KPIs

Attention

In short

Attention is the attention pillar of the E-Score. It is built from Clarity (how concentrated and grouped predicted attention is) and whether your creative's visual complexity sits in a comfortable range. It is highest when the eye is guided to a clear focal area and the creative is neither too bare nor too busy.

What it measures

Attention measures the predicted quality of visual attention on your creative. It tells you how well the layout is likely to capture the eye and guide it to a clear focal area. It is one of the two pillars of the E-Score.

What feeds it

Attention comes from two predictions made on your creative as shown, inside the platform's screen if you picked one:

• the predicted attention map, which shows where people are likely to look (see Heatmaps),
• the predicted visual complexity of the creative.

How it's built

Attention is built from two ingredients:

1. Clarity: whether predicted attention concentrates on a few, well-grouped areas. Clarity itself is built from Focus and Spread.
2. Complexity fit: whether your creative's visual complexity sits in a comfortable range, neither too bare nor too busy.

So Attention is highest when the eye is guided to a clear focal area and the creative's visual complexity is at a comfortable level.

The complexity fit is related to Mental Demand, but it is not the same number. Mental Demand shows where your creative sits on the complexity scale. The complexity fit inside Attention reflects whether that sits in a comfortable range.

Scale and direction

AttentionImageVideo

Predicted quality of visual attention: clear focal areas and a comfortable level of visual complexity.

Scale

0–100

Reading

Higher is better

Higher is better. Unlike the E-Score, it has no colour bands. Its chip shows whether it is above or below the comparison group, when there is one. It is built from components that are positioned against reference creatives of the same media type.

Image vs video

The calculation is the same for images and videos. Images are compared with images, and videos with videos.

For videos, Attention is also shown as a curve on the timeline, so you can see which moments hold the eye well. The overall value summarises the whole video. The scene cards also show the Attention value of each scene ("Att:").

Platforms

Attention is directly affected by the platform you pick. The attention prediction runs on your creative inside the platform's screen. Interface elements and empty areas change where attention gathers and add visual complexity. The model and the calculation stay the same. With a platform selected, no benchmark value is shown. See What changes when you pick a platform.

How to read it

• Low Attention, low Clarity: the eye is pulled in too many directions. Simplify the layout or strengthen one focal element.
• Low Attention, good Clarity: check Mental Demand. The creative is probably visually too bare or too busy.

Common pitfalls

• It is a prediction, not in-market data. Attention does not measure real dwell time or scroll-stopping in a live feed.
• Visual complexity outside a comfortable range lowers Attention through the complexity fit, even when the Mental Demand number looks acceptable.
• Framed and unframed runs are not directly comparable. The same creative can score differently with and without a platform.