Should you still use nofollow to filter facet navigation crawl?

Why is Google still discussing nofollow in the context of facets?

Facet navigation — those multiple filters on e-commerce sites (size, color, price, brand…) — generates nearly infinite combinations of URLs. Each filter applied potentially creates a new page. Without control, a search engine can end up crawling thousands of redundant or SEO-value-less pages.

Mueller reminds us that nofollow allows bots to be informed about which internal links are less important. It’s a historical signal, originally introduced to combat comment spam. However, Google has always tolerated it internally, even if other methods exist.

Is internal nofollow still treated as a crawl signal?

Since March 2020, Google treats nofollow as a “hint” rather than a strict directive. What does that mean? The bot can choose to ignore it. In practice, on internal links, it is still largely respected — especially in contexts where crawl budget is critical.

Mueller does not specify whether this statement reflects guaranteed behavior or mere tolerance. It’s unclear. But on the ground, we observe that nofollow on facets continues to work to limit exploration of unnecessary combinations, provided the website structure is sound otherwise.

What alternatives to nofollow exist for managing facets?

Nofollow is not the only option. The robots.txt file can completely block the crawl of specific URL patterns. Canonicals consolidate variants to a master page. The URL Parameters Handling in Search Console (now deprecated) also allowed for signaling facet parameters.

Each method has its advantages. Robots.txt is radical but can create blind spots if misconfigured. Canonical preserves crawl but requires PageRank. Nofollow is a middle ground: the link exists for the user, but the bot receives a signal of lower priority.

• Internal nofollow: a soft signal, largely respected in practice, easy to implement on filter links
• Robots.txt: total crawl block, risks obscuring legitimate content if poorly calibrated
• Canonical: allows crawl but redirects equity to the main page, consumes budget
• Meta noindex: allows crawl but blocks indexing, a hybrid solution but resource-intensive
• Conditional JavaScript: show facet links only after user interaction, complicating bot rendering

Is this recommendation consistent with on-the-ground observations?

Yes, overall. On e-commerce sites with several thousand references, nofollow on facets continues to significantly reduce the number of pages crawled unnecessarily. Server logs show that Googlebot less frequently follows links marked nofollow, even if it’s not an absolute guarantee.

But be cautious: Mueller refers to “less important” pages or links, not “useless.” Subtlety matters. If a facet page can rank for a niche query (e.g., “blue running shoes size 42”), marking it as nofollow sacrifices an opportunity. [To be verified]: Google does not explicitly state if nofollow harms the ranking potential of a page crawled otherwise.

What are the limitations and risks of this approach?

Internal nofollow can create unintended silos. If all facets are in nofollow, internal linking weakens. Deep product pages may become less accessible to the bot if they are only linked through nofollowed filters.

Furthermore, too much internal nofollow can send a confusing signal: why does your site offer links it deems “less important”? It’s an admission of a shaky structure. A better architecture — for example, stable category pages + facets in canonical — resolves the problem at the source rather than masking it.

When is nofollow not the right solution?

If your site has fewer than 500 pages, crawl budget is not a concern. Adding nofollow everywhere falls into the cargo cult SEO trap. Google will crawl your entire site anyway.

Similarly, if your facets generate unique content (differentiated descriptions, filtered customer reviews, localized stocks), treating them as nofollow wastes long-tail potential. In this case, it's better to index and optimize. Nofollow is a defensive tactic, not an offensive one.

What should you concretely do on a facet site?

Start by auditing your server logs. Identify how many facet pages Googlebot crawls, and compare that with the volume of strategic pages (product sheets, main categories). If the ratio is unbalanced — for example, 70% of the crawl on facets vs. 30% on selling content — you have a problem.

Next, segment your facets. Not all are created equal. A “brand” or “price” filter on a broad category may have ranking potential. However, a “size + color + material” filter is often redundant. Apply nofollow only on low-value combinations.

What mistakes to avoid in implementation?

Do not put all internal links as nofollow out of precaution. It’s counterproductive. Internal linking is one of the few levers you control 100%. Weakening it intentionally is like shooting yourself in the foot.

Another pitfall: nofollow on pagination links. If you paginate your product lists, these links should remain follow so that Googlebot can access the entire catalog. Confusing pagination with facets is a classic mistake that orphaned entire parts of content.

How to check if the strategy is working?

Monitor three metrics in Search Console: pages crawled per day, unindexed discovered pages, and indexed but unvisited pages. If after implementing nofollow, crawl focuses better on your strategic pages and the rate of indexing of unnecessary pages decreases, you’ve succeeded.

At the same time, check in Google Analytics that your organic traffic on facet pages does not collapse. If some facets were indeed ranking, you would see a drop. In this case, you need to arbitrate: saved crawl budget vs. lost long-tail opportunities.

• Audit server logs to measure the distribution of crawl between facets and strategic pages
• Identify low SEO value facet combinations (multiple cumulative filters, redundant parameters)
• Apply nofollow only on non-strategic facet links, not on pagination or main categories
• Ensure product pages remain accessible via at least one follow path (categories, XML sitemap)
• Monitor the evolution of crawl and indexing in Search Console after deployment
• Compare with a mixed approach (canonical + selective nofollow) for fine optimization