Garry Illyes, an analyst on the Google Search team, put faceted navigation at roughly half of all crawling issues reported to the company.
Now, add that to other URL-heavy combinations like action parameters, meaning sort, and others, and you'll quickly see that most crawl issues come back to URL parameters.
Google mentioned this in its Crawling December series in 2024, which went over advice originally posted in a 2014 blog post.
In this document, it suggests that there are two issues that every technical audit should include. This is duplicate content getting overcrawled and new pages taking too long to surface.
How Filter URLs Multiply Very Quickly
For a website with a 5,000-product catalog with 15 filter attributes with 8 values each, there could be millions of URLs. Three problems cause this. Query parameters, session IDs, and static paths. All of these create multiple URLs unnecessarily.
At first, this may not be an issue. However, crawl issues start to arise after 10,000 frequently updated URLs, says Google, something that a mid-sized catalog can easily achieve.
Which Facets Should You Index?
Surprisingly, not all of them. The decision should be commercial rather than technical. A facet earns a URL when people actually search for that combination.
Take a slot game catalog, for instance. Titles can be classified by theme, covering Ancient Egypt, Rome, North Africa, and television shows, and separately by mechanic, like Megaways or progressive jackpot. As a result, one game can sit in several URL paths at once.
Quickly, this can increase the number of URLs on a website. That's why you need to be selective with the type of URLs that are crawled. Not all of the mentioned filters deserve URL parameters.
Instead, only select filters that require it. For example, Egyptian slots is the most searched query. Their volatility, how many reels they have, and whether they are sorted alphabetically or not aren't.
Handling the Rest
Google's position on this is a lot different from what many SEO's think.
They would rather you block low-value filter URLs in robots.txt rather than noindexing them. The reason is that a noindexed page still gets requests and still costs crawl time. Also, never combine the two on the same URL either, because a crawler cannot read a tag on a page it isn't allowed to fetch.
For the facets you want to be crawled, the documentation shared is much more specific.
Use the standard ampersand separator, keep filter order consistent across URLs, and return a genuine 404 for combinations with no results at the URL they were requested rather than redirecting them to a generic error page.
Where Can You Start Today?
Start with your most popular organic pages first. If parameterized URLs are already earning quality traffic, they should remain indexed. For those that are not, they should get blocked so the crawl budget goes back to the pages that are meant to have it.
