Best practice guide to analysing content through Motion.


When using Dragonfly AI Motion it’s important to have a clear idea of your design objectives for the video. For example, are you looking to improve brand recognition/recall, product awareness or click through rates via your calls to action? With this in mind, Dragonfly AI will show you whether the design of your video is supporting these objectives or harming them.

Remember to tell us your objectives in the 'description' field when submitting your request using the Motion Helpdesk.


  • Objective: More visibility of the coffee cup

2.Motion Heatmaps

Our brains have a limit on how much visual information can be processed at any given time. Because of this we have evolved to prioritise where to direct our attention based on specific visual cues. Motion heatmaps use these cues to visualise where the attention of your audience is being directed as they watch your video.

Visual cues used to understand human attention:

Motion Heatmap example:

Heatmap key:

3. Keyframe analysis

In your video there will often be a number of key moments you want to analyse in more depth for example when a new product first appears or when you’re communicating a key message you want your audience to remember.

When requesting keyframe analysis remember to include a time stamp and a description of any regions of interest to help our team provide the most relevant reports.


  • Keyframes: 10s (coffee cup)

3.1 Keyframe analysis - Scene Average

The scene average provides a quick way to measure the level of visual background noise at a specific keyframe in your video. The more elements you have competing for attention in this keyframe, the higher your scene average will be.


High-value elements like product placements or calls to action have a greater chance of standing out in scenes with a low average score.

This can be measured by looking at the difference between the scene average score and the score for a high-value element.

3.2 Keyframe analysis - Hotspots

Hotspots are the areas most likely to win the attention of your audience at a specific keyframe.

Attention clusters

Use the grid to identify/score hotspots in the scene. If several adjacent cells all score highly, they should be counted at a single cluster/hotspot.

Take note of the scene average score. If the average is low (e.g. <60) and the hotspots are high (e.g. 80+) this means the hotspots have a good chance of standing out. If the distance between the two numbers is smaller, messaging may struggle to stand out.

3.3. Keyframe analysis - Regions of Interest

Regions allow you to test how specific, high-value elements are performing at a given keyframe (e.g. A call-to-action button).

Share of Attention (SoA)

Each region gets a Share of Attention (SoA) score which shows the % Share of Attention attributed to that specific region.

If you need any help interpreting your Motion analysis, please contact [email protected] or any member of the Dragonfly AI team.

Want to know more about Motion?

Need more information? Please contact [email protected] or any member of the Dragonfly AI team.

Did this answer your question?