What does Google say about SEO? /
Quick SEO Quiz

Test your SEO knowledge in 5 questions

Less than a minute. Find out how much you really know about Google search.

🕒 ~1 min 🎯 5 questions

Official statement

Google Discover is not solely based on click-through rates but on a combination of signals to encourage a diversity and quality of content that satisfies users.
39:12
🎥 Source video

Extracted from a Google Search Central video

⏱ 59:06 💬 EN 📅 16/10/2019 ✂ 20 statements
Watch on YouTube (39:12) →
Other statements from this video 19
  1. 1:38 Pourquoi les outils SEO et Google Analytics ne montrent-ils pas les mêmes impacts après une Core Update ?
  2. 1:38 Pourquoi les classements post-Core Update évoluent-ils à des vitesses différentes selon vos outils ?
  3. 2:39 Faut-il vraiment s'inquiéter de ses backlinks et utiliser le fichier disavow ?
  4. 2:39 Faut-il vraiment surveiller tous ses backlinks ou Google exagère-t-il le risque ?
  5. 4:10 Le contenu généré par les utilisateurs pèse-t-il vraiment autant que votre contenu éditorial aux yeux de Google ?
  6. 4:11 Le contenu généré par les utilisateurs est-il vraiment traité comme le contenu éditorial par Google ?
  7. 6:51 Faut-il vraiment utiliser noindex pour gérer la visibilité du contenu interne ?
  8. 6:51 Faut-il utiliser le noindex pour tester un contenu avant de l'indexer ?
  9. 6:57 Google a-t-il vraiment un algorithme YMYL spécifique pour la santé et la finance ?
  10. 9:05 Faut-il vraiment isoler les contenus sensibles dans des sous-domaines séparés ?
  11. 10:31 Faut-il cloisonner les sections éditoriales d'un site pour booster sa visibilité dans Google ?
  12. 14:49 Le contenu white label nuit-il vraiment à votre indexation Google ?
  13. 22:02 Faut-il vraiment s'inscrire à Google News pour apparaître dans Discover ?
  14. 32:08 Comment Google News affiche-t-il les extraits de presse française sous la directive droit voisin ?
  15. 34:25 Comment optimiser pour Google Discover sans cibler de mots-clés ?
  16. 49:44 Faut-il vraiment utiliser le code 410 plutôt que le 404 pour accélérer la désindexation ?
  17. 53:59 404 ou 410 : Google fait-il vraiment la différence sur le long terme ?
  18. 54:00 Les balises canoniques locales peuvent-elles vraiment booster votre visibilité sans cannibalisation ?
  19. 57:38 Comment utiliser les balises canoniques pour éviter la cannibalisation entre vos contenus multi-localisations ?
📅
Official statement from (6 years ago)
TL;DR

Google claims that Discover is based not solely on CTR but on a combination of signals to assess content quality. The stated goal: to encourage diversity and serve truly relevant content to users. For SEOs, this means that optimizing for quick clicks without real value is no longer sufficient — it is necessary to focus on deep engagement and user satisfaction.

What you need to understand

What signals does Google combine to evaluate Discover content?

Google does not provide a precise list of the signals used, but the lack of reliance solely on CTR suggests a multi-criteria approach. It likely includes reading time, bounce rates, social shares, return to results, and interactions with the content.

This combination aims to differentiate "clickbait" content from that which genuinely provides value to the user. A sensational title can generate clicks, but if the user leaves the page immediately, Google picks up on that negative signal.

Why does Google emphasize content diversity?

The Discover algorithm seeks to avoid homogenization and filter bubbles. If all the content shown follows the same angle, tone, or sources, user experience deteriorates. Google thus wants to promote a variety of formats, viewpoints, and types of content.

In practical terms, this means that a site dominating a sector will not monopolize Discover. Even a small site can emerge if it offers original and engaging content that aligns with the user's true interests.

Does this statement change the classic SEO quality criteria?

No, it reinforces them. The fundamentals remain the same: E-E-A-T, depth of treatment, relevance. Discover does not create new criteria; it amplifies those already known and applies them in a context of proactive recommendation.

The major difference with traditional Search is that, in Discover, the user has not formulated an explicit query. Google must therefore anticipate their interests and assess content quality even before the click. Hence the importance of a precise title, an impactful image, and content that delivers on its promises from the very first lines.

  • CTR alone is not enough — post-click engagement matters just as much, if not more.
  • Content diversity: Google favors original angles and varied formats.
  • Behavioral signals: reading time, bounce rate, and interactions are scrutinized.
  • Editorial quality: E-E-A-T remains the foundation, even in proactive recommendations.
  • Promise kept: the content must exactly match what the title promises.

SEO Expert opinion

Is this statement consistent with field observations?

Yes and no. In principle, we do see that content with high CTRs but low engagement quickly disappears from Discover. Sites that bet solely on sensationalism have a short lifespan in the feed.

However, the ambiguity remains regarding the exact weighting of signals. Google speaks of a "combination" without specifying the relative weights. [To be verified]: some observe that large media with high domain authority receive disproportionate visibility, even if their content is not always more "diverse" than that of smaller sites. The theory of diversity is beautiful, but the practice sometimes seems less equitable.

What nuances should be applied to this Google statement?

