Clicks as a Ranking Factor: Who Is Surprised?
Google's document calls CTR one of the "pillars" of rankings. How can we continue trusting what Googlers say?
As we are all following Google’s search antitrust trial, a few interesting documents have gone public.
These documents shed some light on Google’s search algorithm and how it has been developing.
One of these documents is called “Life of a Click” giving us the first real confirmation that Google did indeed use the CTR data when adjusting rankings:
I am not sure who this document was created for back in 2016 as it seemed to explain how Google search worked on a very basic level, but it is also the same year Google’s Andrey Lipattsev said that clicks from organic listings were a very gameable and unreliable source of data. In summary, they would like to use this signal but they don’t know how (that link is no longer there, I used the Wayback Machine to re-watch that video).
So did he miss that meeting - the same year - where “clicks” were presented as one of three pillars of ranking?
Or what exactly is meant by “pillars of ranking”?
Now, without pointing any fingers at anyone here, let’s try to get more scientific here:
Clicks are not a direct ranking factor (understandably). They cannot influence anything until a page ranks.
There are lots of patents that were talking about CTR and how that data could have been used in the algorithm. We always knew patent filings had no actual connection to the search algorithm.
How Google was using CTR has been one of the biggest mysteries in SEO, and there’s a great article by Chris Silver Smith at SEL that explains why we were never really convinced by Google’s statements that they were neglecting this huge amount of data they basically owned. And yes, there have been many other SEOs that did tests, read patents, and made educated theories, but none of that has ever been confirmed.
Until now.
Google has helpfully provided an answer to how exactly they were using CTR in a very easy-to-understand 101-style graph:
They added this for us to have no doubts:
"If you search right now, you'll benefit from the billions of past user reactions we've recorded"
Also, not to repeat myself, but seriously, what were these documents and what were they for? I cannot quite imagine a huge corporation using these for any purposes…
But let’s put these arguments aside… Not quite the point here…
So, for all these years that Googlers were repeatedly saying they were not using CTR, they meant they were not using it as a direct signal, but otherwise, this was one of three pillars of ranking?
And what does “pillar of ranking” mean if not something impacting rankings?
Unfortunately, we don’t have access to anything more than those two absurd slides that look almost too embarrassing to be presented in court but the fact is the fact:
User interactions (mostly clicks) impacted rankings.
And there is no reason to believe Google no longer uses CTR as a signal but I am sure they have much more data these days than they had in 2016 when Google Chrome wasn’t that well adopted. I am sure they can track deeper and more complicated journeys, beyond clicks and pogo-sticking.
So what does this mean?
Should we start manipulating CTR data?
Let’s not go there.
Instead, I see two important takeaways from here:
Expect Googlers to never be able to tell us what is really happening. They either don’t know or cannot tell. I mean… We knew this already. This is just another proof of that.
If you needed an additional incentive to make your search snippets more clickable, this is the one. It may still be a pillar. But this is also about your whole site experience because Google has told us many times, this is what they care about. And now we have enough reason to think they may be using the actual user behavior on your pages to better understand how helpful and usable your site is.
Clickability of your search results and usability of your web page should be part of an audit whenever you lost positions. Along with content helpfulness, but that is a different story.