Official statement
Other statements from this video 13 ▾
- □ Pourquoi Google vous pousse-t-il à poster vos problèmes d'indexation dans son forum ?
- 1:06 Pourquoi Google impose-t-il les URLs www plutôt que m-dot comme source principale pour les applications ?
- 2:46 Les pages 404 nuisent-elles vraiment au classement SEO ?
- 3:26 Comment Google Panda juge-t-il vraiment la qualité de votre contenu ?
- 6:08 Pourquoi Panda ne fonctionne-t-il pas en temps réel et qu'est-ce que ça change pour votre site ?
- 10:14 Le budget de crawl dépend-il vraiment de la qualité du contenu ?
- 12:32 La vitesse mobile affecte-t-elle vraiment le classement Google ou est-ce un mythe SEO ?
- 14:16 Le deep linking fonctionne-t-il sans site mobile m-dot ?
- 15:24 La personnalisation des résultats Google repose-t-elle vraiment sur votre historique de navigation ?
- 25:39 AdWords booste-t-il vraiment votre référencement naturel ?
- 33:59 Les liens de faible qualité peuvent-ils vraiment pénaliser votre site ?
- 40:11 Un site hors ligne perd-il son référencement Google ?
- 41:18 Pourquoi Google refuse-t-il de lire un fichier Robots.txt avec une majuscule ?
Google specifies that an incorrect redirect occurs when a site with both m-dot and www versions redirects to a non-equivalent page, particularly from the homepage to internal pages. This definition targets separate mobile configurations that create loops or destination inconsistencies. For an SEO practitioner, this means systematically auditing redirection pairs between versions to avoid contradictory signals that dilute PageRank and disrupt mobile-first indexing.
What you need to understand
What does 'an incorrect redirect' really mean according to Google?
Google defines an incorrect redirect as any redirect that does not lead to a functional equivalent between the m-dot and www versions of a site. This primarily affects architectures where two distinct versions coexist: one for mobile (m.example.com) and one for desktop (www.example.com).
The issue arises when the redirect changes nature: for instance, a mobile user accessing www.example.com/page-a is redirected to m.example.com but not to m.example.com/page-a, instead landing on m.example.com/ (the homepage). This break in equivalence creates a structural inconsistency that Google penalizes.
What problems do homepage to internal page redirects cause?
Google explicitly mentions redirects that change from the homepage to internal pages. This wording targets cases where a user requests example.com but is redirected to example.com/mobile-app or example.com/promo.
This type of redirect forces an undesired destination. For Google, it signals manipulation or a shaky architecture. The engine expects a URL to redirect to its semantic equivalent, not to different content. Arbitrary redirects dilute crawl consistency and fragment ranking signals.
In what contexts does this rule really apply?
This guideline concerns only sites maintaining a separate m-dot or www structure. If your site is responsive with a single URL for all versions, this rule does not apply. Google has reiterated: responsive design is the recommended configuration precisely to avoid such issues.
On the other hand, if you are still managing an m-dot (common in legacy e-commerce or high-traffic media sites), each desktop URL must have its exact mobile equivalent. A redirect from www.example.com/product-x to m.example.com/product-x is correct. Redirecting to m.example.com/ is not.
- Incorrect redirects affect sites with separate mobile and desktop versions (m-dot, www).
- Each URL must redirect to its functional equivalent on the other version, not to a different page.
- Homepage to internal page redirects are explicitly considered incorrect by Google.
- Responsive design eliminates this problem at its core by unifying URLs.
- Ranking signals fragment when redirects create structural inconsistencies.
SEO Expert opinion
Is this statement consistent with real-world observations?
Yes, but Google oversimplifies. In practice, I have observed that incorrect redirects mainly create crawl loops and contradictory indexing signals. A poorly configured m-dot site can send Googlebot Mobile to page A, while Googlebot Desktop sees page B, fragmenting PageRank and creating duplicates.
