URL Encode Space: Six Search Sites Compared

A dated field check of how six site search boxes represent spaces in their query URLs.

Short answer: On October 6, 2026, six tested search boxes wrote spaces as + in their query parameters. YouTube also normalized a tested %20 query to +; the result is specific to these endpoints and this input.

On October 6, 2026, YouTube displayed the same search phrase, url encode space, for search_query=url%20encode%20space and search_query=url+encode+space, then normalized the %20 address to +.

All six site searches—Google, YouTube, Wikipedia, Amazon, GitHub, and eBay—serialized spaces as + in their query parameters on October 6, 2026. This is a dated six-site sample, not an internet-wide count.

That is what these specific search endpoints did during this check. It is not a rule for every URL, path, API, or website.

What appeared in each search URL

SiteSearch parameterAddress-bar form observedResult page
Googleqq=url+encode+spaceOpen the observed search
YouTubesearch_querysearch_query=url+encode+spaceOpen the observed search
Wikipediasearchsearch=url+encode+spaceOpen the observed search
Amazonkk=url+encode+spaceOpen the observed search
GitHubqq=url+encode+spaceOpen the observed repository search
eBay_nkw_nkw=url+encode+spaceOpen the observed search

Did %20 and + show different searches?

We opened each endpoint with the same words, changing only the two spaces:

Search endpointEntered with %20Entered with +Text shown by the site
Google qq=url%20encode%20spaceq=url+encode+spaceurl encode space for both
YouTube search_querysearch_query=url%20encode%20spacesearch_query=url+encode+spaceurl encode space for both; the %20 address was normalized to +
Wikipedia searchsearch=url%20encode%20spacesearch=url+encode+spaceSearch title showed url encode space for both

For this input on these endpoints, either spelling reached the same visible search words. The address-bar behavior was not identical: YouTube normalized %20, while Google and Wikipedia retained the submitted spelling.

What this check shows—and what it does not

How we checked

On October 6, 2026, we entered the exact phrase url encode space into each of the six sites' own search fields and recorded the resulting URL. For the paired check, we opened the same Google, YouTube, and Wikipedia search endpoints twice, once using %20 and once using +, then compared the final address and visible query text.

Reddit required login or a developer token before a search could be submitted, and Stack Overflow showed an anti-bot challenge, so neither was counted.

Download the dated observation CSV with the source URLs and paired test inputs. This small sample does not measure search demand, rankings, or behavior on every site.

For technical context, see the WHATWG form-urlencoded serializer and MDN's URLSearchParams percent-encoding reference.

Related tool: URL Encoder / Decoder — choose the encoding mode expected by the service receiving your URL.