports/tr15/
What can make this go out of date
- none at runtime — the accent table and stop-word list ship with the tool. Google's URL-structure page is revised without notice (edition of 2025-12-10 read for this review).
Pick up where you left off
Stored only in this browser — never sent to our servers.