We all know that ranking well with search engines is imperative for business success in today’s age, but search engines don’t detail the secret sauce for higher ranking. It’s up to SEO professionals to analyze and understand ranking factors.
But, Google recently published a research paper that potentially reveals the ranking factors of the future. While Google didn’t confirm that this new ranking method will be part of its algorithm in the future, it’s definitely worth paying attention to.
Are These Published Algorithms Used by Google?
As stated by Google in the past: “Google research papers in general shouldn’t be assumed to be something that’s actually happening in search.”
However, like we said, Google rarely confirms which algorithms they are using, whether described in research papers or in patents.
How the Current Algorithms Work
According to the research paper, the current algorithms assign a value or a label to each individual document. In the ranking setting, they determine the order of relevance of the whole list of documents.
One of the biggest considerations that new and improved algorithms could implement, according to this paper, is clearer understanding of searcher intent. Most search algorithms already do this, but it could be made stronger by taking the age of the information into account among other factors.
This example straight from the research paper shows how this plays out:
“Consider a search scenario where a user is searching for a name of a musical artist. If all the results returned by the query (e.g., Calvin Harris) are recent, the user may be interested in the latest news or tour information.
If, on the other hand, most of the query results are older (e.g., Frank Sinatra), it is more likely that the user wants to learn about artist discography or biography. Thus, the relevance of each document depends on the distribution of the whole list.”
We can clearly see here that the refinement of search results is aided by the age of the relevant web pages, yielding the searcher the likely best results.
Higher Accuracy by Modeling Human Behavior
Next, the paper reveals that search engine users typically compare the results that they receive to other pages on the web. Understanding this user behavior gives search engines like Google a reason to incorporate comparison into their ranking algorithm. A ranking model with a direct comparison function is much more effective, since it mimics real human behavior.
The Latest Algorithm Does Work
Some past papers show that certain improvements to the algorithms are minimal and that the cost to be successful (including both time and hardware) would be huge. Obviously, this is not an ideal situation.
However, when a research paper shows evidence that its proposed improvements would yield low cost with high improvement, it’s one that should perk our ears as something Google would very likely do.
This particular research is just that: high improvement with little cost in time and hardware. It improves deep neural network and tree-based models.
All of this goes to show the value of being aware of information retrieval research. For example, correlation studies led people in the SEO community to think that likes on Facebook were important to rankings. In reality, though, reading research papers would show that this is very unlikely the case.
How Does This Help Your SEO Efforts?
Google’s ranking factors are becoming less and less “traditional” by the day, and this research paper identifies several creative methods that Google may be implementing in the near future.
As soon as we become comfortable with best practices for ranking, Google changes its algorithm again. This just means that your SEO strategy needs to be adaptable and robust.
Need Someone to Take the Reins?
The findings in this paper are promising and have the potential to spark real change in the world of SEO strategy. Need help taking advantage of these new opportunities for optimization? Our team of SEO experts is ready to help. Contact us today!