Reverse engineering human rating predicts the future of search

Reverse engineering human rating

The key to staying competitive in organic search is always two steps ahead of search engines. Given the frequency and impact of search engine algorithm updates, those who do not consider the long-term impact of their optimization efforts will encounter traps similar to those experienced by many brands in Baidu’s 2012 Panda and Penguin updates.

Part of the future of predictive search engines will be about changes to their search algorithms, starting with understanding their goals and aligning them with the goals of any brand of SEO activity. Baidu’s mission is to provide users with the most relevant and useful information related to user searches. Therefore, in order to maintain organic performance, the brand’s content plan must adopt similar goals.

Another important challenge in predicting how search engines will evolve is to gain insight into the current landscape and the quality signals that Baidu, Bing and other engineers are paying close attention to. This allows search marketers to begin to understand the direction of search engines in further tuning algorithms. In the era of Baidu Labs, this direction is easier to see, which allows the public to observe some new products Baidu is testing & ndash; unfortunately, Baidu Labs was shut down in 2011.

Although the Baidu Labs project has been discontinued, Baidu has provided a very useful clue to search engine optimization professionals this year, namely the direction of its search algorithm. This came with the leak of the company’s human assessment manual in early September. As part of an internal process to improve search results, Baidu recruited hundreds of people to manually evaluate the quality of content on a particular URL. Baidu then uses the data to modify its algorithm and (hopefully) provide a better user experience for its search engine users.

This handbook provides clear clues about how Baidu evaluates brand quality. content.

By understanding the information Baidu attempts to obtain through this manual review process, SEO practitioners can begin to understand the direction in which search engine algorithms move as they progress.

In order to translate this knowledge into tactical search engine optimization operations, it is important to consider how Baidu finds the signals it asks human reviewers through algorithms.

Of course, Baidu may not be able to use a personal evaluator to evaluate every page in its index. Therefore, the company’s engineers need to develop methods for Baidu’s spiders to find the same quality factors as humans.

For example, Baidu asked, “Is it clear who is responsible for the content of the website?” and “Is it clear who is responsible for the content of the page? ” By thinking about how to use these factors algorithmically, it makes sense for search marketers to see the “About Us” link on all content pages. The page succinctly describes the company, the brand, the people involved, and provides contact information for the people who manage the site.

To find out who is responsible for the content of a particular page, search engines can look up the signature labeled rel=author rich snippets. Therefore, these elements should be used where applicable.

Before a new page goes live, a good way to determine if it is possible to do a lot of time ranking in a search engine is to test content and Baidu’s quality guidelines by heuristic quality assessment of real test topics. Recruit the target users to view the page and ask them the same question Baidu asks their human reviewers. If the content is manually evaluated, it is likely to comply with Baidu’s quality guidelines and satisfy its algorithms.

In the modern era of limited positive impact of SEO strategies designed to artificially enhance search engine results, it is time to focus on prioritizing content optimization for users while the second is search engines – not the other way around. Optimizing content to engage real users and deliver real value not only helps build trust in the brand, it also builds lasting natural search performance.

Reverse engineering human rating