I do not work at Google and I do not have a full and complete answer. I sometimes wonder if anyone, even the people working at Google do. I say this because their algorithm is enormous and it may be beyond the mental capacity of any one person to understand every aspect of it.
However, there is one very interesting part of this answer that very few people seem to know about, even though it’s common knowledge.
Google hires human raters who have two main tasks:
Rating Search Engine Results
For the benefit of these human raters, Google publishes a set of instructions, a summary of which is released to the general public.
It’s called the Google Search Quality Rating Guidelines.
The basic idea is that every search query reveals a certain user intention, and relative to that user intention for the search query, how well do the various entries in the search engine result page satisfy that user intent for that search query.
What Google strives for is for the most useful webpages be returned to the person doing the query, relative to the user intent of the query.
The post linked to above provides more details on the Google rating system (it’s really practical).
Testing Algorithm Updates
Prior to Google rolling out an algorithm update, it is tested with human raters. They are presented with a split screen. In one half they see the pre update search results. In the other half they see the post update search results. However the human raters are not informed which is which. They then report to Google which search result is better in their opinion.
Over a large number of human raters, Google learns if the potential algorithm update really does or does not generate better search results.
So how do you rank your site?SEO Needs Link Building