Brazilian television network SBT News wanted to find out what specific content makes visitors stay on its Web site. The team performed A/B tests using Google Optimize to see what elements it could improve and change to attract the audience to the content.

One of the big changes that helped with engagement was positioning of elements like embedded images and video.

“By placing video at the top of the page, we realise we can have more engagement, more retention with our users,” said Editor Cido Coelho, whose team worked with The Guardian on this project.

Testing goals

SBT had four main goals:

1. Increase interaction in sessions from 34 to 45 seconds.
2. Reduce bouncing from 86% to 75%.
3. Increase the number of pages visited from two to three.
4. Increase the number of return visitors from 30% to 40%.

The big lesson learned was that if they wanted to improve in all these categories, Coelho said, they needed to test everything to understand the behaviour of the reader. They also changed a lot of different things on each page to decide what worked and what didn’t.

Accelerated growth

Over the course of the Meta Accelerator project, SBT enjoyed exponential growth for three months. Session interactions increased to 52 seconds, smashing the goal of 45 seconds. Another positive result was the percentage of returning visitors, nearly meeting the goal of a 10% percent increase, from 30% to 40%.

“We highlighted our call to action and included a link on the homepage after the first paragraph so we could increase the recirculation inside the page,” Coelho said. “We saw really good results doing that. In three months, we almost doubled traffic on our home page and the number of pageviews increased in a meaningful way.”

From September to October, during that call to action change, SBT saw its pageviews increase 6.3%. From August to September, SBT saw a 5.2% increase in unique pageviews, Coelho said:

“We could have new experiences, new content, and more information for those who access our Web site.” 

This case study originally appeared in the INMA report, The Benefits and Risks of Media Data Democratisation.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .