How do the different ranking systems work together?
Posted: Wed Feb 19, 2025 6:41 am
First, Google has to compile a thematically appropriate corpus of usually thousands of documents in response to a search query. From these thousands of documents, Google then selects several hundred documents for a special document scoring. To do this, Google uses timeliness signals, page rank, localization signals...
Then the deep learning systems are used for ranking.
Document scoring uses classic information retrieval signals and factors such as keywords, TF-IDF, internal and external links, etc. to determine the objective relevance of a document with respect to the search query.
The ranking can then be done based on deep learning systems saudi arabia cell phone number list that use both user data and quality signals or signals that can be assigned to the EEAT concept for training. Since these signals are used to detect relevance, competence, authority and trust patterns, the ranking is done with a time delay.
User data can be used to determine what type of content users prefer when making a search query. Is it long, comprehensive content or rather short checklists, instructions or definitions? It is therefore important to classify search intents according to micro intents. See my two articles How to use 12 micro intents for SEO and content journey mapping and Micro intents and their role in the customer journey.
I think that iterative re-ranking based on user data is possible after just a few weeks, depending on how quickly Google has collected enough user data. Learning from EEAT signals takes longer because collecting trust patterns and training the algorithms is much more complex. The core updates seem to be responsible for the ranking here.
As I understand it, Google evaluates on three different levels. (Read more in the article The dimensions of Google rankings)
Then the deep learning systems are used for ranking.
Document scoring uses classic information retrieval signals and factors such as keywords, TF-IDF, internal and external links, etc. to determine the objective relevance of a document with respect to the search query.
The ranking can then be done based on deep learning systems saudi arabia cell phone number list that use both user data and quality signals or signals that can be assigned to the EEAT concept for training. Since these signals are used to detect relevance, competence, authority and trust patterns, the ranking is done with a time delay.
User data can be used to determine what type of content users prefer when making a search query. Is it long, comprehensive content or rather short checklists, instructions or definitions? It is therefore important to classify search intents according to micro intents. See my two articles How to use 12 micro intents for SEO and content journey mapping and Micro intents and their role in the customer journey.
I think that iterative re-ranking based on user data is possible after just a few weeks, depending on how quickly Google has collected enough user data. Learning from EEAT signals takes longer because collecting trust patterns and training the algorithms is much more complex. The core updates seem to be responsible for the ranking here.
As I understand it, Google evaluates on three different levels. (Read more in the article The dimensions of Google rankings)