Understand Panda's algorithm

Google's February 24, 2011 page rating change, which takes the name of the engineer who ran it, affected 11.8% of U.S. sites, reducing the presence in page results to content deemed poor, unoriginal, or of little use.
French-language sites were affected in August 2011.

Google panda et la qualité d'un site

"We want to encourage a healthy ecosystem..." 'the Google post said.

The criticism that the company suffered - I recall a joke on April 1 about a yacht named Adsense of the CEO of Demand Media - harmed the search engine, and he had to react.

Panda was a program run manually by Google from time to time to assess the "quality" of sites, which were then integrated into an organic algorithm in January 2012.
It calculates the change ratio for a site to change the initial page rating by other criteria. No other criterion of the algorithm tempers this score.

How Panda's algorithm changes results

In theory, Panda makes the difference between a quality and interest-free site like this:

The algorithm looks for authoritative sites: they often offer new information and innovative content, unlike those who chew five hundred words on a topic in which we do not have special knowledge.

Another more recent quote:

Downgrade sites that do not provide original content or otherwise increase its value.

This solution is self-written. It is she who says what a low-quality site is. But webmasters are looking for more precise criteria .
Here's how Google determines the intended quality of sites, based on patent 8,682,892:

Panda was originally developed as a separate program, as it requires large resources to split the Network into groups of resources and compare them.

With Panda, Google wanted to radically change the very role of the search engine: It no longer wants the results to contribute to the promotion and success of the site, from now on the site must receive its audience elsewhere, from the links it receives, and if it achieves some success, then only the engine can put it forward based on other ranking criteria.

It's about quality, mainly about popularity, because popular sites always get a lot of links, although they publish, often news comes from other sites. Reading the patent also shows that Google does little about originality: fully copied content can be positioned better than the original if it gets more independent links.

How to change the site for Panda

What to do when you're touched by Panda's renewal?

According to Google:

"Low-quality content on part of a site can affect a site's overall ranking, thus removing low-quality pages, combining them or improving the content of individual surface pages to make them useful pages, or moving low-quality pages to another area can ultimately help classify your higher-quality content."

However, all experts agree that it is impossible to cancel the fine without changing the content of existing pages and adding new content.
Merging two pages with banal content will make a new page more with banal content, it will not solve anything.
The webmaster's efforts should focus on obtaining independent feedbacks.

  1. Ideally, the site maximizes the Panda formula if it has a lot of backlinks and no content. The service can match this. But Panda is just a factor that enhances the initial score, this one depends on the content.
  2. For pages whose engine cannot understand the interest in the user, enrich their content. But if they have a lot of feedbacks, changing the content could lead to other penalties (unless it brings them new feedbacks).
  3. Removing all non-feedback pages is undoubtedly an effective way to improve the relationship between them and therefore return to grace. Or do not remove them from the index with the meta noindex tag.
  4. To get backlinks, make sure your content brings something useful, so unreleased (searching for similar content on engines). Always ask yourself what else your page brings.
  5. Customize Content. Use your own words. And, I appeal to bloggers, remember your dissertations, the professor asked you not to reprint either the topic or someone else's answer, but to give your own ideas. Consider different points of view to be authoritative and not seem subjective.
  6. Take care of the user experience, the desire to look at more pages or return to the site.
  7. For pages that have no chance of getting backlinks or being in the SERP, make them dynamic and therefore invisible to engines by answering visitors' questions. This is what we do with the dictionary (button at the top of the page on the right), thanks to the use of Ajax.
  8. Look at the output speed in Analytics or another statistical tool. Pages with a high level of release fine the site. They can be removed or made dynamic if they do not have feedback.
  9. Do not edit pages with multiple backlinks.

Need to know that changing existing content won't be enough to undo Panda's effects because it won't have new backlinks. This is, first of all, new unique content that can do this.
It's all going to take a lot of work, but we're going to console ourselves thinking about content farms that need ten million pages to edit...

Conclusion

The most important and incomprehensible result for webmasters, which was officially confirmed by Google, is that if part of the site is fined, the entire site will be fined. So pages of very good quality will be less well positioned in SERPs than pages of other sites of lower quality!
This is even harder to admit since it became known that even though Google presents its process as a way to choose quality pages, its main effect is to promote already more important sites and further increase their audience.

Date sent from Panda

The first two are official, the following are evaluated:

See also

Evolution de l'algorithme

Algorithm evolution