Back to the Campfire Blog
What Happens When A Bunch of Random Special Characters Sneak Into Your URLs?
Written by: Adrienne Kmetz Tags: technical-seo
Published: Jun 28, 2025
A community member inherited some URLs with a few special characters like commas, parentheses, carrots, asterisks, hashes, and question marks.
What happens to a URL when special characters are inserted into a permalink?
Depending on the character, engines and browsers may treat it differently. Google explains in its guide to URL encoding: "Some characters cannot be part of a URL (for example, the space) and some other characters have a special meaning in a URL."
For example, John Mueller said, the "URLs with "#" in them probably have gotten cut off - the "#" & following is not considered a part of the indexable URL."

To avoid URLs getting cut off or not indexing at all, characters like < > carrots, | pipe characters, % percent signs and #, must be encoded "as HTML variants" using corresponding codes.

Tory Gray, owner of the technical seo agency Gray Dot Company, sheds more detailed light for us on what this means. She explains the impact:
"Using “OK characters” [which are the familiar alphanumeric symbols and a handful of reserved characters] means the URL can be crawled and indexed as is.
Caveat being, a # symbol – if it’s a jump link or internal anchor link it’s PART of a page and not a unique page that’s separately indexable.
Versus a special character that’s not reserved, like a space, would need encoding:
https://thegray.company/services/name this <- non-encoded space
https://thegray.company/services/name%20this <- encoded variant (%20 is a space in HTML syntax)
https://thegray.company/services/name-this <- dashes are different, too, and what I recommend using instead of spaces, to avoid a duplication issue (i.e. it looks like 2 different URLs serving the same content).
Why it matters: Duplicate URLs split link equity
Internal and external links can go to both the encoded and non-encoding variant, splitting the link equity between them:
- Typically the encoded version is what gets indexed
- But typically the non-encoded version is what gets linked to, because that’s what’s visible on a website (or, both get links)
- Resulting in the relative under performance of the indexed URL – it’s not getting the full equity
A canonical tag pointing to the correct variant can solve the issue
...But I prefer updating the URL so that only one exists, & ensuring that internal linking matches that - thus consolidating the equity on a single URL that is therefore more “powerful” and more likely of ranking effectively."
Thank you for this expert answer,
Tory Gray!