First point: the absence of CTR as the sole criterion does not mean it has no weight. It likely remains a strong signal, especially at the start of content. If no one clicks, Google has no way to measure post-click engagement.

Second nuance: "quality" remains a subjective and poorly defined notion. Google measures it by behavioral proxies, but these proxies can be misleading. A dense, long article may have a high reading time without satisfying the user if it is verbose and poorly structured. Google infers satisfaction; it does not measure it directly.

In what cases does this rule not fully apply?

Hot news content partially escapes this logic. When breaking news emerges, Discover may push it massively even without a history of engagement, simply because it corresponds to a detected spike in interest (searches, social trends).

Similarly, very well optimized evergreen content may stagnate in Discover if it does not benefit from a timely "moment". An excellent guide published outside of current events may struggle to break through, even with all quality signals green. Timing and seasonality play an underestimated role.

Warning: Do not overestimate your ability to “hack” Discover with pure technical optimization. The algorithm relies on real behavioral signals that you cannot sustainably simulate. Focus on true editorial value added.

Practical impact and recommendations

What should you concretely do to optimize your content for Discover?

First priority: work on the coherence between title, image, and content. The user who clicks must immediately understand they are in the right place. If the title promises "5 revolutionary strategies" and the article delivers basic advice, the bounce will be immediate, and Google will penalize that.

Next, optimize for real engagement: structure the content to make it scannable (clear subtitles, short paragraphs, lists), integrate rich media (videos, infographics), and most importantly, provide depth. Content that genuinely answers a question or solves a problem generates natural reading time.

What mistakes should be absolutely avoided?

Do not fall into the trap of pure clickbait. A title like "You won't believe what happened next" may generate clicks, but if the content disappoints, Google will record a massive negative signal. In the long term, your domain will lose credibility in Discover.

Another common mistake: neglecting image quality. Discover is a visual feed. A blurry, poorly framed, or generic image will kill your CTR, regardless of the text's quality. Invest in original, high-resolution visuals, and use the appropriate Schema tags (ImageObject, largeFeaturedImage).

How can I check if my site is well positioned for Discover?

Use the Google Search Console, Discover section. Analyze impressions, clicks, but especially trends: content that rises and then collapses quickly likely has an engagement issue. Content that generates few clicks but remains stable may need a better title or image.

Cross-reference this data with your analytics: average time on page, bounce rate, pages per session for Discover traffic. If these metrics are significantly lower than your standard Search traffic, you have a post-click satisfaction issue to address.

  • Align title, image, and content to prevent user disappointment
  • Structure content for quick reading (subtitles, lists, short paragraphs)
  • Use original, high-resolution images with Schema ImageObject
  • Avoid pure clickbait — prioritize accuracy and a promise kept
  • Monitor both Search Console Discover AND behavioral analytics
  • Test different formats (videos, infographics, long guides) to identify what engages
Optimizing for Discover requires a rigorous editorial approach and constant attention to behavioral signals. If you find that your site struggles to emerge despite quality content or if your engagement metrics remain low, an external audit can reveal blind spots. Specialized SEO agencies have tools and cross-industry feedback that allow for quick identification of barriers and adjustments to editorial and technical strategy. Sometimes, an expert external perspective unlocks situations that become invisible when you are entrenched in the day-to-day of content production.

❓ Frequently Asked Questions

Le taux de clics a-t-il encore de l'importance dans Google Discover ?
Oui, mais il n'est plus suffisant à lui seul. Google le combine avec des signaux d'engagement post-clic comme le temps de lecture et le taux de rebond pour évaluer la vraie qualité d'un contenu.
Quels types de contenus Google privilégie-t-il dans Discover ?
Google favorise les contenus originaux, diversifiés et engageants. Les formats variés (articles longs, vidéos, infographies) ont tous leur chance, à condition qu'ils apportent une vraie valeur ajoutée à l'utilisateur.
Un petit site peut-il concurrencer les gros médias dans Discover ?
Oui, en théorie. Google affirme encourager la diversité, ce qui laisse de la place aux sites de niche avec des contenus de qualité. En pratique, l'autorité de domaine joue encore un rôle, mais un contenu vraiment pertinent peut émerger.
Comment mesurer la performance de mes contenus dans Discover ?
Utilisez la section Discover de la Google Search Console pour suivre impressions et clics, puis croisez avec vos analytics pour analyser temps de lecture, rebond et pages par session. C'est le croisement des deux qui révèle les vrais problèmes.
Faut-il optimiser différemment pour Discover et pour la Search classique ?
Pas fondamentalement. Les critères de qualité (E-E-A-T, profondeur, pertinence) restent identiques. La différence principale : Discover exige une cohérence absolue entre titre, image et contenu, car l'utilisateur n'a pas formulé de requête explicite.
🏷 Related Topics
Domain Age & History Content Discover & News AI & SEO

🎥 From the same video 19

Other SEO insights extracted from this same Google Search Central video · duration 59 min · published on 16/10/2019

🎥 Watch the full video on YouTube →

Related statements

💬 Comments (0)

Be the first to comment.

2000 characters remaining
🔔

Get real-time analysis of the latest Google SEO declarations

Be the first to know every time a new official Google statement drops — with full expert analysis.

No spam. Unsubscribe in one click.