Are descriptive URLs really useless for SEO?

What exactly do we mean by speaking URLs?

A speaking URL is a human-readable address that describes the page's content. Instead of having example.com/p?id=12345, you get example.com/mens-running-shoes-2023. This format has been around since the 2000s and has been promoted as a good SEO practice for a long time.

The idea behind this concept is that a readable URL would help Google better understand the page's subject and improve its ranking. This belief is what John Mueller challenges with this straightforward statement.

Why is this statement so surprising to practitioners?

For years, Google's official recommendations have emphasized the importance of clean and descriptive URLs. SEO training, audits, and analysis tools continue to penalize poorly structured URLs. This dissonance creates legitimate confusion.

Mueller clarifies that the impact is indirect: a clear URL helps the user understand where they are clicking in the SERPs, which can influence the click-through rate (CTR). However, the ranking algorithm itself gives no bonus points for a well-formatted URL.

Does Google really use keywords in the URL?

Technically, Google crawls and indexes the words present in the URL. They even appear in bold in search results when they match the query. This suggests a form of semantic processing.

But does it use these words for relevance scoring? Mueller's position is clear: no. The algorithm prioritizes page content, user signals, and backlinks. The URL at best plays a marginal role as a confirmation signal, never as a main lever.

• Speaking URLs do not directly influence ranking in Google's algorithm
• Their usefulness lies in UX and CTR: they make links clearer for the user
• Google indexes keywords in the URL but does not assign any significant weight to them in ranking
• The time spent optimizing URLs should be reallocated to priority levers (content, technical aspects, link building)
• A broken or incomprehensible URL can harm the experience, but a perfect URL does not guarantee any ranking advantage

Is this statement consistent with field observations?

After 15 years of experience, I can confirm that sites with poor URLs (/index.php?id=789) can rank perfectly if their content and link building are solid. Conversely, perfectly optimized URLs on weak sites never perform well.

The challenge lies in the fact that Google continues to recommend clean URLs in its official documentation. This apparent contradiction fuels confusion. My interpretation? Google wants readable URLs for users (and for its crawler, by the way), but this does not weigh anything in the pure ranking algorithm. [To be verified]: no public study has ever properly isolated the URL impact while controlling for all other variables.

In what cases can the URL still influence SEO?

There are measurable indirect effects. A clear and short URL displayed in the SERPs generates a better CTR than a cryptic 150-character URL. A high CTR sends a positive user signal to Google, which can influence ranking in the medium term.

URLs also play a role in internal linking and architecture. A logical structure (/category/subcategory/product) helps Google understand the site's hierarchy. But be careful: it's the architecture that matters, not the keywords in slugs.

Another instance is URLs shared on social media or in articles. A readable URL encourages more clicks and makes it easier to remember. This can indirectly generate more traffic and backlinks, two signals that do impact ranking.

Should we stop optimizing URLs then?

No, but we need to rebalance priorities. Spending 3 hours debating whether the URL should contain the main keyword or the long-tail variant is a waste of time. However, ensuring short, logical URLs without unnecessary parameters remains a good technical hygiene practice.

I recommend: clean and descriptive URLs by default, but no obsession. If revamping the entire URL structure of an existing site requires hundreds of 301 redirects, the effort is probably not worth it. Focus your energy on content, Core Web Vitals, and link building.

What should we do with existing URLs?

If your site is already running with functional and indexed URLs, do not change anything. Even if they are not perfect, the cost of a redesign (redirects, potential bugs, loss of link juice) far outweighs the expected benefit, which according to Mueller is zero in terms of direct ranking.

For new sites or new sections, adopt a simple and coherent structure. Prioritize clarity: /blog/seo-mobile instead of /blog/post-2025-01-15-seo-mobile-complete-guide-2025. Avoid dates in URLs (they age poorly), unnecessary stop words (/the-seo-guide vs /seo-guide), and special characters.

What mistakes should be avoided when creating URLs?

The number one mistake: over-optimizing by stuffing the URL with keywords. /mens-running-shoes-cheap-promo-sales is counterproductive. It smells of manipulation and brings nothing in ranking. A URL should remain natural.

Another common pitfall: dynamic URLs with multiple parameters (?sort=price&color=red&size=42) that create duplicate content and complicate crawling. Use canonical tags and URL parameters in Search Console, but ideally adopt static or properly rewritten URLs.

Avoid overly deep URLs too (/cat1/cat2/cat3/cat4/product). Google can crawl everything, but a flat architecture (maximum 3 clicks from the homepage) facilitates internal PageRank flow and improves indexing.

How can I check if my URLs are clean enough?

Audit your site with Screaming Frog or Oncrawl: identify URLs longer than 100 characters, those containing underscores (prefer hyphens), capital letters (risk of duplication), or improperly encoded accented characters.

Check also for structure consistency: do all your product pages follow the same pattern? A heterogeneous format complicates crawling and automatic internal linking. A serious technical audit should include an analysis of URL patterns and their distribution in the hierarchy.

• Never overhaul already indexed URLs just to add keywords
• For new content, prioritize clarity and brevity over keyword stuffing
• Adopt a coherent and flat structure (maximum 3-4 levels deep)
• Use hyphens to separate words, no underscores or encoded spaces
• Avoid dates, special characters, and unnecessary stop words in slugs
• Properly configure canonical tags and URL parameters to prevent duplicate content