Overview of Realeyes Attention Metrics in Vidmob (Beta)
Vidmob has integrated Realeyes’ attention metrics into its creative data platform, providing you with deeper insights into how your ads engage viewers. Realeyes is a leading provider of attention measurement, using cutting-edge technology to predict viewer attention and emotional responses through webcam-based data. With this integration, you can have predicted data about attention, helping optimize ad strategies and improve overall campaign effectiveness.
Why Realeyes Attention Metrics Matter
Realeyes' attention metrics help advertisers understand how much focus and interest viewers dedicate to their ads. By tracking viewer attention, advertisers can identify which parts of their content capture attention and which parts may cause distraction or disinterest. This allows for a more data-driven approach to ad creation, ultimately improving engagement and driving sales. Realeyes attention data also provides predictive insights into future performance, helping users make informed decisions when creating or optimizing ads.
How to Use Realeyes Attention Metrics in Creative Analytics
Access the Realeyes Attention Tab
To view Realeyes attention metrics, navigate to the ICV (Integrated Creative View) and look for the "Realeyes Attention [Beta]" tab.
Note: This feature is currently available to select users. Talk to your sales representative if you would like access to this feature.
Viewing Overall Attention Score
Once inside the Realeyes Attention tab, you will see an overall attention score for the creative, represented as a numerical value. This score represents the ad's attention potential based on Realeyes' predictive analysis. Hover over the score for more information on how it is calculated.
Review Recommendations and Insights
Along with the attention score, you'll also see Realeyes' recommendations for your ad. These insights suggest changes that could improve attention, such as retaining specific scenes or altering the order of visuals. For instance, you might receive a recommendation to keep scenes with high attention scores (e.g., “First scene scored 85, second scene scored 90”).
Nuances to Be Aware Of:
Multi Asset Ads
If you are analyzing a multi-asset ad, the existing ICV flow remains unchanged, even if Realeyes does not directly analyze all parts of the ad. Filtering by date range and other options continues to apply to all ICV data.Emotional Breakdown
Realeyes offers detailed emotional insights based on viewer facial expressions. For example, “happiness” is determined by specific facial features like smiling, while “disgust” is identified by wrinkling of the nose or downturn of the mouth. Hover over the respective metrics for more in-depth explanations of these emotional responses.
Scene-by-Scene Breakdown
For video ads, you can analyze each scene’s performance in detail. The tab will show a breakdown of attention metrics for every scene, with columns indicating the attention score and Realeyes’ specific recommendations. Hover over each scene’s score to see a breakdown of viewer reactions, such as confusion, distraction, contempt, happiness, surprise, or disgust.
For Image-Based Ads
While video ads provide a scene-by-scene breakdown, image-based ads will display a single row summarizing overall attention and engagement data.
How to Use Realeyes Attention Metrics in Creative Scoring
Access the Realeyes Attention Tab
To view the Realeyes attention scoring, navigate to the ICV and look for the "Realeyes Attention [Beta]" tab, marked with a visual Beta indicator.
Note: This feature is currently available to select users. Talk to your sales representative if you would like access to this feature.
Viewing the Overall Attention Score
Once in the Realeyes Attention tab, you’ll see the overall Attention Potential Value for the selected creative. This number reflects the ad's potential to capture and hold viewer attention, replacing the traditional Adherence by channel and Adherence average. For example, you might see a score like 87.68, representing the ad’s overall effectiveness in gaining attention. Hover over the score to view more details about how this number is calculated.
Reviewing Scene-Level Recommendations
Along with the overall attention score, the tab provides Realeyes’ recommendations or insights. These recommendations offer specific advice on how to optimize your ad for better attention. For example, a recommendation may suggest keeping certain scenes with high attention scores, such as:
“The ad’s overall attention score is 88. We recommend retaining the first scene (attention score: 85) and second scene (attention score: 90), as they are particularly strong in driving performance.”
Scene-by-Scene Breakdown
For video ads, the attention scoring system breaks down each scene, offering individual scores and insights for every section of the ad. Each scene will have its own row with a score, and you can hover over the score to see how viewer reactions like distraction, confusion, contempt, happiness, disgust, or surprise played a role.
Column 1: Scene’s attention score (e.g., 85, 90)
Column 2: Realeyes’ recommendation for improving or maintaining viewer attention
You’ll also see an emotional breakdown for each scene, helping you understand why a particular scene resonated or fell flat with the audience.
For Image-Based Ads
Image ads will display a single row with the overall attention score and any corresponding recommendations, without the scene breakdown that is available for video content.
Nuances to Be Aware Of
Emotional Breakdown
Realeyes offers detailed emotional insights based on viewer facial expressions. For example, “happiness” is determined by specific facial features like smiling, while “disgust” is identified by wrinkling of the nose or downturn of the mouth. Hover over the respective metrics for more in-depth explanations of these emotional responses.
By integrating Realeyes’ attention metrics into Vidmob, you can now optimize your creative content more effectively, ensuring your ads capture and retain viewer focus, driving better engagement and performance.
Ho