Home / Knowledge Base / Glossary / Faceted Navigation: Filters and Indexing

Faceted Navigation: Filters and Indexing

Glossary 2 min read Updated 28.09.2026
Twenty filters create millions of addresses. How to divide them into three categories, what to open, what to close, and why robots.txt is not suitable for this.

Faceted Navigation — filters in the catalog: color, size, price, brand. For the buyer, this is convenience; for search engines, it is a source of almost infinite addresses because filters combine with each other.

What is the Danger

Twenty filters create millions of combinations. The robot spends time crawling them, thousands of nearly identical pages get indexed, and important cards are crawled less frequently. This is a classic reason why a large catalog is partially indexed.

Three Categories of Filters

CategoryExamplesWhat to Do
There is demand, the page is useful“winter tires 17 inches”Open, provide a unique title and text, link from the menu
No demand, but the page is meaningfulCombination of two rare filtersKeep accessible but do not index
Technical parametersSorting, list view, sessionsClose from indexing and do not provide links

How to Organize

  1. Export the list of addresses with parameters from logs or the indexing report.
  2. Check against frequency: what people are actually searching for — see keyword selection.
  3. For in-demand combinations, create static addresses and unique titles.
  4. Close the rest with a noindex meta tag and remove from internal linking.
  5. Check in a month if the crawl coverage of the necessary pages has increased.
Why noindex, not robots.txt
The ban in robots.txt does not remove already known addresses from the index — the robot simply stops seeing what is written on them. To remove a page, it must be accessible to the robot and contain noindex. More details — noindex, nofollow, and robots.txt.

Connection to Promotion

Links to filters make sense only if it is an open, meaningful, and stable page for specific demand. In all other cases, the acceptor should be a section of the catalog or a card — see donor and acceptor.

Where Duplicates Come From

Even filters closed from indexing create a problem if there are regular links to them from catalog pages: the robot still crawls them and spends crawl budget. Therefore, the solution consists of two parts — prohibit indexing and simultaneously remove links from the layout, allowing filters to function without direct links.

A separate source of duplicates is the order of parameters in the address. The same set of filters selected in a different sequence can yield different addresses with identical content. The order of parameters should be normalized on the server side.

Sorting is Not a Filter

Sorting by price or novelty does not change the set of products, only their order. Such addresses should never be indexed: they are guaranteed duplicates of the section.

FAQ

What are the dangers of catalog filters?
Their combinations generate almost infinite numbers of addresses: the robot spends crawl budget on them, while important cards are visited less frequently.
Which filter pages should be opened?
Only those that have real demand: they should have their own title, text, and stable address.
Why can't we just close filters in robots.txt?
A ban in robots.txt does not remove already known addresses from the index. To remove a page, it must be accessible to the robot and contain noindex.
Was this article helpful?
Try it on your project Everything described in the article is available in the dashboard — the registration bonus is already in your balance.
Open dashboard