In short. URL parameters carry query values that can identify content or create duplicate addresses requiring consistent URL and crawl management.

Definition: URL parameters are key-value pairs commonly carried in a URL's query component, the part after the first ? and before any # fragment. Parameters can identify different content or create alternative URLs for the same content.

Key facts

RFC 3986, section 3.4 defines the query component as data that helps identify a resource together with the path.
As of 17 October 2026, Google's URL structure documentation recommends = between a parameter name and its value, and & between parameters.
Google's ecommerce URL guidance advises against repeating the same parameter because Googlebot may ignore one of its values.
Google's faceted navigation documentation explains that combinations of filtering parameters can produce an infinite space of URLs and consume crawling resources.

What does a query component look like?

RFC 3986 describes queries as non-hierarchical data and notes that they often carry key=value pairs. The first question mark introduces the query; a number sign begins a separate fragment component, or the URL ends without a fragment.

Google's faceted navigation example shows a query containing several filters:

https://example.com/items.shtm?products=fish&color=radioactive_green&size=tiny

Here, products, color, and size are parameter names. Their values select the items displayed on the page. Changing a value can therefore change the result set.

Google's URL structure guidance also permits commas to separate multiple values within a single parameter. That convention serves a different purpose from the ampersand separating parameters.

How do parameters create duplicate URL patterns?

Duplicate URLs are different addresses that return the same or very similar content. Google's canonical URL documentation uses a dress-page URL with gclid=ABCD as an example of an alternative address whose signals can be consolidated with a preferred URL.

Google's URL structure documentation identifies referral parameters, shopping sorting parameters and session IDs as causes of excessive URL counts. It recommends trimming unnecessary parameters that leave the content unchanged. Combined filters can also produce many slightly different lists, increasing bandwidth use and complicating complete indexing.

Google's ecommerce guidance adds that a continually changing timestamp can make a site appear to contain an infinite number of pages. It recommends persistent URLs and avoiding internal links containing temporary tracking codes or session IDs.

How should a website handle these URLs?

A canonical URL is the preferred representative of duplicate or very similar pages. Google's canonicalization guidance treats redirects and rel="canonical" annotations as strong signals of that preference. Internal links should consistently point to the canonical URL. Permanent redirects apply when a duplicate page is being retired.

For filtering systems, Google's faceted navigation guidance recommends deciding whether filtered URLs need to be eligible for indexing. When they do not, robots.txt, the file specifying crawler access rules, can prevent crawling of those URL patterns. Canonical annotations may reduce crawling over time, but the documentation describes them as less effective for crawl control.

FAQ

Should I remove every query parameter?

Google's ecommerce documentation accepts query parameters for identifying product variants, including /t-shirt?color=green. You can retain parameters that identify useful content while limiting alternative URLs that return the same page.

Can I repeat a parameter to supply several values?

The same ecommerce guidance recommends ?type=candy,sweet over ?type=candy&type=sweet, because Googlebot may ignore one repeated value. You can represent those values under a single parameter name.

Does blocking a parameter pattern select a canonical URL?

Google's canonicalization documentation says that URLs blocked in robots.txt may still be indexed without their content. You can indicate a preferred URL for duplicate content with rel="canonical".

Sources

Check how a page reads to answer engines. Open the GEO score analyzer