There are an estimated 14,000 features in the documentation , i.e. variables that an algorithm system can use to define its output. But that does not guarantee that a given attribute is used for ranking . It could be used for crawling, indexing, or appearance of results and building the SERP.
It also does not guarantee that it will be used in production , since the documentation also logically includes testing processes.
Finally, even when something appears to be applicable to ranking and production, the weight of that attribute might be minimal or influence only in very specific cases, being irrelevant for most searches or query/URL pairs.
Is all this actionable? Can we apply it and algeria number data change the way we do SEO?
This is the question that almost everyone has.
While I find the leak extremely interesting, and an excellent opportunity to learn or at least broaden the horizons of what is commonly considered SEO theory, I must say that most of what we find will not be directly actionable .
To use a simile, we have a gigantic puzzle with thousands of pieces, but we don't have the reference image that tells us what we're trying to build. Of course, anyone who has been working in SEO for years has an idea, based on their own experience, of what that image should look like. We can take some pieces and match them with others, building what looks like part of the puzzle.