In this article, particular examples and applications are used to illustrate the types of user behavior signals that search engines do – and do not – employ.
It’s no longer enough to know which ranking signals may or may not be at play in the modern era of search.
You should also be aware of the context in which these signals are being sent.
We’ll take a closer look at how search engines utilize user behavior
Unlike the previous three chapters, this will concentrate on particular instances and applications rather than just on ideas.
This chapter will focus on Google since they’re the market leader, but the concepts also apply to Bing.
The lack of use by users sends a message to the engines.
Let’s be clear about what user activity does not affect before we determine how it affects search.
Internet tracking tool
Google Analytics doesn’t have any effect on search results in any manner.
According to Google’s John Mueller, there is no penalty for utilizing or not using Google Analytics in search when questioned about the subject on June 28, 2018.
To put it another way: (@JohnMu) June 28, 2018.
You should not blindly believe whatever Google says, but this is a recurrent reaction, so it makes sense to accept it at its value.
Google Analytics is not used by all website owners, webmasters, and SEO consultants. For example, if Google utilized their analytics tool to measure your site’s success and failure indicators, they would be comparing apples to oranges.
Your unique user metrics (as reported or discovered by Google Analytics) are not utilized for or against you in any particular way.
There is no evidence that Facebook (or Twitter, or Instagram, or any other social media network) counts the number of likes, shares, or follows that a user has. As before, this makes total sense.
Google has indicated that social signals are unreliable due to the difficulty of obtaining all accessible information and data from these networks since they are:
- Too easily manipulated;
I’ll cite a 2018 Hootsuite research that showed a connection as a matter of fairness. In this research, 90 pages of information were separated into three groups, and just Twitter was used.
It’s the “no promotion” camp that’s in charge. For 30 articles, there were no organic tweets or bought marketing.
As a result of organic tweets, we promoted 30 articles.
There were no paid promotions for the next 30 articles; instead, they were promoted via organic tweeting.
The end outcome was as follows:
In other words, the rankings of both groups of pages tweeted improved more quickly than could be attributed to links.
The purpose of this is not to muddy the waters but rather to deliver the most accurate information possible.
As far as we know, Google doesn’t utilize it as a measure. And yet, despite the evidence to the contrary, here we are.
On the one hand, the evidence isn’t conclusive at the size Hootsuite points out in their paper. Our findings here are more likely to result from the reinforcement of entity relevance than the influence of Twitter.
Tagging a thing with a certain hashtag or phrase and then linking it to a specific website is known as “tagging.”
A non-link-based entity reinforcement has resulted in a ranking boost in this case; it doesn’t matter if it was Twitter or any other site.
Something consistent with what we’ve seen in patents and what Google is saying.
Search Engines Can Use User Behavior Signals
So now that we’ve covered the types of user behavior signals that search engines may utilize, let’s look at what the engines don’t use (or at least some of the main components I hear stated regularly).
Behavior once a user clicks on an advertisement.
It is right that Google does not utilize Google Analytics to collect signals for ranking a website; however, this does not imply they do not use post-click data at all, as they have indicated.
Google is aware of the following information:
- Which search results in users click on,
- How long they spent on the destination site before returning to Google, and
- What the users did after leaving the target site.
As a result, we have four primary options for Google to use:
- A searcher visits your site in the search results, quickly returns to the search engine, and clicks on the next link.
- A searcher visits your site in the search results and quickly returns to the search engine, and clicks on the next link. A negative relevance signal would be sent to the engine, indicating that the searcher does not trust the results and feels the query is correct, but the result is incorrect.
- Searchers click on your site, spend some time there, and return to the search engine to search for the same topic again. This would convey to Google that the user has discovered material of interest and is just seeking more information or alternatives, which would be a favorable signal to the search engine.
- When a searcher finds you in the search results and clicks on your site, they may return to the search engine and refine their query before returning to your site. As a result, the search engine may conclude that the user wasn’t very particular about their search terms, which might affect whether or not your site is considered excellent or terrible relevance for the query.
For example, a user clicks on your site in the search results and returns to the engine after some time, and entirely alters their inquiry. Assuming this is the case, it means they’ve completed their mission and are now moving on to the next assignment. With this particular signal, it’d be considered relevant.
This is only one example of the kinds of indications that may be extracted from your click data, and your results will vary depending on the query type and user-activity trends.
There will be an increasing number of them as machine learning systems improve. Even if it isn’t, I wouldn’t be shocked to see it become a primary factor in determining how a site ranks in search results.
The subject of user reviews as a ranking indication cannot be discussed without bringing up user behavior.
When it comes to local search, reviews are the most important signal, according to Matt Southern of S.E.J. in November 2017. Please click here to read the whole story in its entirety.
This is an example of Google directly exploiting a user behavior signal.
If you search for a product or service on the internet, you may then use Android or other systems to verify that you went to the site you were searching for, so completing your goal.
As a review, you are considering whether or not you have passed other companies providing the same service or product and enhancing the value of that signal if you did (e.g., if you’d drive longer to interact with a certain place, it must be better).
Google Analytics may be useful.
Yes, as previously said, Google has stated that they do not utilize data from Google Analytics as a ranking indicator. Although they do not utilize it to build ranking signals, this does not imply they don’t.
- Websites are used, which might be gleaned from Google Analytics.
- Successful websites are more organized than failed ones.
- Users engage with the website and then use that knowledge as global signals.
To be clear, I included “maybe” in the section’s title for a purpose. I don’t know for sure.
It’s possible that Google Analytics is being used based on the language used in their denials.
Machine learning might make it possible for Google to forecast with more accuracy if a new webpage or website would meet a searcher’s goal, even if they don’t use it now because of its complexity.
And to think, this is just the beginning.
The goal of this article has not been to describe particular tactics but rather to provide insight into how search engines are adapting to search and how they rank sites.
Measuring user behavior was made simpler by using examples. Despite everything you’ve just read, don’t stop there. They were only a few of the many instances that might be used.
What you need to keep in mind is this:
The user is the cause for everything.
The Holy Grail has been found if a search engine can determine a user’s purpose and the possibility that that goal will be satisfied by a particular webpage or website.
Using a visitor’s activity as a search signal is something we’ve seen engines use, but it’s important to remember that this is only a means to an end.
We need to figure out how to immediately quantify and measure every user activity. For example, if this can be done in a way that can be applied worldwide, it will probably be utilized as a signal.
Make sure you know what technologies are at play before you start.
- How limited it is in scope.
- In terms of how the signals they produce may be pooled.
- In other words, whether it’s a signal or not
- The cherry on top?
There is always a potential that you might be mistaken in this quest.
Perhaps you’re looking at something that search engines aren’t considering.
Instead of producing an engaging user experience, you’ll be focused on conveying the incorrect signals to your customers.
While the engine may not express its gratitude, your bottom line will.
Isn’t that what we’re all here for?