August 5th, 2020
Hypothesis
Showing the reader their progress in a table of contents on the right side of the article will increase time on site and decrease bounce rate.
Why do this? Because search engines care about dwell time. The more I can keep visitors on the page, the better.
Here's a GIF of what it looks like in action.
Since this table of contents only appears on desktop screens, I'll create a custom report in GA that cuts the data by the AB test, and also filters the results down to Device=Desktop:
Side note: Anybody have any idea how I can make a table of contents also work on mobile?
Once, I'm in the report, I'll put another dimension in called "Landing Page" because I want to look at the metrics at the page level and for users that landed on that page (usually from Google Search).
Results
The table of contents performed significantly better than the control group in terms of time on site.
For this test I ran two control groups, control and control2, so only 33% of users saw the table of contents.
It's so obviously better that I don't even need to plug the data into an AB test calculator.
Here are the results for three popular pages on Starter Story:
As a whole, it also increased the likelihood to subscribe to the Starter Story email list, which was an unexpected benefit:
Showing the reader their progress in a table of contents on the right side of the article will increase time on site and decrease bounce rate.
Why do this? Because search engines care about dwell time. The more I can keep visitors on the page, the better.
Here's a GIF of what it looks like in action.
Since this table of contents only appears on desktop screens, I'll create a custom report in GA that cuts the data by the AB test, and also filters the results down to Device=Desktop:
Side note: Anybody have any idea how I can make a table of contents also work on mobile?
Once, I'm in the report, I'll put another dimension in called "Landing Page" because I want to look at the metrics at the page level and for users that landed on that page (usually from Google Search).
Results
The table of contents performed significantly better than the control group in terms of time on site.
For this test I ran two control groups, control and control2, so only 33% of users saw the table of contents.
It's so obviously better that I don't even need to plug the data into an AB test calculator.
Here are the results for three popular pages on Starter Story:
As a whole, it also increased the likelihood to subscribe to the Starter Story email list, which was an unexpected benefit: