Part of the Crawling & Indexing series. For the full picture, start with Crawling and indexing: how search engines find your pages, and what stops them.
One page can appear at /guides/shoes/, /guides/shoes/?utm_source=email, and /guides/shoes/?sort=price. Should Google crawl all three? There is no universal rule for a question mark in a URL. Classify what the parameter changes first; then decide whether that version deserves discovery, indexing, or neither.
How should you handle URL parameters for SEO?
Use the page’s purpose as the decision rule. Google’s URL structure guidance recommends using as few unnecessary parameters as possible and standard ?key=value&key2=value2 syntax. Its canonical guidance treats duplicate and very similar URLs differently from genuinely distinct pages.
| Parameter type | Example | Usually the right starting point |
|---|---|---|
| Tracking | ?utm_source=newsletter |
Point the variant to the clean URL with a canonical; link internally to the clean URL |
| Sort or view | ?sort=price, ?view=grid |
Canonical to the unsorted page when the primary content is essentially the same |
| Session or temporary ID | ?sid=123 |
Remove from public links; use cookies or another approach if possible |
| Pagination | ?page=2 |
Keep crawlable with its own URL when it exposes different items; see pagination SEO |
| Meaningful filter | ?brand=acme |
Decide whether the filtered result answers a distinct search; see faceted navigation |
| Internal search | ?q=running+shoes |
Usually keep result combinations out of Google’s crawl or index; assess separately from category pages |
These are starting points, not automatic rules. For example, a colour filter with strong demand and a useful product set may be worth a dedicated, internally linked landing page. A random combination of price, stock status and sort order probably is not.
How do parameter URLs become a crawl problem?
The combinations multiply. Ten filters with several values each can produce more URLs than products. Google’s faceted-navigation guidance says this can cause overcrawling and slower discovery of useful pages. Parameters used only to sort, track a campaign or maintain a session can create duplicates even on sites without filters.
This matters most when Google actually spends time fetching those variants or important pages are discovered late. Do not assume that every small site’s query strings are a crawl-budget crisis. Check the evidence: internal links, sitemap entries, crawl data, Search Console indexing examples and, on larger sites, server logs.
There is no longer a Search Console setting for each parameter: Google retired its URL Parameters tool in 2022. The decisions now belong in your URLs and site configuration.
How do you audit URL parameters?
- Inventory parameter keys. Crawl the site and group URLs by path and parameter name. Compare the number of variants with the number of useful base pages. Include links found in navigation, templates and XML sitemaps.
- Sample the actual content. For each key, compare the variant with its clean URL: main content, title, response code, canonical and indexability. A
?page=2variant is not equivalent to a?sort=pricevariant. - Check how variants are discovered. Remove tracking and session URLs from internal links and sitemaps. Link directly to the preferred version wherever possible. A canonical is a hint, not a substitute for clean linking.
- Compare your preference with Google’s choice. Inspect representative URLs in Search Console. If Google chooses a different canonical, investigate content differences and conflicting signals before changing a rule across the site.
- Test one parameter family before rollout. Re-crawl and check indexing over time. A broad wildcard robots rule can accidentally catch useful pages.
Crawlens can help inventory URL variants, their response codes, canonicals and internal links in a site crawl. Use its findings to pick examples for Search Console URL Inspection; a crawler’s own indexability judgment is not Google’s selected canonical.
Canonical, redirect, noindex or robots.txt?
Use a canonical when two URLs remain accessible to users but show duplicate or very similar content. Google can read it only after crawling the URL. Use a permanent redirect when the variant has no reason to exist for users. Use noindex when a page must remain accessible but should not appear in search, and allow crawling so the directive can be seen. Use robots.txt when you want to prevent crawling of a large, unneeded URL space; it is not a reliable way to remove a URL already known to Google from search. Google’s blocking guidance explains the distinction.
Do not combine a robots block with a canonical on the blocked URL and expect Google to see the canonical. Likewise, do not canonical a genuinely distinct paginated page to page one. If you plan to index selected filter pages, keep their URL order consistent, link to them, and return 404 for impossible or empty combinations as Google’s faceted-navigation documentation recommends.
URL parameter audit checklist
- Parameter keys grouped by purpose, not treated as one rule
- Important variants checked for distinct search intent and content
- Tracking and session variants removed from internal links and sitemaps
- Canonical targets return 200 and are not blocked
- Valuable filter and pagination URLs remain discoverable
- Large unneeded URL spaces controlled without blanket blocking
- Representative Google-selected canonicals checked after changes
The goal is not to eliminate ? from every URL. It is to give crawlers a consistent path to the pages people actually need.
Frequently asked questions
Are URL parameters bad for SEO?
No. Parameters can identify useful pages or harmless tracking variants. Problems arise when they create many duplicate, empty or near-identical crawlable URLs, or when a valuable variant is blocked by mistake.
Should I canonical every parameter URL to the base page?
Only when the content is duplicate or very similar. If a parameter creates a distinct page that should appear in search, a canonical to the base page sends the wrong preference.
Should I block parameter URLs in robots.txt?
Consider it for a large URL space that does not need crawling, such as unimportant filter combinations. Do not use it as a blanket rule: Google cannot read a canonical or noindex directive on a blocked URL.
Does Google Search Console still have a URL Parameters tool?
No. Google retired the URL Parameters tool in 2022. Manage URL behavior on your site with crawlable links, canonicals, redirects and, where appropriate, robots.txt.