The URL Parameters tool in Google Search Console no longer exists. Google announced its retirement in March 2022 and switched it off on April 26, 2022, saying that only about 1% of the parameter configurations advertisers and site owners had entered were actually useful for crawling. If you are looking for it in Search Console today, stop looking: it is gone, and there is no replacement tool.
The tool’s disappearance is genuinely good news for most sites, because it was easy to misuse and its worst failure mode was silently removing large sections of a site from Google. What follows is what actually replaced it, in the order you should apply the techniques.
What the tool did and why Google killed it
The URL Parameters tool let you tell Google how to treat specific query string parameters. You could declare that sessionid did not change page content, or that sort merely reordered items, and Google would use that hint to decide which URLs to crawl.
Google’s stated reasoning in the official Search Central announcement was blunt: the vast majority of configurations were not useful, crawlers had gotten much better at working out parameter behavior on their own, and most modern content management systems produce reasonable URL structures without help.
What Google did not emphasize, but every technical SEO who used the tool knows, is the other reason: it was dangerous. A misconfigured entry could tell Google that a parameter was irrelevant when it was the parameter that selected the product. Sites deindexed entire catalogs this way, and the failure was invisible until traffic collapsed.
First, work out whether you have a problem
Most sites do not. Before engineering anything, check whether parameter URLs are actually consuming crawl budget or splitting signals.
In Search Console, open Settings, then Crawl stats. Look at total crawl requests over time and at the file type and response code breakdowns. Then open Indexing, Pages and read the excluded reasons. Two categories matter here: Duplicate without user selected canonical and Alternate page with proper canonical tag. Large numbers in the first are a real signal.
Server logs are better still. If you can get them, count how many crawler requests hit URLs containing a question mark:
# Share of Googlebot requests going to parameter URLs
grep -i googlebot access.log \
| awk '{print $7}' \
| awk -F'?' '{ if (NF > 1) p++; else c++ } END { printf "clean %d param %d (%.1f%% param)\n", c, p, 100*p/(c+p) }'
# Which parameters dominate
grep -i googlebot access.log \
| grep -o '?[^ ]*' \
| tr '&' '\n' \
| cut -d= -f1 \
| sed 's/^?//' \
| sort | uniq -c | sort -rn | head -20If parameter URLs are under about a fifth of crawler requests and your important pages are indexed, you do not have a crawl budget problem and you should leave this alone.
Fix one: canonical tags
The rel canonical link element is the primary replacement, and for most sites it is the only one needed. It tells Google which URL you consider the real one, and it consolidates ranking signals onto that URL.
<!-- On https://example.com/shoes?color=black&sort=price&sessionid=93f2 -->
<link rel="canonical" href="https://example.com/shoes?color=black" />Note what this example does. It drops sort and sessionid, because neither changes what is on the page in a way that deserves its own listing, and it keeps color, because a black shoes page is a genuinely different page that people search for.
That judgment is the whole job. Ask one question about every parameter: would a person plausibly search for this specific combination? If yes, it deserves an indexable URL with its own canonical pointing at itself. If no, canonicalize it away.
Fix two: stop generating and linking the URLs
The cheapest fix is not to create the problem. Three patterns cause most parameter sprawl.
Session identifiers in URLs. There is no reason for this in 2026. Use cookies. If a legacy system still does it, that is the highest priority thing on this list.
Tracking parameters on internal links. Marketing teams add campaign parameters to links inside the site, which creates a parameterized duplicate of every page and can also break session attribution in analytics. Internal links should never carry campaign tags.
Filter and sort links rendered as crawlable anchors. Every combination becomes a URL a crawler will follow. On a category with six filters and four values each, that is thousands of pages describing the same forty products.
For sorting and view controls specifically, the pragmatic answer is to make them not be links: handle them in JavaScript without changing the URL, or use a POST form. A crawler cannot follow what is not there.
Fix three: robots.txt for genuine crawl waste
When a parameter has no legitimate indexable form at all, block the crawl. Google’s robots.txt implementation supports the * wildcard and the $ end anchor.
