It Turns Out CTR Does NOT Affect Search Results
Without a doubt, click through rate (CTR) is one of the most important metrics considered in SEO and internet marketing. CTR is a percentage, showing how many people click on your site when they’re offered the link. The higher the CTR, the more traffic and (hopefully) the more sales.
But CTR doesn’t do everything. For some time, the SEO industry has wondered if Google uses it as a ranking factor—in other words, better CTR means better search rankings. The theory is that people clicking on your site from the search results acts as a vote that your site is valuable. But is that really true?
The debate over CTR has been fierce, and some past experiments suggested it does factor in. But if that was ever true, it’s not anymore. Bartosz Góralewicz just released probably the most scientific CTR/SEO study in history, and he says click through rate does not affect search rank.
The Clearest Study
Bartosz’s approach was refreshingly clean. He knew that isolating the effect of CTR wouldn’t be easy. In past studies, there’s been a question of whether the apparent affect of CTR might actually be caused by something else, so he controlled for as many factors as possible—all the way down to browser history. He also painstakingly designed “clickbots” (automated scripts that act as dummy users) that would truly fool Google. He wanted his wave of fake traffic to look like a post gone viral, and checked in multiple places that Google wasn’t filtering it out.
He succeeded. Over the course of two weeks in March, he flooded the internet with searches and clicks to his site. Google was thoroughly fooled: did the traffic get recorded as real in Analytics and Search Console, but it influenced the data in AdWords Keyword Planner—and even sparked a fake Google Trend(!).
Total “traffic” for the targeted page rose to 21,000 clicks per day, with a staged CTR of nearly 80 percent.
But the result? Bartosz’s rank didn’t improve for any of his keywords.
A Final Answer?
A number of SEO’s have argued that the case still isn’t closed. Perhaps, they say, Google did detect that the traffic was fake, and disregarded it for ranking purposes. They could be right. But that doesn’t explain why they would leave it in other Google products, even their cash cow AdWords. Ultimately, the skeptics have no evidence to support their case.
But I’m inclined to believe Bartosz’s study on common sense grounds. There are just too many reasons why CTR doesn’t make a good ranking signal:
- It’s an echo chamber. The biggest factor influencing CTR is position in the SERPs. The higher up in the ranking you are, the more clicks you get. If clicks themselves were used to decide rankings, it would create a huge echo chamber where the top-ranking sites keep getting farther ahead of the competition. (All ranking factors do this to an extent, but none are as cyclical as CTR would be).
- It’s easy to fake. Google works hard to prevent companies from rigging the SEO game. Good ranking factors are those that reflect real user behavior, insulated from paid incentives, loopholes and dirty tricks. But CTR is highly fakeable, just by launching some automated scripts (“clickbots”). As Bartosz showed, these clickbots have to be built just right to sneak one past Google, but he also proved it’s doable. This tactic has been used repeatedly in CTR studies, and Google knows it’s a weak spot.
- It’s redundant. Most ranking factors reflect specific actions that people take to affirm that a site is valid: social shares, linking out, etc. CTR just means they successfully found it in the search results. In a way, CTR is an aggregate metric of all other ranking signals together.
What is your take on Bartosz’s study? Do you see any holes in his methods, or should we go ahead and nail the coffin shut on CTR?