Listen: Query found to be the biggest predictor of CTR delta.
In a surprise discovery, we found that query itself is the biggest predictor of how much CTR will deviate from the site's position average.
Transcript
What is the strongest predictor of search engine click-through rates? To find out, researchers analyzed over thirty thousand search query and website pairs. They wanted to see how much different parts of a search result, like the title, description, and website address, influence whether someone clicks.
Initially, combining all these features created too much noise. To solve this, they trained four separate machine learning models, each focusing on just one element.
The big discovery was that the search query itself is the absolute strongest signal. This means understanding user intent is far more important for predicting clicks than the actual text in the search snippet. When looking at the snippet itself, titles and descriptions are about equally important, while the website address is the least reliable predictor.
By combining the predictions of all four specialized models, this system can correctly predict whether a search result will perform better or worse than average about eighty percent of the time. In the future, the team plans to expand this system to predict not just whether a result will perform better or worse, but exactly how much traffic will change.