User-agent: *
# Block session and tracking noise
Disallow: /*?*sessionid=
Disallow: /*?*replytocom=
Disallow: /*&utm_
# Block sorting and pagination controls on listing pages
Disallow: /*?*sort=
Disallow: /*?*orderby=
Disallow: /*?*view=
Sitemap: https://example.com/sitemap_index.xmlTest every pattern in a robots.txt tester before deploying it. A missing character here can block your entire site, and the consequences take weeks to reverse.
noindex, then block it later once it has dropped out.Fix four: faceted navigation, properly
Faceted navigation is where this gets genuinely hard, because facets are where the commercial opportunity and the crawl waste live in the same place. The workable approach is to divide facets into three tiers.
| Tier | Examples | Crawl | Index | Treatment |
|---|---|---|---|---|
| Demand facets | Brand, color, size, category | Yes | Yes | Self canonical, in sitemap, unique title and copy |
| Combination facets | Brand plus color plus size | Yes | Usually no | Canonical to the single facet parent |
| Control facets | Sort, price slider, grid or list, per page | No | No | Not links at all, or robots.txt blocked |
Promote a combination facet to tier one only when you have evidence of demand, meaning real impressions in Search Console or real volume in keyword research. Promoting on a hunch is how sites end up with fifty thousand thin category pages.
Two more rules that hold up. Keep parameter order consistent, so ?color=black&size=10 and ?size=10&color=black never both exist. And put demand facets in your sitemap, because that is the clearest possible statement to Google about which variants you consider real pages. Our explainer on the Search Console sitemap report covers how to verify Google is reading it.
Troubleshooting
Parameter URLs are still being indexed after you added canonicals. Check that the canonical target is not itself canonicalized elsewhere, that it returns 200, and that your internal links point at it. Contradictory signals are the usual cause. Use the URL Inspection tool to see Google’s selected canonical versus your declared one.
Traffic dropped after a robots.txt change. Reverse it first, investigate second. Then use the robots.txt tester on a representative sample of real URLs from your logs, not on the examples you had in mind when you wrote the rule.
Crawl stats show heavy crawling of URLs you blocked. Google honors robots.txt on fetch, but crawl stats also count attempts and the report lags. Give it a couple of weeks, and confirm your robots.txt is actually being served at the root of the correct hostname.
You need a URL removed from search results today. Use the Removals tool in Search Console for a temporary block of roughly six months, and fix the underlying cause during that window. Removals is not a substitute for a canonical or a noindex tag.
UTM parameters from campaigns are getting indexed. Those URLs get shared and linked, so they get crawled. Canonicalize them to the clean URL, and make sure your analytics setup does not depend on the parameter surviving, which our guide to using Google Analytics for marketing touches on.
Frequently asked questions
When exactly was the URL Parameters tool removed?
Google announced the deprecation in March 2022 and the tool stopped working on April 26, 2022. Google said only about 1% of the configurations site owners had entered were useful for crawling, and that its crawlers had become good enough at handling parameters without the hints.
Is there a replacement tool in Search Console?
No, and Google has not indicated one is coming. The replacement is a set of practices on your own site: canonical tags, robots.txt patterns, disciplined internal linking, and deliberate decisions about which faceted navigation combinations deserve to be indexed.
Does Google really handle parameters automatically now?
Largely yes for straightforward cases such as session identifiers and tracking tags. It handles large faceted navigation systems less reliably, which is exactly where the crawl budget waste happens. On a big ecommerce site you still have real work to do.
Should I use noindex or robots.txt for parameter URLs?
Use noindex when you want the URL crawled but kept out of the index, which is usually right for facet combinations. Use robots.txt when you want to stop the crawl entirely. Never combine them on the same URL, because a blocked page cannot have its noindex tag read.
What about Bing and other search engines?
Bing Webmaster Tools still offers parameter handling under its configuration settings. Everything described here works across all crawlers though, which is a good reason to fix the site rather than configure each engine separately.
Do I need to do anything if my site has few parameters?
Almost certainly not. Check crawl stats and the Pages report, and if parameter URLs are a small fraction of crawling and your important pages are indexed, leave it alone. Most of the damage in this area comes from unnecessary intervention.
Wrapping up
Losing the URL Parameters tool moved the work from a Search Console setting to your own codebase, and that is the right place for it. A canonical tag, a robots.txt rule and a link that was never generated all work for every crawler, are visible in code review, and cannot be silently forgotten in a console by whoever set them four years ago.
Start by measuring rather than fixing. Look at crawl stats and the Pages report, decide which parameters genuinely represent pages people search for, canonicalize the rest, and only reach for robots.txt where a whole class of URLs has no legitimate indexed form. If you also want to understand where your search traffic is actually coming from while you do this, our guides to finding search queries in Google Analytics 4 and checking backlinks in GA4 cover the measurement side.