What is missing in this statement: no indication of tolerance thresholds. How many incorrect redirects before penalty? Google does not specify. My tests show that a small percentage (less than 5% of URLs) does not trigger a visible sanction, but beyond that, mobile-first indexing becomes erratic.
What nuances should be added to this rule?
Google speaks of equivalence, but reality is blurrier. Take a news site: the mobile version of a long article may be paginated (m.example.com/article/page-1), while the desktop version displays everything on a single URL (www.example.com/article). Technically, this is not a strict equivalence, but Google tolerates this divergence as long as the main content remains identical.
Another nuance: temporary redirects (302) to the mobile homepage while an equivalent page is being created. I have seen e-commerce sites using this technique during product launches. Google does not penalize immediately, but if the 302 becomes permanent, signals degrade. [To be verified]: Google has never published data on the tolerated duration for a 'provisional' 302.
In what cases does this rule not apply strictly?
If your site uses dynamic serving (same URL, different HTML based on user-agent), this rule does not directly apply. There is no redirect, hence no risk of inconsistency. Google then checks for correspondence via the Vary: User-Agent annotation.
Another exception: geolocated redirects. If a mobile user in France accesses www.example.com and is redirected to m.example.fr, Google considers this acceptable if the destination aligns with intent. But beware: a geolocated redirect to the homepage rather than to the equivalent page remains incorrect.
Practical impact and recommendations
What should you prioritize auditing on your site?
Start by extracting all desktop/mobile URL pairs from your sitemaps or server logs. Use Screaming Frog or an equivalent crawler in 'mobile' mode and then 'desktop' to compare final destinations after redirection. Look for discrepancies: a desktop URL that redirects to the mobile homepage while an equivalent page exists.
Then, test cross redirects: a mobile user accessing a www URL must land on the m-dot version of the same page, and vice versa. Use the URL Inspection tool in Search Console to check what Googlebot actually sees. Often, JavaScript or client-side redirects are not followed by the bot.
What mistakes should absolutely be avoided?
Never redirect a specific URL to the homepage of the other version. This is the most common mistake: a product at www.example.com/shoe-x redirected to m.example.com/ instead of m.example.com/shoe-x. This error dilutes PageRank and creates signals of duplicate content.
Also avoid redirect chains (A → B → C). Google follows up to 5 redirects, but each hop dilutes link equity. On mobile, where latency matters, a redirect chain slows down crawling and degrades user experience. Always consolidate into direct redirects (A → C).
How can you check that your configuration is compliant?
Use a Python script or a tool like Netpeak Spider to automatically test 100% of your URL pairs. The script should verify that each desktop URL redirects to its exact mobile equivalent and vice versa. Export exceptions into a CSV for priority correction.
Next, analyze mobile coverage reports in Search Console. Google explicitly reports 'incorrect redirects' in the Errors section. If you see this message, your architecture is already penalized. Immediately correct the listed URLs.
- Extract all desktop/mobile URL pairs from sitemaps and server logs
- Crawl the site in mobile mode then desktop to compare final destinations
- Test cross redirects with the URL Inspection tool in Search Console
- Eliminate redirects to the homepage when an equivalent page exists
- Remove redirect chains and consolidate into direct redirects
- Automate checking URL pairs with a script or dedicated crawler
❓ Frequently Asked Questions
Une redirection 302 vers l'accueil mobile est-elle considérée comme incorrecte ?
Le responsive design élimine-t-il complètement ce risque ?
Combien de redirections incorrectes Google tolère-t-il avant pénalité ?
Les redirections géolocalisées sont-elles concernées par cette règle ?
Comment Search Console signale-t-il ces erreurs ?
🎥 From the same video 13
Other SEO insights extracted from this same Google Search Central video · duration 50 min · published on 21/01/2016
🎥 Watch the full video on YouTube →
💬 Comments (0)
Be the first to comment.