Encode and decode web addresses
Web addresses may only contain a limited set of characters. Spaces, accents and symbols such as &, ?, # and / have to be written as percent codes (a space becomes %20, an ampersand %26) when they are part of a search term, a file name or a parameter value. This is called URL encoding or percent-encoding. Paste plain text on the left to encode it, or an encoded string on the right to make it readable again.
Below the converter you can paste any link to see it taken apart: the site, the path, every parameter after the question mark, decoded and in a table. It is the quickest way to read a long tracking link, a search URL or a redirect, or to check what a link really contains before you click it. Nothing is uploaded.
Which mode do I need?
- A part of a link (the usual one) encodes everything that has a special meaning in a link, including / ? & = and #. Use it for a single value: a search term, a name, an email address or a file name that goes into a link.
- A whole link leaves the link's structure alone and only encodes spaces, accents and characters that are never allowed. Use it to tidy a complete address that contains spaces or non-English letters.
- Form data is the older style used when web forms are sent: spaces become + instead of %20. Decoding in this mode turns + back into spaces.
Common codes
- %20 space, %26 &, %2F /, %3F ?, %3D =, %23 #, %2B +, %25 %, %40 @, %3A :
- Letters outside English become several codes: é is %C3%A9, because it is two bytes in UTF-8, the standard for the web.
When you need it at work
- Building a link by hand, for example a search link or a link to a report with a filter, where a value contains spaces or an ampersand.
- Reading a link someone sent you that is full of %20 and %2F, or a long redirect link from an email, to see where it really goes.
- Spreadsheets and mail merges that build links from a column of names or product titles.
- Testing a web form or an API, where the server receives encoded values and you need to check them.
Good to know
- Double encoding: if you see %2520 in a link, a %20 was encoded a second time. Decode twice to get the original. The tool warns you when the decoded text still looks encoded.
- Broken codes: a lone % or a code that is not valid UTF-8 cannot be decoded; the tool says where the problem is.
- Tracking parameters such as utm_source and fbclid show up clearly in the link breakdown. To remove them in one go, use Clean Tracking Links.
- Encoding is not hiding: anyone can decode a link. It only makes the text safe to travel in an address.
How this tool was tested
Before publishing, I ran 30 automated checks on the finished page in Microsoft Edge, which is built on Chromium like Chrome. Last full run: 21 November 2026.
- Encoding: every special character when encoding part of a link, a whole link keeping its structure, + for spaces in form data, accents and emoji as UTF-8 codes, and the unreserved characters of RFC 3986 left as they are.
- Round trips: 400 random texts with accents, Chinese and emoji came back exactly in both styles. A lone % and codes that are not valid UTF-8 are refused, with the position given, and double encoding is noticed.
- Taking links apart: site, port, decoded path, every parameter in order (including empty and repeated ones), the part after #, a link without https:// and text that is not a link.
- Using the tool: encoding and decoding as you type, switching mode, the warnings and errors, and the link table showing values as text rather than HTML.
What the tests did not cover: Safari and Firefox were not tested. Something not working? Tell me on the contact page.
FAQ
Why does my link have %20 in it?
The original had a space, which links cannot contain. %20 is the space written in a way links accept; browsers show it as a space again.
Is a + in a link a space?
Only in the query part of links sent by forms. Elsewhere a + is a plus. Choose "Form data" to decode + as spaces.
Can I encode emoji?
Yes. Emoji become several percent codes and decode back exactly.