Benchmarks

Image vs video benchmarks

In short

Images and videos share the same KPIs, the same 0–100 scale and the same colour bands, but every score is ranked only among creatives of the same media type, and only images have comparison groups. The same number on an image and on a video is not the same achievement, so compare images with images and videos with videos.

What it is

Foresight measures images and videos with the same KPIs, on the same 0–100 scale and with the same colour bands. Behind the scale, though, each media type has its own reference library.

What differs

• The reference. Every score is compared with creatives of the same media type: an image with images, a video with videos (see how scores are positioned).
• The comparison groups. Images analysed without a platform frame can be compared by creative type and sector; videos have no comparison value. See Sectors and creative types.
• What one score summarises. An image score describes a single view. A video score summarises the whole timeline, so busy and calm moments are averaged together.

What is the same

• The KPI definitions and the way each one is calculated.
• The 0–100 scale (Valence −100 to +100).
• The colour bands and the Mental Demand optimal zone.

Why they are not directly comparable

A score means "how this ranks among images" or "how this ranks among videos". So an E-Score of 62 on an image and 62 on a video are not the same achievement. That is why Foresight's Compare and exports only put images with images and videos with videos.

A GIF is treated as a video.

How to judge an image against a video

Judge each one within its own media type rather than comparing their raw scores: look at its colour band, and for an image at its gap to its comparison group. For example: "the image is 10 points above advertising images in its sector; the video is in the green band among videos".

How to read it

Treat images and videos separately: they share a vocabulary, but not their reference library. Within one media type, scores are directly comparable. Across media types, compare each creative's colour band and, for images, its gap to its comparison group.

Common pitfalls

• Ranking a mixed list of images and videos by E-Score.
• Expecting a platform to make an image and a video comparable. The platform frames both the same way, but each is still ranked only among its own media type.
• Comparing an image's score with a single peak on a video's timeline. The image and the video are still ranked against different libraries, and one moment is not the video's overall score.