How News Website Can Benefit from Attention Analytics
The news readers tend to make a decision whether to read an article or not in a matter of seconds. That is why the news portals should care about users’ attention distribution research.
While A/B testing of catchy article headlines is a quite common technique for the news sites, images are usually selected based on chef editor or journalist gut feeling. Editorial image A/B testing is still quite rare. However, human brain process images 60,000 times faster than text. When the users interact with the news portals’ content, their scrolling is stopped by no other than images.
So, how can you optimise an images’ selection process and make data driven decision on which editorial image to choose? How can attention metrics help news portals guide their readers’ attention through the landing page? Keep on reading.
How to draw readers’ attention to the main news
Let’s imagine your want to direct users’ attention towards Welt.de most important news (the first article) and make sure they notice that it is a video news. To achieve this goal, we focus on selecting the right image that capture readers attention.
To test the performance of different visuals, we use Attention Insight platform. It is an AI-based system trained with previous eye tracking studies data that predicts users’ attention shifts with 90 % accuracy. It enables you to implement a quick test and get insights on how attention distributed in the content.
We have uploaded a screenshot of Welt.de website home page to Attention Insight platform. Within minutes, a platform has generated a heatmap. It is a visual representation of how attention distributed in the website. Besides, it shows the percentage (%) of attention that different objects of the website got.
You can see that image has received 15 % of users attention which is quite good, but can we do better?
To check it out, we have tested 2 other images from video clip.
Once the screenshots with changed images have been uploaded to Attention Insight platform, we saw that:
- An image with Reichstage building has grabbed 19 % of attention. It is a 4 % increase in comparison with the first image.
- The third image with 2 girls has received 26 % of attention.
How to test multiple images at once
Let’s say, your main goal is to select best performing image from your gallery and you are not interested in how it performs in a context of other website items. Just upload a screenshot of your visuals library and select 3 images with the highest percentage (%) of attention.
For example, your goal is to select an image for an article about Global Climate Strike. Upload a screenshot of photo gallery to Attention Insight platform and test which image is the most attention grabbing.
Looking at a heatmap, we can see that the first image has grabbed 20 % of attention. This information empowers an editor to make an objective decision.
To sum up
Thus, usually, journalists allocate an image that best illustrates a story. They select image from their archive that reflects the content and matches the tone of the article. And there is nothing wrong with that, but it is still a subjective decision. While, insights on attention distribution lets you to make an objective, data-based image choice, that ensures a grow in traffic terms.